Our website contains many resources that SVS members can use for help with managing a practice, continuing education, patient education materials and much more. The patient resource pages on the site cover a variety of vascular conditions, tests and treatments. Most recently, we’ve added a page for Transcarotid Artery Revascularization (TCAR). This, and most of our pages, can give patients and/or their loved ones a better understanding of their vascular condition, as well as how it’s being tested and treated. Take a look at our pages and share with your patients today.

Our website contains many resources that SVS members can use for help with managing a practice, continuing education, patient education materials and much more. The patient resource pages on the site cover a variety of vascular conditions, tests and treatments. Most recently, we’ve added a page for Transcarotid Artery Revascularization (TCAR). This, and most of our pages, can give patients and/or their loved ones a better understanding of their vascular condition, as well as how it’s being tested and treated. Take a look at our pages and share with your patients today.

Our website contains many resources that SVS members can use for help with managing a practice, continuing education, patient education materials and much more. The patient resource pages on the site cover a variety of vascular conditions, tests and treatments. Most recently, we’ve added a page for Transcarotid Artery Revascularization (TCAR). This, and most of our pages, can give patients and/or their loved ones a better understanding of their vascular condition, as well as how it’s being tested and treated. Take a look at our pages and share with your patients today.

Another solo-practice neurologist and I were talking last week. He’s understandably worried about the local hospital starting construction on a new “neuroscience center” down the street from us. They have ambitious plans for it, which apparently don’t include those of us who’ve served the community for 20-30 years.

Whatever. I’ve been in a large practice before, and don’t want to be a part of one again.

His concern, which I have, too, is that the hospital center will drive us little guys out of business. This seems to be a common medical practice model these days.

I hope not. I’ve been doing this for a long time, and am happy with my little world. I also believe, perhaps naively, that there’s still a place for a small practice.

My staff and I know my patients. We’re generally tuned in to who needs what, or how much time. We return all calls within a few hours (or less) and try be on top of getting medication refills and records requests done the same day they come in.

While a large practice has some advantages, based on my time with one I’d have to say we didn’t do those things as well there. Messages often weren’t relayed, or were sent to the wrong doctor. Here there’s only me.

I may not make as much, but my appointment times and intervals aren’t dictated by an accountant. This allows me to generally spend as much time as needed with each person and not feel rushed as the day goes on. I hope patients still desire that in a physician, as opposed to a place advertising “20 neurologists, no waiting!” on a sign that would fit in on the Vegas strip.

Obviously, I can’t control what the hospital will do. I can only manage my own little world. I’ll continue doing that as best I can, as long as I’m able.

Time spent worrying about things I can’t change isn’t productive and is bad for one’s blood pressure. So I’ll focus on what I can do, and try not to worry about the rest.
 

Dr. Block has a solo neurology practice in Scottsdale, Ariz.

Another solo-practice neurologist and I were talking last week. He’s understandably worried about the local hospital starting construction on a new “neuroscience center” down the street from us. They have ambitious plans for it, which apparently don’t include those of us who’ve served the community for 20-30 years.

Whatever. I’ve been in a large practice before, and don’t want to be a part of one again.

His concern, which I have, too, is that the hospital center will drive us little guys out of business. This seems to be a common medical practice model these days.

I hope not. I’ve been doing this for a long time, and am happy with my little world. I also believe, perhaps naively, that there’s still a place for a small practice.

My staff and I know my patients. We’re generally tuned in to who needs what, or how much time. We return all calls within a few hours (or less) and try be on top of getting medication refills and records requests done the same day they come in.

While a large practice has some advantages, based on my time with one I’d have to say we didn’t do those things as well there. Messages often weren’t relayed, or were sent to the wrong doctor. Here there’s only me.

I may not make as much, but my appointment times and intervals aren’t dictated by an accountant. This allows me to generally spend as much time as needed with each person and not feel rushed as the day goes on. I hope patients still desire that in a physician, as opposed to a place advertising “20 neurologists, no waiting!” on a sign that would fit in on the Vegas strip.

Obviously, I can’t control what the hospital will do. I can only manage my own little world. I’ll continue doing that as best I can, as long as I’m able.

Time spent worrying about things I can’t change isn’t productive and is bad for one’s blood pressure. So I’ll focus on what I can do, and try not to worry about the rest.
 

Dr. Block has a solo neurology practice in Scottsdale, Ariz.

Another solo-practice neurologist and I were talking last week. He’s understandably worried about the local hospital starting construction on a new “neuroscience center” down the street from us. They have ambitious plans for it, which apparently don’t include those of us who’ve served the community for 20-30 years.

Whatever. I’ve been in a large practice before, and don’t want to be a part of one again.

His concern, which I have, too, is that the hospital center will drive us little guys out of business. This seems to be a common medical practice model these days.

I hope not. I’ve been doing this for a long time, and am happy with my little world. I also believe, perhaps naively, that there’s still a place for a small practice.

My staff and I know my patients. We’re generally tuned in to who needs what, or how much time. We return all calls within a few hours (or less) and try be on top of getting medication refills and records requests done the same day they come in.

While a large practice has some advantages, based on my time with one I’d have to say we didn’t do those things as well there. Messages often weren’t relayed, or were sent to the wrong doctor. Here there’s only me.

I may not make as much, but my appointment times and intervals aren’t dictated by an accountant. This allows me to generally spend as much time as needed with each person and not feel rushed as the day goes on. I hope patients still desire that in a physician, as opposed to a place advertising “20 neurologists, no waiting!” on a sign that would fit in on the Vegas strip.

Obviously, I can’t control what the hospital will do. I can only manage my own little world. I’ll continue doing that as best I can, as long as I’m able.

Time spent worrying about things I can’t change isn’t productive and is bad for one’s blood pressure. So I’ll focus on what I can do, and try not to worry about the rest.
 

Dr. Block has a solo neurology practice in Scottsdale, Ariz.

The practice of dermatology is rife with bedside tools: swabs, smears, and scoring systems. First popularized in specialties such as emergency medicine and internal medicine, clinical scoring systems are now emerging in dermatology. These evidence-based scores can be calculated quickly at the bedside—often through a free smartphone app—to help guide clinical decision-making regarding diagnosis, prognosis, and management. As with any medical tool, scoring systems have limitations and should be used as a supplement, not substitute, for one’s clinical judgement. This article reviews 4 clinical scoring systems practical for dermatology residents.

SCORTEN Prognosticates Cases of Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis

Perhaps the best-known scoring system in dermatology, the SCORTEN is widely used to predict hospital mortality from Stevens-Johnson syndrome/toxic epidermal necrolysis. The SCORTEN includes 7 variables of equal weight—age of 40 years or older, heart rate of 120 beats per minute or more, cancer/hematologic malignancy, involved body surface area (BSA) greater than 10%, serum urea greater than 10 mmol/L, serum bicarbonate less than 20 mmol/L, and serum glucose greater than 14 mmol/L—each contributing 1 point to the overall score if present.¹ The involved BSA is defined as the sum of detached and detachable epidermis.¹

The SCORTEN was developed and prospectively validated to be calculated at the end of the first 24 hours of admission; for this calculation, use the BSA affected at that time, and use the most abnormal values during the first 24 hours of admission for the other variables.¹ In addition, a follow-up study including some of the original coauthors recommends recalculating the SCORTEN at the end of hospital day 3, having found that the score’s predictive value was better on this day than hospital days 1, 2, 4, or 5.² Based on the original study, a SCORTEN of 0 to 1 corresponds to a mortality rate of 3.2%, 2 to 12.1%, 3 to 35.3%, 4 to 58.3%, and 5 or greater to 90.0%.¹

Limitations of the SCORTEN include its ability to overestimate or underestimate mortality as demonstrated by 2 multi-institutional cohorts.^3,4 Recently, the ABCD-10 score was developed as an alternative to the SCORTEN and was found to predict mortality similarly when validated in an internal cohort.⁵

PEST Screens for Psoriatic Arthritis

Dermatologists play an important role in screening for psoriatic arthritis, as an estimated 1 in 5 patients with psoriasis have psoriatic arthritis.⁶ To this end, several screening tools have been developed to help differentiate psoriatic arthritis from other arthritides. Joint guidelines from the American Academy of Dermatology and the National Psoriasis Foundation acknowledge that “. . . these screening tools have tended to perform less well when tested in groups of people other than those for which they were originally developed. As such, their usefulness in routine clinical practice remains controversial.”⁷ Nevertheless, the guidelines state, “[b]ecause screening and early detection of inflammatory arthritis are essential to optimize patient [quality of life] and reduce morbidity, providers may consider using a formal screening tool of their choice.”⁷

With these limitations in mind, I have found the Psoriasis Epidemiology Screening Tool (PEST) to be the most useful psoriatic arthritis screening tool. One study determined that the PEST has the best trade-off between sensitivity and specificity compared to 2 other psoriatic arthritis screening tools, the Psoriatic Arthritis Screening and Evaluation (PASE) and the Early Arthritis for Psoriatic Patients (EARP).⁸

The PEST is comprised of 5 questions: (1) Have you ever had a swollen joint (or joints)? (2) Has a doctor ever told you that you have arthritis? (3) Do your fingernails or toenails have holes or pits? (4) Have you had pain in your heel? (5) Have you had a finger or toe that was completely swollen and painful for no apparent reason? According to the PEST, a referral to a rheumatologist should be considered for patients answering yes to 3 or more questions, which is 97% sensitive and 79% specific for psoriatic arthritis.⁹ Patients who answer yes to fewer than 3 questions should still be referred to a rheumatologist if there is a strong clinical suspicion of psoriatic arthritis.¹⁰

The PEST can be accessed for free in 13 languages via the GRAPPA (Group for Research and Assessment of Psoriasis and Psoriatic Arthritis) app as well as downloaded for free from the National Psoriasis Foundation’s website (https://www.psoriasis.org/psa-screening/providers).

ALT-70 Differentiates Cellulitis From Pseudocellulitis

Overdiagnosing cellulitis in the United States has been estimated to result in up to 130,000 unnecessary hospitalizations and up to $515 million in avoidable health care spending.¹¹ Dermatologists are in a unique position to help fix this issue. In one retrospective study of 1430 inpatient dermatology consultations, 74.32% of inpatients evaluated for presumed cellulitis by a dermatologist were instead diagnosed with a cellulitis mimicker (ie, pseudocellulitis), such as stasis dermatitis or contact dermatitis.¹²

The ALT-70 score was developed and prospectively validated to help differentiate lower extremity cellulitis from pseudocellulitis in adult patients in the emergency department (ED).¹³ In addition, the score has retrospectively been shown to function similarly in the inpatient setting when calculated at 24 and 48 hours after ED presentation.¹⁴ Although the ALT-70 score was designed for use by frontline clinicians prior to dermatology consultation, I also have found it helpful to calculate as a consultant, as it provides an objective measure of risk to communicate to the primary team in support of one diagnosis or another.

ALT-70 is an acronym for the score’s 4 variables: asymmetry, leukocytosis, tachycardia, and age of 70 years or older.¹⁵ If present, each variable confers a certain number of points to the final score: 3 points for asymmetry (defined as unilateral leg involvement), 1 point for leukocytosis (white blood cell count ≥10,000/μL), 1 point for tachycardia (≥90 beats per minute), and 2 points for age of 70 years or older. An ALT-70 score of 0 to 2 corresponds to an 83.3% or greater chance of pseudocellulitis, suggesting that the diagnosis of cellulitis be reconsidered. A score of 3 to 4 is indeterminate, and additional information such as a dermatology consultation should be pursued. A score of 5 to 7 corresponds to an 82.2% or greater chance of cellulitis, signifying that empiric treatment with antibiotics be considered.¹⁵

The ALT-70 score does not apply to cases involving areas other than the lower extremities; intravenous antibiotic use within 48 hours before ED presentation; surgery within the last 30 days; abscess; penetrating trauma; burn; or known history of osteomyelitis, diabetic ulcer, or indwelling hardware at the site of infection.¹⁵ The ALT-70 score is available for free via the MDCalc app and website (https://www.mdcalc.com/alt-70-score-cellulitis).

Mohs AUC Determines the Appropriateness of Mohs Micrographic Surgery

In 2012, the American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic Surgery Association, and American Society for Mohs Surgery published appropriate use criteria (AUC) to guide the decision to pursue Mohs micrographic surgery (MMS) in the United States.¹⁶ Based on various tumor and patient characteristics, the Mohs AUC assign scores to 270 different clinical scenarios. A score of 1 to 3 signifies that MMS is inappropriate and generally not considered acceptable. A score 4 to 6 indicates that the appropriateness of MMS is uncertain. A score 7 to 9 means that MMS is appropriate and generally considered acceptable.¹⁶

Since publication, the Mohs AUC have been criticized for classifying most primary superficial basal cell carcinomas as appropriate for MMS¹⁷ (which an AUC coauthor¹⁸ and others^19,20 have defended), excluding certain reasons for performing MMS (such as operating on multiple tumors on the same day),²¹ including counterintuitive scores,²² and omitting trials from Europe²³ (which AUC coauthors also have defended²⁴). As with any clinical scoring system, the Mohs AUC has limitations; the creators acknowledge that “. . . these criteria should not be interpreted as setting a standard of care, or be deemed inclusive of all proper methods of care nor exclusive of other methods of care reasonably directed to obtaining the same results, even for those indications scored as inappropriate.”¹⁶ The Mohs AUC app (https://www.aad.org/members/aad-apps/mohs-auc) is free and allows users to enter tumor and patient characteristics to determine the score for their specific scenario.

Final Thoughts

Scoring systems are emerging in dermatology as evidence-based bedside tools to help guide clinical decision-making. Despite their limitations, these scores have the potential to make a meaningful impact in dermatology as they have in other specialties.

The practice of dermatology is rife with bedside tools: swabs, smears, and scoring systems. First popularized in specialties such as emergency medicine and internal medicine, clinical scoring systems are now emerging in dermatology. These evidence-based scores can be calculated quickly at the bedside—often through a free smartphone app—to help guide clinical decision-making regarding diagnosis, prognosis, and management. As with any medical tool, scoring systems have limitations and should be used as a supplement, not substitute, for one’s clinical judgement. This article reviews 4 clinical scoring systems practical for dermatology residents.

SCORTEN Prognosticates Cases of Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis

Perhaps the best-known scoring system in dermatology, the SCORTEN is widely used to predict hospital mortality from Stevens-Johnson syndrome/toxic epidermal necrolysis. The SCORTEN includes 7 variables of equal weight—age of 40 years or older, heart rate of 120 beats per minute or more, cancer/hematologic malignancy, involved body surface area (BSA) greater than 10%, serum urea greater than 10 mmol/L, serum bicarbonate less than 20 mmol/L, and serum glucose greater than 14 mmol/L—each contributing 1 point to the overall score if present.¹ The involved BSA is defined as the sum of detached and detachable epidermis.¹

The SCORTEN was developed and prospectively validated to be calculated at the end of the first 24 hours of admission; for this calculation, use the BSA affected at that time, and use the most abnormal values during the first 24 hours of admission for the other variables.¹ In addition, a follow-up study including some of the original coauthors recommends recalculating the SCORTEN at the end of hospital day 3, having found that the score’s predictive value was better on this day than hospital days 1, 2, 4, or 5.² Based on the original study, a SCORTEN of 0 to 1 corresponds to a mortality rate of 3.2%, 2 to 12.1%, 3 to 35.3%, 4 to 58.3%, and 5 or greater to 90.0%.¹

Limitations of the SCORTEN include its ability to overestimate or underestimate mortality as demonstrated by 2 multi-institutional cohorts.^3,4 Recently, the ABCD-10 score was developed as an alternative to the SCORTEN and was found to predict mortality similarly when validated in an internal cohort.⁵

PEST Screens for Psoriatic Arthritis

Dermatologists play an important role in screening for psoriatic arthritis, as an estimated 1 in 5 patients with psoriasis have psoriatic arthritis.⁶ To this end, several screening tools have been developed to help differentiate psoriatic arthritis from other arthritides. Joint guidelines from the American Academy of Dermatology and the National Psoriasis Foundation acknowledge that “. . . these screening tools have tended to perform less well when tested in groups of people other than those for which they were originally developed. As such, their usefulness in routine clinical practice remains controversial.”⁷ Nevertheless, the guidelines state, “[b]ecause screening and early detection of inflammatory arthritis are essential to optimize patient [quality of life] and reduce morbidity, providers may consider using a formal screening tool of their choice.”⁷

With these limitations in mind, I have found the Psoriasis Epidemiology Screening Tool (PEST) to be the most useful psoriatic arthritis screening tool. One study determined that the PEST has the best trade-off between sensitivity and specificity compared to 2 other psoriatic arthritis screening tools, the Psoriatic Arthritis Screening and Evaluation (PASE) and the Early Arthritis for Psoriatic Patients (EARP).⁸

The PEST is comprised of 5 questions: (1) Have you ever had a swollen joint (or joints)? (2) Has a doctor ever told you that you have arthritis? (3) Do your fingernails or toenails have holes or pits? (4) Have you had pain in your heel? (5) Have you had a finger or toe that was completely swollen and painful for no apparent reason? According to the PEST, a referral to a rheumatologist should be considered for patients answering yes to 3 or more questions, which is 97% sensitive and 79% specific for psoriatic arthritis.⁹ Patients who answer yes to fewer than 3 questions should still be referred to a rheumatologist if there is a strong clinical suspicion of psoriatic arthritis.¹⁰

The PEST can be accessed for free in 13 languages via the GRAPPA (Group for Research and Assessment of Psoriasis and Psoriatic Arthritis) app as well as downloaded for free from the National Psoriasis Foundation’s website (https://www.psoriasis.org/psa-screening/providers).

ALT-70 Differentiates Cellulitis From Pseudocellulitis

Overdiagnosing cellulitis in the United States has been estimated to result in up to 130,000 unnecessary hospitalizations and up to $515 million in avoidable health care spending.¹¹ Dermatologists are in a unique position to help fix this issue. In one retrospective study of 1430 inpatient dermatology consultations, 74.32% of inpatients evaluated for presumed cellulitis by a dermatologist were instead diagnosed with a cellulitis mimicker (ie, pseudocellulitis), such as stasis dermatitis or contact dermatitis.¹²

The ALT-70 score was developed and prospectively validated to help differentiate lower extremity cellulitis from pseudocellulitis in adult patients in the emergency department (ED).¹³ In addition, the score has retrospectively been shown to function similarly in the inpatient setting when calculated at 24 and 48 hours after ED presentation.¹⁴ Although the ALT-70 score was designed for use by frontline clinicians prior to dermatology consultation, I also have found it helpful to calculate as a consultant, as it provides an objective measure of risk to communicate to the primary team in support of one diagnosis or another.

ALT-70 is an acronym for the score’s 4 variables: asymmetry, leukocytosis, tachycardia, and age of 70 years or older.¹⁵ If present, each variable confers a certain number of points to the final score: 3 points for asymmetry (defined as unilateral leg involvement), 1 point for leukocytosis (white blood cell count ≥10,000/μL), 1 point for tachycardia (≥90 beats per minute), and 2 points for age of 70 years or older. An ALT-70 score of 0 to 2 corresponds to an 83.3% or greater chance of pseudocellulitis, suggesting that the diagnosis of cellulitis be reconsidered. A score of 3 to 4 is indeterminate, and additional information such as a dermatology consultation should be pursued. A score of 5 to 7 corresponds to an 82.2% or greater chance of cellulitis, signifying that empiric treatment with antibiotics be considered.¹⁵

The ALT-70 score does not apply to cases involving areas other than the lower extremities; intravenous antibiotic use within 48 hours before ED presentation; surgery within the last 30 days; abscess; penetrating trauma; burn; or known history of osteomyelitis, diabetic ulcer, or indwelling hardware at the site of infection.¹⁵ The ALT-70 score is available for free via the MDCalc app and website (https://www.mdcalc.com/alt-70-score-cellulitis).

Mohs AUC Determines the Appropriateness of Mohs Micrographic Surgery

In 2012, the American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic Surgery Association, and American Society for Mohs Surgery published appropriate use criteria (AUC) to guide the decision to pursue Mohs micrographic surgery (MMS) in the United States.¹⁶ Based on various tumor and patient characteristics, the Mohs AUC assign scores to 270 different clinical scenarios. A score of 1 to 3 signifies that MMS is inappropriate and generally not considered acceptable. A score 4 to 6 indicates that the appropriateness of MMS is uncertain. A score 7 to 9 means that MMS is appropriate and generally considered acceptable.¹⁶

Since publication, the Mohs AUC have been criticized for classifying most primary superficial basal cell carcinomas as appropriate for MMS¹⁷ (which an AUC coauthor¹⁸ and others^19,20 have defended), excluding certain reasons for performing MMS (such as operating on multiple tumors on the same day),²¹ including counterintuitive scores,²² and omitting trials from Europe²³ (which AUC coauthors also have defended²⁴). As with any clinical scoring system, the Mohs AUC has limitations; the creators acknowledge that “. . . these criteria should not be interpreted as setting a standard of care, or be deemed inclusive of all proper methods of care nor exclusive of other methods of care reasonably directed to obtaining the same results, even for those indications scored as inappropriate.”¹⁶ The Mohs AUC app (https://www.aad.org/members/aad-apps/mohs-auc) is free and allows users to enter tumor and patient characteristics to determine the score for their specific scenario.

Final Thoughts

Scoring systems are emerging in dermatology as evidence-based bedside tools to help guide clinical decision-making. Despite their limitations, these scores have the potential to make a meaningful impact in dermatology as they have in other specialties.

The practice of dermatology is rife with bedside tools: swabs, smears, and scoring systems. First popularized in specialties such as emergency medicine and internal medicine, clinical scoring systems are now emerging in dermatology. These evidence-based scores can be calculated quickly at the bedside—often through a free smartphone app—to help guide clinical decision-making regarding diagnosis, prognosis, and management. As with any medical tool, scoring systems have limitations and should be used as a supplement, not substitute, for one’s clinical judgement. This article reviews 4 clinical scoring systems practical for dermatology residents.

SCORTEN Prognosticates Cases of Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis

Perhaps the best-known scoring system in dermatology, the SCORTEN is widely used to predict hospital mortality from Stevens-Johnson syndrome/toxic epidermal necrolysis. The SCORTEN includes 7 variables of equal weight—age of 40 years or older, heart rate of 120 beats per minute or more, cancer/hematologic malignancy, involved body surface area (BSA) greater than 10%, serum urea greater than 10 mmol/L, serum bicarbonate less than 20 mmol/L, and serum glucose greater than 14 mmol/L—each contributing 1 point to the overall score if present.¹ The involved BSA is defined as the sum of detached and detachable epidermis.¹

The SCORTEN was developed and prospectively validated to be calculated at the end of the first 24 hours of admission; for this calculation, use the BSA affected at that time, and use the most abnormal values during the first 24 hours of admission for the other variables.¹ In addition, a follow-up study including some of the original coauthors recommends recalculating the SCORTEN at the end of hospital day 3, having found that the score’s predictive value was better on this day than hospital days 1, 2, 4, or 5.² Based on the original study, a SCORTEN of 0 to 1 corresponds to a mortality rate of 3.2%, 2 to 12.1%, 3 to 35.3%, 4 to 58.3%, and 5 or greater to 90.0%.¹

Limitations of the SCORTEN include its ability to overestimate or underestimate mortality as demonstrated by 2 multi-institutional cohorts.^3,4 Recently, the ABCD-10 score was developed as an alternative to the SCORTEN and was found to predict mortality similarly when validated in an internal cohort.⁵

PEST Screens for Psoriatic Arthritis

Dermatologists play an important role in screening for psoriatic arthritis, as an estimated 1 in 5 patients with psoriasis have psoriatic arthritis.⁶ To this end, several screening tools have been developed to help differentiate psoriatic arthritis from other arthritides. Joint guidelines from the American Academy of Dermatology and the National Psoriasis Foundation acknowledge that “. . . these screening tools have tended to perform less well when tested in groups of people other than those for which they were originally developed. As such, their usefulness in routine clinical practice remains controversial.”⁷ Nevertheless, the guidelines state, “[b]ecause screening and early detection of inflammatory arthritis are essential to optimize patient [quality of life] and reduce morbidity, providers may consider using a formal screening tool of their choice.”⁷

With these limitations in mind, I have found the Psoriasis Epidemiology Screening Tool (PEST) to be the most useful psoriatic arthritis screening tool. One study determined that the PEST has the best trade-off between sensitivity and specificity compared to 2 other psoriatic arthritis screening tools, the Psoriatic Arthritis Screening and Evaluation (PASE) and the Early Arthritis for Psoriatic Patients (EARP).⁸

The PEST is comprised of 5 questions: (1) Have you ever had a swollen joint (or joints)? (2) Has a doctor ever told you that you have arthritis? (3) Do your fingernails or toenails have holes or pits? (4) Have you had pain in your heel? (5) Have you had a finger or toe that was completely swollen and painful for no apparent reason? According to the PEST, a referral to a rheumatologist should be considered for patients answering yes to 3 or more questions, which is 97% sensitive and 79% specific for psoriatic arthritis.⁹ Patients who answer yes to fewer than 3 questions should still be referred to a rheumatologist if there is a strong clinical suspicion of psoriatic arthritis.¹⁰

The PEST can be accessed for free in 13 languages via the GRAPPA (Group for Research and Assessment of Psoriasis and Psoriatic Arthritis) app as well as downloaded for free from the National Psoriasis Foundation’s website (https://www.psoriasis.org/psa-screening/providers).

ALT-70 Differentiates Cellulitis From Pseudocellulitis

Overdiagnosing cellulitis in the United States has been estimated to result in up to 130,000 unnecessary hospitalizations and up to $515 million in avoidable health care spending.¹¹ Dermatologists are in a unique position to help fix this issue. In one retrospective study of 1430 inpatient dermatology consultations, 74.32% of inpatients evaluated for presumed cellulitis by a dermatologist were instead diagnosed with a cellulitis mimicker (ie, pseudocellulitis), such as stasis dermatitis or contact dermatitis.¹²

The ALT-70 score was developed and prospectively validated to help differentiate lower extremity cellulitis from pseudocellulitis in adult patients in the emergency department (ED).¹³ In addition, the score has retrospectively been shown to function similarly in the inpatient setting when calculated at 24 and 48 hours after ED presentation.¹⁴ Although the ALT-70 score was designed for use by frontline clinicians prior to dermatology consultation, I also have found it helpful to calculate as a consultant, as it provides an objective measure of risk to communicate to the primary team in support of one diagnosis or another.

ALT-70 is an acronym for the score’s 4 variables: asymmetry, leukocytosis, tachycardia, and age of 70 years or older.¹⁵ If present, each variable confers a certain number of points to the final score: 3 points for asymmetry (defined as unilateral leg involvement), 1 point for leukocytosis (white blood cell count ≥10,000/μL), 1 point for tachycardia (≥90 beats per minute), and 2 points for age of 70 years or older. An ALT-70 score of 0 to 2 corresponds to an 83.3% or greater chance of pseudocellulitis, suggesting that the diagnosis of cellulitis be reconsidered. A score of 3 to 4 is indeterminate, and additional information such as a dermatology consultation should be pursued. A score of 5 to 7 corresponds to an 82.2% or greater chance of cellulitis, signifying that empiric treatment with antibiotics be considered.¹⁵

The ALT-70 score does not apply to cases involving areas other than the lower extremities; intravenous antibiotic use within 48 hours before ED presentation; surgery within the last 30 days; abscess; penetrating trauma; burn; or known history of osteomyelitis, diabetic ulcer, or indwelling hardware at the site of infection.¹⁵ The ALT-70 score is available for free via the MDCalc app and website (https://www.mdcalc.com/alt-70-score-cellulitis).

Mohs AUC Determines the Appropriateness of Mohs Micrographic Surgery

In 2012, the American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic Surgery Association, and American Society for Mohs Surgery published appropriate use criteria (AUC) to guide the decision to pursue Mohs micrographic surgery (MMS) in the United States.¹⁶ Based on various tumor and patient characteristics, the Mohs AUC assign scores to 270 different clinical scenarios. A score of 1 to 3 signifies that MMS is inappropriate and generally not considered acceptable. A score 4 to 6 indicates that the appropriateness of MMS is uncertain. A score 7 to 9 means that MMS is appropriate and generally considered acceptable.¹⁶

Since publication, the Mohs AUC have been criticized for classifying most primary superficial basal cell carcinomas as appropriate for MMS¹⁷ (which an AUC coauthor¹⁸ and others^19,20 have defended), excluding certain reasons for performing MMS (such as operating on multiple tumors on the same day),²¹ including counterintuitive scores,²² and omitting trials from Europe²³ (which AUC coauthors also have defended²⁴). As with any clinical scoring system, the Mohs AUC has limitations; the creators acknowledge that “. . . these criteria should not be interpreted as setting a standard of care, or be deemed inclusive of all proper methods of care nor exclusive of other methods of care reasonably directed to obtaining the same results, even for those indications scored as inappropriate.”¹⁶ The Mohs AUC app (https://www.aad.org/members/aad-apps/mohs-auc) is free and allows users to enter tumor and patient characteristics to determine the score for their specific scenario.

Final Thoughts

Scoring systems are emerging in dermatology as evidence-based bedside tools to help guide clinical decision-making. Despite their limitations, these scores have the potential to make a meaningful impact in dermatology as they have in other specialties.

Jame Abraham, MD, has been appointed chair of the hematology/medical oncology department at Cleveland Clinic in Ohio. In this new role, Dr. Abraham will “recruit and develop staff and guide the department’s focus on patient access and a multidisciplinary approach to care,” according to a statement.

Dr. Abraham is also director of the breast oncology program at Taussig Cancer Institute, codirector of the Cleveland Clinic comprehensive breast cancer program, and a professor of medicine at Cleveland Clinic Lerner College of Medicine. He takes the helm from hematologist Matt Kalaycio, MD. Dr. Kalaycio also serves as editor-in-chief of Hematology News.

In other news, Zhe Ying, PhD, of the Fred Hutchinson Cancer Research Center in Seattle, has received a 5-year Pathway to Independence Award from the National Institute of Dental and Craniofacial Research.

With this award funding, Dr. Ying will investigate oncogene-induced differentiation in PI3K-mutant head and neck squamous cell carcinoma. Specifically, he aims to determine if genetic mutations and niche factors can overcome oncogene-induced differentiation to promote tumorigenesis.

Another grant winner is Gina Mantia-Smaldone, MD, of Fox Chase Cancer Center in Philadelphia. She will receive 3 years of funding from the Gynecologic Oncology Group Foundation and NRG Oncology to study gynecologic malignancies.

This award will also provide Dr. Mantia-Smaldone with research mentorship and opportunities to collaborate with other researchers. Her research is focused on developing targeted therapies for ovarian and endometrial cancers that will, ideally, improve patients’ quality of life.

Lastly, Edna (Eti) Cukierman, PhD, of Fox Chase Cancer Center, received a grant to conduct research with Ashani Weeraratna, PhD, of Johns Hopkins University in Baltimore, and Vivek Shenoy, PhD, and Arjun Raj, PhD, both of the University of Pennsylvania in Philadelphia.

The grant, from the National Cancer Institute, will be used to investigate the link between cell aging and melanoma. Dr. Cukierman, Dr. Weeraratna, Dr. Shenoy, and Dr. Raj will focus their research "on better understanding the deterioration of collagen integrity via cellular aging and its role in melanoma metastasis,” according to a statement.

[email protected]

Jame Abraham, MD, has been appointed chair of the hematology/medical oncology department at Cleveland Clinic in Ohio. In this new role, Dr. Abraham will “recruit and develop staff and guide the department’s focus on patient access and a multidisciplinary approach to care,” according to a statement.

Dr. Abraham is also director of the breast oncology program at Taussig Cancer Institute, codirector of the Cleveland Clinic comprehensive breast cancer program, and a professor of medicine at Cleveland Clinic Lerner College of Medicine. He takes the helm from hematologist Matt Kalaycio, MD. Dr. Kalaycio also serves as editor-in-chief of Hematology News.

In other news, Zhe Ying, PhD, of the Fred Hutchinson Cancer Research Center in Seattle, has received a 5-year Pathway to Independence Award from the National Institute of Dental and Craniofacial Research.

