Applied Evidence

Diabetes mellitus is a complex, progressive disease that affects every family physician’s practice. Major diabetes organizations recommend that treatment be ongoing and progressive in order to control the disease. The American Diabetes Association (ADA), the European Association for the Study of Diabetes (EASD), and the American Association of Clinical Endocrinologists recommend that patients be assessed every 2 to 3 months after diagnosis and that treatment should be intensified if the patient is not meeting treatment goals.^1,2 Using this approach, all people with type 2 diabetes could be on insulin one year after diagnosis.^1,2 

While many family physicians have become comfortable with using once-daily basal insulin such as glargine or detemir, what to do after basal insulin is much more complex. This review builds upon an earlier article in this journal, “Insulin for type 2 diabetes: How and when to get started,”³ by explaining 3 strategies to consider when basal insulin alone isn't enough.

3 main strategies for intensifying treatment

Basal insulin is indicated for patients who have glucose toxicity and persistently elevated hemoglobin A1c (HbA1c) despite using 2 or more oral agents, or for those who have not achieved glucose goals one year into treatment.^3,4 ADA/EASD recommends initiating a weight-based approach for basal insulin therapy based on initial HbA1c levels >7% or >8%.⁴ Instructing and encouraging patients to titrate their own insulin dose based on fasting glucose readings provides greater and faster glucose control.^1,2

Despite these attempts, some patients will not reach their glucose goals with basal insulin. When intensifying treatment beyond basal insulin therapy, patient preference, cost-effectiveness, safety, tolerability, glycemic efficacy, risk of hypoglycemia, effects on cardiovascular risk factors, and other non-glycemic effects should be considered in the shared decision-making process. There are 3 main strategies for intensifying treatment:
1. Basal plus incretin therapy. Add a newer injectable agent such as a glucagon-like peptide 1 receptor agonist (GLP-1RA).
2. Basal plus one strategy. Add prandial insulin prior to the largest meal of the day.
3. Basal-bolus combination. Add insulin prior to all meals.

TABLE 1^5-8 provides details of several studies that have documented the efficacy of these 3 strategies.

CLICK IMAGE TO ENLARGE

Monitoring blood glucose to guide the way
Blood glucose monitoring using either a 7-point glucose monitoring technique or staggered glucose checks should guide insulin intensification. A 7-point glucose profile includes pre-meal and post-meal readings for 3 meals a day and an additional bedtime reading.⁹ This is typically performed for 3 to 7 days prior to an appointment and provides an estimate of a typical full day’s glucose pattern.

Staggered monitoring includes a pair of glucose checks taken immediately before and typically 90 minutes after a meal. This is assigned to a different meal each day in order to obtain the same information as is achieved with 7-point monitoring, but with fewer checks on any given day. It may take up to 2 to 3 weeks to gather the necessary information using the staggered monitoring technique.

In order to optimize insulin strategies for tighter glycemic control, it is important to review blood glucose logs at each office visit with either of the above techniques.

Basal plus incretin therapy

GLP-1RAs are subcutaneously administered injectable incretin agents. They mimic the action of endogenous GLP-1 hormones, which are normally secreted in response to meals by the cells of the small intestine.¹⁰ GLP-1 stimulates glucose-dependent insulin secretion, suppresses postprandial glucagon release from pancreatic alpha cells, signals satiety, and slows gastric emptying.¹⁰ In other words, GLP-1 appears to be a physiologic regulator of appetite and food intake. GLP-1 is rapidly metabolized and inactivated by dipeptidyl peptidase-4 (DPP-4) enzymes.¹⁰ The amplification of insulin secretion elicited by hormones secreted from the gastrointestinal (GI) tract is called the “incretin effect.”¹⁰ Obesity, insulin resistance, and type 2 diabetes greatly reduce the incretin effect.¹⁰

Insulin intensification should be guided by glucose monitoring using the 7-point technique or staggered glucose checks.

