User login
Posttraumatic Stress Treatment That Suits the Symptoms
Effectiveness of Rosiglitazone and Pioglitazone to Reduce Hemoglobin A1c Levels in Veteran Patients With Type 2 Diabetes
Smoking Prevalence and Nicotine Patch Success Rate Within a VA Medical Center
Investigating Infection in the Patient With RA
In 2009, a question of proper practices was brought before the Arizona Regulatory Board of Physician Assistants. The patient in question was seen in an emergency department by a PA/physician health care team for complaints of cough and fever at home. In question was the team’s decision to discharge the patient, who had a history of immunosuppression with an anti–tumor necrosis factor (anti-TNF) medication for a chronic joint disorder, without ordering laboratory tests or blood cultures. Although there is textbook instruction for evaluation of patients with infection who use agents in this drug class,1 many clinicians may be unaware that blood cultures should be drawn as part of a proper workup.
In the United States, medications that have anti-TNF-alpha (anti-TNF-a) properties, including adalimumab, etanercept, and infliximab, are approved by the FDA to treat rheumatoid arthritis (RA).2 Many patients take a second medication; others, a combination of medications. Options include methotrexate, cyclosporine, prednisone, and hydroxychloroquine. Typically, anti-TNF medications represent second- or third-line therapy after unsatisfactory response to disease-modifying antirheumatic drugs (DMARDs).
Although anti-TNF medications are effective for treatment of autoimmune diseases like RA, it should be noted that TNF-a is an important component in the proper functioning of the immune system. Accordingly, it is recognized that patients who use anti-TNF medications are at increased risk for infection.3 Indeed, patients with RA are already at greater risk for contracting infectious illnesses, even before immunosuppressive medications are administered.4
TNF is one cytokine that is responsible for humoral activation and subsequent inflammation.5 In RA and certain other diseases (eg, Crohn’s disease, lupus), TNF-a’s interaction with other cytokines is at least partially responsible for the inflammation that leads to RA-associated joint damage.5 Infliximab and adalimumab are monoclonal antibodies developed specifically to bind with TNF-a, thereby neutralizing a pathway for further release of inflammatory cytokines.
In contrast to recombinant soluble receptors, such as etanercept, the site-specific monoclonal antibodies for TNF-a (ie, infliximab, adalimumab) may play a greater role in suppressing macrophages, which contribute significantly to cellular immunity. Macrophage inhibition may further increase the body’s vulnerability to opportunistic infection.6
Anti-TNF Agents: Clinical Trials
The relationship between treatment with immunosuppressive agents (including infliximab and other anti-TNF medications) and serious infection is evident in the literature. In a 2005 study, Listing et al4 followed patients for 12 months after they were started on etanercept, infliximab, anakinra (another cytokine inhibitor), or DMARDs (controls). Ninety-two (26.5%) of the 346 patients who were given infliximab, compared with 6% of 601 controls, developed infection; of these, 20 patients (21.7%) had a determinable serious infection (ie, that was considered life-threatening or resulted in inpatient treatment, significant disability, or death). Thus, serious infection affected 5.8% of the patient population studied who had received infliximab.
Five of the 512 patients who used etanercept were diagnosed with sepsis, but no episodes of bacteremia were recorded among the infliximab patients. The most prevalent serious infections observed were pneumonia, infective arthritis, and various skin and subcutaneous infections (see table4). Less prevalent respiratory tract ailments included bronchitis, lung abscess, and pleural infection. Overall, small numbers of cases of sepsis were reported, making it difficult for the study authors to comment on the disparity of sepsis rates among treatment groups.4
Another prospective study followed 28 patients (median age, 53) who were treated with infliximab between 1999 and 2002.7 Dosages administered ranged from the standard 3 mg/kg every 8 weeks to 10 mg/kg every 4 weeks, with variation in duration of treatment. A 13% higher rate of infection was reported among these patients, compared with patients who received conventional therapy (DMARDs) over a two-year period.
Six patients were determined to have serious, life-threatening infections (ie, requiring hospitalization or IV antibiotic therapy). Of these, two were given a diagnosis of bacteremia via blood cultures. The cultures grew Streptococcus pneumoniae in one patient, who was also diagnosed incidentally with Pneumocystis carinii through bronchial washings, and the other patient had blood cultures positive for Staphylococcus aureus and bilateral necrotizing pulmonary abscesses.
Case Studies
Individual case studies of infection in patients taking infliximab or other anti-TNF agents for RA reveal specific illnesses, their time lines, and relevant patient management. Crum et al2 reported an example of the common pathogen S pneumoniae in a 47-year-old patient who developed a left lower lobe pneumonia. Three of four blood cultures were positive, CBC with differential showed abnormal results, and the patient’s temperature on presentation was 39.5°C.
In a report of two cases of severe abdominal infection during administration of anti-TNF medication, one 40-year-old woman using infliximab for severe RA presented with a temperature of 39.2°C and a three-day history of abdominal pain. Escherichia coli was identified from blood cultures, although urine cultures were negative. CT led to a diagnosis of pyelonephritis. The patient had become septic but improved after 48 hours’ therapy with IV cefuroxime and gentamicin.8
The second patient was a man, age 60, with a two-week history of abdominal pain and a six-month history of effective etanercept use for persistent psoriatic arthritis. Contrast-enhanced CT revealed a large splenic abscess, and blood cultures identified infection with S aureus. Management began with unsuccessful high-dose IV antibiotics, followed by laparotomy and splenectomy, postsurgical sepsis, and finally, high-dose inotropic therapy that proved effective.8
Infection With Listeria monocytogenes
Listeria monocytogenes, a pathogen sometimes identified in patients who receive anti-TNF-a therapy, has been the focus of several case reports. For example, Gluck et al9 describe a 60-year-old woman with a history of RA and previous treatment with multiple therapies who had recently completed a five-month regimen with infliximab (10 mg/kg every 4 weeks). Fourteen days after receiving her last dose, she presented to a hospital with anemia, fever (measurement undisclosed), and abdominal pain. She was given multiple diagnoses, including a gastric ulcer without hemorrhage, a pulmonary infiltrate, and acute cholecystitis. After she underwent cholecystectomy, her condition deteriorated, and head CT revealed several small subarachnoid hemorrhages and severe brain edema.
The patient experienced multi–organ system failure and died after a brain-stem herniation. What should be noted is that a blood culture taken one day after her surgery was positive for L monocytogenes, as was an intraoperative gallbladder swab-culture for the same organism—this, despite presumed antibiotic therapy.
A second patient described by Gluck et al9 was a 62-year-old woman with RA who developed cholecystitis after her second dose of infliximab, administered over a two-week period (200 mg IV/dose). As in the previous case, the patient manifested cerebral edema and had a fever after undergoing cholecystectomy. Blood cultures performed after the surgery were positive for L monocytogenes. The patient recovered slowly with proper antibiotic therapy, including ampicillin and gentamicin.
A third example of listeriosis involved a 52-year-old patient with RA who presented with symptoms including fever and abdominal pain. CT led to a diagnosis of terminal ileitis. One of two blood cultures taken was positive for L monocytogenes.6
Finally, a 79-year-old patient with a long-standing diagnosis of RA and prosthetic hips reported a fever of three weeks’ duration and left leg pain. As the patient had no fever on initial presentation, blood cultures were not drawn until five days after he was admitted. Three of four blood cultures were positive for L monocytogenes.6
Less Common Infections
Other less common bacterial etiologies have also been reported in the literature, including a fatal case of Salmonella enteritidis septicemia with a pleural empyema after treatment with infliximab.10 Another fatality was reported in a 54-year-old man who had necrotizing fasciitis, with a group A hemolytic streptococcus that grew in blood cultures. The patient, who had been taking infliximab, had no report of fever before or during his hospitalization.11
Discussion
Although infliximab is effective for controlling the symptoms of RA, clinicians must be attentive to fever in patients who use it—even a patient-reported fever that may not be present on physical exam. Due to the potential blunting of the immune response associated with infliximab use, a serious underlying infection can occur without fever. General malaise may be the only symptom a patient reports.
Infliximab and other anti-TNF medications are included in a textbook list of immunosuppressive medications whose use warrants special attention for patients who are evaluated for infection in an emergency setting. According to Burns,1 “Rapid diagnosis and early initiation of therapy are essential in preventing serious morbidity and mortality in immunocompromised patients who often have subtle or unusual presentations and are difficult to diagnose.”1
Active tuberculosis (ATB) is another serious illness that has been reported with infliximab use.12,13 Patients should be screened for TB before taking infliximab, but clinicians must also be aware that ATB can develop as an unwanted result of infliximab therapy. Standard blood cultures may not be an effective means of identifying ATB.
Several types of infection may occur, with or without confirmation by positive blood cultures, including viral and fungal infections. Even bacterial infections, such as a urinary tract infection or bacterial pneumonia, may not yield positive blood cultures immediately, if ever. Each of these has been included among serious illnesses associated with use of anti-TNF medications.4 Nevertheless, blood cultures—Gram-stain, anaerobic, and aerobic blood cultures with sensitivities—should be considered an important component of any comprehensive infectious disease evaluation.
Although the purpose of this article has been to emphasize the importance of blood cultures in the timely diagnosis and treatment of patients taking anti-TNF medications who present with symptoms of infectious disease, it is important to remember that blood cultures are only part of a full infectious disease workup.
Conclusion
The clinical studies and case presentations discussed here serve to remind clinicians that timely acquisition and evaluation of blood cultures are essential to the identification of serious bacteriologic pathogens, and to the chances of recovery in the RA patient with a systemic infection who is immunosuppressed as a result of treatment with infliximab or other anti-TNF agent. Rapid identification of pathogenic organisms will facilitate the implementation of appropriate therapy, thus avoiding the often-deadly outcomes of severe infection and sepsis. Gram-stain, anaerobic, and aerobic blood cultures with sensitivities should be considered the standard of care in any infliximab-treated patient who presents with fever, a patient report of a fever, or any other signs, symptoms, or patient history that could suggest infection.
While this article focuses for the most part on infections associated with infliximab use, the same strategy in standard of practice should apply to all patients who use anti-TNF therapy for treatment of RA.
1. Burns MJ. Chapter 181. Other immunosuppressive medications. In: Marx J, Hockberger R, Walls R, et al, eds. Rosen’s Emergency Medicine: Concepts and Clinical Practice. 7th ed. Philadelphia, PA: Mosby Elsevier; 2009.
2. Crum NF, Lederman ER, Wallace MR. Infections associated with tumor necrosis factor-alpha antagonists. Medicine (Baltimore). 2005; 84(5)291-302.
3. Geraghty EM, Ristow B, Gordon SM, Aronowitz P. Overwhelming parasitemia with Plasmodium falciparum infection in a patient receiving infliximab therapy for rheumatoid arthritis. Clin Infect Dis. 2007;44(10):82-84.
4. Listing J, Strangfeld A, Kary S, et al. Infections in patients with rheumatoid arthritis treated with biologic agents. Arthritis Rheum. 2005;52(11):3403-3412.
5. Louie SG, Park B, Yoon H. Biological response modifiers in the management of rheumatoid arthritis. Am J Health Syst Pharm. 2003;60(4): 346-355.
6. Kesteman T, Yombi JC, Gigi J, Durez P. Listeria infections associated with infliximab: case reports. Clin Rheumatol. 2007;26(12):2173-2175.
7. Kroesen S, Widmer AF, Tyndall A, Hasler P. Serious bacterial infections in patients with rheumatoid arthritis under anti-TNF-alpha therapy. Rheumatology. 2003;42(5):617-621.
8. Goode S, Tierney G, Deighton C. Life threatening intra-abdominal sepsis in patients on anti-TNF-alpha therapy. Gut. 2005;54(4):590-591.
9. Glück T, Linde HJ, Schölmerich J, et al. Anti-tumor necrosis factor therapy and Listeria monocytogenes infection: report of two cases. Arthritis Rheum. 2002;46(8):2255-2257.
10. Rijkeboer A, Voskuyl A, Van Agtmael M. Fatal Salmonella enteritidis septicaemia in a rheumatoid arthritis patient treated with a TNF-alpha antagonist. Scand J Infect Dis. 2007; 39(1):80-83.
11. Chan AT, Cleeve V, Daymond TJ. Necrotising fasciitis in a patient receiving infliximab for rheumatoid arthritis. Postgrad Med J. 2002; 78(915):47-48.
12. Favalli EG, Desiati F, Atzeni F, et al. Serious infections during anti-TNF-alpha treatment in rheumatoid arthritis patients. Autoimmun Rev. 2009;8(3):266-273.
13. Keane J, Gershon S, Wise RP, et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med. 2001;345(15):1098-1104.
In 2009, a question of proper practices was brought before the Arizona Regulatory Board of Physician Assistants. The patient in question was seen in an emergency department by a PA/physician health care team for complaints of cough and fever at home. In question was the team’s decision to discharge the patient, who had a history of immunosuppression with an anti–tumor necrosis factor (anti-TNF) medication for a chronic joint disorder, without ordering laboratory tests or blood cultures. Although there is textbook instruction for evaluation of patients with infection who use agents in this drug class,1 many clinicians may be unaware that blood cultures should be drawn as part of a proper workup.
In the United States, medications that have anti-TNF-alpha (anti-TNF-a) properties, including adalimumab, etanercept, and infliximab, are approved by the FDA to treat rheumatoid arthritis (RA).2 Many patients take a second medication; others, a combination of medications. Options include methotrexate, cyclosporine, prednisone, and hydroxychloroquine. Typically, anti-TNF medications represent second- or third-line therapy after unsatisfactory response to disease-modifying antirheumatic drugs (DMARDs).
Although anti-TNF medications are effective for treatment of autoimmune diseases like RA, it should be noted that TNF-a is an important component in the proper functioning of the immune system. Accordingly, it is recognized that patients who use anti-TNF medications are at increased risk for infection.3 Indeed, patients with RA are already at greater risk for contracting infectious illnesses, even before immunosuppressive medications are administered.4
TNF is one cytokine that is responsible for humoral activation and subsequent inflammation.5 In RA and certain other diseases (eg, Crohn’s disease, lupus), TNF-a’s interaction with other cytokines is at least partially responsible for the inflammation that leads to RA-associated joint damage.5 Infliximab and adalimumab are monoclonal antibodies developed specifically to bind with TNF-a, thereby neutralizing a pathway for further release of inflammatory cytokines.
In contrast to recombinant soluble receptors, such as etanercept, the site-specific monoclonal antibodies for TNF-a (ie, infliximab, adalimumab) may play a greater role in suppressing macrophages, which contribute significantly to cellular immunity. Macrophage inhibition may further increase the body’s vulnerability to opportunistic infection.6
Anti-TNF Agents: Clinical Trials
The relationship between treatment with immunosuppressive agents (including infliximab and other anti-TNF medications) and serious infection is evident in the literature. In a 2005 study, Listing et al4 followed patients for 12 months after they were started on etanercept, infliximab, anakinra (another cytokine inhibitor), or DMARDs (controls). Ninety-two (26.5%) of the 346 patients who were given infliximab, compared with 6% of 601 controls, developed infection; of these, 20 patients (21.7%) had a determinable serious infection (ie, that was considered life-threatening or resulted in inpatient treatment, significant disability, or death). Thus, serious infection affected 5.8% of the patient population studied who had received infliximab.
Five of the 512 patients who used etanercept were diagnosed with sepsis, but no episodes of bacteremia were recorded among the infliximab patients. The most prevalent serious infections observed were pneumonia, infective arthritis, and various skin and subcutaneous infections (see table4). Less prevalent respiratory tract ailments included bronchitis, lung abscess, and pleural infection. Overall, small numbers of cases of sepsis were reported, making it difficult for the study authors to comment on the disparity of sepsis rates among treatment groups.4
Another prospective study followed 28 patients (median age, 53) who were treated with infliximab between 1999 and 2002.7 Dosages administered ranged from the standard 3 mg/kg every 8 weeks to 10 mg/kg every 4 weeks, with variation in duration of treatment. A 13% higher rate of infection was reported among these patients, compared with patients who received conventional therapy (DMARDs) over a two-year period.
Six patients were determined to have serious, life-threatening infections (ie, requiring hospitalization or IV antibiotic therapy). Of these, two were given a diagnosis of bacteremia via blood cultures. The cultures grew Streptococcus pneumoniae in one patient, who was also diagnosed incidentally with Pneumocystis carinii through bronchial washings, and the other patient had blood cultures positive for Staphylococcus aureus and bilateral necrotizing pulmonary abscesses.
Case Studies
Individual case studies of infection in patients taking infliximab or other anti-TNF agents for RA reveal specific illnesses, their time lines, and relevant patient management. Crum et al2 reported an example of the common pathogen S pneumoniae in a 47-year-old patient who developed a left lower lobe pneumonia. Three of four blood cultures were positive, CBC with differential showed abnormal results, and the patient’s temperature on presentation was 39.5°C.
In a report of two cases of severe abdominal infection during administration of anti-TNF medication, one 40-year-old woman using infliximab for severe RA presented with a temperature of 39.2°C and a three-day history of abdominal pain. Escherichia coli was identified from blood cultures, although urine cultures were negative. CT led to a diagnosis of pyelonephritis. The patient had become septic but improved after 48 hours’ therapy with IV cefuroxime and gentamicin.8
The second patient was a man, age 60, with a two-week history of abdominal pain and a six-month history of effective etanercept use for persistent psoriatic arthritis. Contrast-enhanced CT revealed a large splenic abscess, and blood cultures identified infection with S aureus. Management began with unsuccessful high-dose IV antibiotics, followed by laparotomy and splenectomy, postsurgical sepsis, and finally, high-dose inotropic therapy that proved effective.8
Infection With Listeria monocytogenes
Listeria monocytogenes, a pathogen sometimes identified in patients who receive anti-TNF-a therapy, has been the focus of several case reports. For example, Gluck et al9 describe a 60-year-old woman with a history of RA and previous treatment with multiple therapies who had recently completed a five-month regimen with infliximab (10 mg/kg every 4 weeks). Fourteen days after receiving her last dose, she presented to a hospital with anemia, fever (measurement undisclosed), and abdominal pain. She was given multiple diagnoses, including a gastric ulcer without hemorrhage, a pulmonary infiltrate, and acute cholecystitis. After she underwent cholecystectomy, her condition deteriorated, and head CT revealed several small subarachnoid hemorrhages and severe brain edema.
The patient experienced multi–organ system failure and died after a brain-stem herniation. What should be noted is that a blood culture taken one day after her surgery was positive for L monocytogenes, as was an intraoperative gallbladder swab-culture for the same organism—this, despite presumed antibiotic therapy.