With this award funding, Dr. Ying will investigate oncogene-induced differentiation in PI3K-mutant head and neck squamous cell carcinoma. Specifically, he aims to determine if genetic mutations and niche factors can overcome oncogene-induced differentiation to promote tumorigenesis.

Another grant winner is Gina Mantia-Smaldone, MD, of Fox Chase Cancer Center in Philadelphia. She will receive 3 years of funding from the Gynecologic Oncology Group Foundation and NRG Oncology to study gynecologic malignancies.

This award will also provide Dr. Mantia-Smaldone with research mentorship and opportunities to collaborate with other researchers. Her research is focused on developing targeted therapies for ovarian and endometrial cancers that will, ideally, improve patients’ quality of life.

Lastly, Edna (Eti) Cukierman, PhD, of Fox Chase Cancer Center, received a grant to conduct research with Ashani Weeraratna, PhD, of Johns Hopkins University in Baltimore, and Vivek Shenoy, PhD, and Arjun Raj, PhD, both of the University of Pennsylvania in Philadelphia.

The grant, from the National Cancer Institute, will be used to investigate the link between cell aging and melanoma. Dr. Cukierman, Dr. Weeraratna, Dr. Shenoy, and Dr. Raj will focus their research "on better understanding the deterioration of collagen integrity via cellular aging and its role in melanoma metastasis,” according to a statement.

[email protected]

Jame Abraham, MD, has been appointed chair of the hematology/medical oncology department at Cleveland Clinic in Ohio. In this new role, Dr. Abraham will “recruit and develop staff and guide the department’s focus on patient access and a multidisciplinary approach to care,” according to a statement.

Dr. Abraham is also director of the breast oncology program at Taussig Cancer Institute, codirector of the Cleveland Clinic comprehensive breast cancer program, and a professor of medicine at Cleveland Clinic Lerner College of Medicine. He takes the helm from hematologist Matt Kalaycio, MD. Dr. Kalaycio also serves as editor-in-chief of Hematology News.

In other news, Zhe Ying, PhD, of the Fred Hutchinson Cancer Research Center in Seattle, has received a 5-year Pathway to Independence Award from the National Institute of Dental and Craniofacial Research.

With this award funding, Dr. Ying will investigate oncogene-induced differentiation in PI3K-mutant head and neck squamous cell carcinoma. Specifically, he aims to determine if genetic mutations and niche factors can overcome oncogene-induced differentiation to promote tumorigenesis.

Another grant winner is Gina Mantia-Smaldone, MD, of Fox Chase Cancer Center in Philadelphia. She will receive 3 years of funding from the Gynecologic Oncology Group Foundation and NRG Oncology to study gynecologic malignancies.

This award will also provide Dr. Mantia-Smaldone with research mentorship and opportunities to collaborate with other researchers. Her research is focused on developing targeted therapies for ovarian and endometrial cancers that will, ideally, improve patients’ quality of life.

Lastly, Edna (Eti) Cukierman, PhD, of Fox Chase Cancer Center, received a grant to conduct research with Ashani Weeraratna, PhD, of Johns Hopkins University in Baltimore, and Vivek Shenoy, PhD, and Arjun Raj, PhD, both of the University of Pennsylvania in Philadelphia.

The grant, from the National Cancer Institute, will be used to investigate the link between cell aging and melanoma. Dr. Cukierman, Dr. Weeraratna, Dr. Shenoy, and Dr. Raj will focus their research "on better understanding the deterioration of collagen integrity via cellular aging and its role in melanoma metastasis,” according to a statement.

[email protected]

A study of 305 cases of acute flaccid myelitis has found further evidence of a viral etiology but is yet to identify a single pathogen as the primary cause.

Writing in Pediatrics, researchers published an analysis of patients presenting with acute flaccid limb weakness from January 2015 to December 2017 across 43 states.

A total of 25 cases were judged as probable for acute flaccid myelitis (AFM) because they met clinical criteria and had a white blood cell count above 5 cells per mm³ in cerebrospinal fluid, while 193 were judged as confirmed cases based on the additional presence of spinal cord gray matter lesions on MRI.

Overall, 83% of patients had experienced fever, cough, runny nose, vomiting, and/or diarrhea for a median of 5 days before limb weakness began. Two-thirds of patients had experienced a respiratory illness, 62% had experienced a fever, and 29% had experienced gastrointestinal illness.

Overall, 47% of the 193 patients who had specimens tested at a Centers for Disease Control and Prevention or non-CDC laboratory had a pathogen found at any site, 10% had a pathogen detected from a sterile site such as cerebrospinal fluid or sera, and 42% had a pathogen detected from a nonsterile site.

Among 72 patients who had serum specimens tested at the CDC, 2 were positive for enteroviruses. Among the 90 patients who had upper respiratory specimens tested, 36% were positive for either enteroviruses or rhinoviruses.

A number of stool specimens were also tested; 15% were positive for enteroviruses or rhinoviruses and one was positive for parechovirus.

Cerebrospinal fluid was tested in 170 patients, of which 4 were positive for enteroviruses. The testing also found adenovirus, Epstein-Barr virus, human herpesvirus 6, and mycoplasma in six patients. Sera testing of 123 patients found 9 were positive for enteroviruses, West Nile virus, mycoplasma, and coxsackievirus B.

“In our summary of national AFM surveillance from 2015 to 2017, we demonstrate that cases were widely distributed across the United States, the majority of cases occurred in late summer or fall, children were predominantly affected, there is a spectrum of clinical severity, and no single pathogen was identified as the primary cause of AFM,” wrote Tracy Ayers, PhD, from the National Center for Immunization and Respiratory Diseases, and coauthors. “We conclude that symptoms of a viral syndrome within the week before limb weakness, detection of viral pathogens from sterile and nonsterile sites from almost half of patients, and seasonality of AFM incidence, particularly during the 2016 peak year, strongly suggest a viral etiology, including [enteroviruses].”

The authors of an accompanying editorial noted that the clinical syndrome of acute flaccid paralysis caused by myelitis in the gray matter of the spinal cord has previously been associated with a range of viruses, including poliovirus, enteroviruses, and flaviviruses, so a single etiology to explain all cases would not be expected.

“The central question remains: What is driving seasonal biennial nationwide outbreaks of AFM since 2014?” wrote Kevin Messaca, MD, and colleagues from the University of Colorado at Denver, Aurora.

Two authors declared consultancies, grants, and research contracts with the pharmaceutical sector. No other conflicts of interest were declared. One editorial author declared funding from the National Institute of Allergy and Infectious Diseases.

SOURCE: Ayers T et al. Pediatrics. 2019 Oct 7. doi: 10.1542/peds.2019-1619.

*Updated 10/14/2019.

A study of 305 cases of acute flaccid myelitis has found further evidence of a viral etiology but is yet to identify a single pathogen as the primary cause.

Writing in Pediatrics, researchers published an analysis of patients presenting with acute flaccid limb weakness from January 2015 to December 2017 across 43 states.

A total of 25 cases were judged as probable for acute flaccid myelitis (AFM) because they met clinical criteria and had a white blood cell count above 5 cells per mm³ in cerebrospinal fluid, while 193 were judged as confirmed cases based on the additional presence of spinal cord gray matter lesions on MRI.

Overall, 83% of patients had experienced fever, cough, runny nose, vomiting, and/or diarrhea for a median of 5 days before limb weakness began. Two-thirds of patients had experienced a respiratory illness, 62% had experienced a fever, and 29% had experienced gastrointestinal illness.

Overall, 47% of the 193 patients who had specimens tested at a Centers for Disease Control and Prevention or non-CDC laboratory had a pathogen found at any site, 10% had a pathogen detected from a sterile site such as cerebrospinal fluid or sera, and 42% had a pathogen detected from a nonsterile site.

Among 72 patients who had serum specimens tested at the CDC, 2 were positive for enteroviruses. Among the 90 patients who had upper respiratory specimens tested, 36% were positive for either enteroviruses or rhinoviruses.

A number of stool specimens were also tested; 15% were positive for enteroviruses or rhinoviruses and one was positive for parechovirus.

Cerebrospinal fluid was tested in 170 patients, of which 4 were positive for enteroviruses. The testing also found adenovirus, Epstein-Barr virus, human herpesvirus 6, and mycoplasma in six patients. Sera testing of 123 patients found 9 were positive for enteroviruses, West Nile virus, mycoplasma, and coxsackievirus B.

“In our summary of national AFM surveillance from 2015 to 2017, we demonstrate that cases were widely distributed across the United States, the majority of cases occurred in late summer or fall, children were predominantly affected, there is a spectrum of clinical severity, and no single pathogen was identified as the primary cause of AFM,” wrote Tracy Ayers, PhD, from the National Center for Immunization and Respiratory Diseases, and coauthors. “We conclude that symptoms of a viral syndrome within the week before limb weakness, detection of viral pathogens from sterile and nonsterile sites from almost half of patients, and seasonality of AFM incidence, particularly during the 2016 peak year, strongly suggest a viral etiology, including [enteroviruses].”

The authors of an accompanying editorial noted that the clinical syndrome of acute flaccid paralysis caused by myelitis in the gray matter of the spinal cord has previously been associated with a range of viruses, including poliovirus, enteroviruses, and flaviviruses, so a single etiology to explain all cases would not be expected.

“The central question remains: What is driving seasonal biennial nationwide outbreaks of AFM since 2014?” wrote Kevin Messaca, MD, and colleagues from the University of Colorado at Denver, Aurora.

Two authors declared consultancies, grants, and research contracts with the pharmaceutical sector. No other conflicts of interest were declared. One editorial author declared funding from the National Institute of Allergy and Infectious Diseases.

SOURCE: Ayers T et al. Pediatrics. 2019 Oct 7. doi: 10.1542/peds.2019-1619.

*Updated 10/14/2019.

A study of 305 cases of acute flaccid myelitis has found further evidence of a viral etiology but is yet to identify a single pathogen as the primary cause.

Writing in Pediatrics, researchers published an analysis of patients presenting with acute flaccid limb weakness from January 2015 to December 2017 across 43 states.

A total of 25 cases were judged as probable for acute flaccid myelitis (AFM) because they met clinical criteria and had a white blood cell count above 5 cells per mm³ in cerebrospinal fluid, while 193 were judged as confirmed cases based on the additional presence of spinal cord gray matter lesions on MRI.

Overall, 83% of patients had experienced fever, cough, runny nose, vomiting, and/or diarrhea for a median of 5 days before limb weakness began. Two-thirds of patients had experienced a respiratory illness, 62% had experienced a fever, and 29% had experienced gastrointestinal illness.

Overall, 47% of the 193 patients who had specimens tested at a Centers for Disease Control and Prevention or non-CDC laboratory had a pathogen found at any site, 10% had a pathogen detected from a sterile site such as cerebrospinal fluid or sera, and 42% had a pathogen detected from a nonsterile site.

Among 72 patients who had serum specimens tested at the CDC, 2 were positive for enteroviruses. Among the 90 patients who had upper respiratory specimens tested, 36% were positive for either enteroviruses or rhinoviruses.

A number of stool specimens were also tested; 15% were positive for enteroviruses or rhinoviruses and one was positive for parechovirus.

Cerebrospinal fluid was tested in 170 patients, of which 4 were positive for enteroviruses. The testing also found adenovirus, Epstein-Barr virus, human herpesvirus 6, and mycoplasma in six patients. Sera testing of 123 patients found 9 were positive for enteroviruses, West Nile virus, mycoplasma, and coxsackievirus B.

“In our summary of national AFM surveillance from 2015 to 2017, we demonstrate that cases were widely distributed across the United States, the majority of cases occurred in late summer or fall, children were predominantly affected, there is a spectrum of clinical severity, and no single pathogen was identified as the primary cause of AFM,” wrote Tracy Ayers, PhD, from the National Center for Immunization and Respiratory Diseases, and coauthors. “We conclude that symptoms of a viral syndrome within the week before limb weakness, detection of viral pathogens from sterile and nonsterile sites from almost half of patients, and seasonality of AFM incidence, particularly during the 2016 peak year, strongly suggest a viral etiology, including [enteroviruses].”

The authors of an accompanying editorial noted that the clinical syndrome of acute flaccid paralysis caused by myelitis in the gray matter of the spinal cord has previously been associated with a range of viruses, including poliovirus, enteroviruses, and flaviviruses, so a single etiology to explain all cases would not be expected.

“The central question remains: What is driving seasonal biennial nationwide outbreaks of AFM since 2014?” wrote Kevin Messaca, MD, and colleagues from the University of Colorado at Denver, Aurora.

Two authors declared consultancies, grants, and research contracts with the pharmaceutical sector. No other conflicts of interest were declared. One editorial author declared funding from the National Institute of Allergy and Infectious Diseases.

SOURCE: Ayers T et al. Pediatrics. 2019 Oct 7. doi: 10.1542/peds.2019-1619.

*Updated 10/14/2019.

ANSWER

The correct interpretation is atrial fibrillation with aberrantly conducted complexes. The lead I rhythm strip at the bottom of the ECG shows the irregularly irregular rate. There are narrow complexes (see beats 3-7 and 16-18), indicating normal conduction through the atrioventricular node and His-Purkinje system. The remainder of the complexes are wide and aberrantly conducted and are in the same vector as the normally conducted (narrow) complexes.

An important take-away from this case: The computer reading includes a PR interval as well as a QRS duration of 88 ms. There is no PR interval in atrial fibrillation—highlighting the importance of reading the ECG and not relying on the computer’s interpretation. The QRS duration is measured in the normally conducted beats only; it does not include the aberrantly conducted beats.

ANSWER

The correct interpretation is atrial fibrillation with aberrantly conducted complexes. The lead I rhythm strip at the bottom of the ECG shows the irregularly irregular rate. There are narrow complexes (see beats 3-7 and 16-18), indicating normal conduction through the atrioventricular node and His-Purkinje system. The remainder of the complexes are wide and aberrantly conducted and are in the same vector as the normally conducted (narrow) complexes.

An important take-away from this case: The computer reading includes a PR interval as well as a QRS duration of 88 ms. There is no PR interval in atrial fibrillation—highlighting the importance of reading the ECG and not relying on the computer’s interpretation. The QRS duration is measured in the normally conducted beats only; it does not include the aberrantly conducted beats.

ANSWER

The correct interpretation is atrial fibrillation with aberrantly conducted complexes. The lead I rhythm strip at the bottom of the ECG shows the irregularly irregular rate. There are narrow complexes (see beats 3-7 and 16-18), indicating normal conduction through the atrioventricular node and His-Purkinje system. The remainder of the complexes are wide and aberrantly conducted and are in the same vector as the normally conducted (narrow) complexes.

An important take-away from this case: The computer reading includes a PR interval as well as a QRS duration of 88 ms. There is no PR interval in atrial fibrillation—highlighting the importance of reading the ECG and not relying on the computer’s interpretation. The QRS duration is measured in the normally conducted beats only; it does not include the aberrantly conducted beats.

From the Oregon Health and Science University, Portland, OR.

Abstract

• Objective: To describe the current approach to diagnosis and treatment of aplastic anemia.
• Methods: Review of the literature.
• Results: Aplastic anemia can be acquired or associated with an inherited marrow failure syndrome (IMFS), and the treatment and prognosis vary dramatically between these 2 etiologies. Patients may present along a spectrum, ranging from being asymptomatic with incidental findings on peripheral blood testing to life-threatening neutropenic infections or bleeding. Workup and diagnosis involves investigating IMFSs and ruling out malignant or infectious etiologies for pancytopenia.
• Conclusion: Treatment outcomes are excellent with modern supportive care and the current approach to allogeneic transplantation, and therefore referral to a bone marrow transplant program to evaluate for early transplantation is the new standard of care for aplastic anemia.

Keywords: inherited marrow failure syndrome; Fanconi anemia; immunosuppression; transplant; stem cell.

Aplastic anemia is a clinical and pathological entity of bone marrow failure that causes progressive loss of hematopoietic progenitor stem cells (HPSC), resulting in pancytopenia.¹ Patients may present along a spectrum, ranging from being asymptomatic with incidental findings on peripheral blood testing to having life-threatening neutropenic infections or bleeding. Aplastic anemia results from either inherited or acquired causes, and the pathophysiology and treatment approach vary significantly between these 2 causes. Therefore, recognition of inherited marrow failure diseases, such as Fanconi anemia and telomere biology disorders, is critical to establishing the management plan. 

Epidemiology

Aplastic anemia is a rare disorder, with an incidence of approximately 1.5 to 7 cases per million individuals per year.^2,3 A recent Scandinavian study reported that the incidence of aplastic anemia among the Swedish population is 2.3 cases per million individuals per year, with a median age at diagnosis of 60 years and a slight female predominance (52% versus 48%, respectively).² This data is congruent with prior observations made in Barcelona, where the incidence was 2.34 cases per million individuals per year, albeit with a slightly higher incidence in males compared to females (2.54 versus 2.16, respectively).⁴ The incidence of aplastic anemia varies globally, with a disproportionate increase in incidence seen among Asian populations, with rates as high as 8.8 per million individuals per year.^3-5 This variation in incidence in Asia versus other countries has not been well explained. There appears to be a bimodal distribution, with incidence peaks seen in young adults and in older adults.^2,3,6

Pathophysiology

Acquired Aplastic Anemia

The leading hypothesis as to the cause of most cases of acquired aplastic anemia is that a dysregulated immune system destroys HPSCs. Inciting etiologies implicated in the development of acquired aplastic anemia include pregnancy, infection, medications, and exposure to certain chemicals, such as benzene.^1,7 The historical understanding of acquired aplastic anemia implicates cytotoxic T-lymphocyte–mediated destruction of CD34+ hematopoietic stem cells.^1,8,9 This hypothesis served as the basis for treatment of acquired aplastic anemia with immunosuppressive therapy, predominantly anti-thymocyte globulin (ATG) combined with cyclosporine A.^1,8 More recent work has focused on cytokine interactions, particularly the suppressive role of interferon (IFN)-γ on hematopoietic stem cells independent of T-lymphocyte–mediated destruction, which has been demonstrated in a murine model.⁸ The interaction of IFN-γ with the hematopoietic stem cell pool is dynamic. IFN-γ levels are elevated during an acute inflammatory response, such as a viral infection, providing further basis for the immune-mediated nature of the acquired disease.¹⁰ Specifically, in vitro studies suggest the effects of IFN-γ on HPSC may be secondary to interruption of thrombopoietin and its respective signaling pathways, which play a key role in hematopoietic stem cell renewal.¹¹ Eltrombopag, a thrombopoietin receptor antagonist, has shown promise in the treatment of refractory aplastic anemia, with studies indicating that its effectiveness is independent of IFN-γ levels.^11,12

Inherited Aplastic Anemia

The inherited marrow failure syndromes (IMFSs) are a group of disorders characterized by cellular maintenance and repair defects, leading to cytopenias, increased cancer risk, structural defects, and risk of end organ damage, such as liver cirrhosis and pulmonary fibrosis.^13-15 The most common diseases include Fanconi anemia, dyskeratosis congenita/telomere biology disorders, Diamond-Blackfan anemia, and Shwachman-Diamond syndrome, but with the advent of whole exome sequencing, new syndromes continue to be discovered. While classically these disorders present in children, adult presentations are now commonplace. Broadly, the pathophysiology of inherited aplastic anemia relates to the defective HPSCs and an accelerated decline of the hematopoietic stem cell compartment.

The most common IMFSs, Fanconi anemia and telomere biology disorders, are associated with numerous mutations in DNA damage repair pathways and telomere maintenance pathways. TERT, DKC, and TERC mutations are most commonly associated with dyskeratosis congenita, but may also be found infrequently in patients with aplastic anemia presenting at an older age in the absence of the classic phenotypical features.^1,16,17 The recognition of an underlying genetic disorder or telomere biology disorder leading to constitutional aplastic anemia is significant, as these conditions are associated not only with marrow failure, but also with endocrinopathies, organ fibrosis, and and hematopoietic and solid organ malignancies.^13-15 In particular, TERT and TERC gene mutations have been associated with dyskeratosis congenita as well as pulmonary fibrosis and cirrhosis.^18,19 The implications of early diagnosis of an IMFS lie in the approach to treatment and prognosis.

Clonal Disorders and Secondary Malignancies

Myelodysplastic syndrome (MDS) and secondary acute myeloid leukemia (AML) are 2 clonal disorders that may arise from a background of aplastic anemia.^9,20,21 Hypoplastic MDS can be difficult to differentiate from aplastic anemia at diagnosis based on morphology alone, although recent work has demonstrated that molecular testing for somatic mutations in ASXL1, DNMT3A, and BCOR can aid in differentiating a subset of aplastic anemia patients who are more likely to progress to MDS.²¹ Clonal populations of cells harboring 6p uniparental disomy are seen in more than 10% of patients with aplastic anemia on cytogenetic analysis, which can help differentiate the diseases.⁹ Yoshizato and colleagues found lower rates of ASXL1 and DNMT3A mutations in patients with aplastic anemia as compared with patients with MDS or AML. In this study, patients with aplastic anemia had higher rates of mutations in PIGA (reflecting the increased paroxysmal nocturnal hemoglobinuria [PNH] clonality seen in aplastic anemia) and BCOR.⁹ Mutations were also found in genes commonly mutated in MDS and AML, including TET2, RUNX1, TP53, and JAK2, albeit at lower frequencies.⁹ These mutations as a whole have not predicted response to therapy or prognosis. However, when performing survival analysis in patients with specific mutations, those commonly encountered in MDS/AML (ASXL1, DNMT3A, TP53, RUNX1, CSMD1) are associated with faster progression to overt MDS/AML and decreased overall survival (OS),^20,21 suggesting these mutations may represent early clonality that can lead to clonal evolution and the development of secondary malignancies. Conversely, mutations in BCOR and BCORL appear to identify patients who may have a favorable outcome in response to immunosuppressive therapy and, similar to patients with PIGA mutations, improved OS.⁹

Paroxysmal Nocturnal Hemoglobinuria

In addition to having an increased risk of myelodysplasia and malignancy due to the development of a dominant pre-malignant clone, patients with aplastic anemia often harbor progenitor cell clones associated with PNH.^1,17 PNH clones have been identified in more than 50% of patients with aplastic anemia.^22,23 PNH represents a clonal disorder of hematopoiesis in which cells harbor X-linked somatic mutations in the PIGA gene; this gene encodes a protein responsible for the synthesis of glycosylphosphatidylinositol anchors on the cell surface.^22,24 The lack of these cell surface proteins, specifically CD55 (also known as decay accelerating factor) and CD59 (also known as membrane inhibitor of reactive lysis), predisposes red cells to increased complement-mediated lysis.²⁵ The exact mechanism for the development of these clones in patients with aplastic anemia is not fully understood. Current theories hypothesize that the clones are protected from the immune-mediated destruction of normal hematopoietic stem cells due to the absence of the cell surface proteins.^1,20 The role of these clones over time in patients with aplastic anemia is less clear, though recent work demonstrated that despite differences in clonality over the disease course, aplastic anemia patients with small PNH clones are less likely to develop overt hemolysis and larger PNH clones compared to patients harboring larger (≥ 50%) PNH clones at diagnosis.^23,26,27 Additionally, PNH clones in patients with aplastic anemia infrequently become clinically significant.²⁷ It should be noted that these conditions exist along a continuum; that is, patients with aplastic anemia may develop PNH clones, while conversely patients with PNH may develop aplastic anemia.²⁰ Patients with PNH clones should be followed via peripheral blood flow cytometry and complete blood count to track clonal stability and identify clinically significant PNH among aplastic anemia patients.²⁸

Clinical Presentation

Patients with aplastic anemia typically are diagnosed either due to asymptomatic cytopenias found on peripheral blood sampling, symptomatic anemia, bleeding secondary to thrombocytopenia, or wound healing and infectious complications related to neutropenia.²⁹ A thorough history to understand the timing of symptoms, recent infectious symptoms/exposure, habits, and chemical or toxin exposures (including medications, travel, and supplements) helps guide diagnostic testing. Family history is also critical, with attention given to premature graying; pulmonary, renal, and liver disease; and blood disorders.

Patients with an IMFS (eg, Fanconi anemia or dyskeratosis congenita) may have associated phenotypical findings such as urogenital abnormalities or short stature; in addition, those with dyskeratosis congenita may present with the classic triad of oral leukoplakia, lacy skin pigmentation, and dystrophic nails.⁷ However, classic phenotypical findings may be lacking in up to 30% to 40% of patients with an IMFS.⁷ As described previously, while congenital malformations are common in Fanconi anemia and dyskeratosis congenita, a third of patients may have no or only subtle phenotypical abnormalities, including alterations in skin or hair pigmentation, skeletal and growth abnormalities, and endocrine disorders.³⁰ The International Fanconi Anemia Registry identified central nervous system, genitourinary, skin and musculoskeletal, ophthalmic, and gastrointestinal system malformations among children with Fanconi anemia.^31,32 Patients with dyskeratosis congenita may present with pulmonary fibrosis, hepatic cirrhosis, or premature graying, as highlighted in a recent study by DiNardo and colleagues.³³ Therefore, physicians must have a heightened index of suspicion in patients with subtle phenotypical findings and associated cytopenias.

The diagnosis of aplastic anemia should be suspected in any patient presenting with pancytopenia. Aplastic anemia is a diagnosis of exclusion.³⁴ Other conditions associated with peripheral blood pancytopenia should be considered, including infections (HIV, hepatitis, parvovirus B19, cytomegalovirus, Epstein-Barr virus, varicella-zoster virus), nutritional deficiencies (vitamin B12, folate, copper, zinc), autoimmune disease (systemic lupus erythematosus, rheumatoid arthritis, hemophagocytic lymphohistiocytosis), hypersplenism, marrow-occupying diseases (eg, leukemia, lymphoma, MDS), solid malignancies, and fibrosis (Table).⁷

Diagnostic Evaluation

The workup for aplastic anemia should include a thorough history and physical exam to search simultaneously for alternative diagnoses and clues pointing to potential etiologic agents.⁷ Diagnostic tests to be performed include a complete blood count with differential, reticulocyte count, immature platelet fraction, flow cytometry (to rule out lymphoproliferative disorders and atypical myeloid cells and to evaluate for PNH), and bone marrow biopsy with subsequent cytogenetic, immunohistochemical, and molecular testing.³⁵ Typical findings in aplastic anemia include peripheral blood pancytopenia without dysplastic features and bone marrow biopsy demonstrating a hypocellular marrow.⁷ A relative lymphocytosis in the peripheral blood is common.⁷ In patients with a significant PNH clone, a macrocytosis along with elevated lactate dehydrogenase and elevated reticulocyte and granulocyte counts may be present.³⁶

The diagnosis (based on the Camitta criteria³⁷ and modified Camitta criteria³⁸ for severe aplastic anemia) requires 2 of the following findings on peripheral blood samples:

• Absolute neutrophil count (ANC) < 500 cells/µL
• Platelet count < 20,000 cells/µL
• Reticulocyte count < 1% corrected or < 20,000 cells/µL.³⁵

In addition to peripheral blood findings, bone marrow biopsy is essential for the diagnosis, and should demonstrate a markedly hypocellular marrow (cellularity < 25%), occasionally with an increase in T lymphocytes.^7,39 Because marrow cellularity varies with age and can be challenging to assess, additional biopsies may be needed to confirm the diagnosis.²⁹ A 1- to 2-cm core biopsy is necessary to confirm hypocellularity, as small areas of residual hematopoiesis may be present and obscure the diagnosis.³⁵

Excluding Hypocellular MDS and IMFS

Excluding hypocellular MDS is challenging, especially in the older adult presenting with aplastic anemia, as patients with aplastic anemia may have some degree of erythroid dysplasia on bone marrow morphology.³⁶ The presence of a PNH clone on flow cytometry can aid in diagnosing aplastic anemia and excluding MDS,³⁴ although PNH clones can be present in refractory anemia MDS. Patients with aplastic anemia have a lower ratio of CD34+ cells compared to those with hypoplastic MDS, with 1 study demonstrating a mean CD34+ percentage of < 0.5% in aplastic anemia versus 3.7% in hypoplastic MDS.⁴⁰ Cytogenetic and molecular testing can also aid in making this distinction by identifying mutations commonly implicated in MDS.⁷ The presence of monosomy 7 (-7) in aplastic anemia patients is associated with a poor overall prognosis.^34,41

Peripheral blood screening using chromosome breakage analysis (done using either mitomycin C or diepoxybutane as in vitro DNA-crosslinking agents)⁴² and telomere length testing (of peripheral blood leukocytes) is necessary to exclude the main IMFSs, Fanconi anemia and telomere biology disorders, respectively. Ruling out these conditions is imperative, as the approach to treatment varies significantly between IMFS and aplastic anemia. Patients with shortened telomeres should undergo genetic screening for mutations in the telomere maintenance genes to evaluate the underlying defect leading to shortened telomeres. Patients with increased peripheral blood breakage should have genetic testing to detect mutations associated with Fanconi anemia.

Classification

Once the diagnosis of aplastic anemia has been made, the patient should be classified according to the severity of their disease. Disease severity is determined based on peripheral blood ANC: non-severe aplastic anemia (NSAA), ANC > 500 polymorphonuclear neutrophils (PMNs)/µL; severe aplastic anemia (SAA), 200–500 PMNs/µL; and very severe aplastic anemia (VSAA), 0–200 PMNs/µL.^4,34 Disease classification is important, as VSAA is associated with a decreased OS compared to SAA.² Disease classification may affect treatment decisions, as patients with NSAA may be observed for a short period of time, while, conversely, patients with SAA have a worse prognosis with delays in therapy.^43-45

Treatment of Inherited Aplastic Anemia

First-line treatment options for patients with IMFS are androgen therapy and hematopoietic stem cell transplant (HSCT). When evaluating patients for HSCT, it is critical to identify the presence of an IMFS, as the risk and mortality associated with the conditioning regimen, stem cell source, graft-versus-host disease (GVHD), and secondary malignancies differ between patients with IMFS and those with acquired marrow failure syndromes or hematologic malignancies.

Potential sibling donors need to be screened for donor candidacy as well as for the inherited defect. Among patients with Fanconi anemia or a telomere biology disorder, the stem cell source must be considered, with multiple studies in IMFSs and SAA showing superior outcomes with a bone marrow product compared to peripheral blood stem cells.^46-48 In IMFS patients, the donor cell type may affect the choice of conditioning regimen.^5,6 Reduced-intensity conditioning in lieu of myeloablative conditioning without total body irradiation has proved feasible in patients with Fanconi anemia, and is associated with a reduced risk of secondary malignancies.^49,50 Incorporation of fludarabine in the conditioning regimen of patients without a matched sibling donor is associated with superior engraftment and survival^46,49,51 compared to cyclophosphamide conditioning, which was historically used in matched related donors.^50,52  Adding fludarabine appears to be especially beneficial in older patients, in whom its use is associated with lower rates of graft failure, likely due to increased immunosuppression at the time of engraftment.^51,53 Fludarabine has also been incorporated into conditioning regimens for patients with a telomere biology disorder, but outcomes data are limited.⁵

For patients presenting with AML or a high-risk MDS who are subsequently diagnosed with an IMFS, treatment can be more complex, as these patients are at high risk for toxicity from standard chemotherapy. Limited data suggest that induction therapy and transplantation are feasible in this group of patients, and this approach is associated with increased OS, despite lower OS rates than those of IMFS patients who present prior to the development of MDS or AML.^54,55 Further work is needed to determine the optimal induction regimen that balances the risks of treatment-related mortality and complications associated with conditioning regimens, risk of relapse, and risk of secondary malignancies, especially in the cohort of patients diagnosed at an older age.

Treatment of Acquired Aplastic Anemia

Supportive Care

While the workup and treatment plan are being established, attention should be directed at supportive care for prevention of complications. The most common complications leading to death in patients with significant pancytopenia and neutropenia are opportunistic infections and hemorrhagic complications.²

Transfusion support is critical to avoid symptomatic anemia and hemorrhagic complications related to thrombocytopenia, which typically occur with platelet counts lower than 10,000 cells/µL. However, transfusion carries the risk of alloimmunization (which may persist for years following transfusion) and transfusion-related graft versus host disease (trGVHD), and thus use of transfusion should be minimized when possible.^56,57 All blood products given to patients with aplastic anemia should be irradiated and leukoreduced to reduce the risk of both alloimmunization and trGVHD. Guidelines from the British Society for Haematology recommend routine screening for Rh and Kell antibodies to reduce the risk of alloimmunization.⁵⁸ Infectious complications remain a common cause of morbidity and mortality in patients with aplastic anemia who have prolonged neutropenia (defined as an ANC < 500 cells/µL).^59-62 Therefore, patients should receive broad-spectrum antibiotics with antipseudomonal coverage. In a study evaluating the role of granulocyte-colony stimulating factor (G-CSF) in patients with SAA receiving immunosuppressive therapy, 55% of all patient deaths were secondary to infection.⁶³ There was no OS benefit seen in patients who received G-CSF, though a significantly lower rate of infection was observed in the G-CSF arm compared to those not receiving G-CSF (56% versus 81%, P = 0.006). This difference was largely driven by a decrease in infectious episodes in patients with VSAA treated with G-CSF as compared to those who did not receive this therapy (22% versus 48%, P = 0.014).⁶³

Angio-invasive pulmonary aspergillosis and Zygomycetes (eg, Rhizopus, Mucor species) remain major causes of mortality related to opportunistic mycotic infections in patients with aplastic anemia.¹⁸ The infectious risk is directly related to the duration and severity of neutropenia, with one study demonstrating a significant increase in risk in AML patients with neutropenia lasting longer than 3 weeks.⁶⁴ Invasive fungal infections carry a high mortality in patients with severe neutropenia, though due to earlier recognition and empiric antifungal therapy with extended-spectrum azoles, overall mortality secondary to invasive fungal infections is declining.^62,65

While neutropenia related to cytotoxic chemotherapy is commonly associated with gram-negative bacteria due to disruption of mucosal barriers, patients with aplastic anemia have an increased incidence of gram-positive bacteremia with staphylococcal species compared to other neutropenic populations.^61,62 This appears to be changing with time. Valdez et al demonstrated a decrease in prevalence of coagulase-negative staphylococcal infections, increased prevalence of gram-positive bacilli bacteremia, and no change in prevalence of gram-negative bacteremia in patients with aplastic anemia treated between 1989 and 2008.⁶⁵ Gram-negative bacteremia caused by Stenotrophomonas maltophila, Escherichia coli, Klebsiella pneumoniae, Citrobacter, and Proteus has also been reported.⁶² Despite a lack of clinical trials investigating the role of antifungal and antibacterial prophylaxis for patients with aplastic anemia, most centers initiate antifungal prophylaxis in patients with SAA or VSAA with an anti-mold agent such as voriconazole or posaconazole (which has the additional benefit compared to voriconazole of covering Mucor species).^60,66 This is especially true for patients who have received ATG or undergone HSCT. For antimicrobial prophylaxis, a fluoroquinolone antibiotic with a spectrum of activity against Pseudomonas should be considered for patients with an ANC < 500 cells/µL.⁶⁰ Acyclovir or valacyclovir prophylaxis is recommended for varicella-zoster virus and herpes simplex virus. Cytomegalovirus reactivation is minimal in patients with aplastic anemia, unless multiple courses of ATG are used.

Iron overload is another complication the provider must be aware of in the setting of increased transfusions in aplastic anemia patients. Lee and colleagues showed that iron chelation therapy using deferasirox is effective at reducing serum ferritin levels in patients with aplastic anemia (median ferritin level of 3254 ng/mL prior to therapy, 1854 ng/mL following), and is associated with no serious adverse events (most common adverse events included nausea, diarrhea, vomiting, and rash).⁶⁷ Approximately 25% of patients in this trial had an increase in creatinine, with patients taking concomitant cyclosporine affected to a greater degree than those on chelation therapy alone. For patients following HSCT or with improved hematopoiesis following immunosuppressive therapy, phlebotomy can be used to treat iron overload in lieu of chelation therapy.⁵⁸

Approach to Therapy

The main treatment options for SAA and VSAA include allogeneic bone marrow transplant and immunosuppression. The deciding factors as to which treatment is best initially depends on the availability of HLA-matched related donors and age (Figure 1 and Figure 2). Survival is decreased in patients with SAA or VSAA who delay initiation of therapy, and therefore prompt referral for HLA typing and evaluation for bone marrow transplant is a very important first step in managing aplastic anemia.

Matched Sibling Donor Transplant. Current standards of care recommend HLA-matched sibling donor transplant for patients with SAA or VSAA who are younger than 50 years, with the caveat that integration of fludarabine and reduced cyclophosphamide dosing along with ATG shows the best overall outcomes. Locasciulli and colleagues examined outcomes in patients given either immunosuppressive therapy or sibling HSCT between 1991-1996 and 1997-2002, respectively, and found that sibling HSCT was associated with a superior 10-year OS compared to immunosuppressive therapy (73% versus 68%).⁴³ Interestingly in this study, there was no OS improvement seen with immunosuppressive therapy alone (69% versus 73%) between the 2 time periods, despite increased OS in both sibling HSCT (74% and 80%) and MUD HSCT (38% and 65%).⁴³ Though total body irradiation has been used in the past, it is typically not included in current conditioning regimens for matched related donor transplants.⁶⁸

Current conditioning regimens typically use a combination of cyclophosphamide and ATG,^69,70 with or without fludarabine. Fludarabine-based conditioning regimens have shown promise in patients undergoing sibling HSCT. Maury and colleagues evaluated the role of fludarabine in addition to low-dose cyclophosphamide and ATG compared to cyclophosphamide alone or in combination with ATG in patients over age 30 undergoing sibling HSCT.⁵³ There was a nonsignificant improvement in 5-year OS in the fludarabine arm compared to controls (77% ± 8% versus 60% ± 3%, P = 0.14) in the pooled analysis, but when adjusted for age the fludarabine arm had a significantly lower relative risk (RR) of death (0.44; P = 0.04) compared to the control arm. Shin et al reported outcomes with fludarabine/cyclophosphamide/ATG, with excellent overall outcomes and no difference in patients older or younger than 40 years.⁷¹

Kim et al evaluated their experience with patients older than 40 years receiving matched related donors, finding comparable outcomes in those ages 41 to 50 years compared to younger patients. Outcomes declined in those over the age of 50 years.⁷² Long-term data for matched related donor transplant for aplastic anemia show excellent long-term outcomes, with minimal chronic GVHD and good performance status.⁷³ Hence, these factors support the role of matched related donor transplant as the initial treatment in SAA and VSAA.

Regarding the role of transplant for patients who lack a matched related donor, a growing body of literature demonstrating identical outcomes between matched related and MUD transplants for pediatric patients^74,75 supports recent recommendations for upfront unrelated donor transplantation for aplastic anemia.^76,77

Immunosuppressive Therapy. For patients without an HLA-matched sibling donor or those who are older than 50 years of age, immunosuppressive therapy is the first-line therapy. ATG and cyclosporine A are the treatments of choice.⁷⁸ The potential effectiveness of immunosuppressive therapy in treating aplastic anemia was initially observed in patients in whom autologous transplant failed but who still experienced hematopoietic reconstitution despite the failed graft; this observation led to the hypothesis that the conditioning regimen may have an effect on hematopoiesis.^59,78,79

Immunosuppressive therapy with ATG has been used for the treatment of aplastic anemia since the 1980s.⁸⁰ Historically, rabbit ATG had been used, but a 2011 study of horse ATG demonstrated superior hematological response at 6 months compared to rabbit ATG (68% versus 37%).⁵⁹ Superior survival was also seen with horse ATG compared to rabbit ATG (3-year OS, 96% versus 76%). Due to these results, horse ATG is preferred over rabbit ATG. ATG should be used in combination with cyclosporine A to optimize outcomes.

Early studies also demonstrated the efficacy of cyclosporine A in the treatment of aplastic anemia, with response rates equivalent to that of ATG monotherapy.⁸¹ Recent publications still note the efficacy of cyclosporine A in the treatment of aplastic anemia. Its role as an affordable option for single-agent therapy in developing countries is intriguing.⁸¹ The combination of ATG and cyclosporine A was proven superior to either agent alone in a study by Frickhofen et al.⁷⁹ In this study, patients were randomly assigned to a control arm that received ATG plus methylprednisolone or to an arm that received ATG plus cyclosporine A and methylprednisolone. At 6 months, 70% of patients in the cyclosporine A arm had a complete remission (CR) or partial remission compared to 46% in the control arm.⁸² Further work confirmed the long-term efficacy of this regimen, reporting a 7-year OS of 55%.⁸³ Among a pediatric population, immunosuppressive therapy was associated with an 83% 10-year OS.⁸⁴

It is recommended that patients remain on cyclosporine therapy for a minimum of 6 months, after which a gradual taper may be considered, although there is variation among practitioners, with some continuing immunosuppressive therapy for a minimum of 12 months due to a proportion of patients being cyclosporine dependent.^34,84 A study found that within a population of patients who responded to immunosuppressive therapy, 18% became cyclosporine dependent.⁸⁴ The median duration of cyclosporine A treatment at full dose was 12 months, with tapering completed over a median of 19 months after patients had been in a stable CR for a minimum of 3 months. Relapse occurred more often when patients were tapered quickly (decrease ≥ 0.8 mg/kg/month) compared to slowly (0.4-0.7 mg/kg/month) or very slowly (< 0.3 mg/kg/month).

Townsley and colleagues recently investigated incorporating the use of the thrombopoietin receptor agonist eltrombopag with immunosuppressive therapy as first-line therapy in aplastic anemia.⁸⁵ When given at a dose of 150 mg daily in patients ages 12 years and older or 75 mg daily in patients younger than 12 years, in conjunction with cyclosporine A and ATG, patients demonstrated markedly improved hematological response compared to historical treatment with standard immunosuppressive therapy alone.⁴⁵ In the patient cohort administered eltrombopag starting on day 1 and continuing for 6 months, the complete response rate was 58%. Eltrombopag led to improvement in all cell lines among all treatment subgroups, and OS (censored for patients who proceeded to transplant) was 99% at 2 years.¹² Overall, toxicities associated with this therapy were low, with liver enzyme elevations most commonly observed.⁸⁵ Recently, a phase 2 trial of immunosuppressive therapy with or without eltrombopag was reported. Of the 38 patients enrolled, overall response, complete response, and time to response were not statistically different.⁸⁶ With this recent finding, the role of eltrombopag in addition to immunosuppressive therapy is not clearly defined, and further studies are warranted.

OS for patients who do not respond to immunosuppressive therapy is approximately 57% at 5 years, largely due to improved supportive measures among this patient population.^48,65 Therefore, it is important to recognize those patients who have a low chance of response so that second-line therapy can be pursued to improve outcomes.

Matched Unrelated Donor Transplant. For patients with refractory disease following immunosuppressive therapy who lack a matched sibling donor, MUD HSCT is considered standard therapy given the marked improvement in overall outcomes with modulating conditioning regimens and high-resolution HLA typing. A European Society for Blood and Marrow Transplantation (EBMT) analysis comparing matched sibling HSCT to MUD HSCT noted significantly higher rates of acute grade II-IV and grade III-IV GVHD (grade II-IV 13% versus 25%, grade III-IV 5% versus 10%) among patients undergoing MUD transplant.⁴⁷ Chronic GVHD rates were 14% in the sibling group, as compared to 26% in the MUD group. Factors associated with improved survival in this analysis include transplant under age 20 years (84% versus 72%), transplant within 6 months of diagnosis (85% versus 72%), the use of ATG in the conditioning regimen (81% versus 73%), and cytomegalovirus-negative donor and recipient as compared to other combinations (82% versus 76%).⁸⁷ Interestingly, this study demonstrated that OS was not significantly increased when using a sibling HSCT compared to a MUD HSCT, likely as a result of improved understanding of conditioning regimens, GVHD prophylaxis, and supportive care.

Additional studies of MUD HSCT have shown outcomes similar to those seen in sibling HSCT.^34,48 A French study found a significant increase in survival in patients undergoing MUD HSCT compared to historical cohorts (2000-2005: OS 52%; 2006-2012: OS 74%).⁷⁵ The majority of patients underwent conditioning with cyclophosphamide or a combination of busulfan and cyclophosphamide, with or without fludarabine; 81% of patients underwent in vivo T-cell depletion, and a bone marrow donor source was utilized. OS was significantly lower in patients over age 30 years undergoing MUD HSCT (57%) compared to those under age 30 years (70%). Improved OS was also seen when patients underwent transplant within 1 year of diagnosis and when a 10/10 matched donor (compared to a 9/10 mismatched donor) was utilized.⁴⁸

A 2015 study investigated the role of MUD HSCT as frontline therapy instead of immunosuppressive therapy in patients without a matched sibling donor.⁷⁵ The 2-year OS was 96% in the MUD HSCT cohort compared to 91%, 94%, and 74% in historical cohorts of sibling HSCT, frontline immunosuppressive therapy, and second-line MUD HSCT following failed immunosuppressive therapy, respectively. Additionally, event-free survival in the MUD HSCT cohort (defined by the authors as death, lack of response, relapse, occurrence of clonal evolution/clinical PNH, malignancies developing over follow‐up, and transplant for patients receiving immunosuppressive therapy frontline) was similar compared to sibling HSCT and superior to frontline immunosuppressive therapy and second-line MUD HSCT. Furthermore, Samarasinghe et al highlighted the importance of in vivo T-cell depletion with either ATG or alemtuzumab (anti-CD52 monoclonal antibody) in the prevention of acute and chronic GVHD in both sibling HSCT and MUD HSCT.⁸⁸

With continued improvement of less toxic and more immunomodulating conditioning regimens,utilization of bone marrow as a donor cell source, in vivo T-cell depletion, and use of GVHD and antimicrobial prophylaxis, more clinical evidence supports elevating MUD HSCT in the treatment plan for patients without a matched sibling donor.⁸⁹ However, there is still a large population of patients without matched sibling or unrelated donor options. Given the need to expand the transplant pool and thus avoid clonal hematopoiesis, clinically significant PNH, and relapsed aplastic anemia, more work continues to recognize the expanding role of alternative donor transplants (cord blood and haploidentical) as another viable treatment strategy for aplastic anemia after immunosuppressive therapy failure.⁹⁰

Summary

Aplastic anemia is a rare but potentially life-threatening disorder with pancytopenia and a marked reduction in the HSC compartment. It can be acquired or associated with an IMFS, and the treatment and prognosis vary dramatically between these 2 etiologies. Workup and diagnosis involves investigating IMFSs and ruling out malignant or infectious etiologies for pancytopenia. Treatment outcomes are excellent with modern supportive care and the current approach to allogeneic transplantation, and therefore referral to a bone marrow transplant program to evaluate for early transplantation is the new standard of care.

Corresponding author: Gabrielle Meyers, MD, 3181 SW Sam Jackson Park Road, Mail Code UHN73C, Portland, OR 97239.

Financial disclosures: None.

From the Oregon Health and Science University, Portland, OR.

Abstract

• Objective: To describe the current approach to diagnosis and treatment of aplastic anemia.
• Methods: Review of the literature.
• Results: Aplastic anemia can be acquired or associated with an inherited marrow failure syndrome (IMFS), and the treatment and prognosis vary dramatically between these 2 etiologies. Patients may present along a spectrum, ranging from being asymptomatic with incidental findings on peripheral blood testing to life-threatening neutropenic infections or bleeding. Workup and diagnosis involves investigating IMFSs and ruling out malignant or infectious etiologies for pancytopenia.
• Conclusion: Treatment outcomes are excellent with modern supportive care and the current approach to allogeneic transplantation, and therefore referral to a bone marrow transplant program to evaluate for early transplantation is the new standard of care for aplastic anemia.

Keywords: inherited marrow failure syndrome; Fanconi anemia; immunosuppression; transplant; stem cell.

Aplastic anemia is a clinical and pathological entity of bone marrow failure that causes progressive loss of hematopoietic progenitor stem cells (HPSC), resulting in pancytopenia.¹ Patients may present along a spectrum, ranging from being asymptomatic with incidental findings on peripheral blood testing to having life-threatening neutropenic infections or bleeding. Aplastic anemia results from either inherited or acquired causes, and the pathophysiology and treatment approach vary significantly between these 2 causes. Therefore, recognition of inherited marrow failure diseases, such as Fanconi anemia and telomere biology disorders, is critical to establishing the management plan. 

Epidemiology

Aplastic anemia is a rare disorder, with an incidence of approximately 1.5 to 7 cases per million individuals per year.^2,3 A recent Scandinavian study reported that the incidence of aplastic anemia among the Swedish population is 2.3 cases per million individuals per year, with a median age at diagnosis of 60 years and a slight female predominance (52% versus 48%, respectively).² This data is congruent with prior observations made in Barcelona, where the incidence was 2.34 cases per million individuals per year, albeit with a slightly higher incidence in males compared to females (2.54 versus 2.16, respectively).⁴ The incidence of aplastic anemia varies globally, with a disproportionate increase in incidence seen among Asian populations, with rates as high as 8.8 per million individuals per year.^3-5 This variation in incidence in Asia versus other countries has not been well explained. There appears to be a bimodal distribution, with incidence peaks seen in young adults and in older adults.^2,3,6

Pathophysiology

Acquired Aplastic Anemia

The leading hypothesis as to the cause of most cases of acquired aplastic anemia is that a dysregulated immune system destroys HPSCs. Inciting etiologies implicated in the development of acquired aplastic anemia include pregnancy, infection, medications, and exposure to certain chemicals, such as benzene.^1,7 The historical understanding of acquired aplastic anemia implicates cytotoxic T-lymphocyte–mediated destruction of CD34+ hematopoietic stem cells.^1,8,9 This hypothesis served as the basis for treatment of acquired aplastic anemia with immunosuppressive therapy, predominantly anti-thymocyte globulin (ATG) combined with cyclosporine A.^1,8 More recent work has focused on cytokine interactions, particularly the suppressive role of interferon (IFN)-γ on hematopoietic stem cells independent of T-lymphocyte–mediated destruction, which has been demonstrated in a murine model.⁸ The interaction of IFN-γ with the hematopoietic stem cell pool is dynamic. IFN-γ levels are elevated during an acute inflammatory response, such as a viral infection, providing further basis for the immune-mediated nature of the acquired disease.¹⁰ Specifically, in vitro studies suggest the effects of IFN-γ on HPSC may be secondary to interruption of thrombopoietin and its respective signaling pathways, which play a key role in hematopoietic stem cell renewal.¹¹ Eltrombopag, a thrombopoietin receptor antagonist, has shown promise in the treatment of refractory aplastic anemia, with studies indicating that its effectiveness is independent of IFN-γ levels.^11,12

Inherited Aplastic Anemia

The inherited marrow failure syndromes (IMFSs) are a group of disorders characterized by cellular maintenance and repair defects, leading to cytopenias, increased cancer risk, structural defects, and risk of end organ damage, such as liver cirrhosis and pulmonary fibrosis.^13-15 The most common diseases include Fanconi anemia, dyskeratosis congenita/telomere biology disorders, Diamond-Blackfan anemia, and Shwachman-Diamond syndrome, but with the advent of whole exome sequencing, new syndromes continue to be discovered. While classically these disorders present in children, adult presentations are now commonplace. Broadly, the pathophysiology of inherited aplastic anemia relates to the defective HPSCs and an accelerated decline of the hematopoietic stem cell compartment.

The most common IMFSs, Fanconi anemia and telomere biology disorders, are associated with numerous mutations in DNA damage repair pathways and telomere maintenance pathways. TERT, DKC, and TERC mutations are most commonly associated with dyskeratosis congenita, but may also be found infrequently in patients with aplastic anemia presenting at an older age in the absence of the classic phenotypical features.^1,16,17 The recognition of an underlying genetic disorder or telomere biology disorder leading to constitutional aplastic anemia is significant, as these conditions are associated not only with marrow failure, but also with endocrinopathies, organ fibrosis, and and hematopoietic and solid organ malignancies.^13-15 In particular, TERT and TERC gene mutations have been associated with dyskeratosis congenita as well as pulmonary fibrosis and cirrhosis.^18,19 The implications of early diagnosis of an IMFS lie in the approach to treatment and prognosis.

Clonal Disorders and Secondary Malignancies

Myelodysplastic syndrome (MDS) and secondary acute myeloid leukemia (AML) are 2 clonal disorders that may arise from a background of aplastic anemia.^9,20,21 Hypoplastic MDS can be difficult to differentiate from aplastic anemia at diagnosis based on morphology alone, although recent work has demonstrated that molecular testing for somatic mutations in ASXL1, DNMT3A, and BCOR can aid in differentiating a subset of aplastic anemia patients who are more likely to progress to MDS.²¹ Clonal populations of cells harboring 6p uniparental disomy are seen in more than 10% of patients with aplastic anemia on cytogenetic analysis, which can help differentiate the diseases.⁹ Yoshizato and colleagues found lower rates of ASXL1 and DNMT3A mutations in patients with aplastic anemia as compared with patients with MDS or AML. In this study, patients with aplastic anemia had higher rates of mutations in PIGA (reflecting the increased paroxysmal nocturnal hemoglobinuria [PNH] clonality seen in aplastic anemia) and BCOR.⁹ Mutations were also found in genes commonly mutated in MDS and AML, including TET2, RUNX1, TP53, and JAK2, albeit at lower frequencies.⁹ These mutations as a whole have not predicted response to therapy or prognosis. However, when performing survival analysis in patients with specific mutations, those commonly encountered in MDS/AML (ASXL1, DNMT3A, TP53, RUNX1, CSMD1) are associated with faster progression to overt MDS/AML and decreased overall survival (OS),^20,21 suggesting these mutations may represent early clonality that can lead to clonal evolution and the development of secondary malignancies. Conversely, mutations in BCOR and BCORL appear to identify patients who may have a favorable outcome in response to immunosuppressive therapy and, similar to patients with PIGA mutations, improved OS.⁹

Paroxysmal Nocturnal Hemoglobinuria

In addition to having an increased risk of myelodysplasia and malignancy due to the development of a dominant pre-malignant clone, patients with aplastic anemia often harbor progenitor cell clones associated with PNH.^1,17 PNH clones have been identified in more than 50% of patients with aplastic anemia.^22,23 PNH represents a clonal disorder of hematopoiesis in which cells harbor X-linked somatic mutations in the PIGA gene; this gene encodes a protein responsible for the synthesis of glycosylphosphatidylinositol anchors on the cell surface.^22,24 The lack of these cell surface proteins, specifically CD55 (also known as decay accelerating factor) and CD59 (also known as membrane inhibitor of reactive lysis), predisposes red cells to increased complement-mediated lysis.²⁵ The exact mechanism for the development of these clones in patients with aplastic anemia is not fully understood. Current theories hypothesize that the clones are protected from the immune-mediated destruction of normal hematopoietic stem cells due to the absence of the cell surface proteins.^1,20 The role of these clones over time in patients with aplastic anemia is less clear, though recent work demonstrated that despite differences in clonality over the disease course, aplastic anemia patients with small PNH clones are less likely to develop overt hemolysis and larger PNH clones compared to patients harboring larger (≥ 50%) PNH clones at diagnosis.^23,26,27 Additionally, PNH clones in patients with aplastic anemia infrequently become clinically significant.²⁷ It should be noted that these conditions exist along a continuum; that is, patients with aplastic anemia may develop PNH clones, while conversely patients with PNH may develop aplastic anemia.²⁰ Patients with PNH clones should be followed via peripheral blood flow cytometry and complete blood count to track clonal stability and identify clinically significant PNH among aplastic anemia patients.²⁸

Clinical Presentation

Patients with aplastic anemia typically are diagnosed either due to asymptomatic cytopenias found on peripheral blood sampling, symptomatic anemia, bleeding secondary to thrombocytopenia, or wound healing and infectious complications related to neutropenia.²⁹ A thorough history to understand the timing of symptoms, recent infectious symptoms/exposure, habits, and chemical or toxin exposures (including medications, travel, and supplements) helps guide diagnostic testing. Family history is also critical, with attention given to premature graying; pulmonary, renal, and liver disease; and blood disorders.

Patients with an IMFS (eg, Fanconi anemia or dyskeratosis congenita) may have associated phenotypical findings such as urogenital abnormalities or short stature; in addition, those with dyskeratosis congenita may present with the classic triad of oral leukoplakia, lacy skin pigmentation, and dystrophic nails.⁷ However, classic phenotypical findings may be lacking in up to 30% to 40% of patients with an IMFS.⁷ As described previously, while congenital malformations are common in Fanconi anemia and dyskeratosis congenita, a third of patients may have no or only subtle phenotypical abnormalities, including alterations in skin or hair pigmentation, skeletal and growth abnormalities, and endocrine disorders.³⁰ The International Fanconi Anemia Registry identified central nervous system, genitourinary, skin and musculoskeletal, ophthalmic, and gastrointestinal system malformations among children with Fanconi anemia.^31,32 Patients with dyskeratosis congenita may present with pulmonary fibrosis, hepatic cirrhosis, or premature graying, as highlighted in a recent study by DiNardo and colleagues.³³ Therefore, physicians must have a heightened index of suspicion in patients with subtle phenotypical findings and associated cytopenias.

The diagnosis of aplastic anemia should be suspected in any patient presenting with pancytopenia. Aplastic anemia is a diagnosis of exclusion.³⁴ Other conditions associated with peripheral blood pancytopenia should be considered, including infections (HIV, hepatitis, parvovirus B19, cytomegalovirus, Epstein-Barr virus, varicella-zoster virus), nutritional deficiencies (vitamin B12, folate, copper, zinc), autoimmune disease (systemic lupus erythematosus, rheumatoid arthritis, hemophagocytic lymphohistiocytosis), hypersplenism, marrow-occupying diseases (eg, leukemia, lymphoma, MDS), solid malignancies, and fibrosis (Table).⁷

Diagnostic Evaluation

The workup for aplastic anemia should include a thorough history and physical exam to search simultaneously for alternative diagnoses and clues pointing to potential etiologic agents.⁷ Diagnostic tests to be performed include a complete blood count with differential, reticulocyte count, immature platelet fraction, flow cytometry (to rule out lymphoproliferative disorders and atypical myeloid cells and to evaluate for PNH), and bone marrow biopsy with subsequent cytogenetic, immunohistochemical, and molecular testing.³⁵ Typical findings in aplastic anemia include peripheral blood pancytopenia without dysplastic features and bone marrow biopsy demonstrating a hypocellular marrow.⁷ A relative lymphocytosis in the peripheral blood is common.⁷ In patients with a significant PNH clone, a macrocytosis along with elevated lactate dehydrogenase and elevated reticulocyte and granulocyte counts may be present.³⁶

The diagnosis (based on the Camitta criteria³⁷ and modified Camitta criteria³⁸ for severe aplastic anemia) requires 2 of the following findings on peripheral blood samples:

• Absolute neutrophil count (ANC) < 500 cells/µL
• Platelet count < 20,000 cells/µL
• Reticulocyte count < 1% corrected or < 20,000 cells/µL.³⁵

In addition to peripheral blood findings, bone marrow biopsy is essential for the diagnosis, and should demonstrate a markedly hypocellular marrow (cellularity < 25%), occasionally with an increase in T lymphocytes.^7,39 Because marrow cellularity varies with age and can be challenging to assess, additional biopsies may be needed to confirm the diagnosis.²⁹ A 1- to 2-cm core biopsy is necessary to confirm hypocellularity, as small areas of residual hematopoiesis may be present and obscure the diagnosis.³⁵

Excluding Hypocellular MDS and IMFS

Excluding hypocellular MDS is challenging, especially in the older adult presenting with aplastic anemia, as patients with aplastic anemia may have some degree of erythroid dysplasia on bone marrow morphology.³⁶ The presence of a PNH clone on flow cytometry can aid in diagnosing aplastic anemia and excluding MDS,³⁴ although PNH clones can be present in refractory anemia MDS. Patients with aplastic anemia have a lower ratio of CD34+ cells compared to those with hypoplastic MDS, with 1 study demonstrating a mean CD34+ percentage of < 0.5% in aplastic anemia versus 3.7% in hypoplastic MDS.⁴⁰ Cytogenetic and molecular testing can also aid in making this distinction by identifying mutations commonly implicated in MDS.⁷ The presence of monosomy 7 (-7) in aplastic anemia patients is associated with a poor overall prognosis.^34,41

Peripheral blood screening using chromosome breakage analysis (done using either mitomycin C or diepoxybutane as in vitro DNA-crosslinking agents)⁴² and telomere length testing (of peripheral blood leukocytes) is necessary to exclude the main IMFSs, Fanconi anemia and telomere biology disorders, respectively. Ruling out these conditions is imperative, as the approach to treatment varies significantly between IMFS and aplastic anemia. Patients with shortened telomeres should undergo genetic screening for mutations in the telomere maintenance genes to evaluate the underlying defect leading to shortened telomeres. Patients with increased peripheral blood breakage should have genetic testing to detect mutations associated with Fanconi anemia.

Classification

Once the diagnosis of aplastic anemia has been made, the patient should be classified according to the severity of their disease. Disease severity is determined based on peripheral blood ANC: non-severe aplastic anemia (NSAA), ANC > 500 polymorphonuclear neutrophils (PMNs)/µL; severe aplastic anemia (SAA), 200–500 PMNs/µL; and very severe aplastic anemia (VSAA), 0–200 PMNs/µL.^4,34 Disease classification is important, as VSAA is associated with a decreased OS compared to SAA.² Disease classification may affect treatment decisions, as patients with NSAA may be observed for a short period of time, while, conversely, patients with SAA have a worse prognosis with delays in therapy.^43-45

Treatment of Inherited Aplastic Anemia

First-line treatment options for patients with IMFS are androgen therapy and hematopoietic stem cell transplant (HSCT). When evaluating patients for HSCT, it is critical to identify the presence of an IMFS, as the risk and mortality associated with the conditioning regimen, stem cell source, graft-versus-host disease (GVHD), and secondary malignancies differ between patients with IMFS and those with acquired marrow failure syndromes or hematologic malignancies.

Potential sibling donors need to be screened for donor candidacy as well as for the inherited defect. Among patients with Fanconi anemia or a telomere biology disorder, the stem cell source must be considered, with multiple studies in IMFSs and SAA showing superior outcomes with a bone marrow product compared to peripheral blood stem cells.^46-48 In IMFS patients, the donor cell type may affect the choice of conditioning regimen.^5,6 Reduced-intensity conditioning in lieu of myeloablative conditioning without total body irradiation has proved feasible in patients with Fanconi anemia, and is associated with a reduced risk of secondary malignancies.^49,50 Incorporation of fludarabine in the conditioning regimen of patients without a matched sibling donor is associated with superior engraftment and survival^46,49,51 compared to cyclophosphamide conditioning, which was historically used in matched related donors.^50,52  Adding fludarabine appears to be especially beneficial in older patients, in whom its use is associated with lower rates of graft failure, likely due to increased immunosuppression at the time of engraftment.^51,53 Fludarabine has also been incorporated into conditioning regimens for patients with a telomere biology disorder, but outcomes data are limited.⁵

For patients presenting with AML or a high-risk MDS who are subsequently diagnosed with an IMFS, treatment can be more complex, as these patients are at high risk for toxicity from standard chemotherapy. Limited data suggest that induction therapy and transplantation are feasible in this group of patients, and this approach is associated with increased OS, despite lower OS rates than those of IMFS patients who present prior to the development of MDS or AML.^54,55 Further work is needed to determine the optimal induction regimen that balances the risks of treatment-related mortality and complications associated with conditioning regimens, risk of relapse, and risk of secondary malignancies, especially in the cohort of patients diagnosed at an older age.