A second patient described by Gluck et al9 was a 62-year-old woman with RA who developed cholecystitis after her second dose of infliximab, administered over a two-week period (200 mg IV/dose). As in the previous case, the patient manifested cerebral edema and had a fever after undergoing cholecystectomy. Blood cultures performed after the surgery were positive for L monocytogenes. The patient recovered slowly with proper antibiotic therapy, including ampicillin and gentamicin.
A third example of listeriosis involved a 52-year-old patient with RA who presented with symptoms including fever and abdominal pain. CT led to a diagnosis of terminal ileitis. One of two blood cultures taken was positive for L monocytogenes.6
Finally, a 79-year-old patient with a long-standing diagnosis of RA and prosthetic hips reported a fever of three weeks’ duration and left leg pain. As the patient had no fever on initial presentation, blood cultures were not drawn until five days after he was admitted. Three of four blood cultures were positive for L monocytogenes.6
Less Common Infections
Other less common bacterial etiologies have also been reported in the literature, including a fatal case of Salmonella enteritidis septicemia with a pleural empyema after treatment with infliximab.10 Another fatality was reported in a 54-year-old man who had necrotizing fasciitis, with a group A hemolytic streptococcus that grew in blood cultures. The patient, who had been taking infliximab, had no report of fever before or during his hospitalization.11
Discussion
Although infliximab is effective for controlling the symptoms of RA, clinicians must be attentive to fever in patients who use it—even a patient-reported fever that may not be present on physical exam. Due to the potential blunting of the immune response associated with infliximab use, a serious underlying infection can occur without fever. General malaise may be the only symptom a patient reports.
Infliximab and other anti-TNF medications are included in a textbook list of immunosuppressive medications whose use warrants special attention for patients who are evaluated for infection in an emergency setting. According to Burns,1 “Rapid diagnosis and early initiation of therapy are essential in preventing serious morbidity and mortality in immunocompromised patients who often have subtle or unusual presentations and are difficult to diagnose.”1
Active tuberculosis (ATB) is another serious illness that has been reported with infliximab use.12,13 Patients should be screened for TB before taking infliximab, but clinicians must also be aware that ATB can develop as an unwanted result of infliximab therapy. Standard blood cultures may not be an effective means of identifying ATB.
Several types of infection may occur, with or without confirmation by positive blood cultures, including viral and fungal infections. Even bacterial infections, such as a urinary tract infection or bacterial pneumonia, may not yield positive blood cultures immediately, if ever. Each of these has been included among serious illnesses associated with use of anti-TNF medications.4 Nevertheless, blood cultures—Gram-stain, anaerobic, and aerobic blood cultures with sensitivities—should be considered an important component of any comprehensive infectious disease evaluation.
Although the purpose of this article has been to emphasize the importance of blood cultures in the timely diagnosis and treatment of patients taking anti-TNF medications who present with symptoms of infectious disease, it is important to remember that blood cultures are only part of a full infectious disease workup.
Conclusion
The clinical studies and case presentations discussed here serve to remind clinicians that timely acquisition and evaluation of blood cultures are essential to the identification of serious bacteriologic pathogens, and to the chances of recovery in the RA patient with a systemic infection who is immunosuppressed as a result of treatment with infliximab or other anti-TNF agent. Rapid identification of pathogenic organisms will facilitate the implementation of appropriate therapy, thus avoiding the often-deadly outcomes of severe infection and sepsis. Gram-stain, anaerobic, and aerobic blood cultures with sensitivities should be considered the standard of care in any infliximab-treated patient who presents with fever, a patient report of a fever, or any other signs, symptoms, or patient history that could suggest infection.
While this article focuses for the most part on infections associated with infliximab use, the same strategy in standard of practice should apply to all patients who use anti-TNF therapy for treatment of RA.
In 2009, a question of proper practices was brought before the Arizona Regulatory Board of Physician Assistants. The patient in question was seen in an emergency department by a PA/physician health care team for complaints of cough and fever at home. In question was the team’s decision to discharge the patient, who had a history of immunosuppression with an anti–tumor necrosis factor (anti-TNF) medication for a chronic joint disorder, without ordering laboratory tests or blood cultures. Although there is textbook instruction for evaluation of patients with infection who use agents in this drug class,1 many clinicians may be unaware that blood cultures should be drawn as part of a proper workup.
In the United States, medications that have anti-TNF-alpha (anti-TNF-a) properties, including adalimumab, etanercept, and infliximab, are approved by the FDA to treat rheumatoid arthritis (RA).2 Many patients take a second medication; others, a combination of medications. Options include methotrexate, cyclosporine, prednisone, and hydroxychloroquine. Typically, anti-TNF medications represent second- or third-line therapy after unsatisfactory response to disease-modifying antirheumatic drugs (DMARDs).
Although anti-TNF medications are effective for treatment of autoimmune diseases like RA, it should be noted that TNF-a is an important component in the proper functioning of the immune system. Accordingly, it is recognized that patients who use anti-TNF medications are at increased risk for infection.3 Indeed, patients with RA are already at greater risk for contracting infectious illnesses, even before immunosuppressive medications are administered.4
TNF is one cytokine that is responsible for humoral activation and subsequent inflammation.5 In RA and certain other diseases (eg, Crohn’s disease, lupus), TNF-a’s interaction with other cytokines is at least partially responsible for the inflammation that leads to RA-associated joint damage.5 Infliximab and adalimumab are monoclonal antibodies developed specifically to bind with TNF-a, thereby neutralizing a pathway for further release of inflammatory cytokines.
In contrast to recombinant soluble receptors, such as etanercept, the site-specific monoclonal antibodies for TNF-a (ie, infliximab, adalimumab) may play a greater role in suppressing macrophages, which contribute significantly to cellular immunity. Macrophage inhibition may further increase the body’s vulnerability to opportunistic infection.6
Anti-TNF Agents: Clinical Trials
The relationship between treatment with immunosuppressive agents (including infliximab and other anti-TNF medications) and serious infection is evident in the literature. In a 2005 study, Listing et al4 followed patients for 12 months after they were started on etanercept, infliximab, anakinra (another cytokine inhibitor), or DMARDs (controls). Ninety-two (26.5%) of the 346 patients who were given infliximab, compared with 6% of 601 controls, developed infection; of these, 20 patients (21.7%) had a determinable serious infection (ie, that was considered life-threatening or resulted in inpatient treatment, significant disability, or death). Thus, serious infection affected 5.8% of the patient population studied who had received infliximab.
Five of the 512 patients who used etanercept were diagnosed with sepsis, but no episodes of bacteremia were recorded among the infliximab patients. The most prevalent serious infections observed were pneumonia, infective arthritis, and various skin and subcutaneous infections (see table4). Less prevalent respiratory tract ailments included bronchitis, lung abscess, and pleural infection. Overall, small numbers of cases of sepsis were reported, making it difficult for the study authors to comment on the disparity of sepsis rates among treatment groups.4
Another prospective study followed 28 patients (median age, 53) who were treated with infliximab between 1999 and 2002.7 Dosages administered ranged from the standard 3 mg/kg every 8 weeks to 10 mg/kg every 4 weeks, with variation in duration of treatment. A 13% higher rate of infection was reported among these patients, compared with patients who received conventional therapy (DMARDs) over a two-year period.
Six patients were determined to have serious, life-threatening infections (ie, requiring hospitalization or IV antibiotic therapy). Of these, two were given a diagnosis of bacteremia via blood cultures. The cultures grew Streptococcus pneumoniae in one patient, who was also diagnosed incidentally with Pneumocystis carinii through bronchial washings, and the other patient had blood cultures positive for Staphylococcus aureus and bilateral necrotizing pulmonary abscesses.
Case Studies
Individual case studies of infection in patients taking infliximab or other anti-TNF agents for RA reveal specific illnesses, their time lines, and relevant patient management. Crum et al2 reported an example of the common pathogen S pneumoniae in a 47-year-old patient who developed a left lower lobe pneumonia. Three of four blood cultures were positive, CBC with differential showed abnormal results, and the patient’s temperature on presentation was 39.5°C.
In a report of two cases of severe abdominal infection during administration of anti-TNF medication, one 40-year-old woman using infliximab for severe RA presented with a temperature of 39.2°C and a three-day history of abdominal pain. Escherichia coli was identified from blood cultures, although urine cultures were negative. CT led to a diagnosis of pyelonephritis. The patient had become septic but improved after 48 hours’ therapy with IV cefuroxime and gentamicin.8
The second patient was a man, age 60, with a two-week history of abdominal pain and a six-month history of effective etanercept use for persistent psoriatic arthritis. Contrast-enhanced CT revealed a large splenic abscess, and blood cultures identified infection with S aureus. Management began with unsuccessful high-dose IV antibiotics, followed by laparotomy and splenectomy, postsurgical sepsis, and finally, high-dose inotropic therapy that proved effective.8
Infection With Listeria monocytogenes
Listeria monocytogenes, a pathogen sometimes identified in patients who receive anti-TNF-a therapy, has been the focus of several case reports. For example, Gluck et al9 describe a 60-year-old woman with a history of RA and previous treatment with multiple therapies who had recently completed a five-month regimen with infliximab (10 mg/kg every 4 weeks). Fourteen days after receiving her last dose, she presented to a hospital with anemia, fever (measurement undisclosed), and abdominal pain. She was given multiple diagnoses, including a gastric ulcer without hemorrhage, a pulmonary infiltrate, and acute cholecystitis. After she underwent cholecystectomy, her condition deteriorated, and head CT revealed several small subarachnoid hemorrhages and severe brain edema.
The patient experienced multi–organ system failure and died after a brain-stem herniation. What should be noted is that a blood culture taken one day after her surgery was positive for L monocytogenes, as was an intraoperative gallbladder swab-culture for the same organism—this, despite presumed antibiotic therapy.
A second patient described by Gluck et al9 was a 62-year-old woman with RA who developed cholecystitis after her second dose of infliximab, administered over a two-week period (200 mg IV/dose). As in the previous case, the patient manifested cerebral edema and had a fever after undergoing cholecystectomy. Blood cultures performed after the surgery were positive for L monocytogenes. The patient recovered slowly with proper antibiotic therapy, including ampicillin and gentamicin.
A third example of listeriosis involved a 52-year-old patient with RA who presented with symptoms including fever and abdominal pain. CT led to a diagnosis of terminal ileitis. One of two blood cultures taken was positive for L monocytogenes.6
Finally, a 79-year-old patient with a long-standing diagnosis of RA and prosthetic hips reported a fever of three weeks’ duration and left leg pain. As the patient had no fever on initial presentation, blood cultures were not drawn until five days after he was admitted. Three of four blood cultures were positive for L monocytogenes.6
Less Common Infections
Other less common bacterial etiologies have also been reported in the literature, including a fatal case of Salmonella enteritidis septicemia with a pleural empyema after treatment with infliximab.10 Another fatality was reported in a 54-year-old man who had necrotizing fasciitis, with a group A hemolytic streptococcus that grew in blood cultures. The patient, who had been taking infliximab, had no report of fever before or during his hospitalization.11
Discussion
Although infliximab is effective for controlling the symptoms of RA, clinicians must be attentive to fever in patients who use it—even a patient-reported fever that may not be present on physical exam. Due to the potential blunting of the immune response associated with infliximab use, a serious underlying infection can occur without fever. General malaise may be the only symptom a patient reports.
Infliximab and other anti-TNF medications are included in a textbook list of immunosuppressive medications whose use warrants special attention for patients who are evaluated for infection in an emergency setting. According to Burns,1 “Rapid diagnosis and early initiation of therapy are essential in preventing serious morbidity and mortality in immunocompromised patients who often have subtle or unusual presentations and are difficult to diagnose.”1
Active tuberculosis (ATB) is another serious illness that has been reported with infliximab use.12,13 Patients should be screened for TB before taking infliximab, but clinicians must also be aware that ATB can develop as an unwanted result of infliximab therapy. Standard blood cultures may not be an effective means of identifying ATB.
Several types of infection may occur, with or without confirmation by positive blood cultures, including viral and fungal infections. Even bacterial infections, such as a urinary tract infection or bacterial pneumonia, may not yield positive blood cultures immediately, if ever. Each of these has been included among serious illnesses associated with use of anti-TNF medications.4 Nevertheless, blood cultures—Gram-stain, anaerobic, and aerobic blood cultures with sensitivities—should be considered an important component of any comprehensive infectious disease evaluation.
Although the purpose of this article has been to emphasize the importance of blood cultures in the timely diagnosis and treatment of patients taking anti-TNF medications who present with symptoms of infectious disease, it is important to remember that blood cultures are only part of a full infectious disease workup.
Conclusion
The clinical studies and case presentations discussed here serve to remind clinicians that timely acquisition and evaluation of blood cultures are essential to the identification of serious bacteriologic pathogens, and to the chances of recovery in the RA patient with a systemic infection who is immunosuppressed as a result of treatment with infliximab or other anti-TNF agent. Rapid identification of pathogenic organisms will facilitate the implementation of appropriate therapy, thus avoiding the often-deadly outcomes of severe infection and sepsis. Gram-stain, anaerobic, and aerobic blood cultures with sensitivities should be considered the standard of care in any infliximab-treated patient who presents with fever, a patient report of a fever, or any other signs, symptoms, or patient history that could suggest infection.
While this article focuses for the most part on infections associated with infliximab use, the same strategy in standard of practice should apply to all patients who use anti-TNF therapy for treatment of RA.
1. Burns MJ. Chapter 181. Other immunosuppressive medications. In: Marx J, Hockberger R, Walls R, et al, eds. Rosen’s Emergency Medicine: Concepts and Clinical Practice. 7th ed. Philadelphia, PA: Mosby Elsevier; 2009.
2. Crum NF, Lederman ER, Wallace MR. Infections associated with tumor necrosis factor-alpha antagonists. Medicine (Baltimore). 2005; 84(5)291-302.
3. Geraghty EM, Ristow B, Gordon SM, Aronowitz P. Overwhelming parasitemia with Plasmodium falciparum infection in a patient receiving infliximab therapy for rheumatoid arthritis. Clin Infect Dis. 2007;44(10):82-84.
4. Listing J, Strangfeld A, Kary S, et al. Infections in patients with rheumatoid arthritis treated with biologic agents. Arthritis Rheum. 2005;52(11):3403-3412.
5. Louie SG, Park B, Yoon H. Biological response modifiers in the management of rheumatoid arthritis. Am J Health Syst Pharm. 2003;60(4): 346-355.
6. Kesteman T, Yombi JC, Gigi J, Durez P. Listeria infections associated with infliximab: case reports. Clin Rheumatol. 2007;26(12):2173-2175.
7. Kroesen S, Widmer AF, Tyndall A, Hasler P. Serious bacterial infections in patients with rheumatoid arthritis under anti-TNF-alpha therapy. Rheumatology. 2003;42(5):617-621.
8. Goode S, Tierney G, Deighton C. Life threatening intra-abdominal sepsis in patients on anti-TNF-alpha therapy. Gut. 2005;54(4):590-591.
9. Glück T, Linde HJ, Schölmerich J, et al. Anti-tumor necrosis factor therapy and Listeria monocytogenes infection: report of two cases. Arthritis Rheum. 2002;46(8):2255-2257.
10. Rijkeboer A, Voskuyl A, Van Agtmael M. Fatal Salmonella enteritidis septicaemia in a rheumatoid arthritis patient treated with a TNF-alpha antagonist. Scand J Infect Dis. 2007; 39(1):80-83.
11. Chan AT, Cleeve V, Daymond TJ. Necrotising fasciitis in a patient receiving infliximab for rheumatoid arthritis. Postgrad Med J. 2002; 78(915):47-48.
12. Favalli EG, Desiati F, Atzeni F, et al. Serious infections during anti-TNF-alpha treatment in rheumatoid arthritis patients. Autoimmun Rev. 2009;8(3):266-273.
13. Keane J, Gershon S, Wise RP, et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med. 2001;345(15):1098-1104.
1. Burns MJ. Chapter 181. Other immunosuppressive medications. In: Marx J, Hockberger R, Walls R, et al, eds. Rosen’s Emergency Medicine: Concepts and Clinical Practice. 7th ed. Philadelphia, PA: Mosby Elsevier; 2009.
2. Crum NF, Lederman ER, Wallace MR. Infections associated with tumor necrosis factor-alpha antagonists. Medicine (Baltimore). 2005; 84(5)291-302.
3. Geraghty EM, Ristow B, Gordon SM, Aronowitz P. Overwhelming parasitemia with Plasmodium falciparum infection in a patient receiving infliximab therapy for rheumatoid arthritis. Clin Infect Dis. 2007;44(10):82-84.
4. Listing J, Strangfeld A, Kary S, et al. Infections in patients with rheumatoid arthritis treated with biologic agents. Arthritis Rheum. 2005;52(11):3403-3412.
5. Louie SG, Park B, Yoon H. Biological response modifiers in the management of rheumatoid arthritis. Am J Health Syst Pharm. 2003;60(4): 346-355.
6. Kesteman T, Yombi JC, Gigi J, Durez P. Listeria infections associated with infliximab: case reports. Clin Rheumatol. 2007;26(12):2173-2175.
7. Kroesen S, Widmer AF, Tyndall A, Hasler P. Serious bacterial infections in patients with rheumatoid arthritis under anti-TNF-alpha therapy. Rheumatology. 2003;42(5):617-621.
8. Goode S, Tierney G, Deighton C. Life threatening intra-abdominal sepsis in patients on anti-TNF-alpha therapy. Gut. 2005;54(4):590-591.
9. Glück T, Linde HJ, Schölmerich J, et al. Anti-tumor necrosis factor therapy and Listeria monocytogenes infection: report of two cases. Arthritis Rheum. 2002;46(8):2255-2257.
10. Rijkeboer A, Voskuyl A, Van Agtmael M. Fatal Salmonella enteritidis septicaemia in a rheumatoid arthritis patient treated with a TNF-alpha antagonist. Scand J Infect Dis. 2007; 39(1):80-83.
11. Chan AT, Cleeve V, Daymond TJ. Necrotising fasciitis in a patient receiving infliximab for rheumatoid arthritis. Postgrad Med J. 2002; 78(915):47-48.
12. Favalli EG, Desiati F, Atzeni F, et al. Serious infections during anti-TNF-alpha treatment in rheumatoid arthritis patients. Autoimmun Rev. 2009;8(3):266-273.
13. Keane J, Gershon S, Wise RP, et al. Tuberculosis associated with infliximab, a tumor necrosis factor alpha-neutralizing agent. N Engl J Med. 2001;345(15):1098-1104.
Grand Rounds: Man, 72, With Peeling Penile Skin
A 72-year-old man presented to his primary care provider’s office with complaints of peeling skin on his penis and frequent, burning urination. He said he had first noticed redness on his penis about four days earlier, adding that it was growing worse. He was unsure whether he was truly experiencing frequent urination or just more aware of urinating because of the burning pain. He reported no attempts to treat himself, stating that he was “just keeping an eye on it and hoping it would go away.”