Treatment of Acquired Aplastic Anemia

Supportive Care

While the workup and treatment plan are being established, attention should be directed at supportive care for prevention of complications. The most common complications leading to death in patients with significant pancytopenia and neutropenia are opportunistic infections and hemorrhagic complications.²

Transfusion support is critical to avoid symptomatic anemia and hemorrhagic complications related to thrombocytopenia, which typically occur with platelet counts lower than 10,000 cells/µL. However, transfusion carries the risk of alloimmunization (which may persist for years following transfusion) and transfusion-related graft versus host disease (trGVHD), and thus use of transfusion should be minimized when possible.^56,57 All blood products given to patients with aplastic anemia should be irradiated and leukoreduced to reduce the risk of both alloimmunization and trGVHD. Guidelines from the British Society for Haematology recommend routine screening for Rh and Kell antibodies to reduce the risk of alloimmunization.⁵⁸ Infectious complications remain a common cause of morbidity and mortality in patients with aplastic anemia who have prolonged neutropenia (defined as an ANC < 500 cells/µL).^59-62 Therefore, patients should receive broad-spectrum antibiotics with antipseudomonal coverage. In a study evaluating the role of granulocyte-colony stimulating factor (G-CSF) in patients with SAA receiving immunosuppressive therapy, 55% of all patient deaths were secondary to infection.⁶³ There was no OS benefit seen in patients who received G-CSF, though a significantly lower rate of infection was observed in the G-CSF arm compared to those not receiving G-CSF (56% versus 81%, P = 0.006). This difference was largely driven by a decrease in infectious episodes in patients with VSAA treated with G-CSF as compared to those who did not receive this therapy (22% versus 48%, P = 0.014).⁶³

Angio-invasive pulmonary aspergillosis and Zygomycetes (eg, Rhizopus, Mucor species) remain major causes of mortality related to opportunistic mycotic infections in patients with aplastic anemia.¹⁸ The infectious risk is directly related to the duration and severity of neutropenia, with one study demonstrating a significant increase in risk in AML patients with neutropenia lasting longer than 3 weeks.⁶⁴ Invasive fungal infections carry a high mortality in patients with severe neutropenia, though due to earlier recognition and empiric antifungal therapy with extended-spectrum azoles, overall mortality secondary to invasive fungal infections is declining.^62,65

While neutropenia related to cytotoxic chemotherapy is commonly associated with gram-negative bacteria due to disruption of mucosal barriers, patients with aplastic anemia have an increased incidence of gram-positive bacteremia with staphylococcal species compared to other neutropenic populations.^61,62 This appears to be changing with time. Valdez et al demonstrated a decrease in prevalence of coagulase-negative staphylococcal infections, increased prevalence of gram-positive bacilli bacteremia, and no change in prevalence of gram-negative bacteremia in patients with aplastic anemia treated between 1989 and 2008.⁶⁵ Gram-negative bacteremia caused by Stenotrophomonas maltophila, Escherichia coli, Klebsiella pneumoniae, Citrobacter, and Proteus has also been reported.⁶² Despite a lack of clinical trials investigating the role of antifungal and antibacterial prophylaxis for patients with aplastic anemia, most centers initiate antifungal prophylaxis in patients with SAA or VSAA with an anti-mold agent such as voriconazole or posaconazole (which has the additional benefit compared to voriconazole of covering Mucor species).^60,66 This is especially true for patients who have received ATG or undergone HSCT. For antimicrobial prophylaxis, a fluoroquinolone antibiotic with a spectrum of activity against Pseudomonas should be considered for patients with an ANC < 500 cells/µL.⁶⁰ Acyclovir or valacyclovir prophylaxis is recommended for varicella-zoster virus and herpes simplex virus. Cytomegalovirus reactivation is minimal in patients with aplastic anemia, unless multiple courses of ATG are used.

Iron overload is another complication the provider must be aware of in the setting of increased transfusions in aplastic anemia patients. Lee and colleagues showed that iron chelation therapy using deferasirox is effective at reducing serum ferritin levels in patients with aplastic anemia (median ferritin level of 3254 ng/mL prior to therapy, 1854 ng/mL following), and is associated with no serious adverse events (most common adverse events included nausea, diarrhea, vomiting, and rash).⁶⁷ Approximately 25% of patients in this trial had an increase in creatinine, with patients taking concomitant cyclosporine affected to a greater degree than those on chelation therapy alone. For patients following HSCT or with improved hematopoiesis following immunosuppressive therapy, phlebotomy can be used to treat iron overload in lieu of chelation therapy.⁵⁸

Approach to Therapy

The main treatment options for SAA and VSAA include allogeneic bone marrow transplant and immunosuppression. The deciding factors as to which treatment is best initially depends on the availability of HLA-matched related donors and age (Figure 1 and Figure 2). Survival is decreased in patients with SAA or VSAA who delay initiation of therapy, and therefore prompt referral for HLA typing and evaluation for bone marrow transplant is a very important first step in managing aplastic anemia.

Matched Sibling Donor Transplant. Current standards of care recommend HLA-matched sibling donor transplant for patients with SAA or VSAA who are younger than 50 years, with the caveat that integration of fludarabine and reduced cyclophosphamide dosing along with ATG shows the best overall outcomes. Locasciulli and colleagues examined outcomes in patients given either immunosuppressive therapy or sibling HSCT between 1991-1996 and 1997-2002, respectively, and found that sibling HSCT was associated with a superior 10-year OS compared to immunosuppressive therapy (73% versus 68%).⁴³ Interestingly in this study, there was no OS improvement seen with immunosuppressive therapy alone (69% versus 73%) between the 2 time periods, despite increased OS in both sibling HSCT (74% and 80%) and MUD HSCT (38% and 65%).⁴³ Though total body irradiation has been used in the past, it is typically not included in current conditioning regimens for matched related donor transplants.⁶⁸

Current conditioning regimens typically use a combination of cyclophosphamide and ATG,^69,70 with or without fludarabine. Fludarabine-based conditioning regimens have shown promise in patients undergoing sibling HSCT. Maury and colleagues evaluated the role of fludarabine in addition to low-dose cyclophosphamide and ATG compared to cyclophosphamide alone or in combination with ATG in patients over age 30 undergoing sibling HSCT.⁵³ There was a nonsignificant improvement in 5-year OS in the fludarabine arm compared to controls (77% ± 8% versus 60% ± 3%, P = 0.14) in the pooled analysis, but when adjusted for age the fludarabine arm had a significantly lower relative risk (RR) of death (0.44; P = 0.04) compared to the control arm. Shin et al reported outcomes with fludarabine/cyclophosphamide/ATG, with excellent overall outcomes and no difference in patients older or younger than 40 years.⁷¹

Kim et al evaluated their experience with patients older than 40 years receiving matched related donors, finding comparable outcomes in those ages 41 to 50 years compared to younger patients. Outcomes declined in those over the age of 50 years.⁷² Long-term data for matched related donor transplant for aplastic anemia show excellent long-term outcomes, with minimal chronic GVHD and good performance status.⁷³ Hence, these factors support the role of matched related donor transplant as the initial treatment in SAA and VSAA.

Regarding the role of transplant for patients who lack a matched related donor, a growing body of literature demonstrating identical outcomes between matched related and MUD transplants for pediatric patients^74,75 supports recent recommendations for upfront unrelated donor transplantation for aplastic anemia.^76,77

Immunosuppressive Therapy. For patients without an HLA-matched sibling donor or those who are older than 50 years of age, immunosuppressive therapy is the first-line therapy. ATG and cyclosporine A are the treatments of choice.⁷⁸ The potential effectiveness of immunosuppressive therapy in treating aplastic anemia was initially observed in patients in whom autologous transplant failed but who still experienced hematopoietic reconstitution despite the failed graft; this observation led to the hypothesis that the conditioning regimen may have an effect on hematopoiesis.^59,78,79

Immunosuppressive therapy with ATG has been used for the treatment of aplastic anemia since the 1980s.⁸⁰ Historically, rabbit ATG had been used, but a 2011 study of horse ATG demonstrated superior hematological response at 6 months compared to rabbit ATG (68% versus 37%).⁵⁹ Superior survival was also seen with horse ATG compared to rabbit ATG (3-year OS, 96% versus 76%). Due to these results, horse ATG is preferred over rabbit ATG. ATG should be used in combination with cyclosporine A to optimize outcomes.

Early studies also demonstrated the efficacy of cyclosporine A in the treatment of aplastic anemia, with response rates equivalent to that of ATG monotherapy.⁸¹ Recent publications still note the efficacy of cyclosporine A in the treatment of aplastic anemia. Its role as an affordable option for single-agent therapy in developing countries is intriguing.⁸¹ The combination of ATG and cyclosporine A was proven superior to either agent alone in a study by Frickhofen et al.⁷⁹ In this study, patients were randomly assigned to a control arm that received ATG plus methylprednisolone or to an arm that received ATG plus cyclosporine A and methylprednisolone. At 6 months, 70% of patients in the cyclosporine A arm had a complete remission (CR) or partial remission compared to 46% in the control arm.⁸² Further work confirmed the long-term efficacy of this regimen, reporting a 7-year OS of 55%.⁸³ Among a pediatric population, immunosuppressive therapy was associated with an 83% 10-year OS.⁸⁴

It is recommended that patients remain on cyclosporine therapy for a minimum of 6 months, after which a gradual taper may be considered, although there is variation among practitioners, with some continuing immunosuppressive therapy for a minimum of 12 months due to a proportion of patients being cyclosporine dependent.^34,84 A study found that within a population of patients who responded to immunosuppressive therapy, 18% became cyclosporine dependent.⁸⁴ The median duration of cyclosporine A treatment at full dose was 12 months, with tapering completed over a median of 19 months after patients had been in a stable CR for a minimum of 3 months. Relapse occurred more often when patients were tapered quickly (decrease ≥ 0.8 mg/kg/month) compared to slowly (0.4-0.7 mg/kg/month) or very slowly (< 0.3 mg/kg/month).

Townsley and colleagues recently investigated incorporating the use of the thrombopoietin receptor agonist eltrombopag with immunosuppressive therapy as first-line therapy in aplastic anemia.⁸⁵ When given at a dose of 150 mg daily in patients ages 12 years and older or 75 mg daily in patients younger than 12 years, in conjunction with cyclosporine A and ATG, patients demonstrated markedly improved hematological response compared to historical treatment with standard immunosuppressive therapy alone.⁴⁵ In the patient cohort administered eltrombopag starting on day 1 and continuing for 6 months, the complete response rate was 58%. Eltrombopag led to improvement in all cell lines among all treatment subgroups, and OS (censored for patients who proceeded to transplant) was 99% at 2 years.¹² Overall, toxicities associated with this therapy were low, with liver enzyme elevations most commonly observed.⁸⁵ Recently, a phase 2 trial of immunosuppressive therapy with or without eltrombopag was reported. Of the 38 patients enrolled, overall response, complete response, and time to response were not statistically different.⁸⁶ With this recent finding, the role of eltrombopag in addition to immunosuppressive therapy is not clearly defined, and further studies are warranted.

OS for patients who do not respond to immunosuppressive therapy is approximately 57% at 5 years, largely due to improved supportive measures among this patient population.^48,65 Therefore, it is important to recognize those patients who have a low chance of response so that second-line therapy can be pursued to improve outcomes.

Matched Unrelated Donor Transplant. For patients with refractory disease following immunosuppressive therapy who lack a matched sibling donor, MUD HSCT is considered standard therapy given the marked improvement in overall outcomes with modulating conditioning regimens and high-resolution HLA typing. A European Society for Blood and Marrow Transplantation (EBMT) analysis comparing matched sibling HSCT to MUD HSCT noted significantly higher rates of acute grade II-IV and grade III-IV GVHD (grade II-IV 13% versus 25%, grade III-IV 5% versus 10%) among patients undergoing MUD transplant.⁴⁷ Chronic GVHD rates were 14% in the sibling group, as compared to 26% in the MUD group. Factors associated with improved survival in this analysis include transplant under age 20 years (84% versus 72%), transplant within 6 months of diagnosis (85% versus 72%), the use of ATG in the conditioning regimen (81% versus 73%), and cytomegalovirus-negative donor and recipient as compared to other combinations (82% versus 76%).⁸⁷ Interestingly, this study demonstrated that OS was not significantly increased when using a sibling HSCT compared to a MUD HSCT, likely as a result of improved understanding of conditioning regimens, GVHD prophylaxis, and supportive care.

Additional studies of MUD HSCT have shown outcomes similar to those seen in sibling HSCT.^34,48 A French study found a significant increase in survival in patients undergoing MUD HSCT compared to historical cohorts (2000-2005: OS 52%; 2006-2012: OS 74%).⁷⁵ The majority of patients underwent conditioning with cyclophosphamide or a combination of busulfan and cyclophosphamide, with or without fludarabine; 81% of patients underwent in vivo T-cell depletion, and a bone marrow donor source was utilized. OS was significantly lower in patients over age 30 years undergoing MUD HSCT (57%) compared to those under age 30 years (70%). Improved OS was also seen when patients underwent transplant within 1 year of diagnosis and when a 10/10 matched donor (compared to a 9/10 mismatched donor) was utilized.⁴⁸

A 2015 study investigated the role of MUD HSCT as frontline therapy instead of immunosuppressive therapy in patients without a matched sibling donor.⁷⁵ The 2-year OS was 96% in the MUD HSCT cohort compared to 91%, 94%, and 74% in historical cohorts of sibling HSCT, frontline immunosuppressive therapy, and second-line MUD HSCT following failed immunosuppressive therapy, respectively. Additionally, event-free survival in the MUD HSCT cohort (defined by the authors as death, lack of response, relapse, occurrence of clonal evolution/clinical PNH, malignancies developing over follow‐up, and transplant for patients receiving immunosuppressive therapy frontline) was similar compared to sibling HSCT and superior to frontline immunosuppressive therapy and second-line MUD HSCT. Furthermore, Samarasinghe et al highlighted the importance of in vivo T-cell depletion with either ATG or alemtuzumab (anti-CD52 monoclonal antibody) in the prevention of acute and chronic GVHD in both sibling HSCT and MUD HSCT.⁸⁸

With continued improvement of less toxic and more immunomodulating conditioning regimens,utilization of bone marrow as a donor cell source, in vivo T-cell depletion, and use of GVHD and antimicrobial prophylaxis, more clinical evidence supports elevating MUD HSCT in the treatment plan for patients without a matched sibling donor.⁸⁹ However, there is still a large population of patients without matched sibling or unrelated donor options. Given the need to expand the transplant pool and thus avoid clonal hematopoiesis, clinically significant PNH, and relapsed aplastic anemia, more work continues to recognize the expanding role of alternative donor transplants (cord blood and haploidentical) as another viable treatment strategy for aplastic anemia after immunosuppressive therapy failure.⁹⁰

Summary

Aplastic anemia is a rare but potentially life-threatening disorder with pancytopenia and a marked reduction in the HSC compartment. It can be acquired or associated with an IMFS, and the treatment and prognosis vary dramatically between these 2 etiologies. Workup and diagnosis involves investigating IMFSs and ruling out malignant or infectious etiologies for pancytopenia. Treatment outcomes are excellent with modern supportive care and the current approach to allogeneic transplantation, and therefore referral to a bone marrow transplant program to evaluate for early transplantation is the new standard of care.

Corresponding author: Gabrielle Meyers, MD, 3181 SW Sam Jackson Park Road, Mail Code UHN73C, Portland, OR 97239.

Financial disclosures: None.

From the Oregon Health and Science University, Portland, OR.

Abstract

• Objective: To describe the current approach to diagnosis and treatment of aplastic anemia.
• Methods: Review of the literature.
• Results: Aplastic anemia can be acquired or associated with an inherited marrow failure syndrome (IMFS), and the treatment and prognosis vary dramatically between these 2 etiologies. Patients may present along a spectrum, ranging from being asymptomatic with incidental findings on peripheral blood testing to life-threatening neutropenic infections or bleeding. Workup and diagnosis involves investigating IMFSs and ruling out malignant or infectious etiologies for pancytopenia.
• Conclusion: Treatment outcomes are excellent with modern supportive care and the current approach to allogeneic transplantation, and therefore referral to a bone marrow transplant program to evaluate for early transplantation is the new standard of care for aplastic anemia.

Keywords: inherited marrow failure syndrome; Fanconi anemia; immunosuppression; transplant; stem cell.

Aplastic anemia is a clinical and pathological entity of bone marrow failure that causes progressive loss of hematopoietic progenitor stem cells (HPSC), resulting in pancytopenia.¹ Patients may present along a spectrum, ranging from being asymptomatic with incidental findings on peripheral blood testing to having life-threatening neutropenic infections or bleeding. Aplastic anemia results from either inherited or acquired causes, and the pathophysiology and treatment approach vary significantly between these 2 causes. Therefore, recognition of inherited marrow failure diseases, such as Fanconi anemia and telomere biology disorders, is critical to establishing the management plan. 

Epidemiology

Aplastic anemia is a rare disorder, with an incidence of approximately 1.5 to 7 cases per million individuals per year.^2,3 A recent Scandinavian study reported that the incidence of aplastic anemia among the Swedish population is 2.3 cases per million individuals per year, with a median age at diagnosis of 60 years and a slight female predominance (52% versus 48%, respectively).² This data is congruent with prior observations made in Barcelona, where the incidence was 2.34 cases per million individuals per year, albeit with a slightly higher incidence in males compared to females (2.54 versus 2.16, respectively).⁴ The incidence of aplastic anemia varies globally, with a disproportionate increase in incidence seen among Asian populations, with rates as high as 8.8 per million individuals per year.^3-5 This variation in incidence in Asia versus other countries has not been well explained. There appears to be a bimodal distribution, with incidence peaks seen in young adults and in older adults.^2,3,6

Pathophysiology

Acquired Aplastic Anemia

The leading hypothesis as to the cause of most cases of acquired aplastic anemia is that a dysregulated immune system destroys HPSCs. Inciting etiologies implicated in the development of acquired aplastic anemia include pregnancy, infection, medications, and exposure to certain chemicals, such as benzene.^1,7 The historical understanding of acquired aplastic anemia implicates cytotoxic T-lymphocyte–mediated destruction of CD34+ hematopoietic stem cells.^1,8,9 This hypothesis served as the basis for treatment of acquired aplastic anemia with immunosuppressive therapy, predominantly anti-thymocyte globulin (ATG) combined with cyclosporine A.^1,8 More recent work has focused on cytokine interactions, particularly the suppressive role of interferon (IFN)-γ on hematopoietic stem cells independent of T-lymphocyte–mediated destruction, which has been demonstrated in a murine model.⁸ The interaction of IFN-γ with the hematopoietic stem cell pool is dynamic. IFN-γ levels are elevated during an acute inflammatory response, such as a viral infection, providing further basis for the immune-mediated nature of the acquired disease.¹⁰ Specifically, in vitro studies suggest the effects of IFN-γ on HPSC may be secondary to interruption of thrombopoietin and its respective signaling pathways, which play a key role in hematopoietic stem cell renewal.¹¹ Eltrombopag, a thrombopoietin receptor antagonist, has shown promise in the treatment of refractory aplastic anemia, with studies indicating that its effectiveness is independent of IFN-γ levels.^11,12

Inherited Aplastic Anemia

The inherited marrow failure syndromes (IMFSs) are a group of disorders characterized by cellular maintenance and repair defects, leading to cytopenias, increased cancer risk, structural defects, and risk of end organ damage, such as liver cirrhosis and pulmonary fibrosis.^13-15 The most common diseases include Fanconi anemia, dyskeratosis congenita/telomere biology disorders, Diamond-Blackfan anemia, and Shwachman-Diamond syndrome, but with the advent of whole exome sequencing, new syndromes continue to be discovered. While classically these disorders present in children, adult presentations are now commonplace. Broadly, the pathophysiology of inherited aplastic anemia relates to the defective HPSCs and an accelerated decline of the hematopoietic stem cell compartment.

The most common IMFSs, Fanconi anemia and telomere biology disorders, are associated with numerous mutations in DNA damage repair pathways and telomere maintenance pathways. TERT, DKC, and TERC mutations are most commonly associated with dyskeratosis congenita, but may also be found infrequently in patients with aplastic anemia presenting at an older age in the absence of the classic phenotypical features.^1,16,17 The recognition of an underlying genetic disorder or telomere biology disorder leading to constitutional aplastic anemia is significant, as these conditions are associated not only with marrow failure, but also with endocrinopathies, organ fibrosis, and and hematopoietic and solid organ malignancies.^13-15 In particular, TERT and TERC gene mutations have been associated with dyskeratosis congenita as well as pulmonary fibrosis and cirrhosis.^18,19 The implications of early diagnosis of an IMFS lie in the approach to treatment and prognosis.

Clonal Disorders and Secondary Malignancies

Myelodysplastic syndrome (MDS) and secondary acute myeloid leukemia (AML) are 2 clonal disorders that may arise from a background of aplastic anemia.^9,20,21 Hypoplastic MDS can be difficult to differentiate from aplastic anemia at diagnosis based on morphology alone, although recent work has demonstrated that molecular testing for somatic mutations in ASXL1, DNMT3A, and BCOR can aid in differentiating a subset of aplastic anemia patients who are more likely to progress to MDS.²¹ Clonal populations of cells harboring 6p uniparental disomy are seen in more than 10% of patients with aplastic anemia on cytogenetic analysis, which can help differentiate the diseases.⁹ Yoshizato and colleagues found lower rates of ASXL1 and DNMT3A mutations in patients with aplastic anemia as compared with patients with MDS or AML. In this study, patients with aplastic anemia had higher rates of mutations in PIGA (reflecting the increased paroxysmal nocturnal hemoglobinuria [PNH] clonality seen in aplastic anemia) and BCOR.⁹ Mutations were also found in genes commonly mutated in MDS and AML, including TET2, RUNX1, TP53, and JAK2, albeit at lower frequencies.⁹ These mutations as a whole have not predicted response to therapy or prognosis. However, when performing survival analysis in patients with specific mutations, those commonly encountered in MDS/AML (ASXL1, DNMT3A, TP53, RUNX1, CSMD1) are associated with faster progression to overt MDS/AML and decreased overall survival (OS),^20,21 suggesting these mutations may represent early clonality that can lead to clonal evolution and the development of secondary malignancies. Conversely, mutations in BCOR and BCORL appear to identify patients who may have a favorable outcome in response to immunosuppressive therapy and, similar to patients with PIGA mutations, improved OS.⁹

Paroxysmal Nocturnal Hemoglobinuria

In addition to having an increased risk of myelodysplasia and malignancy due to the development of a dominant pre-malignant clone, patients with aplastic anemia often harbor progenitor cell clones associated with PNH.^1,17 PNH clones have been identified in more than 50% of patients with aplastic anemia.^22,23 PNH represents a clonal disorder of hematopoiesis in which cells harbor X-linked somatic mutations in the PIGA gene; this gene encodes a protein responsible for the synthesis of glycosylphosphatidylinositol anchors on the cell surface.^22,24 The lack of these cell surface proteins, specifically CD55 (also known as decay accelerating factor) and CD59 (also known as membrane inhibitor of reactive lysis), predisposes red cells to increased complement-mediated lysis.²⁵ The exact mechanism for the development of these clones in patients with aplastic anemia is not fully understood. Current theories hypothesize that the clones are protected from the immune-mediated destruction of normal hematopoietic stem cells due to the absence of the cell surface proteins.^1,20 The role of these clones over time in patients with aplastic anemia is less clear, though recent work demonstrated that despite differences in clonality over the disease course, aplastic anemia patients with small PNH clones are less likely to develop overt hemolysis and larger PNH clones compared to patients harboring larger (≥ 50%) PNH clones at diagnosis.^23,26,27 Additionally, PNH clones in patients with aplastic anemia infrequently become clinically significant.²⁷ It should be noted that these conditions exist along a continuum; that is, patients with aplastic anemia may develop PNH clones, while conversely patients with PNH may develop aplastic anemia.²⁰ Patients with PNH clones should be followed via peripheral blood flow cytometry and complete blood count to track clonal stability and identify clinically significant PNH among aplastic anemia patients.²⁸

Clinical Presentation

Patients with aplastic anemia typically are diagnosed either due to asymptomatic cytopenias found on peripheral blood sampling, symptomatic anemia, bleeding secondary to thrombocytopenia, or wound healing and infectious complications related to neutropenia.²⁹ A thorough history to understand the timing of symptoms, recent infectious symptoms/exposure, habits, and chemical or toxin exposures (including medications, travel, and supplements) helps guide diagnostic testing. Family history is also critical, with attention given to premature graying; pulmonary, renal, and liver disease; and blood disorders.

Patients with an IMFS (eg, Fanconi anemia or dyskeratosis congenita) may have associated phenotypical findings such as urogenital abnormalities or short stature; in addition, those with dyskeratosis congenita may present with the classic triad of oral leukoplakia, lacy skin pigmentation, and dystrophic nails.⁷ However, classic phenotypical findings may be lacking in up to 30% to 40% of patients with an IMFS.⁷ As described previously, while congenital malformations are common in Fanconi anemia and dyskeratosis congenita, a third of patients may have no or only subtle phenotypical abnormalities, including alterations in skin or hair pigmentation, skeletal and growth abnormalities, and endocrine disorders.³⁰ The International Fanconi Anemia Registry identified central nervous system, genitourinary, skin and musculoskeletal, ophthalmic, and gastrointestinal system malformations among children with Fanconi anemia.^31,32 Patients with dyskeratosis congenita may present with pulmonary fibrosis, hepatic cirrhosis, or premature graying, as highlighted in a recent study by DiNardo and colleagues.³³ Therefore, physicians must have a heightened index of suspicion in patients with subtle phenotypical findings and associated cytopenias.

The diagnosis of aplastic anemia should be suspected in any patient presenting with pancytopenia. Aplastic anemia is a diagnosis of exclusion.³⁴ Other conditions associated with peripheral blood pancytopenia should be considered, including infections (HIV, hepatitis, parvovirus B19, cytomegalovirus, Epstein-Barr virus, varicella-zoster virus), nutritional deficiencies (vitamin B12, folate, copper, zinc), autoimmune disease (systemic lupus erythematosus, rheumatoid arthritis, hemophagocytic lymphohistiocytosis), hypersplenism, marrow-occupying diseases (eg, leukemia, lymphoma, MDS), solid malignancies, and fibrosis (Table).⁷

Diagnostic Evaluation

The workup for aplastic anemia should include a thorough history and physical exam to search simultaneously for alternative diagnoses and clues pointing to potential etiologic agents.⁷ Diagnostic tests to be performed include a complete blood count with differential, reticulocyte count, immature platelet fraction, flow cytometry (to rule out lymphoproliferative disorders and atypical myeloid cells and to evaluate for PNH), and bone marrow biopsy with subsequent cytogenetic, immunohistochemical, and molecular testing.³⁵ Typical findings in aplastic anemia include peripheral blood pancytopenia without dysplastic features and bone marrow biopsy demonstrating a hypocellular marrow.⁷ A relative lymphocytosis in the peripheral blood is common.⁷ In patients with a significant PNH clone, a macrocytosis along with elevated lactate dehydrogenase and elevated reticulocyte and granulocyte counts may be present.³⁶

The diagnosis (based on the Camitta criteria³⁷ and modified Camitta criteria³⁸ for severe aplastic anemia) requires 2 of the following findings on peripheral blood samples:

• Absolute neutrophil count (ANC) < 500 cells/µL
• Platelet count < 20,000 cells/µL
• Reticulocyte count < 1% corrected or < 20,000 cells/µL.³⁵

In addition to peripheral blood findings, bone marrow biopsy is essential for the diagnosis, and should demonstrate a markedly hypocellular marrow (cellularity < 25%), occasionally with an increase in T lymphocytes.^7,39 Because marrow cellularity varies with age and can be challenging to assess, additional biopsies may be needed to confirm the diagnosis.²⁹ A 1- to 2-cm core biopsy is necessary to confirm hypocellularity, as small areas of residual hematopoiesis may be present and obscure the diagnosis.³⁵

Excluding Hypocellular MDS and IMFS

Excluding hypocellular MDS is challenging, especially in the older adult presenting with aplastic anemia, as patients with aplastic anemia may have some degree of erythroid dysplasia on bone marrow morphology.³⁶ The presence of a PNH clone on flow cytometry can aid in diagnosing aplastic anemia and excluding MDS,³⁴ although PNH clones can be present in refractory anemia MDS. Patients with aplastic anemia have a lower ratio of CD34+ cells compared to those with hypoplastic MDS, with 1 study demonstrating a mean CD34+ percentage of < 0.5% in aplastic anemia versus 3.7% in hypoplastic MDS.⁴⁰ Cytogenetic and molecular testing can also aid in making this distinction by identifying mutations commonly implicated in MDS.⁷ The presence of monosomy 7 (-7) in aplastic anemia patients is associated with a poor overall prognosis.^34,41

Peripheral blood screening using chromosome breakage analysis (done using either mitomycin C or diepoxybutane as in vitro DNA-crosslinking agents)⁴² and telomere length testing (of peripheral blood leukocytes) is necessary to exclude the main IMFSs, Fanconi anemia and telomere biology disorders, respectively. Ruling out these conditions is imperative, as the approach to treatment varies significantly between IMFS and aplastic anemia. Patients with shortened telomeres should undergo genetic screening for mutations in the telomere maintenance genes to evaluate the underlying defect leading to shortened telomeres. Patients with increased peripheral blood breakage should have genetic testing to detect mutations associated with Fanconi anemia.

Classification

Once the diagnosis of aplastic anemia has been made, the patient should be classified according to the severity of their disease. Disease severity is determined based on peripheral blood ANC: non-severe aplastic anemia (NSAA), ANC > 500 polymorphonuclear neutrophils (PMNs)/µL; severe aplastic anemia (SAA), 200–500 PMNs/µL; and very severe aplastic anemia (VSAA), 0–200 PMNs/µL.^4,34 Disease classification is important, as VSAA is associated with a decreased OS compared to SAA.² Disease classification may affect treatment decisions, as patients with NSAA may be observed for a short period of time, while, conversely, patients with SAA have a worse prognosis with delays in therapy.^43-45

Treatment of Inherited Aplastic Anemia

First-line treatment options for patients with IMFS are androgen therapy and hematopoietic stem cell transplant (HSCT). When evaluating patients for HSCT, it is critical to identify the presence of an IMFS, as the risk and mortality associated with the conditioning regimen, stem cell source, graft-versus-host disease (GVHD), and secondary malignancies differ between patients with IMFS and those with acquired marrow failure syndromes or hematologic malignancies.

Potential sibling donors need to be screened for donor candidacy as well as for the inherited defect. Among patients with Fanconi anemia or a telomere biology disorder, the stem cell source must be considered, with multiple studies in IMFSs and SAA showing superior outcomes with a bone marrow product compared to peripheral blood stem cells.^46-48 In IMFS patients, the donor cell type may affect the choice of conditioning regimen.^5,6 Reduced-intensity conditioning in lieu of myeloablative conditioning without total body irradiation has proved feasible in patients with Fanconi anemia, and is associated with a reduced risk of secondary malignancies.^49,50 Incorporation of fludarabine in the conditioning regimen of patients without a matched sibling donor is associated with superior engraftment and survival^46,49,51 compared to cyclophosphamide conditioning, which was historically used in matched related donors.^50,52  Adding fludarabine appears to be especially beneficial in older patients, in whom its use is associated with lower rates of graft failure, likely due to increased immunosuppression at the time of engraftment.^51,53 Fludarabine has also been incorporated into conditioning regimens for patients with a telomere biology disorder, but outcomes data are limited.⁵

For patients presenting with AML or a high-risk MDS who are subsequently diagnosed with an IMFS, treatment can be more complex, as these patients are at high risk for toxicity from standard chemotherapy. Limited data suggest that induction therapy and transplantation are feasible in this group of patients, and this approach is associated with increased OS, despite lower OS rates than those of IMFS patients who present prior to the development of MDS or AML.^54,55 Further work is needed to determine the optimal induction regimen that balances the risks of treatment-related mortality and complications associated with conditioning regimens, risk of relapse, and risk of secondary malignancies, especially in the cohort of patients diagnosed at an older age.