The patient’s medical history was limited to hypertension, for which he was taking valsartan, and allergies, for which he took fexofenadine. His surgical history included a tonsillectomy and appendectomy during his early teens. He had no known allergies to any medications.
The patient was married and retired after an executive career. He and his wife split their residence between New York and Florida during seasonal changes and were living in Florida at the time. He reported social drinking (“on rare occasions, these days”) and smoking an occasional cigar. He reported that he showers only once or twice weekly because of dry skin.
The following vital signs were recorded: blood pressure, 110/72 mm Hg; heart rate, 68 beats/min; respirations, 15/min; temperature, 97.8°F; and O2 saturation, 99% on room air. He was 73” tall and weighed 197 lb, with a BMI of 26.
The patient was alert and oriented. His physical exam was overall unremarkable, with the exception of an uncircumcised penis with redness and inflammation on the glans penis and no discharge noted. The reddened area was bright and shiny with a moist appearance and well-defined borders. The man denied any risk for sexually transmitted disease (STD) and denied any penile discharge. He also denied fever, chills, or arthritis.
Urinalysis performed in the office was negative for a urinary tract infection or for elevated glucose. A laboratory report from six months earlier was reviewed; all findings were within normal range, including the blood glucose level, with special attention paid for possible underlying cause; and the prostate-specific antigen (PSA) level, obtained for possible prostatitis or prostate cancer.
The differential diagnosis included eczema or psoriasis, Zoon’s balanitis, penile cancer, balanitis xerotica obliterans (lichen sclerosus), candidiasis balanitis, and circinate balanitis (as occurs in patients with Reiter’s disease; see table1-5). The absence of circumcision and the patient’s report of infrequent bathing raised concern for a hygiene-related etiology; the final diagnosis, made empirically, was candidiasis balanitis. Regarding an underlying cause, the laboratory order included a urine culture, fasting complete blood count, chemistry panel, and PSA level.
The patient was given instructions to wash the affected area twice daily for one week with a lukewarm weak saline solution (1 tablespoon salt/L water),5,6 gently retracting the foreskin; he was also given a topical antifungal cream7 (ketoconazole 2%, although other choices are discussed below), to be applied two to three times daily until his symptoms resolved.6 He was advised to return in one week if the condition did not improve or grew worse5; referral to dermatology would then be considered. The patient was also advised that in the case of a recurrent episode, dermatology would be consulted. The possibility of circumcision was discussed,8 and the patient was given information about the procedure, with referral to a urologist in the area.
Discussion
Balanitis is an inflammation of the glans penis; balanoposthitis involves the foreskin and prepuce.9-11 Balanitis can occur in men of any age, with etiologies varying with a patient’s age. Typical signs and symptoms include redness and swelling of the glans penis or foreskin, itching and/or pain, urethral discharge, phimosis, swollen lymph nodes, ulceration or plaque appearance, and pain on urination.12
In addition to the differential diagnoses mentioned, several additional conditions can be considered in a man with penile lesions. In older men, it is particularly important to investigate such lesions thoroughly, following the patient until the underlying cause is determined and the best treatment choice is selected. Specialists in dermatology and urology can best identify persistent or chronic lesions and make appropriate treatment recommendations, including possible circumcision.
The condition is commonly associated with absence of circumcision, poor hygiene, and phimosis (the inability to retract the foreskin from the glans penis). Accumulation of glandular secretions (smegma) and sloughed epithelial cells under the foreskin can lead to irritation and subsequent infection.
Uncontrolled or poorly controlled diabetes can be implicated in candidiasis infections.1 Other causes and contributing factors include chemical irritants (eg, soaps, lubricating jelly), edematous conditions (including congestive heart failure, cirrhosis, and nephrosis), drug allergies, morbid obesity, and a number of viruses and other pathogens, including those associated with STDs.12
A more detailed laboratory work-up might include the following:
• Serum glucose test (as part of a diabetes screening; in older men, this inflammatory condition can be a presenting sign of diabetes mellitus6)
• Culture of discharge, if any is present
• Serology test for STDs
• Wet mount with potassium hydroxide (for Candida albicans infection)
• Ultrasound, in severe cases or when urinary obstruction is suspected.
Additionally, in chronic cases, the patient should be referred to dermatology or urology for biopsy.5,9 Testing for anaerobes should also be considered for the patient and his sexual partner; if results are positive, treatment with oral metronidazole (400 mg tid for 10 days) is advised.6
In this patient’s case, the test that would best support an in-office diagnosis of candidiasis balanitis is a wet mount with potassium hydroxide. This was not performed at the time of the case patient’s visit, however; the diagnosis was empirically determined.
Management, Including Patient Education
Treatment of candidiasis balanitis involves routinely cleaning the penis and foreskin, as the case patient was instructed; use of soap, an irritant, should be avoided until the condition is resolved.7,10 Appropriate topical antifungal creams include nystatin, ketoconazole, miconazole, clotrimazole, econazole, and terbinafine, applied two to three times daily for at least 10 days; a cream combining an imidazole with 1% hydrocortisone may be effective for patients with significant inflammation.5,6,8,10,13
The patient should be instructed to:
• Keep the area clean and dry
• Wash twice daily with weak saline solution after removing residual medication and before applying fresh medication
• Wear loose cotton underwear
• Avoid sharing towels or cleaning cloths
• Wash personal items and surfaces, if possible, with disinfectant
• Notify sexual partner(s) that they may need treatment
• Discontinue sexual intercourse until infection is resolved
• Continue treatment for 10 to 14 days, even though relief may occur early
• Follow up with the clinician if no improvement is seen within one week
• Consider circumcision, in case of chronic infection.1,2,8,12
Conclusion
It is important to diagnose balanitis correctly, as this condition can affect sexual and urinary function, and its effects should not be underestimated in older men. Differentiating between infectious, noninfectious, premalignant, and malignant lesions will lead to appropriate care and allow early diagnosis or prevention of curable malignancies.
1. Singh S, Bunker C. Male genital dermatoses in old age. Age Ageing. 2008;37(5):500-504.
2. Thompson IM, Teichman JM, Elston DM, Sea J. Noninfectious penile lesions. Am Fam Physician. 2010;81(2):167-174.
3. Lane JE, Johnson J. Persistent penile patch. Am Fam Physician. 2008;78(9):1081-1082.
4. Gupta S, Malhotra AK, Ajith C. Lichen sclerosus: role of occlusion of the genital skin in the pathogenesis. Indian J Dermatol Venereol Leprol. 2010;76(1):56-58.
5. British Association for Sexual Health and HIV, Clinical Effectiveness Group. 2008 UK National Guideline on the Management of Balanoposthitis. www.bashh.org/documents/2062. Accessed September 22, 2010.
6. Ashton R, Leppard B. Differential Diagnosis in Dermatology. 3rd ed. London: Radcliffe Publishing Ltd; 2004:321.
7. NHS Institute for Innovation and Improvement. Clinical Knowledge Summaries: Balanitis (June 2009). www.cks.nhs.uk/balanitis/management/scenario_balanitis_adults#-378526. Accessed September 22, 2010.
8. Parker J. Management of common fungal infections in primary care. Nurs Stand. 2009;23(43):42-46.
9. Green MB, Bailey PP. Infectious processes: urinary tract infections and sexually transmitted diseases. In: Buttaro TM, Trybulski J, Bailey PP, Sandberg-Cook J, eds. Primary Care: A Collaborative Practice. 3rd ed. St. Louis, MO: Mosby Elsevier; 2008:576-590.
10. Singh-Behl D, Tomecki KJ. Common skins infections 2009. www.clevelandclinicmeded .com/medicalpubs/diseasemanagement/dermatol ogy/common-skin-infections. Accessed September 22, 2010.
11. Ko WT, Adal KA, Tomecki KJ. Infectious diseases. Med Clin North Am. 1998;82:(5):1001-1031.
12. Morgan K, McCance, KL. Alterations of the reproductive systems. In: McCance KL, Huether SE, eds. Pathophysiology: The Biologic Basis for Disease in Adults and Children. 5th ed. St. Louis, MO: Elsevier Mosby; 2006:805-807.
13. Waugh MA, Evans EG, Nayyar KC, Fong R. Clotrimazole (Canestan) in the treatment of candidal balanitis in men: with incidental observations on diabetic candidal balanoposthitis. Br J Vener Dis. 1978;54(3):184-186.
A 72-year-old man presented to his primary care provider’s office with complaints of peeling skin on his penis and frequent, burning urination. He said he had first noticed redness on his penis about four days earlier, adding that it was growing worse. He was unsure whether he was truly experiencing frequent urination or just more aware of urinating because of the burning pain. He reported no attempts to treat himself, stating that he was “just keeping an eye on it and hoping it would go away.”
The patient’s medical history was limited to hypertension, for which he was taking valsartan, and allergies, for which he took fexofenadine. His surgical history included a tonsillectomy and appendectomy during his early teens. He had no known allergies to any medications.
The patient was married and retired after an executive career. He and his wife split their residence between New York and Florida during seasonal changes and were living in Florida at the time. He reported social drinking (“on rare occasions, these days”) and smoking an occasional cigar. He reported that he showers only once or twice weekly because of dry skin.
The following vital signs were recorded: blood pressure, 110/72 mm Hg; heart rate, 68 beats/min; respirations, 15/min; temperature, 97.8°F; and O2 saturation, 99% on room air. He was 73” tall and weighed 197 lb, with a BMI of 26.
The patient was alert and oriented. His physical exam was overall unremarkable, with the exception of an uncircumcised penis with redness and inflammation on the glans penis and no discharge noted. The reddened area was bright and shiny with a moist appearance and well-defined borders. The man denied any risk for sexually transmitted disease (STD) and denied any penile discharge. He also denied fever, chills, or arthritis.
Urinalysis performed in the office was negative for a urinary tract infection or for elevated glucose. A laboratory report from six months earlier was reviewed; all findings were within normal range, including the blood glucose level, with special attention paid for possible underlying cause; and the prostate-specific antigen (PSA) level, obtained for possible prostatitis or prostate cancer.
The differential diagnosis included eczema or psoriasis, Zoon’s balanitis, penile cancer, balanitis xerotica obliterans (lichen sclerosus), candidiasis balanitis, and circinate balanitis (as occurs in patients with Reiter’s disease; see table1-5). The absence of circumcision and the patient’s report of infrequent bathing raised concern for a hygiene-related etiology; the final diagnosis, made empirically, was candidiasis balanitis. Regarding an underlying cause, the laboratory order included a urine culture, fasting complete blood count, chemistry panel, and PSA level.
The patient was given instructions to wash the affected area twice daily for one week with a lukewarm weak saline solution (1 tablespoon salt/L water),5,6 gently retracting the foreskin; he was also given a topical antifungal cream7 (ketoconazole 2%, although other choices are discussed below), to be applied two to three times daily until his symptoms resolved.6 He was advised to return in one week if the condition did not improve or grew worse5; referral to dermatology would then be considered. The patient was also advised that in the case of a recurrent episode, dermatology would be consulted. The possibility of circumcision was discussed,8 and the patient was given information about the procedure, with referral to a urologist in the area.
Discussion
Balanitis is an inflammation of the glans penis; balanoposthitis involves the foreskin and prepuce.9-11 Balanitis can occur in men of any age, with etiologies varying with a patient’s age. Typical signs and symptoms include redness and swelling of the glans penis or foreskin, itching and/or pain, urethral discharge, phimosis, swollen lymph nodes, ulceration or plaque appearance, and pain on urination.12
In addition to the differential diagnoses mentioned, several additional conditions can be considered in a man with penile lesions. In older men, it is particularly important to investigate such lesions thoroughly, following the patient until the underlying cause is determined and the best treatment choice is selected. Specialists in dermatology and urology can best identify persistent or chronic lesions and make appropriate treatment recommendations, including possible circumcision.
The condition is commonly associated with absence of circumcision, poor hygiene, and phimosis (the inability to retract the foreskin from the glans penis). Accumulation of glandular secretions (smegma) and sloughed epithelial cells under the foreskin can lead to irritation and subsequent infection.
Uncontrolled or poorly controlled diabetes can be implicated in candidiasis infections.1 Other causes and contributing factors include chemical irritants (eg, soaps, lubricating jelly), edematous conditions (including congestive heart failure, cirrhosis, and nephrosis), drug allergies, morbid obesity, and a number of viruses and other pathogens, including those associated with STDs.12
A more detailed laboratory work-up might include the following:
• Serum glucose test (as part of a diabetes screening; in older men, this inflammatory condition can be a presenting sign of diabetes mellitus6)
• Culture of discharge, if any is present
• Serology test for STDs
• Wet mount with potassium hydroxide (for Candida albicans infection)
• Ultrasound, in severe cases or when urinary obstruction is suspected.
Additionally, in chronic cases, the patient should be referred to dermatology or urology for biopsy.5,9 Testing for anaerobes should also be considered for the patient and his sexual partner; if results are positive, treatment with oral metronidazole (400 mg tid for 10 days) is advised.6
In this patient’s case, the test that would best support an in-office diagnosis of candidiasis balanitis is a wet mount with potassium hydroxide. This was not performed at the time of the case patient’s visit, however; the diagnosis was empirically determined.
Management, Including Patient Education
Treatment of candidiasis balanitis involves routinely cleaning the penis and foreskin, as the case patient was instructed; use of soap, an irritant, should be avoided until the condition is resolved.7,10 Appropriate topical antifungal creams include nystatin, ketoconazole, miconazole, clotrimazole, econazole, and terbinafine, applied two to three times daily for at least 10 days; a cream combining an imidazole with 1% hydrocortisone may be effective for patients with significant inflammation.5,6,8,10,13
The patient should be instructed to:
• Keep the area clean and dry
• Wash twice daily with weak saline solution after removing residual medication and before applying fresh medication
• Wear loose cotton underwear
• Avoid sharing towels or cleaning cloths
• Wash personal items and surfaces, if possible, with disinfectant
• Notify sexual partner(s) that they may need treatment
• Discontinue sexual intercourse until infection is resolved
• Continue treatment for 10 to 14 days, even though relief may occur early
• Follow up with the clinician if no improvement is seen within one week
• Consider circumcision, in case of chronic infection.1,2,8,12
Conclusion
It is important to diagnose balanitis correctly, as this condition can affect sexual and urinary function, and its effects should not be underestimated in older men. Differentiating between infectious, noninfectious, premalignant, and malignant lesions will lead to appropriate care and allow early diagnosis or prevention of curable malignancies.
A 72-year-old man presented to his primary care provider’s office with complaints of peeling skin on his penis and frequent, burning urination. He said he had first noticed redness on his penis about four days earlier, adding that it was growing worse. He was unsure whether he was truly experiencing frequent urination or just more aware of urinating because of the burning pain. He reported no attempts to treat himself, stating that he was “just keeping an eye on it and hoping it would go away.”
The patient’s medical history was limited to hypertension, for which he was taking valsartan, and allergies, for which he took fexofenadine. His surgical history included a tonsillectomy and appendectomy during his early teens. He had no known allergies to any medications.
The patient was married and retired after an executive career. He and his wife split their residence between New York and Florida during seasonal changes and were living in Florida at the time. He reported social drinking (“on rare occasions, these days”) and smoking an occasional cigar. He reported that he showers only once or twice weekly because of dry skin.
The following vital signs were recorded: blood pressure, 110/72 mm Hg; heart rate, 68 beats/min; respirations, 15/min; temperature, 97.8°F; and O2 saturation, 99% on room air. He was 73” tall and weighed 197 lb, with a BMI of 26.
The patient was alert and oriented. His physical exam was overall unremarkable, with the exception of an uncircumcised penis with redness and inflammation on the glans penis and no discharge noted. The reddened area was bright and shiny with a moist appearance and well-defined borders. The man denied any risk for sexually transmitted disease (STD) and denied any penile discharge. He also denied fever, chills, or arthritis.
Urinalysis performed in the office was negative for a urinary tract infection or for elevated glucose. A laboratory report from six months earlier was reviewed; all findings were within normal range, including the blood glucose level, with special attention paid for possible underlying cause; and the prostate-specific antigen (PSA) level, obtained for possible prostatitis or prostate cancer.
The differential diagnosis included eczema or psoriasis, Zoon’s balanitis, penile cancer, balanitis xerotica obliterans (lichen sclerosus), candidiasis balanitis, and circinate balanitis (as occurs in patients with Reiter’s disease; see table1-5). The absence of circumcision and the patient’s report of infrequent bathing raised concern for a hygiene-related etiology; the final diagnosis, made empirically, was candidiasis balanitis. Regarding an underlying cause, the laboratory order included a urine culture, fasting complete blood count, chemistry panel, and PSA level.
The patient was given instructions to wash the affected area twice daily for one week with a lukewarm weak saline solution (1 tablespoon salt/L water),5,6 gently retracting the foreskin; he was also given a topical antifungal cream7 (ketoconazole 2%, although other choices are discussed below), to be applied two to three times daily until his symptoms resolved.6 He was advised to return in one week if the condition did not improve or grew worse5; referral to dermatology would then be considered. The patient was also advised that in the case of a recurrent episode, dermatology would be consulted. The possibility of circumcision was discussed,8 and the patient was given information about the procedure, with referral to a urologist in the area.
Discussion
Balanitis is an inflammation of the glans penis; balanoposthitis involves the foreskin and prepuce.9-11 Balanitis can occur in men of any age, with etiologies varying with a patient’s age. Typical signs and symptoms include redness and swelling of the glans penis or foreskin, itching and/or pain, urethral discharge, phimosis, swollen lymph nodes, ulceration or plaque appearance, and pain on urination.12
In addition to the differential diagnoses mentioned, several additional conditions can be considered in a man with penile lesions. In older men, it is particularly important to investigate such lesions thoroughly, following the patient until the underlying cause is determined and the best treatment choice is selected. Specialists in dermatology and urology can best identify persistent or chronic lesions and make appropriate treatment recommendations, including possible circumcision.
The condition is commonly associated with absence of circumcision, poor hygiene, and phimosis (the inability to retract the foreskin from the glans penis). Accumulation of glandular secretions (smegma) and sloughed epithelial cells under the foreskin can lead to irritation and subsequent infection.
Uncontrolled or poorly controlled diabetes can be implicated in candidiasis infections.1 Other causes and contributing factors include chemical irritants (eg, soaps, lubricating jelly), edematous conditions (including congestive heart failure, cirrhosis, and nephrosis), drug allergies, morbid obesity, and a number of viruses and other pathogens, including those associated with STDs.12
A more detailed laboratory work-up might include the following:
• Serum glucose test (as part of a diabetes screening; in older men, this inflammatory condition can be a presenting sign of diabetes mellitus6)
• Culture of discharge, if any is present
• Serology test for STDs
• Wet mount with potassium hydroxide (for Candida albicans infection)
• Ultrasound, in severe cases or when urinary obstruction is suspected.
Additionally, in chronic cases, the patient should be referred to dermatology or urology for biopsy.5,9 Testing for anaerobes should also be considered for the patient and his sexual partner; if results are positive, treatment with oral metronidazole (400 mg tid for 10 days) is advised.6
In this patient’s case, the test that would best support an in-office diagnosis of candidiasis balanitis is a wet mount with potassium hydroxide. This was not performed at the time of the case patient’s visit, however; the diagnosis was empirically determined.
Management, Including Patient Education
Treatment of candidiasis balanitis involves routinely cleaning the penis and foreskin, as the case patient was instructed; use of soap, an irritant, should be avoided until the condition is resolved.7,10 Appropriate topical antifungal creams include nystatin, ketoconazole, miconazole, clotrimazole, econazole, and terbinafine, applied two to three times daily for at least 10 days; a cream combining an imidazole with 1% hydrocortisone may be effective for patients with significant inflammation.5,6,8,10,13
The patient should be instructed to:
• Keep the area clean and dry
• Wash twice daily with weak saline solution after removing residual medication and before applying fresh medication
• Wear loose cotton underwear
• Avoid sharing towels or cleaning cloths
• Wash personal items and surfaces, if possible, with disinfectant
• Notify sexual partner(s) that they may need treatment
• Discontinue sexual intercourse until infection is resolved
• Continue treatment for 10 to 14 days, even though relief may occur early
• Follow up with the clinician if no improvement is seen within one week
• Consider circumcision, in case of chronic infection.1,2,8,12
Conclusion
It is important to diagnose balanitis correctly, as this condition can affect sexual and urinary function, and its effects should not be underestimated in older men. Differentiating between infectious, noninfectious, premalignant, and malignant lesions will lead to appropriate care and allow early diagnosis or prevention of curable malignancies.
1. Singh S, Bunker C. Male genital dermatoses in old age. Age Ageing. 2008;37(5):500-504.
2. Thompson IM, Teichman JM, Elston DM, Sea J. Noninfectious penile lesions. Am Fam Physician. 2010;81(2):167-174.
3. Lane JE, Johnson J. Persistent penile patch. Am Fam Physician. 2008;78(9):1081-1082.
4. Gupta S, Malhotra AK, Ajith C. Lichen sclerosus: role of occlusion of the genital skin in the pathogenesis. Indian J Dermatol Venereol Leprol. 2010;76(1):56-58.
5. British Association for Sexual Health and HIV, Clinical Effectiveness Group. 2008 UK National Guideline on the Management of Balanoposthitis. www.bashh.org/documents/2062. Accessed September 22, 2010.
6. Ashton R, Leppard B. Differential Diagnosis in Dermatology. 3rd ed. London: Radcliffe Publishing Ltd; 2004:321.
7. NHS Institute for Innovation and Improvement. Clinical Knowledge Summaries: Balanitis (June 2009). www.cks.nhs.uk/balanitis/management/scenario_balanitis_adults#-378526. Accessed September 22, 2010.
8. Parker J. Management of common fungal infections in primary care. Nurs Stand. 2009;23(43):42-46.
9. Green MB, Bailey PP. Infectious processes: urinary tract infections and sexually transmitted diseases. In: Buttaro TM, Trybulski J, Bailey PP, Sandberg-Cook J, eds. Primary Care: A Collaborative Practice. 3rd ed. St. Louis, MO: Mosby Elsevier; 2008:576-590.
10. Singh-Behl D, Tomecki KJ. Common skins infections 2009. www.clevelandclinicmeded .com/medicalpubs/diseasemanagement/dermatol ogy/common-skin-infections. Accessed September 22, 2010.
11. Ko WT, Adal KA, Tomecki KJ. Infectious diseases. Med Clin North Am. 1998;82:(5):1001-1031.
12. Morgan K, McCance, KL. Alterations of the reproductive systems. In: McCance KL, Huether SE, eds. Pathophysiology: The Biologic Basis for Disease in Adults and Children. 5th ed. St. Louis, MO: Elsevier Mosby; 2006:805-807.
13. Waugh MA, Evans EG, Nayyar KC, Fong R. Clotrimazole (Canestan) in the treatment of candidal balanitis in men: with incidental observations on diabetic candidal balanoposthitis. Br J Vener Dis. 1978;54(3):184-186.
1. Singh S, Bunker C. Male genital dermatoses in old age. Age Ageing. 2008;37(5):500-504.
2. Thompson IM, Teichman JM, Elston DM, Sea J. Noninfectious penile lesions. Am Fam Physician. 2010;81(2):167-174.
3. Lane JE, Johnson J. Persistent penile patch. Am Fam Physician. 2008;78(9):1081-1082.
4. Gupta S, Malhotra AK, Ajith C. Lichen sclerosus: role of occlusion of the genital skin in the pathogenesis. Indian J Dermatol Venereol Leprol. 2010;76(1):56-58.
5. British Association for Sexual Health and HIV, Clinical Effectiveness Group. 2008 UK National Guideline on the Management of Balanoposthitis. www.bashh.org/documents/2062. Accessed September 22, 2010.
6. Ashton R, Leppard B. Differential Diagnosis in Dermatology. 3rd ed. London: Radcliffe Publishing Ltd; 2004:321.
7. NHS Institute for Innovation and Improvement. Clinical Knowledge Summaries: Balanitis (June 2009). www.cks.nhs.uk/balanitis/management/scenario_balanitis_adults#-378526. Accessed September 22, 2010.
8. Parker J. Management of common fungal infections in primary care. Nurs Stand. 2009;23(43):42-46.
9. Green MB, Bailey PP. Infectious processes: urinary tract infections and sexually transmitted diseases. In: Buttaro TM, Trybulski J, Bailey PP, Sandberg-Cook J, eds. Primary Care: A Collaborative Practice. 3rd ed. St. Louis, MO: Mosby Elsevier; 2008:576-590.
10. Singh-Behl D, Tomecki KJ. Common skins infections 2009. www.clevelandclinicmeded .com/medicalpubs/diseasemanagement/dermatol ogy/common-skin-infections. Accessed September 22, 2010.
11. Ko WT, Adal KA, Tomecki KJ. Infectious diseases. Med Clin North Am. 1998;82:(5):1001-1031.
12. Morgan K, McCance, KL. Alterations of the reproductive systems. In: McCance KL, Huether SE, eds. Pathophysiology: The Biologic Basis for Disease in Adults and Children. 5th ed. St. Louis, MO: Elsevier Mosby; 2006:805-807.
13. Waugh MA, Evans EG, Nayyar KC, Fong R. Clotrimazole (Canestan) in the treatment of candidal balanitis in men: with incidental observations on diabetic candidal balanoposthitis. Br J Vener Dis. 1978;54(3):184-186.
Hydroxocobalamin for Suspected Cyanide Poisoning
Emergent Management of Pediatric Seizures
Familial Tumoral Calcinosis
UPDATE: PELVIC FLOOR DYSFUNCTION
When a woman has advanced prolapse of the anterior vaginal wall, it is highly likely that she has apical prolapse as well. Consider a study by Rooney and associates that determined that clinically significant vault prolapse is present in most women who have anterior vaginal prolapse of stage II or higher.1 For that reason, suspension of the vaginal apex should be considered whenever surgical treatment of anterior wall defects is planned.
Sacrocolpopexy involves suspension of the vaginal vault from the anterior longitudinal ligament of the sacrum, using Y-shaped mesh to augment native tissue (FIGURE).2 It is an effective, durable treatment for vaginal apical prolapse. With a success rate approaching 93%, this procedure has become the gold standard for repair of vault prolapse. Among its advantages are maximization of vaginal depth and preservation of a normal vaginal axis.
Sacrocolpopexy preserves the vaginal axis
With the vaginal vault suspended from the anterior longitudinal
ligament of the sacrum, the normal vaginal axis is preserved
and vaginal depth is maximized.
Sacrocolpopexy can be performed via the abdominal, laparoscopic, or robotic-assisted approach (TABLE 1). Minimally invasive techniques are attractive because they involve faster recovery than abdominal sacrocolpopexy does. Minimally invasive techniques have also advanced to the point that they are both effective and durable. However, these advantages must be weighed against the effort required to learn the techniques, as well as their higher cost.
TABLE 1
How the 3 approaches to sacrocolpopexy compare
| Approach | Advantages and disadvantages |
|---|---|
| Abdominal | Shortest operative time No significant Trendelenburg position required Highest estimated blood loss Longest length of stay Low rate of complications Longest postoperative recovery Well-established long-term durability |
| Laparoscopic | Longer operative time Moderate Trendelenburg position required Lower estimated blood loss Shorter length of stay Surgical technique least similar to abdominal procedure Low rate of complications Shorter postoperative recovery Long-term durability less firmly established |
| Robotic-assisted | Longest operative time Steep Trendelenburg position required Lower estimated blood loss Shorter length of stay Surgical technique resembles that of abdominal approach Low rate of complications Shorter postoperative recovery Long-term durability appears to be good |
In this article, we highlight:
- a comparison of the laparoscopic and abdominal approaches to sacrocolpopexy
- an investigation of the learning curve associated with robotic-assisted sacrocolpopexy
- a study exploring the durability of robotic-assisted repair
- an estimate of the costs associated with each route of operation.
Laparoscopic vs abdominal sacrocolpopexy—how do they compare?
Paraiso MF, Walters MD, Rackley RR, Melek S, Hugney C. Laparoscopic and abdominal sacral colpopexies: a comparative cohort study. Am J Obstet Gynecol. 2005;192(5):1752–1758.
When surgeons at the Cleveland Clinic performed a retrospective cohort study to compare laparoscopic and abdominal sacrocolpopexy, they found significantly longer operative time with the laparoscopic route, with an average difference of 51 minutes (P < .0001). However, the laparoscopic approach was associated with lower blood loss (although there was no difference between groups in hematocrit on postoperative day 1); shorter hospital stay (average of 1.8 days versus 4 days [P < .001]); and comparable rates of intraoperative and postoperative complications.
Details of the trial
Paraiso and colleagues reviewed the medical charts of 56 consecutive patients who had undergone laparoscopic sacrocolpopexy, comparing them with the charts of 61 consecutive patients who had undergone the procedure using the abdominal approach. The operations had been performed between 1998 and 2003 for treatment of posthysterectomy vaginal prolapse.
The groups underwent similar rates of concurrent procedures. The laparotomy group had a significantly higher number of Burch procedures (P = .007), and the laparoscopic group had a significantly higher rate of adhesiolysis (P = .002).
Among the complications noted— which occurred at comparable rates between groups—were cystotomy, enterotomy, need for transfusion, deep-vein thrombosis, ileus, small bowel obstruction, wound infection, ventral hernia, mesh erosion, and recurrent prolapse. One laparoscopic case was converted to laparotomy because of excessive bleeding during the rectopexy portion of the operation.
Laparoscopy may have taken longer than this trial suggests
This study is one of very few well-designed trials comparing laparoscopic sacrocolpopexy to the historical gold standard of abdominal sacrocolpopexy for vault prolapse.
Twenty-eight percent of laparoscopic procedures in this study used tacking devices in lieu of suturing. Had suturing been performed universally, an even greater difference in surgical time may have been observed.
There may also be differences between groups in the durability of the two types of repair, an outcome not included in this particular study.
The laparoscopic approach offers a shorter hospital stay with no increase in intraoperative or postoperative complications, compared with abdominal sacrocolpopexy. However, it entails a significantly longer operative time than the abdominal approach does.
How steep is the learning curve for robotic-assisted sacrocolpopexy?
Akl MN, Long JB, Giles DL, et al. Robotic-assisted sacrocolpopexy: technique and learning curve. Surg Endosc. 2009;23(10):2390–2394.
Akl and coworkers reviewed the medical records of all patients who had undergone robotic-assisted sacrocolpopexy at the Mayo Clinics in Arizona and Florida between 2004 and 2007. All operations were performed by the same four urogynecologists, with an average operative time of 197.9 minutes (standard deviation, ± 66.8 minutes). However, after the first 10 cases, the operative time decreased by 64.3 minutes—a decline of 25.4% (P < .01; 95% confidence interval [CI], 16.1–112.4 minutes).
Details of the trial
Researchers collected baseline information on participants’ age, stage of prolapse, and concomitant procedures. They also gathered data on average operative time, estimated blood loss, intraoperative and postoperative complications, conversion to laparotomy, and length of hospitalization.
Of 80 women who had advanced pelvic organ prolapse (stage III/IV) who underwent robotic-assisted sacrocolpopexy, 88% underwent concomitant robotic and vaginal procedures, including robotic supracervical hysterectomy, Burch procedure, paravaginal repair, lysis of adhesions, bilateral salpingooophorectomy, vaginal cystocele or rectocele repair, and placement of a midurethral sling.
Estimated blood loss for the robotic-assisted approach ranged from 25 mL to 300 mL, with a mean loss of 96.8 mL. Average length of hospitalization was 2.6 days. Four cases (5%) were converted to laparotomy because of limited exposure and one intraoperative bladder injury. Other intraoperative complications included small-bowel injury during trocar placement and one ureteral injury. Postoperative complications included one case of ileus and five (6%) vaginal mesh erosions. Three patients developed recurrent prolapse and underwent subsequent correction.
Learning curve could have been measured more precisely
The authors did not specifically measure the learning curve for robotic-assisted sacrocolpopexy, as they took into account the concomitant procedures. For this reason, the decrease in operative time observed after 10 cases may not accurately reflect an improvement in the performance of sacrocolpopexy.
Akl and colleagues consider this detail to be a strength of the study because most women who undergo prolapse surgery have concomitant procedures. However, recording the length of time it took to perform the sacrocolpopexy portion of the procedure would have been more accurate.
The average length of stay approached that of the abdominal route. Length of stay may decline as a surgeon gains experience with the robotic-assisted approach.
Robotic-assisted sacrocolpopexy has a steep learning curve with respect to technique and surgical time.
Does robotic-assisted sacrocolpopexy provide durable support?
Elliott DS, Krambeck AE, Chow GK. Long-term results of robotic assisted laparoscopic sacrocolpopexy for the treatment of high grade vaginal vault prolapse. J Urol. 2006;176(2):655–659.
Among the few recent series reporting long-term outcomes after robotic-assisted sacrocolpopexy is this observational study from the Mayo Clinic. It involved 30 women who underwent the operation for the treatment of Baden Walker grade 4/4 posthysterectomy vaginal vault prolapse. The authors concluded that advanced prolapse can be treated with robotic-assisted sacrocolpopexy with long-term success and minimal complications.
Details of the trial
Of 30 women in this trial, 52% underwent an anti-incontinence procedure at the time of sacrocolpopexy. Women who had multiple vaginal defects or a history of abdominal surgery were excluded from the study.
Average operative time was 3.1 hours (range, 2.15–4.75 hours) in the early phase of development of operative technique (described in the manuscript) but diminished over time to an average of 2.5 hours.
Twenty-nine patients were discharged from the hospital after an overnight stay. Very few immediate postoperative complications were observed. Two patients experienced mild port-site infections that required outpatient treatment, and one patient had persistent vaginal bleeding from the incision made during the anti-incontinence procedure.
Most patients were followed for at least 1 year
The mean follow-up in this study was 24 months (range, 16–39 months). During this period, 21 women were followed for a full year. Long-term observation revealed that the repair of vault prolapse remained successful in 19 of these women.
One patient experienced recurrent prolapse 7 months after surgery. Another developed a rectocele 9 months after sacrocolpopexy. Vaginal mesh erosions occurred in two patients within 6 months after the procedure; both patients were treated with outpatient resection of the exposed mesh, with no recurrence of the prolapse.
Although a larger sample size and longer follow-up would be ideal, this study demonstrates a low rate of recurrent prolapse 1 year after the procedure.
Robotic sacrocolpopexy appears to provide long-term durability for the treatment of advanced vaginal vault prolapse.
Depending on where you practice, you may have as many as three options: abdominal, laparoscopic, or robotic-assisted. Here are basic questions you should address when choosing one:
- How familiar are you with the technique? if the answer is “not much,” you can anticipate that the cost and time required to perform it will be significantly higher.
- Are the appropriate instruments and surgical team available?
- Does the patient have comorbidities? Consider, for example, the fact that she may not be able to tolerate a steep Trendelenberg position—required for the robotic-assisted approach—if she has severe cardiac or pulmonary disease. However, if she has a risk of poor wound healing, a large abdominal incision may not be advisable and postoperative immobility can be risky. if she is obese, laparoscopic or robotic port placement is challenging, but visualization and retraction will be easier. The need for anticoagulation is another consideration, as it will affect estimated blood loss and the choice of an incision, among other things.
- Let’s not forget the patient. Given the pros and cons, what approach does she prefer?
How much do laparoscopic, abdominal, and robotic-assisted sacrocolpopexy cost?
Judd JP, Siddiqui NY, Barnett JC, et al. Cost-minimization analysis of robotic-assisted, laparoscopic, and abdominal sacrocolpopexy. J Minim Invasive Gynecol. 2010;17: 493–499.
This cost-minimization analysis concluded that robotic-assisted sacrocolpopexy incurs the highest hospital charges but is reimbursed by Medicare at a rate similar to reimbursement for the abdominal and laparoscopic routes (TABLE 2).
TABLE 2
Cost of sacrocolpopexy is significant—especially using the robotic approach
| Approach | Cost of a procedure | Operative time, min (range) |
|---|---|---|
| Robotic-assisted | $8,508 | 328 (130–383) |
| Laparoscopic | $7,353 | 269 (97–334) |
| Abdominal | $5,792 | 170 (110–286) |
| Source: Judd JP, Siddiqui NY, Barnett JC, Visco AG, Havrilesky LJ, Wu JM. Cost-minimization analysis of robotic-assisted, laparoscopic, and abdominal sacrocolpopexy. J Minim invasive Gynecol. 2010;17(4):493–499. | ||
The analysis accounted for realistic practices, such as the inclusion of concurrent hysterectomy and other procedures.
Details of the trial
Surgeons from Duke University developed a decision-analysis model in which a hypothetical group of women with advanced vaginal prolapse could choose between one of the three routes of sacrocolpopexy: abdominal, laparoscopic, or robotic-assisted. Researchers postulated two different scenarios:
- the hospital had ownership of a robotic system
- the hospital invested in the initial purchase and maintenance of such a system.