Treatment of Acquired Aplastic Anemia

Supportive Care

While the workup and treatment plan are being established, attention should be directed at supportive care for prevention of complications. The most common complications leading to death in patients with significant pancytopenia and neutropenia are opportunistic infections and hemorrhagic complications.²

Transfusion support is critical to avoid symptomatic anemia and hemorrhagic complications related to thrombocytopenia, which typically occur with platelet counts lower than 10,000 cells/µL. However, transfusion carries the risk of alloimmunization (which may persist for years following transfusion) and transfusion-related graft versus host disease (trGVHD), and thus use of transfusion should be minimized when possible.^56,57 All blood products given to patients with aplastic anemia should be irradiated and leukoreduced to reduce the risk of both alloimmunization and trGVHD. Guidelines from the British Society for Haematology recommend routine screening for Rh and Kell antibodies to reduce the risk of alloimmunization.⁵⁸ Infectious complications remain a common cause of morbidity and mortality in patients with aplastic anemia who have prolonged neutropenia (defined as an ANC < 500 cells/µL).^59-62 Therefore, patients should receive broad-spectrum antibiotics with antipseudomonal coverage. In a study evaluating the role of granulocyte-colony stimulating factor (G-CSF) in patients with SAA receiving immunosuppressive therapy, 55% of all patient deaths were secondary to infection.⁶³ There was no OS benefit seen in patients who received G-CSF, though a significantly lower rate of infection was observed in the G-CSF arm compared to those not receiving G-CSF (56% versus 81%, P = 0.006). This difference was largely driven by a decrease in infectious episodes in patients with VSAA treated with G-CSF as compared to those who did not receive this therapy (22% versus 48%, P = 0.014).⁶³

Angio-invasive pulmonary aspergillosis and Zygomycetes (eg, Rhizopus, Mucor species) remain major causes of mortality related to opportunistic mycotic infections in patients with aplastic anemia.¹⁸ The infectious risk is directly related to the duration and severity of neutropenia, with one study demonstrating a significant increase in risk in AML patients with neutropenia lasting longer than 3 weeks.⁶⁴ Invasive fungal infections carry a high mortality in patients with severe neutropenia, though due to earlier recognition and empiric antifungal therapy with extended-spectrum azoles, overall mortality secondary to invasive fungal infections is declining.^62,65

While neutropenia related to cytotoxic chemotherapy is commonly associated with gram-negative bacteria due to disruption of mucosal barriers, patients with aplastic anemia have an increased incidence of gram-positive bacteremia with staphylococcal species compared to other neutropenic populations.^61,62 This appears to be changing with time. Valdez et al demonstrated a decrease in prevalence of coagulase-negative staphylococcal infections, increased prevalence of gram-positive bacilli bacteremia, and no change in prevalence of gram-negative bacteremia in patients with aplastic anemia treated between 1989 and 2008.⁶⁵ Gram-negative bacteremia caused by Stenotrophomonas maltophila, Escherichia coli, Klebsiella pneumoniae, Citrobacter, and Proteus has also been reported.⁶² Despite a lack of clinical trials investigating the role of antifungal and antibacterial prophylaxis for patients with aplastic anemia, most centers initiate antifungal prophylaxis in patients with SAA or VSAA with an anti-mold agent such as voriconazole or posaconazole (which has the additional benefit compared to voriconazole of covering Mucor species).^60,66 This is especially true for patients who have received ATG or undergone HSCT. For antimicrobial prophylaxis, a fluoroquinolone antibiotic with a spectrum of activity against Pseudomonas should be considered for patients with an ANC < 500 cells/µL.⁶⁰ Acyclovir or valacyclovir prophylaxis is recommended for varicella-zoster virus and herpes simplex virus. Cytomegalovirus reactivation is minimal in patients with aplastic anemia, unless multiple courses of ATG are used.

Iron overload is another complication the provider must be aware of in the setting of increased transfusions in aplastic anemia patients. Lee and colleagues showed that iron chelation therapy using deferasirox is effective at reducing serum ferritin levels in patients with aplastic anemia (median ferritin level of 3254 ng/mL prior to therapy, 1854 ng/mL following), and is associated with no serious adverse events (most common adverse events included nausea, diarrhea, vomiting, and rash).⁶⁷ Approximately 25% of patients in this trial had an increase in creatinine, with patients taking concomitant cyclosporine affected to a greater degree than those on chelation therapy alone. For patients following HSCT or with improved hematopoiesis following immunosuppressive therapy, phlebotomy can be used to treat iron overload in lieu of chelation therapy.⁵⁸

Approach to Therapy

The main treatment options for SAA and VSAA include allogeneic bone marrow transplant and immunosuppression. The deciding factors as to which treatment is best initially depends on the availability of HLA-matched related donors and age (Figure 1 and Figure 2). Survival is decreased in patients with SAA or VSAA who delay initiation of therapy, and therefore prompt referral for HLA typing and evaluation for bone marrow transplant is a very important first step in managing aplastic anemia.

Matched Sibling Donor Transplant. Current standards of care recommend HLA-matched sibling donor transplant for patients with SAA or VSAA who are younger than 50 years, with the caveat that integration of fludarabine and reduced cyclophosphamide dosing along with ATG shows the best overall outcomes. Locasciulli and colleagues examined outcomes in patients given either immunosuppressive therapy or sibling HSCT between 1991-1996 and 1997-2002, respectively, and found that sibling HSCT was associated with a superior 10-year OS compared to immunosuppressive therapy (73% versus 68%).⁴³ Interestingly in this study, there was no OS improvement seen with immunosuppressive therapy alone (69% versus 73%) between the 2 time periods, despite increased OS in both sibling HSCT (74% and 80%) and MUD HSCT (38% and 65%).⁴³ Though total body irradiation has been used in the past, it is typically not included in current conditioning regimens for matched related donor transplants.⁶⁸

Current conditioning regimens typically use a combination of cyclophosphamide and ATG,^69,70 with or without fludarabine. Fludarabine-based conditioning regimens have shown promise in patients undergoing sibling HSCT. Maury and colleagues evaluated the role of fludarabine in addition to low-dose cyclophosphamide and ATG compared to cyclophosphamide alone or in combination with ATG in patients over age 30 undergoing sibling HSCT.⁵³ There was a nonsignificant improvement in 5-year OS in the fludarabine arm compared to controls (77% ± 8% versus 60% ± 3%, P = 0.14) in the pooled analysis, but when adjusted for age the fludarabine arm had a significantly lower relative risk (RR) of death (0.44; P = 0.04) compared to the control arm. Shin et al reported outcomes with fludarabine/cyclophosphamide/ATG, with excellent overall outcomes and no difference in patients older or younger than 40 years.⁷¹

Kim et al evaluated their experience with patients older than 40 years receiving matched related donors, finding comparable outcomes in those ages 41 to 50 years compared to younger patients. Outcomes declined in those over the age of 50 years.⁷² Long-term data for matched related donor transplant for aplastic anemia show excellent long-term outcomes, with minimal chronic GVHD and good performance status.⁷³ Hence, these factors support the role of matched related donor transplant as the initial treatment in SAA and VSAA.

Regarding the role of transplant for patients who lack a matched related donor, a growing body of literature demonstrating identical outcomes between matched related and MUD transplants for pediatric patients^74,75 supports recent recommendations for upfront unrelated donor transplantation for aplastic anemia.^76,77

Immunosuppressive Therapy. For patients without an HLA-matched sibling donor or those who are older than 50 years of age, immunosuppressive therapy is the first-line therapy. ATG and cyclosporine A are the treatments of choice.⁷⁸ The potential effectiveness of immunosuppressive therapy in treating aplastic anemia was initially observed in patients in whom autologous transplant failed but who still experienced hematopoietic reconstitution despite the failed graft; this observation led to the hypothesis that the conditioning regimen may have an effect on hematopoiesis.^59,78,79

Immunosuppressive therapy with ATG has been used for the treatment of aplastic anemia since the 1980s.⁸⁰ Historically, rabbit ATG had been used, but a 2011 study of horse ATG demonstrated superior hematological response at 6 months compared to rabbit ATG (68% versus 37%).⁵⁹ Superior survival was also seen with horse ATG compared to rabbit ATG (3-year OS, 96% versus 76%). Due to these results, horse ATG is preferred over rabbit ATG. ATG should be used in combination with cyclosporine A to optimize outcomes.

Early studies also demonstrated the efficacy of cyclosporine A in the treatment of aplastic anemia, with response rates equivalent to that of ATG monotherapy.⁸¹ Recent publications still note the efficacy of cyclosporine A in the treatment of aplastic anemia. Its role as an affordable option for single-agent therapy in developing countries is intriguing.⁸¹ The combination of ATG and cyclosporine A was proven superior to either agent alone in a study by Frickhofen et al.⁷⁹ In this study, patients were randomly assigned to a control arm that received ATG plus methylprednisolone or to an arm that received ATG plus cyclosporine A and methylprednisolone. At 6 months, 70% of patients in the cyclosporine A arm had a complete remission (CR) or partial remission compared to 46% in the control arm.⁸² Further work confirmed the long-term efficacy of this regimen, reporting a 7-year OS of 55%.⁸³ Among a pediatric population, immunosuppressive therapy was associated with an 83% 10-year OS.⁸⁴

It is recommended that patients remain on cyclosporine therapy for a minimum of 6 months, after which a gradual taper may be considered, although there is variation among practitioners, with some continuing immunosuppressive therapy for a minimum of 12 months due to a proportion of patients being cyclosporine dependent.^34,84 A study found that within a population of patients who responded to immunosuppressive therapy, 18% became cyclosporine dependent.⁸⁴ The median duration of cyclosporine A treatment at full dose was 12 months, with tapering completed over a median of 19 months after patients had been in a stable CR for a minimum of 3 months. Relapse occurred more often when patients were tapered quickly (decrease ≥ 0.8 mg/kg/month) compared to slowly (0.4-0.7 mg/kg/month) or very slowly (< 0.3 mg/kg/month).

Townsley and colleagues recently investigated incorporating the use of the thrombopoietin receptor agonist eltrombopag with immunosuppressive therapy as first-line therapy in aplastic anemia.⁸⁵ When given at a dose of 150 mg daily in patients ages 12 years and older or 75 mg daily in patients younger than 12 years, in conjunction with cyclosporine A and ATG, patients demonstrated markedly improved hematological response compared to historical treatment with standard immunosuppressive therapy alone.⁴⁵ In the patient cohort administered eltrombopag starting on day 1 and continuing for 6 months, the complete response rate was 58%. Eltrombopag led to improvement in all cell lines among all treatment subgroups, and OS (censored for patients who proceeded to transplant) was 99% at 2 years.¹² Overall, toxicities associated with this therapy were low, with liver enzyme elevations most commonly observed.⁸⁵ Recently, a phase 2 trial of immunosuppressive therapy with or without eltrombopag was reported. Of the 38 patients enrolled, overall response, complete response, and time to response were not statistically different.⁸⁶ With this recent finding, the role of eltrombopag in addition to immunosuppressive therapy is not clearly defined, and further studies are warranted.

OS for patients who do not respond to immunosuppressive therapy is approximately 57% at 5 years, largely due to improved supportive measures among this patient population.^48,65 Therefore, it is important to recognize those patients who have a low chance of response so that second-line therapy can be pursued to improve outcomes.

Matched Unrelated Donor Transplant. For patients with refractory disease following immunosuppressive therapy who lack a matched sibling donor, MUD HSCT is considered standard therapy given the marked improvement in overall outcomes with modulating conditioning regimens and high-resolution HLA typing. A European Society for Blood and Marrow Transplantation (EBMT) analysis comparing matched sibling HSCT to MUD HSCT noted significantly higher rates of acute grade II-IV and grade III-IV GVHD (grade II-IV 13% versus 25%, grade III-IV 5% versus 10%) among patients undergoing MUD transplant.⁴⁷ Chronic GVHD rates were 14% in the sibling group, as compared to 26% in the MUD group. Factors associated with improved survival in this analysis include transplant under age 20 years (84% versus 72%), transplant within 6 months of diagnosis (85% versus 72%), the use of ATG in the conditioning regimen (81% versus 73%), and cytomegalovirus-negative donor and recipient as compared to other combinations (82% versus 76%).⁸⁷ Interestingly, this study demonstrated that OS was not significantly increased when using a sibling HSCT compared to a MUD HSCT, likely as a result of improved understanding of conditioning regimens, GVHD prophylaxis, and supportive care.

Additional studies of MUD HSCT have shown outcomes similar to those seen in sibling HSCT.^34,48 A French study found a significant increase in survival in patients undergoing MUD HSCT compared to historical cohorts (2000-2005: OS 52%; 2006-2012: OS 74%).⁷⁵ The majority of patients underwent conditioning with cyclophosphamide or a combination of busulfan and cyclophosphamide, with or without fludarabine; 81% of patients underwent in vivo T-cell depletion, and a bone marrow donor source was utilized. OS was significantly lower in patients over age 30 years undergoing MUD HSCT (57%) compared to those under age 30 years (70%). Improved OS was also seen when patients underwent transplant within 1 year of diagnosis and when a 10/10 matched donor (compared to a 9/10 mismatched donor) was utilized.⁴⁸

A 2015 study investigated the role of MUD HSCT as frontline therapy instead of immunosuppressive therapy in patients without a matched sibling donor.⁷⁵ The 2-year OS was 96% in the MUD HSCT cohort compared to 91%, 94%, and 74% in historical cohorts of sibling HSCT, frontline immunosuppressive therapy, and second-line MUD HSCT following failed immunosuppressive therapy, respectively. Additionally, event-free survival in the MUD HSCT cohort (defined by the authors as death, lack of response, relapse, occurrence of clonal evolution/clinical PNH, malignancies developing over follow‐up, and transplant for patients receiving immunosuppressive therapy frontline) was similar compared to sibling HSCT and superior to frontline immunosuppressive therapy and second-line MUD HSCT. Furthermore, Samarasinghe et al highlighted the importance of in vivo T-cell depletion with either ATG or alemtuzumab (anti-CD52 monoclonal antibody) in the prevention of acute and chronic GVHD in both sibling HSCT and MUD HSCT.⁸⁸

With continued improvement of less toxic and more immunomodulating conditioning regimens,utilization of bone marrow as a donor cell source, in vivo T-cell depletion, and use of GVHD and antimicrobial prophylaxis, more clinical evidence supports elevating MUD HSCT in the treatment plan for patients without a matched sibling donor.⁸⁹ However, there is still a large population of patients without matched sibling or unrelated donor options. Given the need to expand the transplant pool and thus avoid clonal hematopoiesis, clinically significant PNH, and relapsed aplastic anemia, more work continues to recognize the expanding role of alternative donor transplants (cord blood and haploidentical) as another viable treatment strategy for aplastic anemia after immunosuppressive therapy failure.⁹⁰

Summary

Aplastic anemia is a rare but potentially life-threatening disorder with pancytopenia and a marked reduction in the HSC compartment. It can be acquired or associated with an IMFS, and the treatment and prognosis vary dramatically between these 2 etiologies. Workup and diagnosis involves investigating IMFSs and ruling out malignant or infectious etiologies for pancytopenia. Treatment outcomes are excellent with modern supportive care and the current approach to allogeneic transplantation, and therefore referral to a bone marrow transplant program to evaluate for early transplantation is the new standard of care.

Corresponding author: Gabrielle Meyers, MD, 3181 SW Sam Jackson Park Road, Mail Code UHN73C, Portland, OR 97239.

Financial disclosures: None.

From the University of Texas MD Anderson Cancer Center, Houston, TX (Drs. Narayanan, Lopez, Chaoul, Liu, Milbury, and Cohen, and Ms. Mallaiah); the University of Texas Health Science Center at Tyler (Dr. Meegada); and Texas Tech University Health Sciences Center, Lubbock, TX (Ms. Francisco).

Abstract

• Objective: To review the effects of yoga as an adjunct supportive care modality alongside conventional cancer treatment on quality of life (QOL), physical and mental health outcomes, and physiological and biological measures of cancer survivors.
• Methods: Nonsystematic review of the literature.
• Results: Yoga therapy, one of the most frequently used mind-body modalities, has been studied extensively in cancer survivors (from the time of diagnosis through long-term recovery). Yoga affects human physiology on multiple levels, including psychological outcomes, immune and endocrine function, and cardiovascular parameters, as well as multiple areas of QOL. It has been found to reduce psychological stress and fatigue and improve QOL in cancer patients and survivors. Yoga has also been used to manage symptoms such as arthralgia, fatigue, and insomnia. In addition, yoga offers benefits not only for cancer survivors but also for their caregivers.
• Conclusion: As part of an integrative, evidence-informed approach to cancer care, yoga may provide benefits that support the health of cancer survivors and caregivers.

Keywords: fatigue; cancer; proinflammatory cytokines; integrative; mind-body practices; meditation; DNA damage; stress; psychoneuro-immunoendocrine axis; lymphedema; insomnia.

A diagnosis of cancer and adverse effects related to its treatment may have negative effects on quality of life (QOL), contributing to emotional and physical distress in patients and caregivers. Many patients express an interest in pursuing nonpharmacological options, alone or as an adjunct to conventional therapy, to help manage symptoms. The use of complementary medicine approaches to health, including nonpharmacological approaches to symptom management, is highest among individuals with cancer.¹ According to a published expert consensus, integrative oncology is defined as a “patient-centered, evidence-informed field of cancer care that utilizes mind and body practices, natural products, and/or lifestyle modifications from different traditions alongside conventional cancer treatments. Integrative oncology aims to optimize health, QOL, and clinical outcomes across the cancer care continuum and to empower people to prevent cancer and become active participants before, during, and beyond cancer treatment.”² A key component of this definition, often misunderstood in the field of oncology, is that these modalities and treatments are used alongside conventional cancer treatments and not as an alternative. In an attempt to meet patients’ needs and appropriately use these approaches, integrative oncology programs are now part of most cancer centers in the United States.^3-6

Because of their overall safety, mind-body therapies are commonly used by patients and recommended by clinicians. Mind-body therapies include yoga, tai chi, qigong, meditation, and relaxation. Expressive arts such as journaling and music, art, and dance therapies also fall in the mind-body category.⁷ Yoga is a movement-based mind-body practice that focuses on synchronizing body, breath, and mind. Yoga has been increasingly used by patients for health benefits,⁸ and numerous studies have evaluated yoga as a complementary intervention for individuals with cancer.^9-14 Here, we review the physiological basis of yoga in oncology and the effects of yoga on biological processes, QOL, and symptoms during and after cancer treatment.

Physiological Basis

Many patients may use mind-body programs such as yoga to help manage the psychological and physiological consequences of unmanaged chronic stress and improve their overall QOL. The central nervous system, endocrine system, and immune system influence and interact with each other in a complex manner in response to chronic stress.^15,16 In a stressful situation, the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system (SNS) are activated. HPA axis stimulation leads to adrenocorticotrophic hormone production by the pituitary gland, which releases glucocorticoid hormones. SNS axis stimulation leads to epinephrine and norepinephrine production by the adrenal gland.^17,18 Recently, studies have explored modulation of signal transduction between the nervous and immune systems and how that may impact tumor growth and metastasis.¹⁹ Multiple studies, controlled for prognosis, disease stage, and other factors, have shown that patients experiencing more distress or higher levels of depressive symptoms do not live as long as their counterparts with low distress or depression levels.²⁰ Both the meditative and physical components of yoga can lead to enhanced relaxation, reduced SNS activation, and greater parasympathetic tone, countering the negative physiological effects of chronic stress. The effects of yoga on the HPA axis and SNS, proinflammatory cytokines, immune function, and DNA damage are discussed below.

Biological Processes

Nervous System

The effects of yoga and other forms of meditation on brain functions have been established through several studies. Yoga seems to influence basal ganglia function by improving circuits that are involved in complex cognitive functions, motor coordination, and somatosensory and emotional processes.^21,22 Additionally, changes in neurotransmitter levels have been observed after yoga practice. For instance, in a 12-week yoga intervention in healthy subjects, increased levels of thalamic gamma-aminobutyric acid (GABA) in the yoga group were reported to have a positive correlation with improved mood and decreased anxiety compared with a group who did metabolically matched walking exercise.²³ Levels of GABA, an inhibitory neurotransmitter, are decreased in conditions such as anxiety, depression, and epilepsy.²⁴ Yoga therapy has been shown to improve symptoms of mood disorders and epilepsy, which leads to the hypothesis that the mechanism driving the benefits of yoga may work through stimulation of vagal efferents and an increase in GABA-mediated cortical-inhibitory tone.^24,25

Stress activates the HPA/SNS axis, which releases hormones such as cortisol and norepinephrine. These hormones may play a role in angiogenesis, inflammation, immune suppression, and other physiological functions, and may even reduce the effect of chemotherapeutic agents.^26,27 Regular yoga practice has been shown to reduce SNS and HPA axis activity, most likely by increasing parasympathetic dominance through vagal stimulation, as demonstrated through increases in heart rate variability.²⁸ One indicator of HPA axis dysregulation, diurnal salivary cortisol rhythm, was shown to predict survival in patients with advanced breast and renal cancer.^29-33 Yoga has been shown to lead to less cortisol dysregulation due to radiotherapy and to reductions in mean cortisol levels and early morning cortisol levels in breast cancer patients undergoing radiotherapy.³⁴ This lends support to the hypothesis that yoga helps restore HPA axis balance.

Proinflammatory Cytokines

Cancer patients tend to have increased levels of inflammatory markers such as interleukin (IL)-4, IL-10, tumor necrosis factor (TNF), interferon-γ, and C-reactive protein. This increase in inflammation is associated with worse outcomes in cancer.³⁵ This association becomes highly relevant because the effect of inflammation on host cells in the tumor microenvironment is connected to disease progression.²⁶ Inflammatory cytokines are also implicated in cancer-related symptoms such as fatigue, cognitive dysfunction, peripheral neuropathy, and sleep disturbances.³⁶

Yoga is known to reduce stress and may directly or indirectly decrease inflammatory cytokines. A randomized clinical trial of a 12-week hatha yoga intervention among breast cancer survivors demonstrated decreases in IL-6, IL-1β, TNF, corticotropin-releasing factor, and cognitive complaints in the yoga group compared with those in the standard care group after 3 months.^37,38 Furthermore, Carlson et al showed that, after mindfulness-based stress reduction involving a combination of gentle yoga, meditation techniques, and relaxation exercises, breast and prostate cancer patients had reduced levels of proinflammatory cytokines and cortisol.³⁹ These reductions translated into patients reporting decreased stress levels and enhanced QOL.

Immune Function

The effects of yoga practice on the immune system have been studied in both healthy individuals and individuals with cancer. The effects on T and B lymphocytes, natural killer (NK) cells, and other immune effector cells demonstrate that meditation and yoga have beneficial effects on immune activity.⁴⁰ Hormones such as catecholamines and glucocorticoids are thought to influence the availability and function of NK cells, and, as noted above, yoga has been shown to modulate stress hormones and lead to reduced immune suppression in patients with early-stage breast cancer undergoing chemotherapy.⁴¹ Additional evidence supports the ability of yoga to reduce immune suppression in the postsurgical setting, with no observed decrease in NK cell percentage after surgery for those in a yoga group compared with a control group.⁴² This finding is relevant to patients undergoing surgical management of their cancer and highlights the impact of yoga on the immune system.

DNA Damage

Radiation damages DNA in the peripheral blood lymphocytes of patients undergoing treatment.^43,44 This damage is significant in breast cancer patients undergoing radiotherapy.⁴⁵ Stress additionally causes DNA damage⁴⁶ and is correlated to impaired DNA repair capacity.^47,48 In a study conducted by Banerjee et al, breast cancer patients were randomly assigned to a yoga group or a supportive therapy group for 6 weeks during radiotherapy.⁴⁹ Prior to the intervention, patients in the study had significant genomic instability. After treatment, patients in the yoga group experienced not only a significant reduction in anxiety and depression levels, but also a reduction in DNA damage due to radiotherapy.

There is evidence showing that yoga therapy improves multiple aspects of QOL, including physical functioning, emotional health outcomes, and the symptoms cancer patients may experience, such as sleep disturbances, fatigue, and pain. Danhauer et al systematically reviewed both nonrandomized trials and randomized controlled trials involving yoga during cancer treatment.⁵⁰ They found that yoga improved depression and anxiety as well as sleep and fatigue. Benefits of yoga in cancer based on randomized controlled trials are summarized in the Table. The role of yoga in improving QOL and managing symptoms patients experience during and after treatment is discussed in the following sections.

Quality of Life

Danhauer et al’s systematic review of trials involving yoga during cancer treatment found that yoga improved multiple aspects of QOL.⁵⁰ For example, yoga has been shown to improve QOL in breast cancer patients undergoing radiotherapy. In a study by Chandwani et al, yoga (60-minute sessions twice a week for 6 weeks) was associated with better general health perception and physical functioning scores as well as greater benefit finding, or finding meaning in their experience, after radiotherapy compared with a wait-list group.⁵¹ The yoga group had an increase in intrusive thoughts, believed to be due to a more thorough processing of the cancer experience, which helps to improve patients’ outlook on life.⁵² The benefits of yoga extend beyond psychological measures during radiation treatment. Yoga was found to increase physical functioning compared with stretching in breast cancer patients undergoing radiotherapy.⁵³

Cognitive Function

Cancer-related cognitive impairment commonly occurs during cancer treatments (eg, chemotherapy, radiotherapy, surgery, hormone therapy) and persists for months or years in survivors.⁵⁴ Impairment of memory, executive function, attention, and concentration are commonly reported. In a trial of a combined hatha and restorative yoga program called Yoga for Cancer Survivors (YOCAS), which was designed by researchers at the University of Rochester, patients in the yoga arm had less memory difficulty than did patients in the standard care arm.⁵⁵ However, the primary aim of the trial was to treat insomnia, so this secondary outcome needs to be interpreted with caution. Deficits in attention, memory, and executive function are often seen in cancer-related cognitive impairment, and the meditative aspect of yoga may have behavioral and neurophysical benefits that could improve cognitive functions.⁵⁶ More evidence is needed to understand the role of yoga in improving cognitive functioning.

Emotional Health

Psychosocial stress is high among breast cancer patients and survivors.^57,58 This causes circadian rhythm and cortisol regulation abnormalities, which are reported in women with breast cancer.^59-64 Yoga is known to help stress and psychosocial and physical functioning in patients with cancer.⁶⁵ Yoga was also shown to be equivalent to cognitive behavioral therapy in stress management in a population of patients without cancer.⁶⁶ Daily yoga sessions lasting 60 minutes were shown to reduce reactive anxiety and trait anxiety in early-stage breast cancer patients undergoing conventional radiotherapy and chemotherapy compared with patients receiving supportive therapy, highlighting the role of yoga in managing anxiety related to treatment.⁶⁷ In a study done by Culos-Reed et al, 20 cancer survivors who did 75 minutes of yoga per week for 7 weeks were compared with 18 cancer survivors who served as a control group.⁶⁸ The intervention group reported significant improvement in emotional well-being, depression, concentration, and mood disturbances. In a longitudinal study by Mackenzie et al, 66 cancer survivors completed a 7-week yoga program and were assessed at baseline, immediately after the final yoga session, and at 3 and 6 months after the final session.⁶⁹ Participants had significantly improved energy levels and affect. They also had moderate improvement in mindfulness and a moderate decrease in stress. Breast cancer patients who underwent restorative yoga sessions found improvements in mental health, depression, positive affect, and spirituality (peace/meaning).⁷⁰ This was more pronounced in women with higher negative affect and lower emotional well-being at baseline. In a study of patients with ovarian cancer receiving chemotherapy, patients were instructed to perform up to 15-minute sessions including awareness, body movement, and breathing.⁷¹ Even with just 1 session of yoga intervention, patients experienced decreased anxiety.

Fatigue

Studies on yoga show improvement in fatigue both during and after treatment. In breast cancer patients undergoing chemotherapy, yoga was shown to benefit cognitive fatigue.⁷² Older cancer survivors also seem to benefit from yoga interventions.⁷³ In a trial of a DVD-based yoga program, the benefits of yoga were similar to those of strengthening exercises, and both interventions helped decrease fatigue and improve QOL during the first year after diagnosis in early-stage breast cancer patients with cancer-related fatigue.⁷⁴ Bower et al also showed that, for breast cancer survivors experiencing persistent chronic fatigue, a targeted yoga intervention led to significant improvements in fatigue and vigor over a 3-month follow-up compared with controls.⁷⁵ Fatigue is commonly seen in breast cancer patients who are receiving adjuvant chemotherapy. In a study by Taso et al, women with breast cancer receiving chemotherapy were assigned to 60-minute yoga sessions incorporating Anusara yoga, gentle stretching, and relaxation twice a week for 8 weeks.⁷⁶ By week 4, patients with low pretest fatigue in the yoga group experienced a reduction in fatigue. By week 8, all patients in the yoga group experienced a reduction in fatigue. Four weeks after the yoga intervention, patients in the group maintained the reduction in fatigue. This study shows the feasibility of an 8-week yoga program for women undergoing breast cancer therapy by improving fatigue. Yoga recently was added to National Comprehensive Cancer Network (NCCN) guidelines for management of cancer-related fatigue (level 1 evidence).⁷⁷ However, the evidence was based on studies in women with breast cancer and survivors; therefore, more studies are needed in men and women with other cancers.

Distress surrounding surgery in patients with breast cancer can impact postoperative outcomes. Yoga interventions, including breathing exercises, regulated breathing, and yogic relaxation techniques, improved several postsurgical measures such as length of hospital stay, drain retention, and suture removal.⁷⁸ In this study, patients who practiced yoga also experienced a decrease in plasma TNF and better wound healing. Symptoms of anxiety and distress that occur preoperatively can lead to impaired immune function in addition to decreased QOL. In a study of yoga in early-stage breast cancer patients undergoing surgery, the benefit of yoga was seen not only with stress reduction but also with immune enhancement.⁴²

Yoga has been shown to help alleviate acute pain and distress in women undergoing major surgery for gynecological cancer. A regimen of 3 15-minute sessions of yoga, including awareness meditation, coordination of breath with movement, and relaxation breathing, was shown to reduce acute pain and distress in such patients in an inpatient setting.⁷⁹

Menopausal Symptoms

Breast cancer survivors have more severe menopausal symptoms compared with women without cancer.^80,81 Hot flashes cause sleep disturbances and worsen fatigue and QOL.⁸² Tamoxifen and aromatase inhibitors significantly worsen menopausal symptoms such as hot flashes.⁸¹ Carson et al conducted a study of yoga that included postures, breathing techniques, didactic presentations, and group discussions.⁸³ The yoga awareness regimen consisted of 8 weekly 120-minute group classes. Patients in the yoga arm had statistically significant improvements in the frequency, severity, and number of hot flashes. There were also improvements in arthralgia (joint pain), fatigue, sleep disturbance, vigor, and acceptance.

Arthralgia

Joint pain can be a major side effect that interferes with daily functions and activities in postmenopausal breast cancer survivors who receive aromatase inhibitor therapy.⁸⁴ Arthralgia is reported in up to 50% of patients treated with aromatase inhibitors.^84,85 It can affect functional status and lead to discontinuation of aromatase inhibitor therapy, jeopardizing clinical outcomes.⁸⁶ Yoga as a complementary therapy has been shown to improve conditions such as low back pain⁸⁷ and knee osteoarthritis⁸⁸ in patients who do not have cancer. In a single-arm pilot trial by Galantino et al, breast cancer patients with aromatase inhibitor–related joint pain were provided with twice-weekly yoga sessions for 8 weeks. There were statistically significant improvements in balance, flexibility, pain severity, and health-related QOL.⁸⁹ As noted above, improvement in arthralgia was also found in the study conducted by Carson et al.⁸³

Insomnia

Insomnia is common among cancer patients and survivors^90,91 and leads to increased fatigue and depression, decreased adherence to cancer treatments, and poor physical function and QOL.^90-92 Management of insomnia consists of pharmacologic therapies such as benzodiazepines^93,94 and nonpharmacologic options such as cognitive behavioral therapy.⁹⁵

The first study of yoga found to improve sleep quality was conducted at MD Anderson Cancer Center in lymphoma patients.⁹⁶ The effects of Tibetan yoga practices incorporating controlled breathing and visualization, mindfulness techniques, and low-impact postures were studied. Patients in the Tibetan yoga group had better subjective sleep quality, faster sleep latency, longer sleep duration, and less use of sleep medications. Mustian et al conducted a large yoga study in cancer survivors in which patients reporting chronic sleep disturbances were randomly assigned to the YOCAS program, which consisted of pranayama (breath control), 16 gentle hatha and restorative yoga postures, and meditation, or to usual care.⁹² The study reported improvements in global sleep quality, subjective sleep quality, actigraphy measures (wake after sleep onset, sleep efficiency), daytime dysfunction, and use of sleep medication after the yoga intervention compared with participants who received standard care.