Researchers reviewed the literature to formulate their estimates of operative time, rate of conversion to laparotomy, rate of transfusion, and length of hospital stay. In addition, the costs of initial anesthesia setup, professional fees, per-minute intraoperative fees, and postanesthesia care were applied to each approach. Operating room costs per minute and the cost of disposable items such as drapes, gowns, gloves, and single-use instruments were added. For the robotic approach, the costs of reusable instruments were distributed across 10 operations. Reusable instruments for laparoscopic and abdominal surgery were assumed to incur no additional investment. Last, postoperative care—including laboratory tests, pharmacy usage, and the need for a hospital room—were individualized for each route of surgery and applied to the cost.
Costs were estimated in 2008 US dollars, based on procedure costs incurred at Duke University Medical Center.
Physician reimbursement data were obtained from Medicare reimbursement rates for anesthesia and from surgeon Current Procedural Terminology (CPT) codes specific to each procedure.
Quality-of-life assessments were not measured. Nor was the cost to society of the postoperative loss of productivity and wages for each surgical route. Had these losses been recognized, the authors observed, the cost of robotic surgery may have been lower.
The cost of robotic surgery was equivalent to the cost of laparoscopy in only two instances:
- when the operative time of robotic surgery was reduced to 149 minutes
- when the cost of robotic disposable items was less than $2,132 (reduced from a baseline cost of $3,293).
Robotic sacrocolpopexy is costly. this is an important consideration when implementing new technology. cost-saving scenarios are useful to maximize patient benefit and minimize financial burden.
We want to hear from you! Tell us what you think.
When a woman has advanced prolapse of the anterior vaginal wall, it is highly likely that she has apical prolapse as well. Consider a study by Rooney and associates that determined that clinically significant vault prolapse is present in most women who have anterior vaginal prolapse of stage II or higher.1 For that reason, suspension of the vaginal apex should be considered whenever surgical treatment of anterior wall defects is planned.
Sacrocolpopexy involves suspension of the vaginal vault from the anterior longitudinal ligament of the sacrum, using Y-shaped mesh to augment native tissue (FIGURE).2 It is an effective, durable treatment for vaginal apical prolapse. With a success rate approaching 93%, this procedure has become the gold standard for repair of vault prolapse. Among its advantages are maximization of vaginal depth and preservation of a normal vaginal axis.
Sacrocolpopexy preserves the vaginal axis
With the vaginal vault suspended from the anterior longitudinal
ligament of the sacrum, the normal vaginal axis is preserved
and vaginal depth is maximized.
Sacrocolpopexy can be performed via the abdominal, laparoscopic, or robotic-assisted approach (TABLE 1). Minimally invasive techniques are attractive because they involve faster recovery than abdominal sacrocolpopexy does. Minimally invasive techniques have also advanced to the point that they are both effective and durable. However, these advantages must be weighed against the effort required to learn the techniques, as well as their higher cost.
TABLE 1
How the 3 approaches to sacrocolpopexy compare
| Approach | Advantages and disadvantages |
|---|---|
| Abdominal | Shortest operative time No significant Trendelenburg position required Highest estimated blood loss Longest length of stay Low rate of complications Longest postoperative recovery Well-established long-term durability |
| Laparoscopic | Longer operative time Moderate Trendelenburg position required Lower estimated blood loss Shorter length of stay Surgical technique least similar to abdominal procedure Low rate of complications Shorter postoperative recovery Long-term durability less firmly established |
| Robotic-assisted | Longest operative time Steep Trendelenburg position required Lower estimated blood loss Shorter length of stay Surgical technique resembles that of abdominal approach Low rate of complications Shorter postoperative recovery Long-term durability appears to be good |
In this article, we highlight:
- a comparison of the laparoscopic and abdominal approaches to sacrocolpopexy
- an investigation of the learning curve associated with robotic-assisted sacrocolpopexy
- a study exploring the durability of robotic-assisted repair
- an estimate of the costs associated with each route of operation.
Laparoscopic vs abdominal sacrocolpopexy—how do they compare?
Paraiso MF, Walters MD, Rackley RR, Melek S, Hugney C. Laparoscopic and abdominal sacral colpopexies: a comparative cohort study. Am J Obstet Gynecol. 2005;192(5):1752–1758.
When surgeons at the Cleveland Clinic performed a retrospective cohort study to compare laparoscopic and abdominal sacrocolpopexy, they found significantly longer operative time with the laparoscopic route, with an average difference of 51 minutes (P < .0001). However, the laparoscopic approach was associated with lower blood loss (although there was no difference between groups in hematocrit on postoperative day 1); shorter hospital stay (average of 1.8 days versus 4 days [P < .001]); and comparable rates of intraoperative and postoperative complications.
Details of the trial
Paraiso and colleagues reviewed the medical charts of 56 consecutive patients who had undergone laparoscopic sacrocolpopexy, comparing them with the charts of 61 consecutive patients who had undergone the procedure using the abdominal approach. The operations had been performed between 1998 and 2003 for treatment of posthysterectomy vaginal prolapse.
The groups underwent similar rates of concurrent procedures. The laparotomy group had a significantly higher number of Burch procedures (P = .007), and the laparoscopic group had a significantly higher rate of adhesiolysis (P = .002).
Among the complications noted— which occurred at comparable rates between groups—were cystotomy, enterotomy, need for transfusion, deep-vein thrombosis, ileus, small bowel obstruction, wound infection, ventral hernia, mesh erosion, and recurrent prolapse. One laparoscopic case was converted to laparotomy because of excessive bleeding during the rectopexy portion of the operation.
Laparoscopy may have taken longer than this trial suggests
This study is one of very few well-designed trials comparing laparoscopic sacrocolpopexy to the historical gold standard of abdominal sacrocolpopexy for vault prolapse.
Twenty-eight percent of laparoscopic procedures in this study used tacking devices in lieu of suturing. Had suturing been performed universally, an even greater difference in surgical time may have been observed.
There may also be differences between groups in the durability of the two types of repair, an outcome not included in this particular study.
The laparoscopic approach offers a shorter hospital stay with no increase in intraoperative or postoperative complications, compared with abdominal sacrocolpopexy. However, it entails a significantly longer operative time than the abdominal approach does.
How steep is the learning curve for robotic-assisted sacrocolpopexy?
Akl MN, Long JB, Giles DL, et al. Robotic-assisted sacrocolpopexy: technique and learning curve. Surg Endosc. 2009;23(10):2390–2394.
Akl and coworkers reviewed the medical records of all patients who had undergone robotic-assisted sacrocolpopexy at the Mayo Clinics in Arizona and Florida between 2004 and 2007. All operations were performed by the same four urogynecologists, with an average operative time of 197.9 minutes (standard deviation, ± 66.8 minutes). However, after the first 10 cases, the operative time decreased by 64.3 minutes—a decline of 25.4% (P < .01; 95% confidence interval [CI], 16.1–112.4 minutes).
Details of the trial
Researchers collected baseline information on participants’ age, stage of prolapse, and concomitant procedures. They also gathered data on average operative time, estimated blood loss, intraoperative and postoperative complications, conversion to laparotomy, and length of hospitalization.
Of 80 women who had advanced pelvic organ prolapse (stage III/IV) who underwent robotic-assisted sacrocolpopexy, 88% underwent concomitant robotic and vaginal procedures, including robotic supracervical hysterectomy, Burch procedure, paravaginal repair, lysis of adhesions, bilateral salpingooophorectomy, vaginal cystocele or rectocele repair, and placement of a midurethral sling.
Estimated blood loss for the robotic-assisted approach ranged from 25 mL to 300 mL, with a mean loss of 96.8 mL. Average length of hospitalization was 2.6 days. Four cases (5%) were converted to laparotomy because of limited exposure and one intraoperative bladder injury. Other intraoperative complications included small-bowel injury during trocar placement and one ureteral injury. Postoperative complications included one case of ileus and five (6%) vaginal mesh erosions. Three patients developed recurrent prolapse and underwent subsequent correction.
Learning curve could have been measured more precisely
The authors did not specifically measure the learning curve for robotic-assisted sacrocolpopexy, as they took into account the concomitant procedures. For this reason, the decrease in operative time observed after 10 cases may not accurately reflect an improvement in the performance of sacrocolpopexy.
Akl and colleagues consider this detail to be a strength of the study because most women who undergo prolapse surgery have concomitant procedures. However, recording the length of time it took to perform the sacrocolpopexy portion of the procedure would have been more accurate.
The average length of stay approached that of the abdominal route. Length of stay may decline as a surgeon gains experience with the robotic-assisted approach.
Robotic-assisted sacrocolpopexy has a steep learning curve with respect to technique and surgical time.
Does robotic-assisted sacrocolpopexy provide durable support?
Elliott DS, Krambeck AE, Chow GK. Long-term results of robotic assisted laparoscopic sacrocolpopexy for the treatment of high grade vaginal vault prolapse. J Urol. 2006;176(2):655–659.
Among the few recent series reporting long-term outcomes after robotic-assisted sacrocolpopexy is this observational study from the Mayo Clinic. It involved 30 women who underwent the operation for the treatment of Baden Walker grade 4/4 posthysterectomy vaginal vault prolapse. The authors concluded that advanced prolapse can be treated with robotic-assisted sacrocolpopexy with long-term success and minimal complications.
Details of the trial
Of 30 women in this trial, 52% underwent an anti-incontinence procedure at the time of sacrocolpopexy. Women who had multiple vaginal defects or a history of abdominal surgery were excluded from the study.
Average operative time was 3.1 hours (range, 2.15–4.75 hours) in the early phase of development of operative technique (described in the manuscript) but diminished over time to an average of 2.5 hours.
Twenty-nine patients were discharged from the hospital after an overnight stay. Very few immediate postoperative complications were observed. Two patients experienced mild port-site infections that required outpatient treatment, and one patient had persistent vaginal bleeding from the incision made during the anti-incontinence procedure.
Most patients were followed for at least 1 year
The mean follow-up in this study was 24 months (range, 16–39 months). During this period, 21 women were followed for a full year. Long-term observation revealed that the repair of vault prolapse remained successful in 19 of these women.
One patient experienced recurrent prolapse 7 months after surgery. Another developed a rectocele 9 months after sacrocolpopexy. Vaginal mesh erosions occurred in two patients within 6 months after the procedure; both patients were treated with outpatient resection of the exposed mesh, with no recurrence of the prolapse.
Although a larger sample size and longer follow-up would be ideal, this study demonstrates a low rate of recurrent prolapse 1 year after the procedure.
Robotic sacrocolpopexy appears to provide long-term durability for the treatment of advanced vaginal vault prolapse.
Depending on where you practice, you may have as many as three options: abdominal, laparoscopic, or robotic-assisted. Here are basic questions you should address when choosing one:
- How familiar are you with the technique? if the answer is “not much,” you can anticipate that the cost and time required to perform it will be significantly higher.
- Are the appropriate instruments and surgical team available?
- Does the patient have comorbidities? Consider, for example, the fact that she may not be able to tolerate a steep Trendelenberg position—required for the robotic-assisted approach—if she has severe cardiac or pulmonary disease. However, if she has a risk of poor wound healing, a large abdominal incision may not be advisable and postoperative immobility can be risky. if she is obese, laparoscopic or robotic port placement is challenging, but visualization and retraction will be easier. The need for anticoagulation is another consideration, as it will affect estimated blood loss and the choice of an incision, among other things.
- Let’s not forget the patient. Given the pros and cons, what approach does she prefer?
How much do laparoscopic, abdominal, and robotic-assisted sacrocolpopexy cost?
Judd JP, Siddiqui NY, Barnett JC, et al. Cost-minimization analysis of robotic-assisted, laparoscopic, and abdominal sacrocolpopexy. J Minim Invasive Gynecol. 2010;17: 493–499.
This cost-minimization analysis concluded that robotic-assisted sacrocolpopexy incurs the highest hospital charges but is reimbursed by Medicare at a rate similar to reimbursement for the abdominal and laparoscopic routes (TABLE 2).
TABLE 2
Cost of sacrocolpopexy is significant—especially using the robotic approach
| Approach | Cost of a procedure | Operative time, min (range) |
|---|---|---|
| Robotic-assisted | $8,508 | 328 (130–383) |
| Laparoscopic | $7,353 | 269 (97–334) |
| Abdominal | $5,792 | 170 (110–286) |
| Source: Judd JP, Siddiqui NY, Barnett JC, Visco AG, Havrilesky LJ, Wu JM. Cost-minimization analysis of robotic-assisted, laparoscopic, and abdominal sacrocolpopexy. J Minim invasive Gynecol. 2010;17(4):493–499. | ||
The analysis accounted for realistic practices, such as the inclusion of concurrent hysterectomy and other procedures.
Details of the trial
Surgeons from Duke University developed a decision-analysis model in which a hypothetical group of women with advanced vaginal prolapse could choose between one of the three routes of sacrocolpopexy: abdominal, laparoscopic, or robotic-assisted. Researchers postulated two different scenarios:
- the hospital had ownership of a robotic system
- the hospital invested in the initial purchase and maintenance of such a system.
Researchers reviewed the literature to formulate their estimates of operative time, rate of conversion to laparotomy, rate of transfusion, and length of hospital stay. In addition, the costs of initial anesthesia setup, professional fees, per-minute intraoperative fees, and postanesthesia care were applied to each approach. Operating room costs per minute and the cost of disposable items such as drapes, gowns, gloves, and single-use instruments were added. For the robotic approach, the costs of reusable instruments were distributed across 10 operations. Reusable instruments for laparoscopic and abdominal surgery were assumed to incur no additional investment. Last, postoperative care—including laboratory tests, pharmacy usage, and the need for a hospital room—were individualized for each route of surgery and applied to the cost.
Costs were estimated in 2008 US dollars, based on procedure costs incurred at Duke University Medical Center.
Physician reimbursement data were obtained from Medicare reimbursement rates for anesthesia and from surgeon Current Procedural Terminology (CPT) codes specific to each procedure.
Quality-of-life assessments were not measured. Nor was the cost to society of the postoperative loss of productivity and wages for each surgical route. Had these losses been recognized, the authors observed, the cost of robotic surgery may have been lower.
The cost of robotic surgery was equivalent to the cost of laparoscopy in only two instances:
- when the operative time of robotic surgery was reduced to 149 minutes
- when the cost of robotic disposable items was less than $2,132 (reduced from a baseline cost of $3,293).
Robotic sacrocolpopexy is costly. this is an important consideration when implementing new technology. cost-saving scenarios are useful to maximize patient benefit and minimize financial burden.
We want to hear from you! Tell us what you think.
When a woman has advanced prolapse of the anterior vaginal wall, it is highly likely that she has apical prolapse as well. Consider a study by Rooney and associates that determined that clinically significant vault prolapse is present in most women who have anterior vaginal prolapse of stage II or higher.1 For that reason, suspension of the vaginal apex should be considered whenever surgical treatment of anterior wall defects is planned.
Sacrocolpopexy involves suspension of the vaginal vault from the anterior longitudinal ligament of the sacrum, using Y-shaped mesh to augment native tissue (FIGURE).2 It is an effective, durable treatment for vaginal apical prolapse. With a success rate approaching 93%, this procedure has become the gold standard for repair of vault prolapse. Among its advantages are maximization of vaginal depth and preservation of a normal vaginal axis.
Sacrocolpopexy preserves the vaginal axis
With the vaginal vault suspended from the anterior longitudinal
ligament of the sacrum, the normal vaginal axis is preserved
and vaginal depth is maximized.
Sacrocolpopexy can be performed via the abdominal, laparoscopic, or robotic-assisted approach (TABLE 1). Minimally invasive techniques are attractive because they involve faster recovery than abdominal sacrocolpopexy does. Minimally invasive techniques have also advanced to the point that they are both effective and durable. However, these advantages must be weighed against the effort required to learn the techniques, as well as their higher cost.
TABLE 1
How the 3 approaches to sacrocolpopexy compare
| Approach | Advantages and disadvantages |
|---|---|
| Abdominal | Shortest operative time No significant Trendelenburg position required Highest estimated blood loss Longest length of stay Low rate of complications Longest postoperative recovery Well-established long-term durability |
| Laparoscopic | Longer operative time Moderate Trendelenburg position required Lower estimated blood loss Shorter length of stay Surgical technique least similar to abdominal procedure Low rate of complications Shorter postoperative recovery Long-term durability less firmly established |
| Robotic-assisted | Longest operative time Steep Trendelenburg position required Lower estimated blood loss Shorter length of stay Surgical technique resembles that of abdominal approach Low rate of complications Shorter postoperative recovery Long-term durability appears to be good |
In this article, we highlight:
- a comparison of the laparoscopic and abdominal approaches to sacrocolpopexy
- an investigation of the learning curve associated with robotic-assisted sacrocolpopexy
- a study exploring the durability of robotic-assisted repair
- an estimate of the costs associated with each route of operation.
Laparoscopic vs abdominal sacrocolpopexy—how do they compare?
Paraiso MF, Walters MD, Rackley RR, Melek S, Hugney C. Laparoscopic and abdominal sacral colpopexies: a comparative cohort study. Am J Obstet Gynecol. 2005;192(5):1752–1758.
When surgeons at the Cleveland Clinic performed a retrospective cohort study to compare laparoscopic and abdominal sacrocolpopexy, they found significantly longer operative time with the laparoscopic route, with an average difference of 51 minutes (P < .0001). However, the laparoscopic approach was associated with lower blood loss (although there was no difference between groups in hematocrit on postoperative day 1); shorter hospital stay (average of 1.8 days versus 4 days [P < .001]); and comparable rates of intraoperative and postoperative complications.
Details of the trial
Paraiso and colleagues reviewed the medical charts of 56 consecutive patients who had undergone laparoscopic sacrocolpopexy, comparing them with the charts of 61 consecutive patients who had undergone the procedure using the abdominal approach. The operations had been performed between 1998 and 2003 for treatment of posthysterectomy vaginal prolapse.
The groups underwent similar rates of concurrent procedures. The laparotomy group had a significantly higher number of Burch procedures (P = .007), and the laparoscopic group had a significantly higher rate of adhesiolysis (P = .002).
Among the complications noted— which occurred at comparable rates between groups—were cystotomy, enterotomy, need for transfusion, deep-vein thrombosis, ileus, small bowel obstruction, wound infection, ventral hernia, mesh erosion, and recurrent prolapse. One laparoscopic case was converted to laparotomy because of excessive bleeding during the rectopexy portion of the operation.
Laparoscopy may have taken longer than this trial suggests
This study is one of very few well-designed trials comparing laparoscopic sacrocolpopexy to the historical gold standard of abdominal sacrocolpopexy for vault prolapse.
Twenty-eight percent of laparoscopic procedures in this study used tacking devices in lieu of suturing. Had suturing been performed universally, an even greater difference in surgical time may have been observed.
There may also be differences between groups in the durability of the two types of repair, an outcome not included in this particular study.
The laparoscopic approach offers a shorter hospital stay with no increase in intraoperative or postoperative complications, compared with abdominal sacrocolpopexy. However, it entails a significantly longer operative time than the abdominal approach does.
How steep is the learning curve for robotic-assisted sacrocolpopexy?
Akl MN, Long JB, Giles DL, et al. Robotic-assisted sacrocolpopexy: technique and learning curve. Surg Endosc. 2009;23(10):2390–2394.
Akl and coworkers reviewed the medical records of all patients who had undergone robotic-assisted sacrocolpopexy at the Mayo Clinics in Arizona and Florida between 2004 and 2007. All operations were performed by the same four urogynecologists, with an average operative time of 197.9 minutes (standard deviation, ± 66.8 minutes). However, after the first 10 cases, the operative time decreased by 64.3 minutes—a decline of 25.4% (P < .01; 95% confidence interval [CI], 16.1–112.4 minutes).
Details of the trial
Researchers collected baseline information on participants’ age, stage of prolapse, and concomitant procedures. They also gathered data on average operative time, estimated blood loss, intraoperative and postoperative complications, conversion to laparotomy, and length of hospitalization.
Of 80 women who had advanced pelvic organ prolapse (stage III/IV) who underwent robotic-assisted sacrocolpopexy, 88% underwent concomitant robotic and vaginal procedures, including robotic supracervical hysterectomy, Burch procedure, paravaginal repair, lysis of adhesions, bilateral salpingooophorectomy, vaginal cystocele or rectocele repair, and placement of a midurethral sling.
Estimated blood loss for the robotic-assisted approach ranged from 25 mL to 300 mL, with a mean loss of 96.8 mL. Average length of hospitalization was 2.6 days. Four cases (5%) were converted to laparotomy because of limited exposure and one intraoperative bladder injury. Other intraoperative complications included small-bowel injury during trocar placement and one ureteral injury. Postoperative complications included one case of ileus and five (6%) vaginal mesh erosions. Three patients developed recurrent prolapse and underwent subsequent correction.
Learning curve could have been measured more precisely
The authors did not specifically measure the learning curve for robotic-assisted sacrocolpopexy, as they took into account the concomitant procedures. For this reason, the decrease in operative time observed after 10 cases may not accurately reflect an improvement in the performance of sacrocolpopexy.
Akl and colleagues consider this detail to be a strength of the study because most women who undergo prolapse surgery have concomitant procedures. However, recording the length of time it took to perform the sacrocolpopexy portion of the procedure would have been more accurate.
The average length of stay approached that of the abdominal route. Length of stay may decline as a surgeon gains experience with the robotic-assisted approach.
Robotic-assisted sacrocolpopexy has a steep learning curve with respect to technique and surgical time.
Does robotic-assisted sacrocolpopexy provide durable support?
Elliott DS, Krambeck AE, Chow GK. Long-term results of robotic assisted laparoscopic sacrocolpopexy for the treatment of high grade vaginal vault prolapse. J Urol. 2006;176(2):655–659.
Among the few recent series reporting long-term outcomes after robotic-assisted sacrocolpopexy is this observational study from the Mayo Clinic. It involved 30 women who underwent the operation for the treatment of Baden Walker grade 4/4 posthysterectomy vaginal vault prolapse. The authors concluded that advanced prolapse can be treated with robotic-assisted sacrocolpopexy with long-term success and minimal complications.
Details of the trial
Of 30 women in this trial, 52% underwent an anti-incontinence procedure at the time of sacrocolpopexy. Women who had multiple vaginal defects or a history of abdominal surgery were excluded from the study.
Average operative time was 3.1 hours (range, 2.15–4.75 hours) in the early phase of development of operative technique (described in the manuscript) but diminished over time to an average of 2.5 hours.
Twenty-nine patients were discharged from the hospital after an overnight stay. Very few immediate postoperative complications were observed. Two patients experienced mild port-site infections that required outpatient treatment, and one patient had persistent vaginal bleeding from the incision made during the anti-incontinence procedure.
Most patients were followed for at least 1 year
The mean follow-up in this study was 24 months (range, 16–39 months). During this period, 21 women were followed for a full year. Long-term observation revealed that the repair of vault prolapse remained successful in 19 of these women.
One patient experienced recurrent prolapse 7 months after surgery. Another developed a rectocele 9 months after sacrocolpopexy. Vaginal mesh erosions occurred in two patients within 6 months after the procedure; both patients were treated with outpatient resection of the exposed mesh, with no recurrence of the prolapse.
Although a larger sample size and longer follow-up would be ideal, this study demonstrates a low rate of recurrent prolapse 1 year after the procedure.
Robotic sacrocolpopexy appears to provide long-term durability for the treatment of advanced vaginal vault prolapse.
Depending on where you practice, you may have as many as three options: abdominal, laparoscopic, or robotic-assisted. Here are basic questions you should address when choosing one:
- How familiar are you with the technique? if the answer is “not much,” you can anticipate that the cost and time required to perform it will be significantly higher.
- Are the appropriate instruments and surgical team available?
- Does the patient have comorbidities? Consider, for example, the fact that she may not be able to tolerate a steep Trendelenberg position—required for the robotic-assisted approach—if she has severe cardiac or pulmonary disease. However, if she has a risk of poor wound healing, a large abdominal incision may not be advisable and postoperative immobility can be risky. if she is obese, laparoscopic or robotic port placement is challenging, but visualization and retraction will be easier. The need for anticoagulation is another consideration, as it will affect estimated blood loss and the choice of an incision, among other things.
- Let’s not forget the patient. Given the pros and cons, what approach does she prefer?
How much do laparoscopic, abdominal, and robotic-assisted sacrocolpopexy cost?
Judd JP, Siddiqui NY, Barnett JC, et al. Cost-minimization analysis of robotic-assisted, laparoscopic, and abdominal sacrocolpopexy. J Minim Invasive Gynecol. 2010;17: 493–499.
This cost-minimization analysis concluded that robotic-assisted sacrocolpopexy incurs the highest hospital charges but is reimbursed by Medicare at a rate similar to reimbursement for the abdominal and laparoscopic routes (TABLE 2).
TABLE 2
Cost of sacrocolpopexy is significant—especially using the robotic approach
| Approach | Cost of a procedure | Operative time, min (range) |
|---|---|---|
| Robotic-assisted | $8,508 | 328 (130–383) |
| Laparoscopic | $7,353 | 269 (97–334) |
| Abdominal | $5,792 | 170 (110–286) |
| Source: Judd JP, Siddiqui NY, Barnett JC, Visco AG, Havrilesky LJ, Wu JM. Cost-minimization analysis of robotic-assisted, laparoscopic, and abdominal sacrocolpopexy. J Minim invasive Gynecol. 2010;17(4):493–499. | ||
The analysis accounted for realistic practices, such as the inclusion of concurrent hysterectomy and other procedures.
Details of the trial
Surgeons from Duke University developed a decision-analysis model in which a hypothetical group of women with advanced vaginal prolapse could choose between one of the three routes of sacrocolpopexy: abdominal, laparoscopic, or robotic-assisted. Researchers postulated two different scenarios:
- the hospital had ownership of a robotic system
- the hospital invested in the initial purchase and maintenance of such a system.
Researchers reviewed the literature to formulate their estimates of operative time, rate of conversion to laparotomy, rate of transfusion, and length of hospital stay. In addition, the costs of initial anesthesia setup, professional fees, per-minute intraoperative fees, and postanesthesia care were applied to each approach. Operating room costs per minute and the cost of disposable items such as drapes, gowns, gloves, and single-use instruments were added. For the robotic approach, the costs of reusable instruments were distributed across 10 operations. Reusable instruments for laparoscopic and abdominal surgery were assumed to incur no additional investment. Last, postoperative care—including laboratory tests, pharmacy usage, and the need for a hospital room—were individualized for each route of surgery and applied to the cost.
Costs were estimated in 2008 US dollars, based on procedure costs incurred at Duke University Medical Center.
Physician reimbursement data were obtained from Medicare reimbursement rates for anesthesia and from surgeon Current Procedural Terminology (CPT) codes specific to each procedure.
Quality-of-life assessments were not measured. Nor was the cost to society of the postoperative loss of productivity and wages for each surgical route. Had these losses been recognized, the authors observed, the cost of robotic surgery may have been lower.
The cost of robotic surgery was equivalent to the cost of laparoscopy in only two instances:
- when the operative time of robotic surgery was reduced to 149 minutes
- when the cost of robotic disposable items was less than $2,132 (reduced from a baseline cost of $3,293).
Robotic sacrocolpopexy is costly. this is an important consideration when implementing new technology. cost-saving scenarios are useful to maximize patient benefit and minimize financial burden.
We want to hear from you! Tell us what you think.
Skilled US imaging of the adnexae: The non-neoplastic mass
Part 1: A Starting Point (September 2010)
Part 3: Ovarian neoplasms (November 2010)
Part 4: The fallopian tubes (December 2010)
Scanning the ovaries is no simple task. As we mentioned in Part 1 of this four-part series, the practitioner must use the right equipment, take basic preparatory steps, be watchful for clues in the history, and reach a conclusion about what he or she sees. Not only that: The ultrasonographer must be extraordinarily vigilant, paying close attention to multiple characteristics of any mass, from thickness of the wall to the presence of papillations or a blood supply—signs of potential malignancy.
In this article, we detail the traits of various types of non-neoplastic ovarian masses, including:
- functional cysts—follicles, the corpus luteum, and theca lutein cysts
- nonfunctional cysts—serous masses and endometriomas
- cystadenofibromas. Although these masses are usually categorized histo-logically as neoplasms, we include them here due to their almost daily appearance in a busy gynecologic ultrasonographic (US) facility.
In Part 3, we will cover ovarian neoplasms, and in Part 4, our focus will be tubal entities such as ectopic pregnancy and torsion.
FIGURE 1 What is a mass made of? 6 morphologic building blocks
Take an inventory of the mass
Any adnexal mass should be assessed in light of its essential characteristics (Figure 1).
Wall structure. Pay attention to thickness. We use an arbitrary cutoff of 4 mm, giving extra scrutiny to thicknesses exceeding that measurement. In our experience, the thicker the wall, the more likely the mass is to be malignant.
Septation and loculation. A mass is typically unilocular or multilocular. Multilocularity is more common in tumors of low malignant potential and malignant neoplasms.
Papillation. Any internal or external papillae or excrescences should draw your attention. Papillarity in an ovarian mass renders that mass suspicious for malignancy.
Measure (height and width) any papillae that are identified, and document them. Because papillae are associated with ovarian malignancy, further assessment is warranted immediately. The first step is determining whether the papillations contain blood vessels—a task for which color and power Doppler are helpful. We prefer power Doppler because it is more sensitive, detecting blood-flow velocity in the lowest detectable range of 2 cm/s, and because it is not directionally influenced.
Papillae that contain blood vessels with detectable flow are suspicious for malignancy.
Exacoustos and colleagues found that papillae as large as 15 mm in height and 10 mm in width (base) were present in 48% of borderline ovarian tumors but in only 4% of benign and 4% of malignant tumors. However, when the intracystic solid tissue exceeded those dimensions, the lesions were present in 48% of invasive ovarian tumors, 18% of borderline ovarian tumors, and 7% of benign masses.1
Internal echo-structure. A mass can be anechoic, a finding that usually indicates the presence of clear fluid. Mostly solid masses are echogenic. And masses that contain particulate matter, such as blood, cellular matter, or even mucous material, usually have echogenicity of a low level, often described as a “ground-glass” appearance. A mass can also have mixed echo-genicity, a finding usually found in cases involving teratoma or malignancy.
Shadowing. If it is present, it may signify the presence of an extremely dense, solid tissue, such as bone or calcification. The diagnosis of a benign teratoma (i.e., dermoid cyst) should be entertained if shadowing is present in a hyperechoic nodule or mass. Malignant masses very rarely, if ever, display frank shadowing.
Overall appearance. On rare occasions, a bizarre shape or “complex” appearance (as it is termed in most radiology reports) may indicate a malignant mass. More likely it indicates the presence of a teratoma, cystadenoma, or even an atypical corpus luteum. In some reports generated by US laboratories, the term “complex” is applied to all structures other than simple cysts.
Size. The size of a mass can be misleading, as small ovarian lesions with the appropriate sonographic characteristics may be malignant and some larger ones without those characteristics may not be. However, it is understood that the larger an ovarian lesion, the more likely it is a tumor. One important distinction: The amount of fluid in a cystic structure or the amount of old blood in an endometrioma is not the disease process…it is the byproduct of the process. So an 8-cm endometrioma may create fewer pain or fertility issues than a 2- or 3-cm endometrioma. Similarly, the amount of “chocolate” fluid is not automatically indicative of the amount of active endometriotic glands or their sequelae!2
Ascites. If it is present, it should be recorded and investigated further because it may be caused by a malignant intra-abdominal tumor.
Motion tenderness. If the to-and-fro movement of the vaginal probe elicits any motion tenderness, it, too, should be documented. It may be a sign of pelvic peritonitis. In such cases, an “ominous appearing” adnexal finding may represent an inflammatory, rather than malignant, mass.
One of the components of extensive evaluation of the adnexae in general and ovaries in particular is color or power Doppler interrogation—or both.
Tumors contain a relatively large number of pathologic blood vessels that lack the muscular layer found in normal blood vessels and, as a result, demonstrate lower resistance to flow. Diastolic flow is high in these vessels, and resistance and pulsatility indices are low.
We also pay attention when these blood vessels have a tortuous appearance, changes in caliber, anastomoses, and vascular lakes.3 The more tortuous the vessels, with multiple inter-vessel connections and dilatations with changing calibers, the greater the risk of malignancy.4 No less important is the presence of a vessel within a “complex” ovarian mass. A centrally located vessel (also called a “lead vessel”) is suspicious for malignancy.5
A gallery of non-neoplastic ovarian masses
Non-neoplastic cysts are, by far, the most common structures of the ovary. They may be functional, as in the case of the follicles, corpus luteum, and theca lutein cysts, or they may be nonfunctional, as in serous cysts and endometriomas. (As we noted in Part 1, do not call the follicles and corpus luteum “cysts” because this designation suggests pathology.)6
FIGURE 2 Simple cyst
This cyst is anechoic and unilocular with thin walls and no papillae.
Functional cysts
Functional cysts, also known as “simple” cysts, may grow as large as 4 to 5 cm in diameter (Figure 2). They are typically unilocular, anechoic, and thin-walled, with no papillae, and almost never malignant. They usually resolve and require no treatment unless rupture or torsion occurs. Except for the corpus luteum, they have no increased blood flow, and need be viewed only by transvaginal ultrasonography (TVS).
The corpus luteum also can be recognized by TVS. It can exhibit any of a variety of internal structures and echo patterns, due to the multitude of shapes of the blood or clot that can be seen within it (Figure 3).
FIGURE 3 Corpus luteum
A–C. Gray-scale, color Doppler, and power Doppler images, respectively, of a typical corpus luteum. B and C show the enveloping vessels, or “ring of fire.” D. A rather typical gray-scale appearance with a mesh-like, linear internal texture. E. A common feature of the corpus luteum is a linear interphase (arrow) between the clot (c) and the liquified serum (s).
The corpus luteum is typically enveloped by blood vessels, visible on color Doppler as what is called a “ring of fire.” It regresses without intervention. In hyperstimulated ovaries, however, more than one may be present; this poses a real diagnostic challenge when ectopic pregnancy is suspected because it is difficult to differentiate the two entities.
Because the corpus luteum can sometimes resemble some types of ovarian tumors on TVS, imaging during the secretory phase of the cycle in a woman of reproductive age is not ideal. Instead, she should be scanned (or rescanned) between days 5 and 9 of the cycle.
FIGURE 4 Hormonally stimulated ovaries
A, B. The right and left ovaries stimulated by follicle-stimulating hormone preparation (arrow points to hilus). C. An ovary stimulated by clomiphene.
Lutein cysts may reach 5 to 10 cm in diameter. They generally have a thick wall, are multilocular, and typically occur after hormonal induction of ovulation (Figure 4). They also can occur in diabetes, molar pregnancy, and hydrops fetalis. We have seen a unilateral theca lutein cyst in a normal pregnancy (Figure 5). No treatment is necessary unless rupture or torsion occurs.
FIGURE 5 Lutein cysts
A–C. The typical “stained glass” appearance of three lutein cysts of the right ovary in a pregnant patient. D. Color Doppler image of the ovary demonstrating high-velocity flow (peak systolic velocity of 20.4 cm/s).
Serous cysts
These cysts can reach 4 cm in diameter, have smooth walls with no papillae, are unilocular, and occur most often during menopause. No pathological blood flow is visible in their walls. Most gynecologists follow them (Figure 6).1,7
FIGURE 6 Serous cyst
A. Right ovary containing the cyst. B. Normal left ovary. C. Power Doppler interrogation showing no particular flow in the walls of a serous cyst.
Endometriomas
After the simple cyst, the endometrioma is the most prevalent ovarian or adnexal cyst (Figure 7). It usually has a thick wall and is filled with homogeneous fluid with low-level echo-genicity. It can reach 10 cm in size, and many are bilateral. It is sometimes called a “chocolate” cyst because of its dark blood content.
FIGURE 7 Endometriomas
Endometriomas have low echogenicity. A. Unilateral, unilocular cyst with thin walls. B. Bilateral endometriomas. C. Blood flow in a solid or papillary component of the endometrioma is an occasional finding. It should be investigated further because of the risk that it represents endometrioid cancer.
Endometriomas do not resolve; they usually require surgical excision, although very small ones wholly contained within an ovary are often managed medically or expectantly.
These masses rarely (<1%) give rise to endometrioid carcinoma. Should an endometrioma contain papillae with blood vessels, it is extremely suspicious for endometrioid cancer.
FIGURE 8 Cystic fibromas
A. Sonographic image shows a thin wall and hyperechoic, small mural nodules. B. Macroscopic appearance of an area of internal papillary excrescences. C. Measurement of the small, mural nodules. D. Lack of blood flow in the small papillae, a typical finding on color or power Doppler. E, F. Blood flow in the wall of the cyst and in the mural nodules.
Ovarian fibromas
A fibroma is a slow-growing, benign, solid ovarian tumor. It usually has a cystic component and then is called a cystadenofibroma.
The cystic variety is filled with anechoic fluid and has a thin wall. However, its pathognomonic feature is the small (2–3 mm), extremely hyperechoic mural nodules (papillae) it contains (Figure 8A–C). In the overwhelming majority of cases, no blood vessels are detectable, and the mass is unilocular (Figure 8D–E). It can be recognized in the ovary by the semilunar shape of the tissue surrounding it (crescent sign). The differential diagnosis includes the simple (serous) cyst.
The solid fibroma has a myometrium-like texture, with few or no detectable blood vessels in the stroma. The differential diagnosis includes the Brenner tumor and the Krukenberg tumor.