Yoga to Address Other Symptoms

There is preliminary evidence supporting yoga as an integrative therapy for other symptoms unique to cancer survivors. For example, in head and neck cancer survivors, soft tissue damage involving the jaw, neck, shoulders, and chest results in swallowing issues, trismus, and aspiration, which are more pronounced in patients treated with conventional radiotherapy than in those treated with intensity-modulated radiotherapy.⁹⁷ Some late effects of radiotherapy for head and neck cancer—such as pain, anxiety, and impaired shoulder function—were shown to be improved through the practice of hatha yoga in 1 study.⁹⁸ Similarly, in a randomized controlled pilot study of patients with stage I to III breast cancer 6 months after treatment, participants in an 8-week yoga program experienced a reduction in arm induration and improvement in a QOL subscale of lymphedema symptoms. However, more evidence is needed to support the use of yoga as a therapeutic measure for breast cancer lymphedema.^99,100

Yoga for Caregivers

Along with cancer patients, caregivers face psychological and physical burdens as well as deterioration in their QOL. Caregivers tend to report clinical levels of anxiety, depression, sleep disturbance, and fatigue and have similar or in fact higher levels than those of the patients for whom they are caring.^101,102 Yoga has been found to help caregivers of patients with cancer. Recently, MD Anderson researchers conducted a trial in patients with high-grade glioma and their caregivers as dyads.^103,104 Each dyad attended 2 or 3 60-minute weekly Vivekananda yoga sessions involving breathing exercises, physical exercises, relaxation, and meditation. The researchers found that the yoga program was safe, feasible, acceptable, and subjectively useful for patients with high-grade glioma and their caregivers. Preliminary evidence of QOL improvement for both patients and caregivers was noted. An improvement in QOL was also demonstrated in another preliminary study of yoga in patients undergoing thoracic radiotherapy and their caregivers.¹⁰⁵

Another study by the group at MD Anderson evaluated a couple-based Tibetan yoga program that emphasized breathing exercises, gentle movements, guided visualizations, and emotional connectedness during radiotherapy for lung cancer.¹⁰⁶ This study included 10 patient‐caregiver dyads and found the program to be feasible, safe, and acceptable. The researchers also found preliminary evidence of improved QOL by the end of radiotherapy relative to baseline—specifically in the areas of spiritual well‐being for patients, fatigue for caregivers, and sleep disturbances and mental health issues such as anxiety and depressive symptoms for both patients and caregivers. This is noteworthy, as QOL typically deteriorates during the course of radiotherapy, and the yoga program was able to buffer these changes.

Conclusion

Yoga therapy has been used successfully as an adjunct modality to improve QOL and cancer-related symptoms. As a part of an integrative medicine approach, yoga is commonly recommended for patients undergoing cancer treatment. Danhauer et al reviewed randomized controlled trials during and after treatment and concluded that the evidence is clearly positive for QOL, fatigue, and perceived stress.¹⁰⁷ Results are less consistent but supportive for psychosocial outcomes such as benefit finding and spirituality. Evidence is mixed for sleep, anxiety, and depression. Post-treatment studies demonstrate improvements in fatigue, sleep, and multiple QOL domains. Yoga has been included in NCCN guidelines for fatigue management. Yoga, if approved by a physician, is also included among the behavioral therapies for anticipatory emesis and prevention and treatment of nausea in the recent update of the NCCN guidelines.¹⁰⁸ The Society for Integrative Oncology guidelines include yoga for anxiety/stress reduction as a part of integrative treatment in breast cancer patients during and after therapy, which was endorsed by the American Society of Clinical Oncology.¹⁰⁹

Because of the strong evidence for its benefits and a low side-effect profile, yoga is offered in group-class settings for patients during and after treatment and/or for caregivers in our institution. We often prescribe yoga as a therapeutic modality for selected groups of patients in our clinical practice. However, some patients may have restrictions after surgery that must be considered. In general, yoga has an excellent safety profile, the evidence base is strong, and we recommend that yoga therapy should be part of the standard of care as an integrative approach for patients with cancer undergoing active treatment as well as for cancer survivors and caregivers.

Acknowledgement: The authors thank Bryan Tutt for providing editorial assistance.

Corresponding author: Santhosshi Narayanan, MD, Department of Palliative, Rehabilitation, and Integrative Medicine, Unit 1414, The University of Texas MD Anderson Cancer Center, 1400 Pressler St., Houston, TX 77030; [email protected].

Financial disclosures: None.

From the University of Texas MD Anderson Cancer Center, Houston, TX (Drs. Narayanan, Lopez, Chaoul, Liu, Milbury, and Cohen, and Ms. Mallaiah); the University of Texas Health Science Center at Tyler (Dr. Meegada); and Texas Tech University Health Sciences Center, Lubbock, TX (Ms. Francisco).

Abstract

• Objective: To review the effects of yoga as an adjunct supportive care modality alongside conventional cancer treatment on quality of life (QOL), physical and mental health outcomes, and physiological and biological measures of cancer survivors.
• Methods: Nonsystematic review of the literature.
• Results: Yoga therapy, one of the most frequently used mind-body modalities, has been studied extensively in cancer survivors (from the time of diagnosis through long-term recovery). Yoga affects human physiology on multiple levels, including psychological outcomes, immune and endocrine function, and cardiovascular parameters, as well as multiple areas of QOL. It has been found to reduce psychological stress and fatigue and improve QOL in cancer patients and survivors. Yoga has also been used to manage symptoms such as arthralgia, fatigue, and insomnia. In addition, yoga offers benefits not only for cancer survivors but also for their caregivers.
• Conclusion: As part of an integrative, evidence-informed approach to cancer care, yoga may provide benefits that support the health of cancer survivors and caregivers.

Keywords: fatigue; cancer; proinflammatory cytokines; integrative; mind-body practices; meditation; DNA damage; stress; psychoneuro-immunoendocrine axis; lymphedema; insomnia.

A diagnosis of cancer and adverse effects related to its treatment may have negative effects on quality of life (QOL), contributing to emotional and physical distress in patients and caregivers. Many patients express an interest in pursuing nonpharmacological options, alone or as an adjunct to conventional therapy, to help manage symptoms. The use of complementary medicine approaches to health, including nonpharmacological approaches to symptom management, is highest among individuals with cancer.¹ According to a published expert consensus, integrative oncology is defined as a “patient-centered, evidence-informed field of cancer care that utilizes mind and body practices, natural products, and/or lifestyle modifications from different traditions alongside conventional cancer treatments. Integrative oncology aims to optimize health, QOL, and clinical outcomes across the cancer care continuum and to empower people to prevent cancer and become active participants before, during, and beyond cancer treatment.”² A key component of this definition, often misunderstood in the field of oncology, is that these modalities and treatments are used alongside conventional cancer treatments and not as an alternative. In an attempt to meet patients’ needs and appropriately use these approaches, integrative oncology programs are now part of most cancer centers in the United States.^3-6

Because of their overall safety, mind-body therapies are commonly used by patients and recommended by clinicians. Mind-body therapies include yoga, tai chi, qigong, meditation, and relaxation. Expressive arts such as journaling and music, art, and dance therapies also fall in the mind-body category.⁷ Yoga is a movement-based mind-body practice that focuses on synchronizing body, breath, and mind. Yoga has been increasingly used by patients for health benefits,⁸ and numerous studies have evaluated yoga as a complementary intervention for individuals with cancer.^9-14 Here, we review the physiological basis of yoga in oncology and the effects of yoga on biological processes, QOL, and symptoms during and after cancer treatment.

Physiological Basis

Many patients may use mind-body programs such as yoga to help manage the psychological and physiological consequences of unmanaged chronic stress and improve their overall QOL. The central nervous system, endocrine system, and immune system influence and interact with each other in a complex manner in response to chronic stress.^15,16 In a stressful situation, the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system (SNS) are activated. HPA axis stimulation leads to adrenocorticotrophic hormone production by the pituitary gland, which releases glucocorticoid hormones. SNS axis stimulation leads to epinephrine and norepinephrine production by the adrenal gland.^17,18 Recently, studies have explored modulation of signal transduction between the nervous and immune systems and how that may impact tumor growth and metastasis.¹⁹ Multiple studies, controlled for prognosis, disease stage, and other factors, have shown that patients experiencing more distress or higher levels of depressive symptoms do not live as long as their counterparts with low distress or depression levels.²⁰ Both the meditative and physical components of yoga can lead to enhanced relaxation, reduced SNS activation, and greater parasympathetic tone, countering the negative physiological effects of chronic stress. The effects of yoga on the HPA axis and SNS, proinflammatory cytokines, immune function, and DNA damage are discussed below.

Biological Processes

Nervous System

The effects of yoga and other forms of meditation on brain functions have been established through several studies. Yoga seems to influence basal ganglia function by improving circuits that are involved in complex cognitive functions, motor coordination, and somatosensory and emotional processes.^21,22 Additionally, changes in neurotransmitter levels have been observed after yoga practice. For instance, in a 12-week yoga intervention in healthy subjects, increased levels of thalamic gamma-aminobutyric acid (GABA) in the yoga group were reported to have a positive correlation with improved mood and decreased anxiety compared with a group who did metabolically matched walking exercise.²³ Levels of GABA, an inhibitory neurotransmitter, are decreased in conditions such as anxiety, depression, and epilepsy.²⁴ Yoga therapy has been shown to improve symptoms of mood disorders and epilepsy, which leads to the hypothesis that the mechanism driving the benefits of yoga may work through stimulation of vagal efferents and an increase in GABA-mediated cortical-inhibitory tone.^24,25

Stress activates the HPA/SNS axis, which releases hormones such as cortisol and norepinephrine. These hormones may play a role in angiogenesis, inflammation, immune suppression, and other physiological functions, and may even reduce the effect of chemotherapeutic agents.^26,27 Regular yoga practice has been shown to reduce SNS and HPA axis activity, most likely by increasing parasympathetic dominance through vagal stimulation, as demonstrated through increases in heart rate variability.²⁸ One indicator of HPA axis dysregulation, diurnal salivary cortisol rhythm, was shown to predict survival in patients with advanced breast and renal cancer.^29-33 Yoga has been shown to lead to less cortisol dysregulation due to radiotherapy and to reductions in mean cortisol levels and early morning cortisol levels in breast cancer patients undergoing radiotherapy.³⁴ This lends support to the hypothesis that yoga helps restore HPA axis balance.

Proinflammatory Cytokines

Cancer patients tend to have increased levels of inflammatory markers such as interleukin (IL)-4, IL-10, tumor necrosis factor (TNF), interferon-γ, and C-reactive protein. This increase in inflammation is associated with worse outcomes in cancer.³⁵ This association becomes highly relevant because the effect of inflammation on host cells in the tumor microenvironment is connected to disease progression.²⁶ Inflammatory cytokines are also implicated in cancer-related symptoms such as fatigue, cognitive dysfunction, peripheral neuropathy, and sleep disturbances.³⁶

Yoga is known to reduce stress and may directly or indirectly decrease inflammatory cytokines. A randomized clinical trial of a 12-week hatha yoga intervention among breast cancer survivors demonstrated decreases in IL-6, IL-1β, TNF, corticotropin-releasing factor, and cognitive complaints in the yoga group compared with those in the standard care group after 3 months.^37,38 Furthermore, Carlson et al showed that, after mindfulness-based stress reduction involving a combination of gentle yoga, meditation techniques, and relaxation exercises, breast and prostate cancer patients had reduced levels of proinflammatory cytokines and cortisol.³⁹ These reductions translated into patients reporting decreased stress levels and enhanced QOL.

Immune Function

The effects of yoga practice on the immune system have been studied in both healthy individuals and individuals with cancer. The effects on T and B lymphocytes, natural killer (NK) cells, and other immune effector cells demonstrate that meditation and yoga have beneficial effects on immune activity.⁴⁰ Hormones such as catecholamines and glucocorticoids are thought to influence the availability and function of NK cells, and, as noted above, yoga has been shown to modulate stress hormones and lead to reduced immune suppression in patients with early-stage breast cancer undergoing chemotherapy.⁴¹ Additional evidence supports the ability of yoga to reduce immune suppression in the postsurgical setting, with no observed decrease in NK cell percentage after surgery for those in a yoga group compared with a control group.⁴² This finding is relevant to patients undergoing surgical management of their cancer and highlights the impact of yoga on the immune system.

DNA Damage

Radiation damages DNA in the peripheral blood lymphocytes of patients undergoing treatment.^43,44 This damage is significant in breast cancer patients undergoing radiotherapy.⁴⁵ Stress additionally causes DNA damage⁴⁶ and is correlated to impaired DNA repair capacity.^47,48 In a study conducted by Banerjee et al, breast cancer patients were randomly assigned to a yoga group or a supportive therapy group for 6 weeks during radiotherapy.⁴⁹ Prior to the intervention, patients in the study had significant genomic instability. After treatment, patients in the yoga group experienced not only a significant reduction in anxiety and depression levels, but also a reduction in DNA damage due to radiotherapy.

There is evidence showing that yoga therapy improves multiple aspects of QOL, including physical functioning, emotional health outcomes, and the symptoms cancer patients may experience, such as sleep disturbances, fatigue, and pain. Danhauer et al systematically reviewed both nonrandomized trials and randomized controlled trials involving yoga during cancer treatment.⁵⁰ They found that yoga improved depression and anxiety as well as sleep and fatigue. Benefits of yoga in cancer based on randomized controlled trials are summarized in the Table. The role of yoga in improving QOL and managing symptoms patients experience during and after treatment is discussed in the following sections.

Quality of Life

Danhauer et al’s systematic review of trials involving yoga during cancer treatment found that yoga improved multiple aspects of QOL.⁵⁰ For example, yoga has been shown to improve QOL in breast cancer patients undergoing radiotherapy. In a study by Chandwani et al, yoga (60-minute sessions twice a week for 6 weeks) was associated with better general health perception and physical functioning scores as well as greater benefit finding, or finding meaning in their experience, after radiotherapy compared with a wait-list group.⁵¹ The yoga group had an increase in intrusive thoughts, believed to be due to a more thorough processing of the cancer experience, which helps to improve patients’ outlook on life.⁵² The benefits of yoga extend beyond psychological measures during radiation treatment. Yoga was found to increase physical functioning compared with stretching in breast cancer patients undergoing radiotherapy.⁵³

Cognitive Function

Cancer-related cognitive impairment commonly occurs during cancer treatments (eg, chemotherapy, radiotherapy, surgery, hormone therapy) and persists for months or years in survivors.⁵⁴ Impairment of memory, executive function, attention, and concentration are commonly reported. In a trial of a combined hatha and restorative yoga program called Yoga for Cancer Survivors (YOCAS), which was designed by researchers at the University of Rochester, patients in the yoga arm had less memory difficulty than did patients in the standard care arm.⁵⁵ However, the primary aim of the trial was to treat insomnia, so this secondary outcome needs to be interpreted with caution. Deficits in attention, memory, and executive function are often seen in cancer-related cognitive impairment, and the meditative aspect of yoga may have behavioral and neurophysical benefits that could improve cognitive functions.⁵⁶ More evidence is needed to understand the role of yoga in improving cognitive functioning.

Emotional Health

Psychosocial stress is high among breast cancer patients and survivors.^57,58 This causes circadian rhythm and cortisol regulation abnormalities, which are reported in women with breast cancer.^59-64 Yoga is known to help stress and psychosocial and physical functioning in patients with cancer.⁶⁵ Yoga was also shown to be equivalent to cognitive behavioral therapy in stress management in a population of patients without cancer.⁶⁶ Daily yoga sessions lasting 60 minutes were shown to reduce reactive anxiety and trait anxiety in early-stage breast cancer patients undergoing conventional radiotherapy and chemotherapy compared with patients receiving supportive therapy, highlighting the role of yoga in managing anxiety related to treatment.⁶⁷ In a study done by Culos-Reed et al, 20 cancer survivors who did 75 minutes of yoga per week for 7 weeks were compared with 18 cancer survivors who served as a control group.⁶⁸ The intervention group reported significant improvement in emotional well-being, depression, concentration, and mood disturbances. In a longitudinal study by Mackenzie et al, 66 cancer survivors completed a 7-week yoga program and were assessed at baseline, immediately after the final yoga session, and at 3 and 6 months after the final session.⁶⁹ Participants had significantly improved energy levels and affect. They also had moderate improvement in mindfulness and a moderate decrease in stress. Breast cancer patients who underwent restorative yoga sessions found improvements in mental health, depression, positive affect, and spirituality (peace/meaning).⁷⁰ This was more pronounced in women with higher negative affect and lower emotional well-being at baseline. In a study of patients with ovarian cancer receiving chemotherapy, patients were instructed to perform up to 15-minute sessions including awareness, body movement, and breathing.⁷¹ Even with just 1 session of yoga intervention, patients experienced decreased anxiety.

Fatigue

Studies on yoga show improvement in fatigue both during and after treatment. In breast cancer patients undergoing chemotherapy, yoga was shown to benefit cognitive fatigue.⁷² Older cancer survivors also seem to benefit from yoga interventions.⁷³ In a trial of a DVD-based yoga program, the benefits of yoga were similar to those of strengthening exercises, and both interventions helped decrease fatigue and improve QOL during the first year after diagnosis in early-stage breast cancer patients with cancer-related fatigue.⁷⁴ Bower et al also showed that, for breast cancer survivors experiencing persistent chronic fatigue, a targeted yoga intervention led to significant improvements in fatigue and vigor over a 3-month follow-up compared with controls.⁷⁵ Fatigue is commonly seen in breast cancer patients who are receiving adjuvant chemotherapy. In a study by Taso et al, women with breast cancer receiving chemotherapy were assigned to 60-minute yoga sessions incorporating Anusara yoga, gentle stretching, and relaxation twice a week for 8 weeks.⁷⁶ By week 4, patients with low pretest fatigue in the yoga group experienced a reduction in fatigue. By week 8, all patients in the yoga group experienced a reduction in fatigue. Four weeks after the yoga intervention, patients in the group maintained the reduction in fatigue. This study shows the feasibility of an 8-week yoga program for women undergoing breast cancer therapy by improving fatigue. Yoga recently was added to National Comprehensive Cancer Network (NCCN) guidelines for management of cancer-related fatigue (level 1 evidence).⁷⁷ However, the evidence was based on studies in women with breast cancer and survivors; therefore, more studies are needed in men and women with other cancers.

Distress surrounding surgery in patients with breast cancer can impact postoperative outcomes. Yoga interventions, including breathing exercises, regulated breathing, and yogic relaxation techniques, improved several postsurgical measures such as length of hospital stay, drain retention, and suture removal.⁷⁸ In this study, patients who practiced yoga also experienced a decrease in plasma TNF and better wound healing. Symptoms of anxiety and distress that occur preoperatively can lead to impaired immune function in addition to decreased QOL. In a study of yoga in early-stage breast cancer patients undergoing surgery, the benefit of yoga was seen not only with stress reduction but also with immune enhancement.⁴²

Yoga has been shown to help alleviate acute pain and distress in women undergoing major surgery for gynecological cancer. A regimen of 3 15-minute sessions of yoga, including awareness meditation, coordination of breath with movement, and relaxation breathing, was shown to reduce acute pain and distress in such patients in an inpatient setting.⁷⁹

Menopausal Symptoms

Breast cancer survivors have more severe menopausal symptoms compared with women without cancer.^80,81 Hot flashes cause sleep disturbances and worsen fatigue and QOL.⁸² Tamoxifen and aromatase inhibitors significantly worsen menopausal symptoms such as hot flashes.⁸¹ Carson et al conducted a study of yoga that included postures, breathing techniques, didactic presentations, and group discussions.⁸³ The yoga awareness regimen consisted of 8 weekly 120-minute group classes. Patients in the yoga arm had statistically significant improvements in the frequency, severity, and number of hot flashes. There were also improvements in arthralgia (joint pain), fatigue, sleep disturbance, vigor, and acceptance.

Arthralgia

Joint pain can be a major side effect that interferes with daily functions and activities in postmenopausal breast cancer survivors who receive aromatase inhibitor therapy.⁸⁴ Arthralgia is reported in up to 50% of patients treated with aromatase inhibitors.^84,85 It can affect functional status and lead to discontinuation of aromatase inhibitor therapy, jeopardizing clinical outcomes.⁸⁶ Yoga as a complementary therapy has been shown to improve conditions such as low back pain⁸⁷ and knee osteoarthritis⁸⁸ in patients who do not have cancer. In a single-arm pilot trial by Galantino et al, breast cancer patients with aromatase inhibitor–related joint pain were provided with twice-weekly yoga sessions for 8 weeks. There were statistically significant improvements in balance, flexibility, pain severity, and health-related QOL.⁸⁹ As noted above, improvement in arthralgia was also found in the study conducted by Carson et al.⁸³

Insomnia

Insomnia is common among cancer patients and survivors^90,91 and leads to increased fatigue and depression, decreased adherence to cancer treatments, and poor physical function and QOL.^90-92 Management of insomnia consists of pharmacologic therapies such as benzodiazepines^93,94 and nonpharmacologic options such as cognitive behavioral therapy.⁹⁵

The first study of yoga found to improve sleep quality was conducted at MD Anderson Cancer Center in lymphoma patients.⁹⁶ The effects of Tibetan yoga practices incorporating controlled breathing and visualization, mindfulness techniques, and low-impact postures were studied. Patients in the Tibetan yoga group had better subjective sleep quality, faster sleep latency, longer sleep duration, and less use of sleep medications. Mustian et al conducted a large yoga study in cancer survivors in which patients reporting chronic sleep disturbances were randomly assigned to the YOCAS program, which consisted of pranayama (breath control), 16 gentle hatha and restorative yoga postures, and meditation, or to usual care.⁹² The study reported improvements in global sleep quality, subjective sleep quality, actigraphy measures (wake after sleep onset, sleep efficiency), daytime dysfunction, and use of sleep medication after the yoga intervention compared with participants who received standard care.

Yoga to Address Other Symptoms

There is preliminary evidence supporting yoga as an integrative therapy for other symptoms unique to cancer survivors. For example, in head and neck cancer survivors, soft tissue damage involving the jaw, neck, shoulders, and chest results in swallowing issues, trismus, and aspiration, which are more pronounced in patients treated with conventional radiotherapy than in those treated with intensity-modulated radiotherapy.⁹⁷ Some late effects of radiotherapy for head and neck cancer—such as pain, anxiety, and impaired shoulder function—were shown to be improved through the practice of hatha yoga in 1 study.⁹⁸ Similarly, in a randomized controlled pilot study of patients with stage I to III breast cancer 6 months after treatment, participants in an 8-week yoga program experienced a reduction in arm induration and improvement in a QOL subscale of lymphedema symptoms. However, more evidence is needed to support the use of yoga as a therapeutic measure for breast cancer lymphedema.^99,100

Yoga for Caregivers

Along with cancer patients, caregivers face psychological and physical burdens as well as deterioration in their QOL. Caregivers tend to report clinical levels of anxiety, depression, sleep disturbance, and fatigue and have similar or in fact higher levels than those of the patients for whom they are caring.^101,102 Yoga has been found to help caregivers of patients with cancer. Recently, MD Anderson researchers conducted a trial in patients with high-grade glioma and their caregivers as dyads.^103,104 Each dyad attended 2 or 3 60-minute weekly Vivekananda yoga sessions involving breathing exercises, physical exercises, relaxation, and meditation. The researchers found that the yoga program was safe, feasible, acceptable, and subjectively useful for patients with high-grade glioma and their caregivers. Preliminary evidence of QOL improvement for both patients and caregivers was noted. An improvement in QOL was also demonstrated in another preliminary study of yoga in patients undergoing thoracic radiotherapy and their caregivers.¹⁰⁵

Another study by the group at MD Anderson evaluated a couple-based Tibetan yoga program that emphasized breathing exercises, gentle movements, guided visualizations, and emotional connectedness during radiotherapy for lung cancer.¹⁰⁶ This study included 10 patient‐caregiver dyads and found the program to be feasible, safe, and acceptable. The researchers also found preliminary evidence of improved QOL by the end of radiotherapy relative to baseline—specifically in the areas of spiritual well‐being for patients, fatigue for caregivers, and sleep disturbances and mental health issues such as anxiety and depressive symptoms for both patients and caregivers. This is noteworthy, as QOL typically deteriorates during the course of radiotherapy, and the yoga program was able to buffer these changes.

Conclusion

Yoga therapy has been used successfully as an adjunct modality to improve QOL and cancer-related symptoms. As a part of an integrative medicine approach, yoga is commonly recommended for patients undergoing cancer treatment. Danhauer et al reviewed randomized controlled trials during and after treatment and concluded that the evidence is clearly positive for QOL, fatigue, and perceived stress.¹⁰⁷ Results are less consistent but supportive for psychosocial outcomes such as benefit finding and spirituality. Evidence is mixed for sleep, anxiety, and depression. Post-treatment studies demonstrate improvements in fatigue, sleep, and multiple QOL domains. Yoga has been included in NCCN guidelines for fatigue management. Yoga, if approved by a physician, is also included among the behavioral therapies for anticipatory emesis and prevention and treatment of nausea in the recent update of the NCCN guidelines.¹⁰⁸ The Society for Integrative Oncology guidelines include yoga for anxiety/stress reduction as a part of integrative treatment in breast cancer patients during and after therapy, which was endorsed by the American Society of Clinical Oncology.¹⁰⁹

Because of the strong evidence for its benefits and a low side-effect profile, yoga is offered in group-class settings for patients during and after treatment and/or for caregivers in our institution. We often prescribe yoga as a therapeutic modality for selected groups of patients in our clinical practice. However, some patients may have restrictions after surgery that must be considered. In general, yoga has an excellent safety profile, the evidence base is strong, and we recommend that yoga therapy should be part of the standard of care as an integrative approach for patients with cancer undergoing active treatment as well as for cancer survivors and caregivers.

Acknowledgement: The authors thank Bryan Tutt for providing editorial assistance.

Corresponding author: Santhosshi Narayanan, MD, Department of Palliative, Rehabilitation, and Integrative Medicine, Unit 1414, The University of Texas MD Anderson Cancer Center, 1400 Pressler St., Houston, TX 77030; [email protected].

Financial disclosures: None.

From the University of Texas MD Anderson Cancer Center, Houston, TX (Drs. Narayanan, Lopez, Chaoul, Liu, Milbury, and Cohen, and Ms. Mallaiah); the University of Texas Health Science Center at Tyler (Dr. Meegada); and Texas Tech University Health Sciences Center, Lubbock, TX (Ms. Francisco).

Abstract

• Objective: To review the effects of yoga as an adjunct supportive care modality alongside conventional cancer treatment on quality of life (QOL), physical and mental health outcomes, and physiological and biological measures of cancer survivors.
• Methods: Nonsystematic review of the literature.
• Results: Yoga therapy, one of the most frequently used mind-body modalities, has been studied extensively in cancer survivors (from the time of diagnosis through long-term recovery). Yoga affects human physiology on multiple levels, including psychological outcomes, immune and endocrine function, and cardiovascular parameters, as well as multiple areas of QOL. It has been found to reduce psychological stress and fatigue and improve QOL in cancer patients and survivors. Yoga has also been used to manage symptoms such as arthralgia, fatigue, and insomnia. In addition, yoga offers benefits not only for cancer survivors but also for their caregivers.
• Conclusion: As part of an integrative, evidence-informed approach to cancer care, yoga may provide benefits that support the health of cancer survivors and caregivers.

Keywords: fatigue; cancer; proinflammatory cytokines; integrative; mind-body practices; meditation; DNA damage; stress; psychoneuro-immunoendocrine axis; lymphedema; insomnia.

A diagnosis of cancer and adverse effects related to its treatment may have negative effects on quality of life (QOL), contributing to emotional and physical distress in patients and caregivers. Many patients express an interest in pursuing nonpharmacological options, alone or as an adjunct to conventional therapy, to help manage symptoms. The use of complementary medicine approaches to health, including nonpharmacological approaches to symptom management, is highest among individuals with cancer.¹ According to a published expert consensus, integrative oncology is defined as a “patient-centered, evidence-informed field of cancer care that utilizes mind and body practices, natural products, and/or lifestyle modifications from different traditions alongside conventional cancer treatments. Integrative oncology aims to optimize health, QOL, and clinical outcomes across the cancer care continuum and to empower people to prevent cancer and become active participants before, during, and beyond cancer treatment.”² A key component of this definition, often misunderstood in the field of oncology, is that these modalities and treatments are used alongside conventional cancer treatments and not as an alternative. In an attempt to meet patients’ needs and appropriately use these approaches, integrative oncology programs are now part of most cancer centers in the United States.^3-6

Because of their overall safety, mind-body therapies are commonly used by patients and recommended by clinicians. Mind-body therapies include yoga, tai chi, qigong, meditation, and relaxation. Expressive arts such as journaling and music, art, and dance therapies also fall in the mind-body category.⁷ Yoga is a movement-based mind-body practice that focuses on synchronizing body, breath, and mind. Yoga has been increasingly used by patients for health benefits,⁸ and numerous studies have evaluated yoga as a complementary intervention for individuals with cancer.^9-14 Here, we review the physiological basis of yoga in oncology and the effects of yoga on biological processes, QOL, and symptoms during and after cancer treatment.

Physiological Basis

Many patients may use mind-body programs such as yoga to help manage the psychological and physiological consequences of unmanaged chronic stress and improve their overall QOL. The central nervous system, endocrine system, and immune system influence and interact with each other in a complex manner in response to chronic stress.^15,16 In a stressful situation, the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system (SNS) are activated. HPA axis stimulation leads to adrenocorticotrophic hormone production by the pituitary gland, which releases glucocorticoid hormones. SNS axis stimulation leads to epinephrine and norepinephrine production by the adrenal gland.^17,18 Recently, studies have explored modulation of signal transduction between the nervous and immune systems and how that may impact tumor growth and metastasis.¹⁹ Multiple studies, controlled for prognosis, disease stage, and other factors, have shown that patients experiencing more distress or higher levels of depressive symptoms do not live as long as their counterparts with low distress or depression levels.²⁰ Both the meditative and physical components of yoga can lead to enhanced relaxation, reduced SNS activation, and greater parasympathetic tone, countering the negative physiological effects of chronic stress. The effects of yoga on the HPA axis and SNS, proinflammatory cytokines, immune function, and DNA damage are discussed below.

Biological Processes

Nervous System

The effects of yoga and other forms of meditation on brain functions have been established through several studies. Yoga seems to influence basal ganglia function by improving circuits that are involved in complex cognitive functions, motor coordination, and somatosensory and emotional processes.^21,22 Additionally, changes in neurotransmitter levels have been observed after yoga practice. For instance, in a 12-week yoga intervention in healthy subjects, increased levels of thalamic gamma-aminobutyric acid (GABA) in the yoga group were reported to have a positive correlation with improved mood and decreased anxiety compared with a group who did metabolically matched walking exercise.²³ Levels of GABA, an inhibitory neurotransmitter, are decreased in conditions such as anxiety, depression, and epilepsy.²⁴ Yoga therapy has been shown to improve symptoms of mood disorders and epilepsy, which leads to the hypothesis that the mechanism driving the benefits of yoga may work through stimulation of vagal efferents and an increase in GABA-mediated cortical-inhibitory tone.^24,25

Stress activates the HPA/SNS axis, which releases hormones such as cortisol and norepinephrine. These hormones may play a role in angiogenesis, inflammation, immune suppression, and other physiological functions, and may even reduce the effect of chemotherapeutic agents.^26,27 Regular yoga practice has been shown to reduce SNS and HPA axis activity, most likely by increasing parasympathetic dominance through vagal stimulation, as demonstrated through increases in heart rate variability.²⁸ One indicator of HPA axis dysregulation, diurnal salivary cortisol rhythm, was shown to predict survival in patients with advanced breast and renal cancer.^29-33 Yoga has been shown to lead to less cortisol dysregulation due to radiotherapy and to reductions in mean cortisol levels and early morning cortisol levels in breast cancer patients undergoing radiotherapy.³⁴ This lends support to the hypothesis that yoga helps restore HPA axis balance.