According to a technology assessment from the Agency for Healthcare Research and Quality (AHRQ), “conventional gray-scale ultrasonography is the most common imaging modality used to differentiate benign from malignant adnexal masses. Especially with the advent of high-frequency transvaginal probes, the quality of the images allows description of the gross anatomic features of the lesion.”8 This descriptive ability is limited, however, “by the great variability of macroscopic characteristics of both benign and malignant masses. Furthermore, the technique is operator dependent.”8
To overcome these challenges, some experts have developed ultrasonographic (US) morphologic scoring systems, which assign a value to individual characteristics. Lerner and colleagues devised a 4-point system:
| Characteristic | Points | |||
|---|---|---|---|---|
| 0 | 1 | 2 | 3 | |
| Wall structure | Smooth or small irregularities (<3 mm) | Solid or not applicable | Papillarities larger than 3 mm | |
| Shadowing | Yes | No | ||
| Septation | None or thin (<3 mm) | Thick (≥3 mm) | ||
| Echogenicity | Sonolucent or low-level echo or echogenic core | Mixed or high | ||
The mean point value for benign masses was 1.8; for tumors of low malignant potential it was 3.9; and for malignant tumors it was 5.6 (P < .0005). Lerner and associates proposed a cutoff of 3. A score of 3 or higher, they felt, would be most predictive of malignancy, with sensitivity of 96.8% and specificity of 77%. Positive and negative predictive values were 29.4% and 99.6%, respectively.9
Almost all published scoring systems are based upon or derived from one reported by Sassone and coworkers.10 The most important and practical feature of all scoring systems is their ability to rule out malignancy.
Morphology and Doppler: A synergistic combination
As the same AHRQ report points out, “all of the diagnostic tests and scoring systems we evaluated exhibited a trade-off between sensitivity and specificity—studies of a given test that reported higher sensitivity had lower specificity, and vice versa.”8 Among evaluation methods, the combination of US morphology scores and Doppler imaging achieved the highest pooled sensitivity and specificity scores in distinguishing benign and malignant adnexal masses in postmenopausal women: 86% and 91%, respectively, according to the AHRQ report.8
Compare these figures with those of:
- Bimanual pelvic examination (45% and 90%, respectively)
- Doppler resistance index (72% and 90%)
- Doppler pulsatility index (80% and 73%)
- presence of blood vessels (88% and 78%).
The combination of US morphology scores and Doppler was comparable to the pooled sensitivity and specificity of magnetic resonance imaging (91% and 88%, respectively) and superior to computed tomography (90% and 75%, respectively).
Why the need to know?
Discrimination between benign and malignant masses serves a number of purposes, depending on the setting.
For example, if a symptomatic woman is found to have an adnexal mass, it is important to identify the type of mass causing the symptoms to determine the best course of treatment. And because surgery may be one of the treatment options, it is helpful to know whether a mass is likely to be malignant so that the patient can be referred to a specialist or center that has optimal surgical expertise.8
Some asymptomatic masses may be identified during the annual bimanual pelvic examination recommended by ACOG or during pregnancy-related US imaging. In this setting, it is important to ascertain whether the mass is likely to be malignant so that the patient can be referred to a specialist, if necessary. In addition, thorough assessment of the mass can help “avoid unnecessary diagnostic procedures, including surgery, and anxiety in women with asymptomatic, nonmalignant conditions. In some cases, there may be a rationale for removing certain asymptomatic benign lesions, including prevention of malignant transformation; prevention of ovarian torsion”; and prevention of rupture. Surgery may also be appropriate to avert the need for more complicated surgery in the future or to enhance fertility.8 —Janelle Yates, Senior Editor
Stay tuned!
Next issue, in Part 3 of this series, we will review the use of imaging in the investigation of ovarian neoplasms, both benign and malignant, with an abundance of US images to accompany our discussion.
We want to hear from you! Tell us what you think.
1. Exacoustos C, Romanini ME, Rinaldo D, et al. Preoperative sonographic features of borderline ovarian tumors. Ultrasound Obstet Gynecol. 2004;25(1):50-59.
2. Rulin MC, Preston AL. Adnexal masses in postmenopausal women. Obstet Gynecol. 1987;70(4):578-581.
3. Timor-Tritsch IE, Goldstein SR. The complexity of a complex mass and the simplicity of a simple cyst. J Ultraound Med. 2005;24(3):255-258.
4. Sladkevicius P, Jokubkiene L, Valentin L. Contribution of morphological assessment of the vessel tree by three-dimensional ultrasound to a correct diagnos is of malignancy in ovarian masses. Ultrasound Obstet Gynecol. 2007;30(6):874-882.
5. Testa AC, Mancari R, Di Legge A, et al. The “lead vessel”: a vascular ultrasound feature of metastasis in the ovaries. Ultrasound Obstet Gynecol. 2008;31:218-221.
6. Goldstein SR. Postmenopausal adnexal cysts: how clinical management has evolved. Am J Obstet Gynecol. 1996;175(6):1496-1501.
7. Levine D, Gosink BB, Wolf S, Feldesman MR, Pretorius D. Simple adnexal cysts: the natural history in postmenopausal women. Radiology. 1992;184(3):653-659.
8. Myers ER, Bastian LA, Havrilesky LJ, et al. Management of adnexal mass. Evidence Report Technol Assess. 2006;Feb;(130):1-145.
9. Lerner JP, Timor-Tritsch IE, Federman A, Abramovich G. Transvaginal ultrasonographic characterization of ovarian masses with an improved, weighted scoring system. Am J Obstet Gynecol. 1994;170(1 Pt 1):81-85.
10. Sassone AM, Timor-Tritsch IE, Artner A, et al. Transvaginal sonographic characterization of ovarian disease: evaluation of a new scoring system to predict ovarian malignancy. Obstet Gynecol. 2001;78:70-76.
Part 1: A Starting Point (September 2010)
Part 3: Ovarian neoplasms (November 2010)
Part 4: The fallopian tubes (December 2010)
Scanning the ovaries is no simple task. As we mentioned in Part 1 of this four-part series, the practitioner must use the right equipment, take basic preparatory steps, be watchful for clues in the history, and reach a conclusion about what he or she sees. Not only that: The ultrasonographer must be extraordinarily vigilant, paying close attention to multiple characteristics of any mass, from thickness of the wall to the presence of papillations or a blood supply—signs of potential malignancy.
In this article, we detail the traits of various types of non-neoplastic ovarian masses, including:
- functional cysts—follicles, the corpus luteum, and theca lutein cysts
- nonfunctional cysts—serous masses and endometriomas
- cystadenofibromas. Although these masses are usually categorized histo-logically as neoplasms, we include them here due to their almost daily appearance in a busy gynecologic ultrasonographic (US) facility.
In Part 3, we will cover ovarian neoplasms, and in Part 4, our focus will be tubal entities such as ectopic pregnancy and torsion.
FIGURE 1 What is a mass made of? 6 morphologic building blocks
Take an inventory of the mass
Any adnexal mass should be assessed in light of its essential characteristics (Figure 1).
Wall structure. Pay attention to thickness. We use an arbitrary cutoff of 4 mm, giving extra scrutiny to thicknesses exceeding that measurement. In our experience, the thicker the wall, the more likely the mass is to be malignant.
Septation and loculation. A mass is typically unilocular or multilocular. Multilocularity is more common in tumors of low malignant potential and malignant neoplasms.
Papillation. Any internal or external papillae or excrescences should draw your attention. Papillarity in an ovarian mass renders that mass suspicious for malignancy.
Measure (height and width) any papillae that are identified, and document them. Because papillae are associated with ovarian malignancy, further assessment is warranted immediately. The first step is determining whether the papillations contain blood vessels—a task for which color and power Doppler are helpful. We prefer power Doppler because it is more sensitive, detecting blood-flow velocity in the lowest detectable range of 2 cm/s, and because it is not directionally influenced.
Papillae that contain blood vessels with detectable flow are suspicious for malignancy.
Exacoustos and colleagues found that papillae as large as 15 mm in height and 10 mm in width (base) were present in 48% of borderline ovarian tumors but in only 4% of benign and 4% of malignant tumors. However, when the intracystic solid tissue exceeded those dimensions, the lesions were present in 48% of invasive ovarian tumors, 18% of borderline ovarian tumors, and 7% of benign masses.1
Internal echo-structure. A mass can be anechoic, a finding that usually indicates the presence of clear fluid. Mostly solid masses are echogenic. And masses that contain particulate matter, such as blood, cellular matter, or even mucous material, usually have echogenicity of a low level, often described as a “ground-glass” appearance. A mass can also have mixed echo-genicity, a finding usually found in cases involving teratoma or malignancy.
Shadowing. If it is present, it may signify the presence of an extremely dense, solid tissue, such as bone or calcification. The diagnosis of a benign teratoma (i.e., dermoid cyst) should be entertained if shadowing is present in a hyperechoic nodule or mass. Malignant masses very rarely, if ever, display frank shadowing.
Overall appearance. On rare occasions, a bizarre shape or “complex” appearance (as it is termed in most radiology reports) may indicate a malignant mass. More likely it indicates the presence of a teratoma, cystadenoma, or even an atypical corpus luteum. In some reports generated by US laboratories, the term “complex” is applied to all structures other than simple cysts.
Size. The size of a mass can be misleading, as small ovarian lesions with the appropriate sonographic characteristics may be malignant and some larger ones without those characteristics may not be. However, it is understood that the larger an ovarian lesion, the more likely it is a tumor. One important distinction: The amount of fluid in a cystic structure or the amount of old blood in an endometrioma is not the disease process…it is the byproduct of the process. So an 8-cm endometrioma may create fewer pain or fertility issues than a 2- or 3-cm endometrioma. Similarly, the amount of “chocolate” fluid is not automatically indicative of the amount of active endometriotic glands or their sequelae!2
Ascites. If it is present, it should be recorded and investigated further because it may be caused by a malignant intra-abdominal tumor.
Motion tenderness. If the to-and-fro movement of the vaginal probe elicits any motion tenderness, it, too, should be documented. It may be a sign of pelvic peritonitis. In such cases, an “ominous appearing” adnexal finding may represent an inflammatory, rather than malignant, mass.
One of the components of extensive evaluation of the adnexae in general and ovaries in particular is color or power Doppler interrogation—or both.
Tumors contain a relatively large number of pathologic blood vessels that lack the muscular layer found in normal blood vessels and, as a result, demonstrate lower resistance to flow. Diastolic flow is high in these vessels, and resistance and pulsatility indices are low.
We also pay attention when these blood vessels have a tortuous appearance, changes in caliber, anastomoses, and vascular lakes.3 The more tortuous the vessels, with multiple inter-vessel connections and dilatations with changing calibers, the greater the risk of malignancy.4 No less important is the presence of a vessel within a “complex” ovarian mass. A centrally located vessel (also called a “lead vessel”) is suspicious for malignancy.5
A gallery of non-neoplastic ovarian masses
Non-neoplastic cysts are, by far, the most common structures of the ovary. They may be functional, as in the case of the follicles, corpus luteum, and theca lutein cysts, or they may be nonfunctional, as in serous cysts and endometriomas. (As we noted in Part 1, do not call the follicles and corpus luteum “cysts” because this designation suggests pathology.)6
FIGURE 2 Simple cyst
This cyst is anechoic and unilocular with thin walls and no papillae.
Functional cysts
Functional cysts, also known as “simple” cysts, may grow as large as 4 to 5 cm in diameter (Figure 2). They are typically unilocular, anechoic, and thin-walled, with no papillae, and almost never malignant. They usually resolve and require no treatment unless rupture or torsion occurs. Except for the corpus luteum, they have no increased blood flow, and need be viewed only by transvaginal ultrasonography (TVS).
The corpus luteum also can be recognized by TVS. It can exhibit any of a variety of internal structures and echo patterns, due to the multitude of shapes of the blood or clot that can be seen within it (Figure 3).
FIGURE 3 Corpus luteum
A–C. Gray-scale, color Doppler, and power Doppler images, respectively, of a typical corpus luteum. B and C show the enveloping vessels, or “ring of fire.” D. A rather typical gray-scale appearance with a mesh-like, linear internal texture. E. A common feature of the corpus luteum is a linear interphase (arrow) between the clot (c) and the liquified serum (s).
The corpus luteum is typically enveloped by blood vessels, visible on color Doppler as what is called a “ring of fire.” It regresses without intervention. In hyperstimulated ovaries, however, more than one may be present; this poses a real diagnostic challenge when ectopic pregnancy is suspected because it is difficult to differentiate the two entities.
Because the corpus luteum can sometimes resemble some types of ovarian tumors on TVS, imaging during the secretory phase of the cycle in a woman of reproductive age is not ideal. Instead, she should be scanned (or rescanned) between days 5 and 9 of the cycle.
FIGURE 4 Hormonally stimulated ovaries
A, B. The right and left ovaries stimulated by follicle-stimulating hormone preparation (arrow points to hilus). C. An ovary stimulated by clomiphene.
Lutein cysts may reach 5 to 10 cm in diameter. They generally have a thick wall, are multilocular, and typically occur after hormonal induction of ovulation (Figure 4). They also can occur in diabetes, molar pregnancy, and hydrops fetalis. We have seen a unilateral theca lutein cyst in a normal pregnancy (Figure 5). No treatment is necessary unless rupture or torsion occurs.
FIGURE 5 Lutein cysts
A–C. The typical “stained glass” appearance of three lutein cysts of the right ovary in a pregnant patient. D. Color Doppler image of the ovary demonstrating high-velocity flow (peak systolic velocity of 20.4 cm/s).
Serous cysts
These cysts can reach 4 cm in diameter, have smooth walls with no papillae, are unilocular, and occur most often during menopause. No pathological blood flow is visible in their walls. Most gynecologists follow them (Figure 6).1,7
FIGURE 6 Serous cyst
A. Right ovary containing the cyst. B. Normal left ovary. C. Power Doppler interrogation showing no particular flow in the walls of a serous cyst.
Endometriomas
After the simple cyst, the endometrioma is the most prevalent ovarian or adnexal cyst (Figure 7). It usually has a thick wall and is filled with homogeneous fluid with low-level echo-genicity. It can reach 10 cm in size, and many are bilateral. It is sometimes called a “chocolate” cyst because of its dark blood content.
FIGURE 7 Endometriomas
Endometriomas have low echogenicity. A. Unilateral, unilocular cyst with thin walls. B. Bilateral endometriomas. C. Blood flow in a solid or papillary component of the endometrioma is an occasional finding. It should be investigated further because of the risk that it represents endometrioid cancer.
Endometriomas do not resolve; they usually require surgical excision, although very small ones wholly contained within an ovary are often managed medically or expectantly.
These masses rarely (<1%) give rise to endometrioid carcinoma. Should an endometrioma contain papillae with blood vessels, it is extremely suspicious for endometrioid cancer.
FIGURE 8 Cystic fibromas
A. Sonographic image shows a thin wall and hyperechoic, small mural nodules. B. Macroscopic appearance of an area of internal papillary excrescences. C. Measurement of the small, mural nodules. D. Lack of blood flow in the small papillae, a typical finding on color or power Doppler. E, F. Blood flow in the wall of the cyst and in the mural nodules.
Ovarian fibromas
A fibroma is a slow-growing, benign, solid ovarian tumor. It usually has a cystic component and then is called a cystadenofibroma.
The cystic variety is filled with anechoic fluid and has a thin wall. However, its pathognomonic feature is the small (2–3 mm), extremely hyperechoic mural nodules (papillae) it contains (Figure 8A–C). In the overwhelming majority of cases, no blood vessels are detectable, and the mass is unilocular (Figure 8D–E). It can be recognized in the ovary by the semilunar shape of the tissue surrounding it (crescent sign). The differential diagnosis includes the simple (serous) cyst.
The solid fibroma has a myometrium-like texture, with few or no detectable blood vessels in the stroma. The differential diagnosis includes the Brenner tumor and the Krukenberg tumor.
According to a technology assessment from the Agency for Healthcare Research and Quality (AHRQ), “conventional gray-scale ultrasonography is the most common imaging modality used to differentiate benign from malignant adnexal masses. Especially with the advent of high-frequency transvaginal probes, the quality of the images allows description of the gross anatomic features of the lesion.”8 This descriptive ability is limited, however, “by the great variability of macroscopic characteristics of both benign and malignant masses. Furthermore, the technique is operator dependent.”8
To overcome these challenges, some experts have developed ultrasonographic (US) morphologic scoring systems, which assign a value to individual characteristics. Lerner and colleagues devised a 4-point system:
| Characteristic | Points | |||
|---|---|---|---|---|
| 0 | 1 | 2 | 3 | |
| Wall structure | Smooth or small irregularities (<3 mm) | Solid or not applicable | Papillarities larger than 3 mm | |
| Shadowing | Yes | No | ||
| Septation | None or thin (<3 mm) | Thick (≥3 mm) | ||
| Echogenicity | Sonolucent or low-level echo or echogenic core | Mixed or high | ||
The mean point value for benign masses was 1.8; for tumors of low malignant potential it was 3.9; and for malignant tumors it was 5.6 (P < .0005). Lerner and associates proposed a cutoff of 3. A score of 3 or higher, they felt, would be most predictive of malignancy, with sensitivity of 96.8% and specificity of 77%. Positive and negative predictive values were 29.4% and 99.6%, respectively.9
Almost all published scoring systems are based upon or derived from one reported by Sassone and coworkers.10 The most important and practical feature of all scoring systems is their ability to rule out malignancy.
Morphology and Doppler: A synergistic combination
As the same AHRQ report points out, “all of the diagnostic tests and scoring systems we evaluated exhibited a trade-off between sensitivity and specificity—studies of a given test that reported higher sensitivity had lower specificity, and vice versa.”8 Among evaluation methods, the combination of US morphology scores and Doppler imaging achieved the highest pooled sensitivity and specificity scores in distinguishing benign and malignant adnexal masses in postmenopausal women: 86% and 91%, respectively, according to the AHRQ report.8
Compare these figures with those of:
- Bimanual pelvic examination (45% and 90%, respectively)
- Doppler resistance index (72% and 90%)
- Doppler pulsatility index (80% and 73%)
- presence of blood vessels (88% and 78%).
The combination of US morphology scores and Doppler was comparable to the pooled sensitivity and specificity of magnetic resonance imaging (91% and 88%, respectively) and superior to computed tomography (90% and 75%, respectively).
Why the need to know?
Discrimination between benign and malignant masses serves a number of purposes, depending on the setting.
For example, if a symptomatic woman is found to have an adnexal mass, it is important to identify the type of mass causing the symptoms to determine the best course of treatment. And because surgery may be one of the treatment options, it is helpful to know whether a mass is likely to be malignant so that the patient can be referred to a specialist or center that has optimal surgical expertise.8
Some asymptomatic masses may be identified during the annual bimanual pelvic examination recommended by ACOG or during pregnancy-related US imaging. In this setting, it is important to ascertain whether the mass is likely to be malignant so that the patient can be referred to a specialist, if necessary. In addition, thorough assessment of the mass can help “avoid unnecessary diagnostic procedures, including surgery, and anxiety in women with asymptomatic, nonmalignant conditions. In some cases, there may be a rationale for removing certain asymptomatic benign lesions, including prevention of malignant transformation; prevention of ovarian torsion”; and prevention of rupture. Surgery may also be appropriate to avert the need for more complicated surgery in the future or to enhance fertility.8 —Janelle Yates, Senior Editor
Stay tuned!
Next issue, in Part 3 of this series, we will review the use of imaging in the investigation of ovarian neoplasms, both benign and malignant, with an abundance of US images to accompany our discussion.
We want to hear from you! Tell us what you think.
Part 1: A Starting Point (September 2010)
Part 3: Ovarian neoplasms (November 2010)
Part 4: The fallopian tubes (December 2010)
Scanning the ovaries is no simple task. As we mentioned in Part 1 of this four-part series, the practitioner must use the right equipment, take basic preparatory steps, be watchful for clues in the history, and reach a conclusion about what he or she sees. Not only that: The ultrasonographer must be extraordinarily vigilant, paying close attention to multiple characteristics of any mass, from thickness of the wall to the presence of papillations or a blood supply—signs of potential malignancy.
In this article, we detail the traits of various types of non-neoplastic ovarian masses, including:
- functional cysts—follicles, the corpus luteum, and theca lutein cysts
- nonfunctional cysts—serous masses and endometriomas
- cystadenofibromas. Although these masses are usually categorized histo-logically as neoplasms, we include them here due to their almost daily appearance in a busy gynecologic ultrasonographic (US) facility.
In Part 3, we will cover ovarian neoplasms, and in Part 4, our focus will be tubal entities such as ectopic pregnancy and torsion.
FIGURE 1 What is a mass made of? 6 morphologic building blocks
Take an inventory of the mass
Any adnexal mass should be assessed in light of its essential characteristics (Figure 1).
Wall structure. Pay attention to thickness. We use an arbitrary cutoff of 4 mm, giving extra scrutiny to thicknesses exceeding that measurement. In our experience, the thicker the wall, the more likely the mass is to be malignant.
Septation and loculation. A mass is typically unilocular or multilocular. Multilocularity is more common in tumors of low malignant potential and malignant neoplasms.
Papillation. Any internal or external papillae or excrescences should draw your attention. Papillarity in an ovarian mass renders that mass suspicious for malignancy.
Measure (height and width) any papillae that are identified, and document them. Because papillae are associated with ovarian malignancy, further assessment is warranted immediately. The first step is determining whether the papillations contain blood vessels—a task for which color and power Doppler are helpful. We prefer power Doppler because it is more sensitive, detecting blood-flow velocity in the lowest detectable range of 2 cm/s, and because it is not directionally influenced.
Papillae that contain blood vessels with detectable flow are suspicious for malignancy.
Exacoustos and colleagues found that papillae as large as 15 mm in height and 10 mm in width (base) were present in 48% of borderline ovarian tumors but in only 4% of benign and 4% of malignant tumors. However, when the intracystic solid tissue exceeded those dimensions, the lesions were present in 48% of invasive ovarian tumors, 18% of borderline ovarian tumors, and 7% of benign masses.1
Internal echo-structure. A mass can be anechoic, a finding that usually indicates the presence of clear fluid. Mostly solid masses are echogenic. And masses that contain particulate matter, such as blood, cellular matter, or even mucous material, usually have echogenicity of a low level, often described as a “ground-glass” appearance. A mass can also have mixed echo-genicity, a finding usually found in cases involving teratoma or malignancy.
Shadowing. If it is present, it may signify the presence of an extremely dense, solid tissue, such as bone or calcification. The diagnosis of a benign teratoma (i.e., dermoid cyst) should be entertained if shadowing is present in a hyperechoic nodule or mass. Malignant masses very rarely, if ever, display frank shadowing.
Overall appearance. On rare occasions, a bizarre shape or “complex” appearance (as it is termed in most radiology reports) may indicate a malignant mass. More likely it indicates the presence of a teratoma, cystadenoma, or even an atypical corpus luteum. In some reports generated by US laboratories, the term “complex” is applied to all structures other than simple cysts.
Size. The size of a mass can be misleading, as small ovarian lesions with the appropriate sonographic characteristics may be malignant and some larger ones without those characteristics may not be. However, it is understood that the larger an ovarian lesion, the more likely it is a tumor. One important distinction: The amount of fluid in a cystic structure or the amount of old blood in an endometrioma is not the disease process…it is the byproduct of the process. So an 8-cm endometrioma may create fewer pain or fertility issues than a 2- or 3-cm endometrioma. Similarly, the amount of “chocolate” fluid is not automatically indicative of the amount of active endometriotic glands or their sequelae!2
Ascites. If it is present, it should be recorded and investigated further because it may be caused by a malignant intra-abdominal tumor.
Motion tenderness. If the to-and-fro movement of the vaginal probe elicits any motion tenderness, it, too, should be documented. It may be a sign of pelvic peritonitis. In such cases, an “ominous appearing” adnexal finding may represent an inflammatory, rather than malignant, mass.
One of the components of extensive evaluation of the adnexae in general and ovaries in particular is color or power Doppler interrogation—or both.
Tumors contain a relatively large number of pathologic blood vessels that lack the muscular layer found in normal blood vessels and, as a result, demonstrate lower resistance to flow. Diastolic flow is high in these vessels, and resistance and pulsatility indices are low.
We also pay attention when these blood vessels have a tortuous appearance, changes in caliber, anastomoses, and vascular lakes.3 The more tortuous the vessels, with multiple inter-vessel connections and dilatations with changing calibers, the greater the risk of malignancy.4 No less important is the presence of a vessel within a “complex” ovarian mass. A centrally located vessel (also called a “lead vessel”) is suspicious for malignancy.5
A gallery of non-neoplastic ovarian masses
Non-neoplastic cysts are, by far, the most common structures of the ovary. They may be functional, as in the case of the follicles, corpus luteum, and theca lutein cysts, or they may be nonfunctional, as in serous cysts and endometriomas. (As we noted in Part 1, do not call the follicles and corpus luteum “cysts” because this designation suggests pathology.)6
FIGURE 2 Simple cyst
This cyst is anechoic and unilocular with thin walls and no papillae.
Functional cysts
Functional cysts, also known as “simple” cysts, may grow as large as 4 to 5 cm in diameter (Figure 2). They are typically unilocular, anechoic, and thin-walled, with no papillae, and almost never malignant. They usually resolve and require no treatment unless rupture or torsion occurs. Except for the corpus luteum, they have no increased blood flow, and need be viewed only by transvaginal ultrasonography (TVS).
The corpus luteum also can be recognized by TVS. It can exhibit any of a variety of internal structures and echo patterns, due to the multitude of shapes of the blood or clot that can be seen within it (Figure 3).
FIGURE 3 Corpus luteum
A–C. Gray-scale, color Doppler, and power Doppler images, respectively, of a typical corpus luteum. B and C show the enveloping vessels, or “ring of fire.” D. A rather typical gray-scale appearance with a mesh-like, linear internal texture. E. A common feature of the corpus luteum is a linear interphase (arrow) between the clot (c) and the liquified serum (s).
The corpus luteum is typically enveloped by blood vessels, visible on color Doppler as what is called a “ring of fire.” It regresses without intervention. In hyperstimulated ovaries, however, more than one may be present; this poses a real diagnostic challenge when ectopic pregnancy is suspected because it is difficult to differentiate the two entities.
Because the corpus luteum can sometimes resemble some types of ovarian tumors on TVS, imaging during the secretory phase of the cycle in a woman of reproductive age is not ideal. Instead, she should be scanned (or rescanned) between days 5 and 9 of the cycle.
FIGURE 4 Hormonally stimulated ovaries
A, B. The right and left ovaries stimulated by follicle-stimulating hormone preparation (arrow points to hilus). C. An ovary stimulated by clomiphene.
Lutein cysts may reach 5 to 10 cm in diameter. They generally have a thick wall, are multilocular, and typically occur after hormonal induction of ovulation (Figure 4). They also can occur in diabetes, molar pregnancy, and hydrops fetalis. We have seen a unilateral theca lutein cyst in a normal pregnancy (Figure 5). No treatment is necessary unless rupture or torsion occurs.
FIGURE 5 Lutein cysts
A–C. The typical “stained glass” appearance of three lutein cysts of the right ovary in a pregnant patient. D. Color Doppler image of the ovary demonstrating high-velocity flow (peak systolic velocity of 20.4 cm/s).
Serous cysts
These cysts can reach 4 cm in diameter, have smooth walls with no papillae, are unilocular, and occur most often during menopause. No pathological blood flow is visible in their walls. Most gynecologists follow them (Figure 6).1,7
FIGURE 6 Serous cyst
A. Right ovary containing the cyst. B. Normal left ovary. C. Power Doppler interrogation showing no particular flow in the walls of a serous cyst.
Endometriomas
After the simple cyst, the endometrioma is the most prevalent ovarian or adnexal cyst (Figure 7). It usually has a thick wall and is filled with homogeneous fluid with low-level echo-genicity. It can reach 10 cm in size, and many are bilateral. It is sometimes called a “chocolate” cyst because of its dark blood content.
FIGURE 7 Endometriomas
Endometriomas have low echogenicity. A. Unilateral, unilocular cyst with thin walls. B. Bilateral endometriomas. C. Blood flow in a solid or papillary component of the endometrioma is an occasional finding. It should be investigated further because of the risk that it represents endometrioid cancer.
Endometriomas do not resolve; they usually require surgical excision, although very small ones wholly contained within an ovary are often managed medically or expectantly.
These masses rarely (<1%) give rise to endometrioid carcinoma. Should an endometrioma contain papillae with blood vessels, it is extremely suspicious for endometrioid cancer.
FIGURE 8 Cystic fibromas
A. Sonographic image shows a thin wall and hyperechoic, small mural nodules. B. Macroscopic appearance of an area of internal papillary excrescences. C. Measurement of the small, mural nodules. D. Lack of blood flow in the small papillae, a typical finding on color or power Doppler. E, F. Blood flow in the wall of the cyst and in the mural nodules.
Ovarian fibromas
A fibroma is a slow-growing, benign, solid ovarian tumor. It usually has a cystic component and then is called a cystadenofibroma.
The cystic variety is filled with anechoic fluid and has a thin wall. However, its pathognomonic feature is the small (2–3 mm), extremely hyperechoic mural nodules (papillae) it contains (Figure 8A–C). In the overwhelming majority of cases, no blood vessels are detectable, and the mass is unilocular (Figure 8D–E). It can be recognized in the ovary by the semilunar shape of the tissue surrounding it (crescent sign). The differential diagnosis includes the simple (serous) cyst.
The solid fibroma has a myometrium-like texture, with few or no detectable blood vessels in the stroma. The differential diagnosis includes the Brenner tumor and the Krukenberg tumor.
According to a technology assessment from the Agency for Healthcare Research and Quality (AHRQ), “conventional gray-scale ultrasonography is the most common imaging modality used to differentiate benign from malignant adnexal masses. Especially with the advent of high-frequency transvaginal probes, the quality of the images allows description of the gross anatomic features of the lesion.”8 This descriptive ability is limited, however, “by the great variability of macroscopic characteristics of both benign and malignant masses. Furthermore, the technique is operator dependent.”8
To overcome these challenges, some experts have developed ultrasonographic (US) morphologic scoring systems, which assign a value to individual characteristics. Lerner and colleagues devised a 4-point system:
| Characteristic | Points | |||
|---|---|---|---|---|
| 0 | 1 | 2 | 3 | |
| Wall structure | Smooth or small irregularities (<3 mm) | Solid or not applicable | Papillarities larger than 3 mm | |
| Shadowing | Yes | No | ||
| Septation | None or thin (<3 mm) | Thick (≥3 mm) | ||
| Echogenicity | Sonolucent or low-level echo or echogenic core | Mixed or high | ||
The mean point value for benign masses was 1.8; for tumors of low malignant potential it was 3.9; and for malignant tumors it was 5.6 (P < .0005). Lerner and associates proposed a cutoff of 3. A score of 3 or higher, they felt, would be most predictive of malignancy, with sensitivity of 96.8% and specificity of 77%. Positive and negative predictive values were 29.4% and 99.6%, respectively.9
Almost all published scoring systems are based upon or derived from one reported by Sassone and coworkers.10 The most important and practical feature of all scoring systems is their ability to rule out malignancy.
Morphology and Doppler: A synergistic combination
As the same AHRQ report points out, “all of the diagnostic tests and scoring systems we evaluated exhibited a trade-off between sensitivity and specificity—studies of a given test that reported higher sensitivity had lower specificity, and vice versa.”8 Among evaluation methods, the combination of US morphology scores and Doppler imaging achieved the highest pooled sensitivity and specificity scores in distinguishing benign and malignant adnexal masses in postmenopausal women: 86% and 91%, respectively, according to the AHRQ report.8
Compare these figures with those of:
- Bimanual pelvic examination (45% and 90%, respectively)
- Doppler resistance index (72% and 90%)
- Doppler pulsatility index (80% and 73%)
- presence of blood vessels (88% and 78%).
The combination of US morphology scores and Doppler was comparable to the pooled sensitivity and specificity of magnetic resonance imaging (91% and 88%, respectively) and superior to computed tomography (90% and 75%, respectively).
Why the need to know?
Discrimination between benign and malignant masses serves a number of purposes, depending on the setting.
For example, if a symptomatic woman is found to have an adnexal mass, it is important to identify the type of mass causing the symptoms to determine the best course of treatment. And because surgery may be one of the treatment options, it is helpful to know whether a mass is likely to be malignant so that the patient can be referred to a specialist or center that has optimal surgical expertise.8
Some asymptomatic masses may be identified during the annual bimanual pelvic examination recommended by ACOG or during pregnancy-related US imaging. In this setting, it is important to ascertain whether the mass is likely to be malignant so that the patient can be referred to a specialist, if necessary. In addition, thorough assessment of the mass can help “avoid unnecessary diagnostic procedures, including surgery, and anxiety in women with asymptomatic, nonmalignant conditions. In some cases, there may be a rationale for removing certain asymptomatic benign lesions, including prevention of malignant transformation; prevention of ovarian torsion”; and prevention of rupture. Surgery may also be appropriate to avert the need for more complicated surgery in the future or to enhance fertility.8 —Janelle Yates, Senior Editor
Stay tuned!
Next issue, in Part 3 of this series, we will review the use of imaging in the investigation of ovarian neoplasms, both benign and malignant, with an abundance of US images to accompany our discussion.
We want to hear from you! Tell us what you think.
1. Exacoustos C, Romanini ME, Rinaldo D, et al. Preoperative sonographic features of borderline ovarian tumors. Ultrasound Obstet Gynecol. 2004;25(1):50-59.
2. Rulin MC, Preston AL. Adnexal masses in postmenopausal women. Obstet Gynecol. 1987;70(4):578-581.
3. Timor-Tritsch IE, Goldstein SR. The complexity of a complex mass and the simplicity of a simple cyst. J Ultraound Med. 2005;24(3):255-258.
4. Sladkevicius P, Jokubkiene L, Valentin L. Contribution of morphological assessment of the vessel tree by three-dimensional ultrasound to a correct diagnos is of malignancy in ovarian masses. Ultrasound Obstet Gynecol. 2007;30(6):874-882.
5. Testa AC, Mancari R, Di Legge A, et al. The “lead vessel”: a vascular ultrasound feature of metastasis in the ovaries. Ultrasound Obstet Gynecol. 2008;31:218-221.
6. Goldstein SR. Postmenopausal adnexal cysts: how clinical management has evolved. Am J Obstet Gynecol. 1996;175(6):1496-1501.
7. Levine D, Gosink BB, Wolf S, Feldesman MR, Pretorius D. Simple adnexal cysts: the natural history in postmenopausal women. Radiology. 1992;184(3):653-659.
8. Myers ER, Bastian LA, Havrilesky LJ, et al. Management of adnexal mass. Evidence Report Technol Assess. 2006;Feb;(130):1-145.
9. Lerner JP, Timor-Tritsch IE, Federman A, Abramovich G. Transvaginal ultrasonographic characterization of ovarian masses with an improved, weighted scoring system. Am J Obstet Gynecol. 1994;170(1 Pt 1):81-85.
10. Sassone AM, Timor-Tritsch IE, Artner A, et al. Transvaginal sonographic characterization of ovarian disease: evaluation of a new scoring system to predict ovarian malignancy. Obstet Gynecol. 2001;78:70-76.
1. Exacoustos C, Romanini ME, Rinaldo D, et al. Preoperative sonographic features of borderline ovarian tumors. Ultrasound Obstet Gynecol. 2004;25(1):50-59.
2. Rulin MC, Preston AL. Adnexal masses in postmenopausal women. Obstet Gynecol. 1987;70(4):578-581.
3. Timor-Tritsch IE, Goldstein SR. The complexity of a complex mass and the simplicity of a simple cyst. J Ultraound Med. 2005;24(3):255-258.
4. Sladkevicius P, Jokubkiene L, Valentin L. Contribution of morphological assessment of the vessel tree by three-dimensional ultrasound to a correct diagnos is of malignancy in ovarian masses. Ultrasound Obstet Gynecol. 2007;30(6):874-882.
5. Testa AC, Mancari R, Di Legge A, et al. The “lead vessel”: a vascular ultrasound feature of metastasis in the ovaries. Ultrasound Obstet Gynecol. 2008;31:218-221.
6. Goldstein SR. Postmenopausal adnexal cysts: how clinical management has evolved. Am J Obstet Gynecol. 1996;175(6):1496-1501.
7. Levine D, Gosink BB, Wolf S, Feldesman MR, Pretorius D. Simple adnexal cysts: the natural history in postmenopausal women. Radiology. 1992;184(3):653-659.
8. Myers ER, Bastian LA, Havrilesky LJ, et al. Management of adnexal mass. Evidence Report Technol Assess. 2006;Feb;(130):1-145.
9. Lerner JP, Timor-Tritsch IE, Federman A, Abramovich G. Transvaginal ultrasonographic characterization of ovarian masses with an improved, weighted scoring system. Am J Obstet Gynecol. 1994;170(1 Pt 1):81-85.
10. Sassone AM, Timor-Tritsch IE, Artner A, et al. Transvaginal sonographic characterization of ovarian disease: evaluation of a new scoring system to predict ovarian malignancy. Obstet Gynecol. 2001;78:70-76.