Proinflammatory Cytokines

Cancer patients tend to have increased levels of inflammatory markers such as interleukin (IL)-4, IL-10, tumor necrosis factor (TNF), interferon-γ, and C-reactive protein. This increase in inflammation is associated with worse outcomes in cancer.³⁵ This association becomes highly relevant because the effect of inflammation on host cells in the tumor microenvironment is connected to disease progression.²⁶ Inflammatory cytokines are also implicated in cancer-related symptoms such as fatigue, cognitive dysfunction, peripheral neuropathy, and sleep disturbances.³⁶

Yoga is known to reduce stress and may directly or indirectly decrease inflammatory cytokines. A randomized clinical trial of a 12-week hatha yoga intervention among breast cancer survivors demonstrated decreases in IL-6, IL-1β, TNF, corticotropin-releasing factor, and cognitive complaints in the yoga group compared with those in the standard care group after 3 months.^37,38 Furthermore, Carlson et al showed that, after mindfulness-based stress reduction involving a combination of gentle yoga, meditation techniques, and relaxation exercises, breast and prostate cancer patients had reduced levels of proinflammatory cytokines and cortisol.³⁹ These reductions translated into patients reporting decreased stress levels and enhanced QOL.

Immune Function

The effects of yoga practice on the immune system have been studied in both healthy individuals and individuals with cancer. The effects on T and B lymphocytes, natural killer (NK) cells, and other immune effector cells demonstrate that meditation and yoga have beneficial effects on immune activity.⁴⁰ Hormones such as catecholamines and glucocorticoids are thought to influence the availability and function of NK cells, and, as noted above, yoga has been shown to modulate stress hormones and lead to reduced immune suppression in patients with early-stage breast cancer undergoing chemotherapy.⁴¹ Additional evidence supports the ability of yoga to reduce immune suppression in the postsurgical setting, with no observed decrease in NK cell percentage after surgery for those in a yoga group compared with a control group.⁴² This finding is relevant to patients undergoing surgical management of their cancer and highlights the impact of yoga on the immune system.

DNA Damage

Radiation damages DNA in the peripheral blood lymphocytes of patients undergoing treatment.^43,44 This damage is significant in breast cancer patients undergoing radiotherapy.⁴⁵ Stress additionally causes DNA damage⁴⁶ and is correlated to impaired DNA repair capacity.^47,48 In a study conducted by Banerjee et al, breast cancer patients were randomly assigned to a yoga group or a supportive therapy group for 6 weeks during radiotherapy.⁴⁹ Prior to the intervention, patients in the study had significant genomic instability. After treatment, patients in the yoga group experienced not only a significant reduction in anxiety and depression levels, but also a reduction in DNA damage due to radiotherapy.

There is evidence showing that yoga therapy improves multiple aspects of QOL, including physical functioning, emotional health outcomes, and the symptoms cancer patients may experience, such as sleep disturbances, fatigue, and pain. Danhauer et al systematically reviewed both nonrandomized trials and randomized controlled trials involving yoga during cancer treatment.⁵⁰ They found that yoga improved depression and anxiety as well as sleep and fatigue. Benefits of yoga in cancer based on randomized controlled trials are summarized in the Table. The role of yoga in improving QOL and managing symptoms patients experience during and after treatment is discussed in the following sections.

Quality of Life

Danhauer et al’s systematic review of trials involving yoga during cancer treatment found that yoga improved multiple aspects of QOL.⁵⁰ For example, yoga has been shown to improve QOL in breast cancer patients undergoing radiotherapy. In a study by Chandwani et al, yoga (60-minute sessions twice a week for 6 weeks) was associated with better general health perception and physical functioning scores as well as greater benefit finding, or finding meaning in their experience, after radiotherapy compared with a wait-list group.⁵¹ The yoga group had an increase in intrusive thoughts, believed to be due to a more thorough processing of the cancer experience, which helps to improve patients’ outlook on life.⁵² The benefits of yoga extend beyond psychological measures during radiation treatment. Yoga was found to increase physical functioning compared with stretching in breast cancer patients undergoing radiotherapy.⁵³

Cognitive Function

Cancer-related cognitive impairment commonly occurs during cancer treatments (eg, chemotherapy, radiotherapy, surgery, hormone therapy) and persists for months or years in survivors.⁵⁴ Impairment of memory, executive function, attention, and concentration are commonly reported. In a trial of a combined hatha and restorative yoga program called Yoga for Cancer Survivors (YOCAS), which was designed by researchers at the University of Rochester, patients in the yoga arm had less memory difficulty than did patients in the standard care arm.⁵⁵ However, the primary aim of the trial was to treat insomnia, so this secondary outcome needs to be interpreted with caution. Deficits in attention, memory, and executive function are often seen in cancer-related cognitive impairment, and the meditative aspect of yoga may have behavioral and neurophysical benefits that could improve cognitive functions.⁵⁶ More evidence is needed to understand the role of yoga in improving cognitive functioning.

Emotional Health

Psychosocial stress is high among breast cancer patients and survivors.^57,58 This causes circadian rhythm and cortisol regulation abnormalities, which are reported in women with breast cancer.^59-64 Yoga is known to help stress and psychosocial and physical functioning in patients with cancer.⁶⁵ Yoga was also shown to be equivalent to cognitive behavioral therapy in stress management in a population of patients without cancer.⁶⁶ Daily yoga sessions lasting 60 minutes were shown to reduce reactive anxiety and trait anxiety in early-stage breast cancer patients undergoing conventional radiotherapy and chemotherapy compared with patients receiving supportive therapy, highlighting the role of yoga in managing anxiety related to treatment.⁶⁷ In a study done by Culos-Reed et al, 20 cancer survivors who did 75 minutes of yoga per week for 7 weeks were compared with 18 cancer survivors who served as a control group.⁶⁸ The intervention group reported significant improvement in emotional well-being, depression, concentration, and mood disturbances. In a longitudinal study by Mackenzie et al, 66 cancer survivors completed a 7-week yoga program and were assessed at baseline, immediately after the final yoga session, and at 3 and 6 months after the final session.⁶⁹ Participants had significantly improved energy levels and affect. They also had moderate improvement in mindfulness and a moderate decrease in stress. Breast cancer patients who underwent restorative yoga sessions found improvements in mental health, depression, positive affect, and spirituality (peace/meaning).⁷⁰ This was more pronounced in women with higher negative affect and lower emotional well-being at baseline. In a study of patients with ovarian cancer receiving chemotherapy, patients were instructed to perform up to 15-minute sessions including awareness, body movement, and breathing.⁷¹ Even with just 1 session of yoga intervention, patients experienced decreased anxiety.

Fatigue

Studies on yoga show improvement in fatigue both during and after treatment. In breast cancer patients undergoing chemotherapy, yoga was shown to benefit cognitive fatigue.⁷² Older cancer survivors also seem to benefit from yoga interventions.⁷³ In a trial of a DVD-based yoga program, the benefits of yoga were similar to those of strengthening exercises, and both interventions helped decrease fatigue and improve QOL during the first year after diagnosis in early-stage breast cancer patients with cancer-related fatigue.⁷⁴ Bower et al also showed that, for breast cancer survivors experiencing persistent chronic fatigue, a targeted yoga intervention led to significant improvements in fatigue and vigor over a 3-month follow-up compared with controls.⁷⁵ Fatigue is commonly seen in breast cancer patients who are receiving adjuvant chemotherapy. In a study by Taso et al, women with breast cancer receiving chemotherapy were assigned to 60-minute yoga sessions incorporating Anusara yoga, gentle stretching, and relaxation twice a week for 8 weeks.⁷⁶ By week 4, patients with low pretest fatigue in the yoga group experienced a reduction in fatigue. By week 8, all patients in the yoga group experienced a reduction in fatigue. Four weeks after the yoga intervention, patients in the group maintained the reduction in fatigue. This study shows the feasibility of an 8-week yoga program for women undergoing breast cancer therapy by improving fatigue. Yoga recently was added to National Comprehensive Cancer Network (NCCN) guidelines for management of cancer-related fatigue (level 1 evidence).⁷⁷ However, the evidence was based on studies in women with breast cancer and survivors; therefore, more studies are needed in men and women with other cancers.

Distress surrounding surgery in patients with breast cancer can impact postoperative outcomes. Yoga interventions, including breathing exercises, regulated breathing, and yogic relaxation techniques, improved several postsurgical measures such as length of hospital stay, drain retention, and suture removal.⁷⁸ In this study, patients who practiced yoga also experienced a decrease in plasma TNF and better wound healing. Symptoms of anxiety and distress that occur preoperatively can lead to impaired immune function in addition to decreased QOL. In a study of yoga in early-stage breast cancer patients undergoing surgery, the benefit of yoga was seen not only with stress reduction but also with immune enhancement.⁴²

Yoga has been shown to help alleviate acute pain and distress in women undergoing major surgery for gynecological cancer. A regimen of 3 15-minute sessions of yoga, including awareness meditation, coordination of breath with movement, and relaxation breathing, was shown to reduce acute pain and distress in such patients in an inpatient setting.⁷⁹

Menopausal Symptoms

Breast cancer survivors have more severe menopausal symptoms compared with women without cancer.^80,81 Hot flashes cause sleep disturbances and worsen fatigue and QOL.⁸² Tamoxifen and aromatase inhibitors significantly worsen menopausal symptoms such as hot flashes.⁸¹ Carson et al conducted a study of yoga that included postures, breathing techniques, didactic presentations, and group discussions.⁸³ The yoga awareness regimen consisted of 8 weekly 120-minute group classes. Patients in the yoga arm had statistically significant improvements in the frequency, severity, and number of hot flashes. There were also improvements in arthralgia (joint pain), fatigue, sleep disturbance, vigor, and acceptance.

Arthralgia

Joint pain can be a major side effect that interferes with daily functions and activities in postmenopausal breast cancer survivors who receive aromatase inhibitor therapy.⁸⁴ Arthralgia is reported in up to 50% of patients treated with aromatase inhibitors.^84,85 It can affect functional status and lead to discontinuation of aromatase inhibitor therapy, jeopardizing clinical outcomes.⁸⁶ Yoga as a complementary therapy has been shown to improve conditions such as low back pain⁸⁷ and knee osteoarthritis⁸⁸ in patients who do not have cancer. In a single-arm pilot trial by Galantino et al, breast cancer patients with aromatase inhibitor–related joint pain were provided with twice-weekly yoga sessions for 8 weeks. There were statistically significant improvements in balance, flexibility, pain severity, and health-related QOL.⁸⁹ As noted above, improvement in arthralgia was also found in the study conducted by Carson et al.⁸³

Insomnia

Insomnia is common among cancer patients and survivors^90,91 and leads to increased fatigue and depression, decreased adherence to cancer treatments, and poor physical function and QOL.^90-92 Management of insomnia consists of pharmacologic therapies such as benzodiazepines^93,94 and nonpharmacologic options such as cognitive behavioral therapy.⁹⁵

The first study of yoga found to improve sleep quality was conducted at MD Anderson Cancer Center in lymphoma patients.⁹⁶ The effects of Tibetan yoga practices incorporating controlled breathing and visualization, mindfulness techniques, and low-impact postures were studied. Patients in the Tibetan yoga group had better subjective sleep quality, faster sleep latency, longer sleep duration, and less use of sleep medications. Mustian et al conducted a large yoga study in cancer survivors in which patients reporting chronic sleep disturbances were randomly assigned to the YOCAS program, which consisted of pranayama (breath control), 16 gentle hatha and restorative yoga postures, and meditation, or to usual care.⁹² The study reported improvements in global sleep quality, subjective sleep quality, actigraphy measures (wake after sleep onset, sleep efficiency), daytime dysfunction, and use of sleep medication after the yoga intervention compared with participants who received standard care.

Yoga to Address Other Symptoms

There is preliminary evidence supporting yoga as an integrative therapy for other symptoms unique to cancer survivors. For example, in head and neck cancer survivors, soft tissue damage involving the jaw, neck, shoulders, and chest results in swallowing issues, trismus, and aspiration, which are more pronounced in patients treated with conventional radiotherapy than in those treated with intensity-modulated radiotherapy.⁹⁷ Some late effects of radiotherapy for head and neck cancer—such as pain, anxiety, and impaired shoulder function—were shown to be improved through the practice of hatha yoga in 1 study.⁹⁸ Similarly, in a randomized controlled pilot study of patients with stage I to III breast cancer 6 months after treatment, participants in an 8-week yoga program experienced a reduction in arm induration and improvement in a QOL subscale of lymphedema symptoms. However, more evidence is needed to support the use of yoga as a therapeutic measure for breast cancer lymphedema.^99,100

Yoga for Caregivers

Along with cancer patients, caregivers face psychological and physical burdens as well as deterioration in their QOL. Caregivers tend to report clinical levels of anxiety, depression, sleep disturbance, and fatigue and have similar or in fact higher levels than those of the patients for whom they are caring.^101,102 Yoga has been found to help caregivers of patients with cancer. Recently, MD Anderson researchers conducted a trial in patients with high-grade glioma and their caregivers as dyads.^103,104 Each dyad attended 2 or 3 60-minute weekly Vivekananda yoga sessions involving breathing exercises, physical exercises, relaxation, and meditation. The researchers found that the yoga program was safe, feasible, acceptable, and subjectively useful for patients with high-grade glioma and their caregivers. Preliminary evidence of QOL improvement for both patients and caregivers was noted. An improvement in QOL was also demonstrated in another preliminary study of yoga in patients undergoing thoracic radiotherapy and their caregivers.¹⁰⁵

Another study by the group at MD Anderson evaluated a couple-based Tibetan yoga program that emphasized breathing exercises, gentle movements, guided visualizations, and emotional connectedness during radiotherapy for lung cancer.¹⁰⁶ This study included 10 patient‐caregiver dyads and found the program to be feasible, safe, and acceptable. The researchers also found preliminary evidence of improved QOL by the end of radiotherapy relative to baseline—specifically in the areas of spiritual well‐being for patients, fatigue for caregivers, and sleep disturbances and mental health issues such as anxiety and depressive symptoms for both patients and caregivers. This is noteworthy, as QOL typically deteriorates during the course of radiotherapy, and the yoga program was able to buffer these changes.

Conclusion

Yoga therapy has been used successfully as an adjunct modality to improve QOL and cancer-related symptoms. As a part of an integrative medicine approach, yoga is commonly recommended for patients undergoing cancer treatment. Danhauer et al reviewed randomized controlled trials during and after treatment and concluded that the evidence is clearly positive for QOL, fatigue, and perceived stress.¹⁰⁷ Results are less consistent but supportive for psychosocial outcomes such as benefit finding and spirituality. Evidence is mixed for sleep, anxiety, and depression. Post-treatment studies demonstrate improvements in fatigue, sleep, and multiple QOL domains. Yoga has been included in NCCN guidelines for fatigue management. Yoga, if approved by a physician, is also included among the behavioral therapies for anticipatory emesis and prevention and treatment of nausea in the recent update of the NCCN guidelines.¹⁰⁸ The Society for Integrative Oncology guidelines include yoga for anxiety/stress reduction as a part of integrative treatment in breast cancer patients during and after therapy, which was endorsed by the American Society of Clinical Oncology.¹⁰⁹

Because of the strong evidence for its benefits and a low side-effect profile, yoga is offered in group-class settings for patients during and after treatment and/or for caregivers in our institution. We often prescribe yoga as a therapeutic modality for selected groups of patients in our clinical practice. However, some patients may have restrictions after surgery that must be considered. In general, yoga has an excellent safety profile, the evidence base is strong, and we recommend that yoga therapy should be part of the standard of care as an integrative approach for patients with cancer undergoing active treatment as well as for cancer survivors and caregivers.

Acknowledgement: The authors thank Bryan Tutt for providing editorial assistance.

Corresponding author: Santhosshi Narayanan, MD, Department of Palliative, Rehabilitation, and Integrative Medicine, Unit 1414, The University of Texas MD Anderson Cancer Center, 1400 Pressler St., Houston, TX 77030; [email protected].

Financial disclosures: None.

From The Communication in Healthcare Group, Seattle, WA.

Abstract

• Objective: To review established approaches to disclosure and resolution following adverse medical outcomes and highlight barriers that may hinder universal implementation of effective disclosure/resolution practices.
• Methods: An overview of established approaches to disclosure and resolution of adverse medical outcomes is presented.
• Results: Clinicians must be equipped to manage situations where adverse medical outcomes occur even though the care provided was reasonable, within the standard, as well as in situations where preventable problems in the care provided were likely the cause of patient harm. Established approaches that have proven useful for investigating, disclosing, and resolving situations, captured in the acronyms AIDR, ALEE, and TEAM, can assist clinicians in the disclosure and ultimate resolution of these 2 types of situations.
• Conclusion: Health care organizations with a solid commitment and a reliable structure for ensuring adherence to full disclosure and fair resolution of adverse outcomes have demonstrated sustainable progress in ethically and effectively resolving situations where patients are harmed by medical care.

Keywords: safety; medical error; adverse outcomes; resolution; communication.

Much has been learned over the 20 years since the Institute of Medicine’s (IOM) report To Err Is Human¹ was published. At the time it was published, the IOM report made it clear that only a minority of preventable patient harms were being acknowledged, investigated, and reported. In the face of adverse outcomes “dissemble, deny, and defend” was a common strategy of many clinicians, institutions, and liability carriers.² The health care system appeared to place a priority on protecting itself from reputational and financial harm over the rights of injured patients to be given an accurate understanding of what had happened in their care and to pursue restitution, if appropriate.^3-5

The emerging quality improvement movement was accompanied by calls for increased patient advocacy. This included the goal of greater transparency and more timely and equitable resolutions with patients who have been harmed by problems in care. Health care systems pressed for confidentiality protections in exchange for increased focus on quality improvement.⁶ Applying medical ethics of autonomy, no-maleficence, beneficence, and justice initially took a backseat, as risk management was given priority.⁷ Insurance carriers have no ethical obligation, and a clear disincentive, to assure that harmed patients are fully informed and offered restitution. Some self-insured health systems, however, began experimenting with more proactive and transparent approaches to disclosure and resolution. In contrast to the often-reported fear of a liability explosion, they reported reduced claims and suits, shorter time to resolution, and reduced overall financial cost,^8-10 providing some evidence that perhaps greater openness could work after all.

But for providers and staff to allow transparency and candor to become the norm, institutions needed to create a more “just culture” for managing errors. Individual impairment or willful disregard of safe practice would need to be handled differently from the slips and lapses that more often contributed to preventable harm.¹¹ For example, the nurse who was inadequately oriented to the equipment on an unfamiliar unit where she was asked to work a double shift due to a staffing shortage should not be held as accountable as an employee who knowingly violated agreed upon safe practices, even though patient harm resulted in each situation. It became clear that patient harm was usually the result of multiple factors involving individuals, communication, procedures, systems, and equipment. Blaming and disciplining individuals at the sharp end would not reliably reduce adverse outcomes.

Since the 1999 IOM report, we have developed general agreement on best practices for investigating, disclosing, and resolving situations where patients are harmed by medical care.^12,13 This article reviews the perspectives and practices that appear necessary for effective disclosure and resolution after an adverse outcome and highlights barriers to reliably enacting them in practice.

Elements of Effective Disclosure

Effective disclosure to patients and families hinges on determining and providing an accurate understanding of what happened in the patient’s care. It should be the care providers’ and their institution’s responsibility to determine causation and disclose it. This should not require only the most upset patients and families initiating a legal process taking 3 years or more to complete. The most consequential question must be answered, “Was the care provided reasonable?” That is, was everything done within the standard, as would have been expected by similarly trained clinicians with the information and resources available at that time? It follows that if care was reasonable, then the adverse outcome could not normally have been prevented, no correction in care processes is called for, and no financial compensation is required. If the care review reveals deficiencies in care that were linked to patient harm, then achieving a satisfying resolution would be more complex and difficult.¹³ First, individuals would have to accept that they have contributed to patient harm, itself an often-contentious process and psychologically devastating realization. Then they must have this difficult conversation with patient and family, creating liability risk for themselves in the process. They must commit to correcting the problems that contributed to the harm. They must facilitate, rather than obstruct, a path to a restitution that addresses the medical, practical, and financial harms that have resulted. Given the challenges inherent to disclosure and resolution, it is no wonder that dissembling, denying, and defending was the common practice for the preceding decades.¹⁴

Disclosure and Resolution Pathways

I was the co-developer of an approach to disclosure and resolution which is now widely accepted and that has been taught across the United States and Canada to more than 50,000 health care providers and administrators over 18 years.^15,16 We learned that resolving adverse medical outcomes is a 4-part process (anticipate, investigate, disclose, resolve [AIDR]). Most adverse or simply disappointing outcomes occur despite reasonable care (eg, due to biological variability, the imprecision of the science and limitations and risks of the procedures). The minority of harms are associated with deficiencies in the care (ie, unreasonable care). We need to equip ourselves to manage both situations effectively. The approach we developed can be captured in 3 acronyms: AIDR, ALEE, and TEAM,

AIDR

This acronym encapsulates the overview guidance for clinicians after an adverse event or outcome, regardless of the cause.

Anticipate the thoughts and feelings of the harmed/disappointed patient and family and reach out immediately with an expression of sympathy.

Investigate sufficiently to address questions about most likely causation and do not conjecture prior to investigation. Ask for patience—“You deserve more than a guess”—and keep in regular contact to reinforce the promise that there will be a full reporting when the review is complete.

Disclose (in a planned and coordinated manner) what has been learned in the investigation.

Resolve the situation with the patient and family consistent with our ethical principles.

If our failure caused the harm (care unreasonable/breached the standard), then working toward a fair restitution and taking corrective actions are appropriate. If the care was found to have been reasonable, then compensation would not be offered and corrective action is unwarranted. The organization would defend reasonable care if a claim was still pursued.

This process involves ethical clarity, emotional intelligence, and discipline. Clinicians must first acknowledge that a disappointing outcome or event has occurred. Clinicians involved in the care, usually led by the attending provider, then immediately reach out to the patient and family with sympathy, a plan of care to address the medical issues, and the promise to investigate and follow-up with the patient and family when the harm and its causes are more clearly determined. To disclose simply means to provide an accurate understanding (ie, the understanding determined by the investigation we conducted) of what happened, its causes, and consequences. Depending on the extent of the harm and the complexity and time needed for the investigation, a “coach” or “disclosure coordinator” who has advanced training in managing these situations is brought in to guide the process. The disclosure coach/coordinator provides a consistent and steady hand throughout the process of investigation, disclosure, and ultimately resolution with patient and family. Patients and families often move across settings during the time of the AIDR process, and it is easy to lose track of them unless someone is following the entire process until resolved.

ALEE

When the investigation of an adverse/disappointing outcome determines the care was reasonable and therefore the adverse outcome could not have been prevented, we use the ALEE pathway to guide the disclosure conversation (Step 3 in AIDR) with the patient and family:

Anticipate. What are the questions, thoughts, and feelings we would expect the patient and family will have? On this track, there is nothing to apologize for since the care was reasonable, yet expressing compassion and sympathy for the patient’s experience is essential. “I/we really sympathize with how differently this has turned out than we had hoped.”

Listen. Invite and listen for their questions and concerns, how they are seeing the situation, and where and what they are finding most upsetting and in need of explanation.

Empathize. There are 2 kinds of empathy required here. Cognitive empathy means showing that we understand their thinking from their perspective, separate from whether we fully agree. Emotional empathy involves demonstrating that their emotions are understandable given the situation, even if those emotions are painful for clinicians to experience. Listening in step 2 is how we learned their perspective and emotions. Now we can show accurate empathy: I/we can understand how upsetting it is to be facing another set of procedures to treat the unfortunate complications from your last surgery.

Explain. Even when care is reasonable, questions and perhaps suspicions are to be expected. Listening and empathizing sets us up to focus our explanations on the patient’s and family’s key questions with a level of thoughtfulness and transparency that conveys credibility. We should not assume, however, that they have accepted our explanation. Instead, solicit their reactions and unresolved questions as part of the disclosure discussion. It is normal for additional concerns to emerge in the days after the disclosure discussion, and we should be ready to address these concerns until resolved. In some instances, the patient and family will not be satisfied and it may be helpful to offer an independent review of the care. If the unresolved patient and family engages an attorney, that will be the first step taken anyway. Proactively offering an independent review signals confidence in your objectivity and sensitivity to the importance of fairness for the patient and family: Your questions and concerns are completely normal in light of the disappointing experience you have had. Let me see if I/we can address those now to your satisfaction.

TEAM

If the investigation determines that aspects of the care were unreasonable (breached the standard) and the adverse outcome/harm was related to the deficiencies in the care, then we use the TEAM pathway to disclose and resolve the situation with the patient and family

Truthful and Transparent and Teamwork. We should be offering our most accurate understanding of how the adverse outcome occurred, with sufficient depth and clarity that the patient and family can see how we reached that conclusion. In straightforward situations involving minor harm (eg, an allergic reaction to a medication that the clinician overlooked and that resulted in an urgent care center visit), a very limited investigation may clarify the situation sufficiently that the prescribing provider, accompanied by an office or staff nurse as support and witness, may be able to complete an effective disclosure in a single discussion, and simply writing off a bill or arranging to reimburse the urgent care center visit cost may satisfy the affected patient.

In more complex situations involving greater harm, a number of people must be involved to accomplish TEAM tasks: to offer an explanation, to answer questions, to make apologies, to explain changes intended to reduce the chance of harm to others in the future, and to work through any restitution that may be appropriate. Appointing a disclosure coach/coordinator/facilitator who has had extended training in the disclosure process can help guide these more complex situations. Risk management, insurance carriers, and legal counsel should be aware and advising throughout the process and participating directly in meetings with the patient and family, as appropriate. Since on the TEAM track we are admitting liability, offering a path to financial restitution may be warranted and the disclosure process may trigger reporting requirements with regulatory as well as human resource implications.

The patient and family may want to include other people on their “team” as well. Since complex disclosure meetings need to be carefully planned in advance, we should clarify who will be attending from the health care side and who the family intends to involve. We should anticipate potential requests and questions such as: Would it be OK to record this meeting? Can we ask our attorney to attend? Who are all these people and why are they in this meeting? (We should introduce all team members and clarify how their involvement is necessary to help reach the most satisfying resolution for all involved.)

Empathize. Admitting that deficient aspects in the care contributed to the harm will trigger thoughts, emotions, and expectations for the patient and family. Empathizing involves seeing the whole situation from their perspective and acknowledging their emotions as understandable. Empathizing is not the same as fully agreeing with the patient’s and family’s perspective, but we will not be able to effectively address concerns and expectations that we have not understood. Organizations should have supports in place for staff who are involved in these difficult situations. Nonetheless, we must prioritize the patient’s and family‘s feelings in a disclosure meeting.

Apologize and be Accountable. This calls for both expressions of sympathy as well as a genuine apology for having caused harm by failure in some aspect of care: We are very sorry you are going through this difficult situation. We are especially sorry to tell you that we now recognize that problems in the care we provided are the most likely cause of this harm. Would this be a good time to explain what we learned?

Having the responsible clinicians present increases the chances of achieving the most complete resolution in a single planned and well facilitated meeting. The tasks for that meeting include: offering an explanation that reveals the problems in care that contributed to the adverse outcome, making sincere apologies, and explaining changes to reduce chance of harm to others. The disclosure coach can work with individuals to help them understand how and why their involvement can be important and to help staff members become ready to participate constructively in the disclosure meeting. When individuals appear unable or unwilling to contribute constructively, a plan is needed for how their part can be replaced (eg, a charge nurse or department chair might need to step in to explain and apologize for the care of a subordinate). Managers/administrators can explain contributory factors for what may at first appear to have been simply individual negligence. Administrators can describe the actions that the organization is taking to correct problems that contributed to the patient harm: As nursing executive, it is my responsibility to see that all our staff have been adequately trained on the equipment we are asking them to use. We now recognize that the nurse’s lack of familiarity with that equipment contributed to the harm you experienced and I am very sorry for that. It is my responsibility to get that problem corrected, and we are already taking steps to assure that. Patients and families often have ideas for improving care processes and appreciate being invited to share these ideas as a service to future patients.

Manage until resolved. On the “care unreasonable” track, we must signal openness to helping with the patient’s and family’s immediate and longer-term needs, as well as their expectations about financial and other forms of restitution. Someone should be in the meeting who can describe the next steps in working towards a fair restitution and how that process will take place following the conclusion of the disclosure meeting. The close involvement of risk and claims professionals throughout the process of investigation through to the disclosure discussion itself will assure a more satisfactory handoff to questions about around financial compensation

Psychological Barriers to Implementation of Disclosure Pathways

Many organizations and researchers agree that disclosure and resolution pathways as just described are the most ethical and effective ways for all parties to resolve these painful situations. So why isn’t this approach universally practiced? In concluding this article, it may be helpful to point out some of the human dynamics that make resolution more difficult and how they might be addressed.

A key issue is the “urge to self-preservation.” Health care organizations have often been accused of disclosing only what they cannot hide. We have repeatedly observed how individuals and organizations are often initially motivated to do whatever is needed to protect themselves, even when those behaviors are frankly deceptive. This is almost to be expected. By age 4 children have learned to use deception as a defensive strategy when confronted with misbehavior. Research shows that children and adults continue the strategy to escape censure or punishment and simply get better at hiding their tracks.¹⁸ Because people want to preserve their image as ethical individuals, they have also learned to rationalize/justify this deception as necessary for self-preservation (“My dad would have killed me,” “I will lose my license,” “It is not fair that I take the blame when others have done the same thing and gotten away with it.”). Imagining the most extreme, and therefore “unfair” consequences, helps justify the individual’s use of dissembling and frank deception in order to avoid them. Clinicians and organizations may convince themselves that they are the victims entitled to protection rather than the injured patient. Patients and families often accept explanations that are less than candid, as doctors and nurses remain among the most trusted of professionals. Sufficiently understanding the complexities of the care is beyond the capability of most lay people. Successfully challenging the clinician’s or institution’s exculpatory explanation for an adverse outcome is very difficult, even though many clinicians believe that the tort system is stacked against them.

As a result, even the most sensible of best practices, toolkits, and trainings will not make full disclosure and fair resolution of adverse outcomes more likely without a counterweight of solid ethical commitment and a reliable structure for ensuring adherence. Sustainable progress has been demonstrated in those institutions^8,10,17 where: (1) institutional values and ethics around disclosure were elevated above self-protection, (2) efficient processes for recognizing and objectively reviewing care involving an adverse outcome were developed and followed, (3) salaried and institutionally insured staff and providers were required to participate in and accept a fair path to resolution in the context of a just culture, and (4) the institution was able to deliver on any commitments (eg, financial, corrective actions) it has made. Conversely, disclosure and resolution programs have struggled in the following situations: where values and ethics are not clarified and made primary; where the processes for reviewing adverse outcomes are slow, inconsistent, and open to political interference; where independent providers have latitude to insist on self-protective behaviors; and where liability carriers who place highest priority on avoiding financial exposure are involved.

Conclusion

The challenge of effectively disclosing and resolving adverse medical outcomes will continue to be most formidable for health care systems with independent medical staffs with separate liability carriers. Can these systems get a firm consensus on the ethics that are paramount in disclosure situations? Can they create care review systems that are efficient and objective and reach conclusions that are binding on those involved? Are they willing to provide explanations to patients and families regardless of the consequences to themselves? Can they coordinate an efficient path to financial and other forms of restitution in those situations where problems in the care contributed to the patient being harmed? And can they enforce these practices despite the self-concerns of all the involved parties? The good news is we now know how to disclose and resolve adverse medical outcomes with patients and families in a way that is fair to providers, staff, and institutions and will not break the bank. For health care organizations, implementing effective disclosure and resolution practices starts with a commitment to both build consensus for this process and consistently enforce it.

Corresponding author: Daniel O’Connell, PhD, 2212 Queen Anne Ave. N. #810, Seattle, WA 98109; [email protected].

Financial disclosures: None.

From The Communication in Healthcare Group, Seattle, WA.

Abstract

• Objective: To review established approaches to disclosure and resolution following adverse medical outcomes and highlight barriers that may hinder universal implementation of effective disclosure/resolution practices.
• Methods: An overview of established approaches to disclosure and resolution of adverse medical outcomes is presented.
• Results: Clinicians must be equipped to manage situations where adverse medical outcomes occur even though the care provided was reasonable, within the standard, as well as in situations where preventable problems in the care provided were likely the cause of patient harm. Established approaches that have proven useful for investigating, disclosing, and resolving situations, captured in the acronyms AIDR, ALEE, and TEAM, can assist clinicians in the disclosure and ultimate resolution of these 2 types of situations.
• Conclusion: Health care organizations with a solid commitment and a reliable structure for ensuring adherence to full disclosure and fair resolution of adverse outcomes have demonstrated sustainable progress in ethically and effectively resolving situations where patients are harmed by medical care.

Keywords: safety; medical error; adverse outcomes; resolution; communication.

Much has been learned over the 20 years since the Institute of Medicine’s (IOM) report To Err Is Human¹ was published. At the time it was published, the IOM report made it clear that only a minority of preventable patient harms were being acknowledged, investigated, and reported. In the face of adverse outcomes “dissemble, deny, and defend” was a common strategy of many clinicians, institutions, and liability carriers.² The health care system appeared to place a priority on protecting itself from reputational and financial harm over the rights of injured patients to be given an accurate understanding of what had happened in their care and to pursue restitution, if appropriate.^3-5

The emerging quality improvement movement was accompanied by calls for increased patient advocacy. This included the goal of greater transparency and more timely and equitable resolutions with patients who have been harmed by problems in care. Health care systems pressed for confidentiality protections in exchange for increased focus on quality improvement.⁶ Applying medical ethics of autonomy, no-maleficence, beneficence, and justice initially took a backseat, as risk management was given priority.⁷ Insurance carriers have no ethical obligation, and a clear disincentive, to assure that harmed patients are fully informed and offered restitution. Some self-insured health systems, however, began experimenting with more proactive and transparent approaches to disclosure and resolution. In contrast to the often-reported fear of a liability explosion, they reported reduced claims and suits, shorter time to resolution, and reduced overall financial cost,^8-10 providing some evidence that perhaps greater openness could work after all.

But for providers and staff to allow transparency and candor to become the norm, institutions needed to create a more “just culture” for managing errors. Individual impairment or willful disregard of safe practice would need to be handled differently from the slips and lapses that more often contributed to preventable harm.¹¹ For example, the nurse who was inadequately oriented to the equipment on an unfamiliar unit where she was asked to work a double shift due to a staffing shortage should not be held as accountable as an employee who knowingly violated agreed upon safe practices, even though patient harm resulted in each situation. It became clear that patient harm was usually the result of multiple factors involving individuals, communication, procedures, systems, and equipment. Blaming and disciplining individuals at the sharp end would not reliably reduce adverse outcomes.

Since the 1999 IOM report, we have developed general agreement on best practices for investigating, disclosing, and resolving situations where patients are harmed by medical care.^12,13 This article reviews the perspectives and practices that appear necessary for effective disclosure and resolution after an adverse outcome and highlights barriers to reliably enacting them in practice.

Elements of Effective Disclosure

Effective disclosure to patients and families hinges on determining and providing an accurate understanding of what happened in the patient’s care. It should be the care providers’ and their institution’s responsibility to determine causation and disclose it. This should not require only the most upset patients and families initiating a legal process taking 3 years or more to complete. The most consequential question must be answered, “Was the care provided reasonable?” That is, was everything done within the standard, as would have been expected by similarly trained clinicians with the information and resources available at that time? It follows that if care was reasonable, then the adverse outcome could not normally have been prevented, no correction in care processes is called for, and no financial compensation is required. If the care review reveals deficiencies in care that were linked to patient harm, then achieving a satisfying resolution would be more complex and difficult.¹³ First, individuals would have to accept that they have contributed to patient harm, itself an often-contentious process and psychologically devastating realization. Then they must have this difficult conversation with patient and family, creating liability risk for themselves in the process. They must commit to correcting the problems that contributed to the harm. They must facilitate, rather than obstruct, a path to a restitution that addresses the medical, practical, and financial harms that have resulted. Given the challenges inherent to disclosure and resolution, it is no wonder that dissembling, denying, and defending was the common practice for the preceding decades.¹⁴

Disclosure and Resolution Pathways

I was the co-developer of an approach to disclosure and resolution which is now widely accepted and that has been taught across the United States and Canada to more than 50,000 health care providers and administrators over 18 years.^15,16 We learned that resolving adverse medical outcomes is a 4-part process (anticipate, investigate, disclose, resolve [AIDR]). Most adverse or simply disappointing outcomes occur despite reasonable care (eg, due to biological variability, the imprecision of the science and limitations and risks of the procedures). The minority of harms are associated with deficiencies in the care (ie, unreasonable care). We need to equip ourselves to manage both situations effectively. The approach we developed can be captured in 3 acronyms: AIDR, ALEE, and TEAM,

AIDR

This acronym encapsulates the overview guidance for clinicians after an adverse event or outcome, regardless of the cause.

Anticipate the thoughts and feelings of the harmed/disappointed patient and family and reach out immediately with an expression of sympathy.

Investigate sufficiently to address questions about most likely causation and do not conjecture prior to investigation. Ask for patience—“You deserve more than a guess”—and keep in regular contact to reinforce the promise that there will be a full reporting when the review is complete.

Disclose (in a planned and coordinated manner) what has been learned in the investigation.

Resolve the situation with the patient and family consistent with our ethical principles.

If our failure caused the harm (care unreasonable/breached the standard), then working toward a fair restitution and taking corrective actions are appropriate. If the care was found to have been reasonable, then compensation would not be offered and corrective action is unwarranted. The organization would defend reasonable care if a claim was still pursued.

This process involves ethical clarity, emotional intelligence, and discipline. Clinicians must first acknowledge that a disappointing outcome or event has occurred. Clinicians involved in the care, usually led by the attending provider, then immediately reach out to the patient and family with sympathy, a plan of care to address the medical issues, and the promise to investigate and follow-up with the patient and family when the harm and its causes are more clearly determined. To disclose simply means to provide an accurate understanding (ie, the understanding determined by the investigation we conducted) of what happened, its causes, and consequences. Depending on the extent of the harm and the complexity and time needed for the investigation, a “coach” or “disclosure coordinator” who has advanced training in managing these situations is brought in to guide the process. The disclosure coach/coordinator provides a consistent and steady hand throughout the process of investigation, disclosure, and ultimately resolution with patient and family. Patients and families often move across settings during the time of the AIDR process, and it is easy to lose track of them unless someone is following the entire process until resolved.

ALEE

When the investigation of an adverse/disappointing outcome determines the care was reasonable and therefore the adverse outcome could not have been prevented, we use the ALEE pathway to guide the disclosure conversation (Step 3 in AIDR) with the patient and family:

Anticipate. What are the questions, thoughts, and feelings we would expect the patient and family will have? On this track, there is nothing to apologize for since the care was reasonable, yet expressing compassion and sympathy for the patient’s experience is essential. “I/we really sympathize with how differently this has turned out than we had hoped.”

Listen. Invite and listen for their questions and concerns, how they are seeing the situation, and where and what they are finding most upsetting and in need of explanation.

Empathize. There are 2 kinds of empathy required here. Cognitive empathy means showing that we understand their thinking from their perspective, separate from whether we fully agree. Emotional empathy involves demonstrating that their emotions are understandable given the situation, even if those emotions are painful for clinicians to experience. Listening in step 2 is how we learned their perspective and emotions. Now we can show accurate empathy: I/we can understand how upsetting it is to be facing another set of procedures to treat the unfortunate complications from your last surgery.

Explain. Even when care is reasonable, questions and perhaps suspicions are to be expected. Listening and empathizing sets us up to focus our explanations on the patient’s and family’s key questions with a level of thoughtfulness and transparency that conveys credibility. We should not assume, however, that they have accepted our explanation. Instead, solicit their reactions and unresolved questions as part of the disclosure discussion. It is normal for additional concerns to emerge in the days after the disclosure discussion, and we should be ready to address these concerns until resolved. In some instances, the patient and family will not be satisfied and it may be helpful to offer an independent review of the care. If the unresolved patient and family engages an attorney, that will be the first step taken anyway. Proactively offering an independent review signals confidence in your objectivity and sensitivity to the importance of fairness for the patient and family: Your questions and concerns are completely normal in light of the disappointing experience you have had. Let me see if I/we can address those now to your satisfaction.

TEAM

If the investigation determines that aspects of the care were unreasonable (breached the standard) and the adverse outcome/harm was related to the deficiencies in the care, then we use the TEAM pathway to disclose and resolve the situation with the patient and family

Truthful and Transparent and Teamwork. We should be offering our most accurate understanding of how the adverse outcome occurred, with sufficient depth and clarity that the patient and family can see how we reached that conclusion. In straightforward situations involving minor harm (eg, an allergic reaction to a medication that the clinician overlooked and that resulted in an urgent care center visit), a very limited investigation may clarify the situation sufficiently that the prescribing provider, accompanied by an office or staff nurse as support and witness, may be able to complete an effective disclosure in a single discussion, and simply writing off a bill or arranging to reimburse the urgent care center visit cost may satisfy the affected patient.

In more complex situations involving greater harm, a number of people must be involved to accomplish TEAM tasks: to offer an explanation, to answer questions, to make apologies, to explain changes intended to reduce the chance of harm to others in the future, and to work through any restitution that may be appropriate. Appointing a disclosure coach/coordinator/facilitator who has had extended training in the disclosure process can help guide these more complex situations. Risk management, insurance carriers, and legal counsel should be aware and advising throughout the process and participating directly in meetings with the patient and family, as appropriate. Since on the TEAM track we are admitting liability, offering a path to financial restitution may be warranted and the disclosure process may trigger reporting requirements with regulatory as well as human resource implications.

The patient and family may want to include other people on their “team” as well. Since complex disclosure meetings need to be carefully planned in advance, we should clarify who will be attending from the health care side and who the family intends to involve. We should anticipate potential requests and questions such as: Would it be OK to record this meeting? Can we ask our attorney to attend? Who are all these people and why are they in this meeting? (We should introduce all team members and clarify how their involvement is necessary to help reach the most satisfying resolution for all involved.)

Empathize. Admitting that deficient aspects in the care contributed to the harm will trigger thoughts, emotions, and expectations for the patient and family. Empathizing involves seeing the whole situation from their perspective and acknowledging their emotions as understandable. Empathizing is not the same as fully agreeing with the patient’s and family’s perspective, but we will not be able to effectively address concerns and expectations that we have not understood. Organizations should have supports in place for staff who are involved in these difficult situations. Nonetheless, we must prioritize the patient’s and family‘s feelings in a disclosure meeting.

Apologize and be Accountable. This calls for both expressions of sympathy as well as a genuine apology for having caused harm by failure in some aspect of care: We are very sorry you are going through this difficult situation. We are especially sorry to tell you that we now recognize that problems in the care we provided are the most likely cause of this harm. Would this be a good time to explain what we learned?

Having the responsible clinicians present increases the chances of achieving the most complete resolution in a single planned and well facilitated meeting. The tasks for that meeting include: offering an explanation that reveals the problems in care that contributed to the adverse outcome, making sincere apologies, and explaining changes to reduce chance of harm to others. The disclosure coach can work with individuals to help them understand how and why their involvement can be important and to help staff members become ready to participate constructively in the disclosure meeting. When individuals appear unable or unwilling to contribute constructively, a plan is needed for how their part can be replaced (eg, a charge nurse or department chair might need to step in to explain and apologize for the care of a subordinate). Managers/administrators can explain contributory factors for what may at first appear to have been simply individual negligence. Administrators can describe the actions that the organization is taking to correct problems that contributed to the patient harm: As nursing executive, it is my responsibility to see that all our staff have been adequately trained on the equipment we are asking them to use. We now recognize that the nurse’s lack of familiarity with that equipment contributed to the harm you experienced and I am very sorry for that. It is my responsibility to get that problem corrected, and we are already taking steps to assure that. Patients and families often have ideas for improving care processes and appreciate being invited to share these ideas as a service to future patients.

Manage until resolved. On the “care unreasonable” track, we must signal openness to helping with the patient’s and family’s immediate and longer-term needs, as well as their expectations about financial and other forms of restitution. Someone should be in the meeting who can describe the next steps in working towards a fair restitution and how that process will take place following the conclusion of the disclosure meeting. The close involvement of risk and claims professionals throughout the process of investigation through to the disclosure discussion itself will assure a more satisfactory handoff to questions about around financial compensation

Psychological Barriers to Implementation of Disclosure Pathways

Many organizations and researchers agree that disclosure and resolution pathways as just described are the most ethical and effective ways for all parties to resolve these painful situations. So why isn’t this approach universally practiced? In concluding this article, it may be helpful to point out some of the human dynamics that make resolution more difficult and how they might be addressed.

A key issue is the “urge to self-preservation.” Health care organizations have often been accused of disclosing only what they cannot hide. We have repeatedly observed how individuals and organizations are often initially motivated to do whatever is needed to protect themselves, even when those behaviors are frankly deceptive. This is almost to be expected. By age 4 children have learned to use deception as a defensive strategy when confronted with misbehavior. Research shows that children and adults continue the strategy to escape censure or punishment and simply get better at hiding their tracks.¹⁸ Because people want to preserve their image as ethical individuals, they have also learned to rationalize/justify this deception as necessary for self-preservation (“My dad would have killed me,” “I will lose my license,” “It is not fair that I take the blame when others have done the same thing and gotten away with it.”). Imagining the most extreme, and therefore “unfair” consequences, helps justify the individual’s use of dissembling and frank deception in order to avoid them. Clinicians and organizations may convince themselves that they are the victims entitled to protection rather than the injured patient. Patients and families often accept explanations that are less than candid, as doctors and nurses remain among the most trusted of professionals. Sufficiently understanding the complexities of the care is beyond the capability of most lay people. Successfully challenging the clinician’s or institution’s exculpatory explanation for an adverse outcome is very difficult, even though many clinicians believe that the tort system is stacked against them.

As a result, even the most sensible of best practices, toolkits, and trainings will not make full disclosure and fair resolution of adverse outcomes more likely without a counterweight of solid ethical commitment and a reliable structure for ensuring adherence. Sustainable progress has been demonstrated in those institutions^8,10,17 where: (1) institutional values and ethics around disclosure were elevated above self-protection, (2) efficient processes for recognizing and objectively reviewing care involving an adverse outcome were developed and followed, (3) salaried and institutionally insured staff and providers were required to participate in and accept a fair path to resolution in the context of a just culture, and (4) the institution was able to deliver on any commitments (eg, financial, corrective actions) it has made. Conversely, disclosure and resolution programs have struggled in the following situations: where values and ethics are not clarified and made primary; where the processes for reviewing adverse outcomes are slow, inconsistent, and open to political interference; where independent providers have latitude to insist on self-protective behaviors; and where liability carriers who place highest priority on avoiding financial exposure are involved.

Conclusion

The challenge of effectively disclosing and resolving adverse medical outcomes will continue to be most formidable for health care systems with independent medical staffs with separate liability carriers. Can these systems get a firm consensus on the ethics that are paramount in disclosure situations? Can they create care review systems that are efficient and objective and reach conclusions that are binding on those involved? Are they willing to provide explanations to patients and families regardless of the consequences to themselves? Can they coordinate an efficient path to financial and other forms of restitution in those situations where problems in the care contributed to the patient being harmed? And can they enforce these practices despite the self-concerns of all the involved parties? The good news is we now know how to disclose and resolve adverse medical outcomes with patients and families in a way that is fair to providers, staff, and institutions and will not break the bank. For health care organizations, implementing effective disclosure and resolution practices starts with a commitment to both build consensus for this process and consistently enforce it.

Corresponding author: Daniel O’Connell, PhD, 2212 Queen Anne Ave. N. #810, Seattle, WA 98109; [email protected].

Financial disclosures: None.

From The Communication in Healthcare Group, Seattle, WA.

Abstract

• Objective: To review established approaches to disclosure and resolution following adverse medical outcomes and highlight barriers that may hinder universal implementation of effective disclosure/resolution practices.
• Methods: An overview of established approaches to disclosure and resolution of adverse medical outcomes is presented.
• Results: Clinicians must be equipped to manage situations where adverse medical outcomes occur even though the care provided was reasonable, within the standard, as well as in situations where preventable problems in the care provided were likely the cause of patient harm. Established approaches that have proven useful for investigating, disclosing, and resolving situations, captured in the acronyms AIDR, ALEE, and TEAM, can assist clinicians in the disclosure and ultimate resolution of these 2 types of situations.
• Conclusion: Health care organizations with a solid commitment and a reliable structure for ensuring adherence to full disclosure and fair resolution of adverse outcomes have demonstrated sustainable progress in ethically and effectively resolving situations where patients are harmed by medical care.

Keywords: safety; medical error; adverse outcomes; resolution; communication.

Much has been learned over the 20 years since the Institute of Medicine’s (IOM) report To Err Is Human¹ was published. At the time it was published, the IOM report made it clear that only a minority of preventable patient harms were being acknowledged, investigated, and reported. In the face of adverse outcomes “dissemble, deny, and defend” was a common strategy of many clinicians, institutions, and liability carriers.² The health care system appeared to place a priority on protecting itself from reputational and financial harm over the rights of injured patients to be given an accurate understanding of what had happened in their care and to pursue restitution, if appropriate.^3-5

The emerging quality improvement movement was accompanied by calls for increased patient advocacy. This included the goal of greater transparency and more timely and equitable resolutions with patients who have been harmed by problems in care. Health care systems pressed for confidentiality protections in exchange for increased focus on quality improvement.⁶ Applying medical ethics of autonomy, no-maleficence, beneficence, and justice initially took a backseat, as risk management was given priority.⁷ Insurance carriers have no ethical obligation, and a clear disincentive, to assure that harmed patients are fully informed and offered restitution. Some self-insured health systems, however, began experimenting with more proactive and transparent approaches to disclosure and resolution. In contrast to the often-reported fear of a liability explosion, they reported reduced claims and suits, shorter time to resolution, and reduced overall financial cost,^8-10 providing some evidence that perhaps greater openness could work after all.

But for providers and staff to allow transparency and candor to become the norm, institutions needed to create a more “just culture” for managing errors. Individual impairment or willful disregard of safe practice would need to be handled differently from the slips and lapses that more often contributed to preventable harm.¹¹ For example, the nurse who was inadequately oriented to the equipment on an unfamiliar unit where she was asked to work a double shift due to a staffing shortage should not be held as accountable as an employee who knowingly violated agreed upon safe practices, even though patient harm resulted in each situation. It became clear that patient harm was usually the result of multiple factors involving individuals, communication, procedures, systems, and equipment. Blaming and disciplining individuals at the sharp end would not reliably reduce adverse outcomes.

Since the 1999 IOM report, we have developed general agreement on best practices for investigating, disclosing, and resolving situations where patients are harmed by medical care.^12,13 This article reviews the perspectives and practices that appear necessary for effective disclosure and resolution after an adverse outcome and highlights barriers to reliably enacting them in practice.

Elements of Effective Disclosure

Effective disclosure to patients and families hinges on determining and providing an accurate understanding of what happened in the patient’s care. It should be the care providers’ and their institution’s responsibility to determine causation and disclose it. This should not require only the most upset patients and families initiating a legal process taking 3 years or more to complete. The most consequential question must be answered, “Was the care provided reasonable?” That is, was everything done within the standard, as would have been expected by similarly trained clinicians with the information and resources available at that time? It follows that if care was reasonable, then the adverse outcome could not normally have been prevented, no correction in care processes is called for, and no financial compensation is required. If the care review reveals deficiencies in care that were linked to patient harm, then achieving a satisfying resolution would be more complex and difficult.¹³ First, individuals would have to accept that they have contributed to patient harm, itself an often-contentious process and psychologically devastating realization. Then they must have this difficult conversation with patient and family, creating liability risk for themselves in the process. They must commit to correcting the problems that contributed to the harm. They must facilitate, rather than obstruct, a path to a restitution that addresses the medical, practical, and financial harms that have resulted. Given the challenges inherent to disclosure and resolution, it is no wonder that dissembling, denying, and defending was the common practice for the preceding decades.¹⁴

Disclosure and Resolution Pathways

I was the co-developer of an approach to disclosure and resolution which is now widely accepted and that has been taught across the United States and Canada to more than 50,000 health care providers and administrators over 18 years.^15,16 We learned that resolving adverse medical outcomes is a 4-part process (anticipate, investigate, disclose, resolve [AIDR]). Most adverse or simply disappointing outcomes occur despite reasonable care (eg, due to biological variability, the imprecision of the science and limitations and risks of the procedures). The minority of harms are associated with deficiencies in the care (ie, unreasonable care). We need to equip ourselves to manage both situations effectively. The approach we developed can be captured in 3 acronyms: AIDR, ALEE, and TEAM,

AIDR

This acronym encapsulates the overview guidance for clinicians after an adverse event or outcome, regardless of the cause.

Anticipate the thoughts and feelings of the harmed/disappointed patient and family and reach out immediately with an expression of sympathy.

Investigate sufficiently to address questions about most likely causation and do not conjecture prior to investigation. Ask for patience—“You deserve more than a guess”—and keep in regular contact to reinforce the promise that there will be a full reporting when the review is complete.

Disclose (in a planned and coordinated manner) what has been learned in the investigation.

Resolve the situation with the patient and family consistent with our ethical principles.

If our failure caused the harm (care unreasonable/breached the standard), then working toward a fair restitution and taking corrective actions are appropriate. If the care was found to have been reasonable, then compensation would not be offered and corrective action is unwarranted. The organization would defend reasonable care if a claim was still pursued.

This process involves ethical clarity, emotional intelligence, and discipline. Clinicians must first acknowledge that a disappointing outcome or event has occurred. Clinicians involved in the care, usually led by the attending provider, then immediately reach out to the patient and family with sympathy, a plan of care to address the medical issues, and the promise to investigate and follow-up with the patient and family when the harm and its causes are more clearly determined. To disclose simply means to provide an accurate understanding (ie, the understanding determined by the investigation we conducted) of what happened, its causes, and consequences. Depending on the extent of the harm and the complexity and time needed for the investigation, a “coach” or “disclosure coordinator” who has advanced training in managing these situations is brought in to guide the process. The disclosure coach/coordinator provides a consistent and steady hand throughout the process of investigation, disclosure, and ultimately resolution with patient and family. Patients and families often move across settings during the time of the AIDR process, and it is easy to lose track of them unless someone is following the entire process until resolved.

ALEE

When the investigation of an adverse/disappointing outcome determines the care was reasonable and therefore the adverse outcome could not have been prevented, we use the ALEE pathway to guide the disclosure conversation (Step 3 in AIDR) with the patient and family:

Anticipate. What are the questions, thoughts, and feelings we would expect the patient and family will have? On this track, there is nothing to apologize for since the care was reasonable, yet expressing compassion and sympathy for the patient’s experience is essential. “I/we really sympathize with how differently this has turned out than we had hoped.”

Listen. Invite and listen for their questions and concerns, how they are seeing the situation, and where and what they are finding most upsetting and in need of explanation.

Empathize. There are 2 kinds of empathy required here. Cognitive empathy means showing that we understand their thinking from their perspective, separate from whether we fully agree. Emotional empathy involves demonstrating that their emotions are understandable given the situation, even if those emotions are painful for clinicians to experience. Listening in step 2 is how we learned their perspective and emotions. Now we can show accurate empathy: I/we can understand how upsetting it is to be facing another set of procedures to treat the unfortunate complications from your last surgery.

Explain. Even when care is reasonable, questions and perhaps suspicions are to be expected. Listening and empathizing sets us up to focus our explanations on the patient’s and family’s key questions with a level of thoughtfulness and transparency that conveys credibility. We should not assume, however, that they have accepted our explanation. Instead, solicit their reactions and unresolved questions as part of the disclosure discussion. It is normal for additional concerns to emerge in the days after the disclosure discussion, and we should be ready to address these concerns until resolved. In some instances, the patient and family will not be satisfied and it may be helpful to offer an independent review of the care. If the unresolved patient and family engages an attorney, that will be the first step taken anyway. Proactively offering an independent review signals confidence in your objectivity and sensitivity to the importance of fairness for the patient and family: Your questions and concerns are completely normal in light of the disappointing experience you have had. Let me see if I/we can address those now to your satisfaction.

TEAM

If the investigation determines that aspects of the care were unreasonable (breached the standard) and the adverse outcome/harm was related to the deficiencies in the care, then we use the TEAM pathway to disclose and resolve the situation with the patient and family

Truthful and Transparent and Teamwork. We should be offering our most accurate understanding of how the adverse outcome occurred, with sufficient depth and clarity that the patient and family can see how we reached that conclusion. In straightforward situations involving minor harm (eg, an allergic reaction to a medication that the clinician overlooked and that resulted in an urgent care center visit), a very limited investigation may clarify the situation sufficiently that the prescribing provider, accompanied by an office or staff nurse as support and witness, may be able to complete an effective disclosure in a single discussion, and simply writing off a bill or arranging to reimburse the urgent care center visit cost may satisfy the affected patient.

In more complex situations involving greater harm, a number of people must be involved to accomplish TEAM tasks: to offer an explanation, to answer questions, to make apologies, to explain changes intended to reduce the chance of harm to others in the future, and to work through any restitution that may be appropriate. Appointing a disclosure coach/coordinator/facilitator who has had extended training in the disclosure process can help guide these more complex situations. Risk management, insurance carriers, and legal counsel should be aware and advising throughout the process and participating directly in meetings with the patient and family, as appropriate. Since on the TEAM track we are admitting liability, offering a path to financial restitution may be warranted and the disclosure process may trigger reporting requirements with regulatory as well as human resource implications.

The patient and family may want to include other people on their “team” as well. Since complex disclosure meetings need to be carefully planned in advance, we should clarify who will be attending from the health care side and who the family intends to involve. We should anticipate potential requests and questions such as: Would it be OK to record this meeting? Can we ask our attorney to attend? Who are all these people and why are they in this meeting? (We should introduce all team members and clarify how their involvement is necessary to help reach the most satisfying resolution for all involved.)

Empathize. Admitting that deficient aspects in the care contributed to the harm will trigger thoughts, emotions, and expectations for the patient and family. Empathizing involves seeing the whole situation from their perspective and acknowledging their emotions as understandable. Empathizing is not the same as fully agreeing with the patient’s and family’s perspective, but we will not be able to effectively address concerns and expectations that we have not understood. Organizations should have supports in place for staff who are involved in these difficult situations. Nonetheless, we must prioritize the patient’s and family‘s feelings in a disclosure meeting.

Apologize and be Accountable. This calls for both expressions of sympathy as well as a genuine apology for having caused harm by failure in some aspect of care: We are very sorry you are going through this difficult situation. We are especially sorry to tell you that we now recognize that problems in the care we provided are the most likely cause of this harm. Would this be a good time to explain what we learned?

Having the responsible clinicians present increases the chances of achieving the most complete resolution in a single planned and well facilitated meeting. The tasks for that meeting include: offering an explanation that reveals the problems in care that contributed to the adverse outcome, making sincere apologies, and explaining changes to reduce chance of harm to others. The disclosure coach can work with individuals to help them understand how and why their involvement can be important and to help staff members become ready to participate constructively in the disclosure meeting. When individuals appear unable or unwilling to contribute constructively, a plan is needed for how their part can be replaced (eg, a charge nurse or department chair might need to step in to explain and apologize for the care of a subordinate). Managers/administrators can explain contributory factors for what may at first appear to have been simply individual negligence. Administrators can describe the actions that the organization is taking to correct problems that contributed to the patient harm: As nursing executive, it is my responsibility to see that all our staff have been adequately trained on the equipment we are asking them to use. We now recognize that the nurse’s lack of familiarity with that equipment contributed to the harm you experienced and I am very sorry for that. It is my responsibility to get that problem corrected, and we are already taking steps to assure that. Patients and families often have ideas for improving care processes and appreciate being invited to share these ideas as a service to future patients.

Manage until resolved. On the “care unreasonable” track, we must signal openness to helping with the patient’s and family’s immediate and longer-term needs, as well as their expectations about financial and other forms of restitution. Someone should be in the meeting who can describe the next steps in working towards a fair restitution and how that process will take place following the conclusion of the disclosure meeting. The close involvement of risk and claims professionals throughout the process of investigation through to the disclosure discussion itself will assure a more satisfactory handoff to questions about around financial compensation

Psychological Barriers to Implementation of Disclosure Pathways

Many organizations and researchers agree that disclosure and resolution pathways as just described are the most ethical and effective ways for all parties to resolve these painful situations. So why isn’t this approach universally practiced? In concluding this article, it may be helpful to point out some of the human dynamics that make resolution more difficult and how they might be addressed.

A key issue is the “urge to self-preservation.” Health care organizations have often been accused of disclosing only what they cannot hide. We have repeatedly observed how individuals and organizations are often initially motivated to do whatever is needed to protect themselves, even when those behaviors are frankly deceptive. This is almost to be expected. By age 4 children have learned to use deception as a defensive strategy when confronted with misbehavior. Research shows that children and adults continue the strategy to escape censure or punishment and simply get better at hiding their tracks.¹⁸ Because people want to preserve their image as ethical individuals, they have also learned to rationalize/justify this deception as necessary for self-preservation (“My dad would have killed me,” “I will lose my license,” “It is not fair that I take the blame when others have done the same thing and gotten away with it.”). Imagining the most extreme, and therefore “unfair” consequences, helps justify the individual’s use of dissembling and frank deception in order to avoid them. Clinicians and organizations may convince themselves that they are the victims entitled to protection rather than the injured patient. Patients and families often accept explanations that are less than candid, as doctors and nurses remain among the most trusted of professionals. Sufficiently understanding the complexities of the care is beyond the capability of most lay people. Successfully challenging the clinician’s or institution’s exculpatory explanation for an adverse outcome is very difficult, even though many clinicians believe that the tort system is stacked against them.

As a result, even the most sensible of best practices, toolkits, and trainings will not make full disclosure and fair resolution of adverse outcomes more likely without a counterweight of solid ethical commitment and a reliable structure for ensuring adherence. Sustainable progress has been demonstrated in those institutions^8,10,17 where: (1) institutional values and ethics around disclosure were elevated above self-protection, (2) efficient processes for recognizing and objectively reviewing care involving an adverse outcome were developed and followed, (3) salaried and institutionally insured staff and providers were required to participate in and accept a fair path to resolution in the context of a just culture, and (4) the institution was able to deliver on any commitments (eg, financial, corrective actions) it has made. Conversely, disclosure and resolution programs have struggled in the following situations: where values and ethics are not clarified and made primary; where the processes for reviewing adverse outcomes are slow, inconsistent, and open to political interference; where independent providers have latitude to insist on self-protective behaviors; and where liability carriers who place highest priority on avoiding financial exposure are involved.

Conclusion

The challenge of effectively disclosing and resolving adverse medical outcomes will continue to be most formidable for health care systems with independent medical staffs with separate liability carriers. Can these systems get a firm consensus on the ethics that are paramount in disclosure situations? Can they create care review systems that are efficient and objective and reach conclusions that are binding on those involved? Are they willing to provide explanations to patients and families regardless of the consequences to themselves? Can they coordinate an efficient path to financial and other forms of restitution in those situations where problems in the care contributed to the patient being harmed? And can they enforce these practices despite the self-concerns of all the involved parties? The good news is we now know how to disclose and resolve adverse medical outcomes with patients and families in a way that is fair to providers, staff, and institutions and will not break the bank. For health care organizations, implementing effective disclosure and resolution practices starts with a commitment to both build consensus for this process and consistently enforce it.

Corresponding author: Daniel O’Connell, PhD, 2212 Queen Anne Ave. N. #810, Seattle, WA 98109; [email protected].

Financial disclosures: None.