GLP-1RAs mimic the incretin effect and are not degraded by endogenous DPP-4 enzymes.¹⁰ They provide a pharmacologic level of GLP-1 activity, including beneficial glucose effects (via insulin secretion and glucagon suppression), but they also increase GI adverse effects, such as nausea and vomiting.^11-15 Further, they can suppress appetite and contribute to weight loss.^11-15

GLP-1RAs can be considered as an add-on therapy for patients whose HbA1c exceeds 7% and whose fasting blood glucose ranges from 80 to 130 mg/dL, or for patients with a basal insulin dose >0.5 unit/kg/d. The 5 currently available GLP-1RAs (exenatide, exenatide extended-release, liraglutide, albiglutide, and dulaglutide) are compared in TABLE 2.^11-15

Dosing varies with each agent and includes twice daily before meals for exenatide, once daily (independent of meals) for liraglutide, and once weekly for exenatide extended-release, albiglutide, and dulaglutide. These agents should not be used for patients with a history of pancreatitis or a personal or family history of medullary thyroid cancer or multiple endocrine neoplasia type 2. Because exenatide is cleared through the kidneys, its use is contraindicated in patients with a creatinine clearance <30 mL/min or end-stage renal disease. Caution is advised for its use in patients with a creatinine clearance of 30 to 50 mL/min.¹¹

Basal plus one strategy

To best utilize prandial insulin, it is important to know what the patient’s glucose readings are before and after meals as assessed by the 7-point or staggered blood glucose monitoring techniques described earlier. Once you have clarified which meal(s) are raising the patient’s glucose levels, selecting appropriate treatment becomes easier. To reduce the glucose-monitoring burden for the patient, it may be acceptable to allow the patient to omit the fasting glucose measurement (if stable).

The first major decision is whether to treat one meal per day (basal plus one) or all meals (basal-bolus). Adding a rapid-acting insulin prior to one meal a day (usually the largest meal) is a reasonable starting point.¹⁶

The meal that produces the highest postprandial glucose readings can be considered the meal of greatest glycemic impact. The “delta” value—the difference between pre-meal glucose and 2-hour postprandial glucose readings—also helps to determine the largest meal of the day.¹⁷ The average physiologic delta is ≤50 mg/dL.¹⁷ If the delta for a meal is >75 mg/dL, consider initiating prandial insulin prior to that meal and titrating the dose to achieve a target glucose level of <130 mg/dL before the next meal.

A GLP-1RA can be considered for patients with an HbA1c >7% and a fasting glucose ranging from 80 to 130 mg/dL.

Using 4 to 6 units of a rapid-acting insulin per meal is a good initial regimen for a basal plus one (as well as for a basal-bolus) approach.¹⁶ If the patient experiences significantly increased insulin demands as indicated by glucose patterns where the post-meal glucose is still consistently above 180 mg/dL, the initial regimen may be modified to 0.1 unit per kg per meal,^17-19 and then titrated up to a maximum of 50% of the total daily insulin dose (TDD) for basal plus one¹⁶ (or 10%-20% of TDD per meal for basal-bolus).

Consider the timing of administration. Rapid-acting insulin analogs exhibit peak pharmacodynamic activity 60 minutes after injection (TABLE 3).²⁰

Peak carbohydrate absorption following a meal occurs approximately 75 to 90 minutes after eating begins.^17,21 Thus, to synchronize the action of insulin with carbohydrate digestion, the analog should be injected 15 minutes before meals. This can be increased by titrating prandial insulin by 1 unit/d to a goal of either a 90-minute to 2-hour postprandial glucose of <140 to 180 mg/dL or the next preprandial glucose of <130 mg/dL.¹⁶ The goal is to obtain a near-normal physiologic delta of <50 mg/dL. The drop in delta noted with every unit of insulin added to the current dose can provide a rough approximation of how many additional insulin titrations will be needed to achieve a delta of <50 mg/dL.

Basal-bolus combination

A gradual increase from one injection before a single meal each day to as-needed multiple daily injections (MDIs) is the next step in hyperglycemia management. Starting slow and building up to insulin therapy prior to each meal offers structure, simplicity, and physician-patient confidence in diabetes management. The slow progression from basal plus one to basal-bolus combination allows the patient ease into a complex, labor-intensive regimen of MDIs. Additionally, the stepwise reduction of postprandial hyperglycemia with this slow approach often reduces the incidence of hypoglycemia (more on this in a moment).⁸

Advanced insulin users can calculate an “insulin-to-carbohydrate ratio” (ICR) to estimate the amount of insulin they need to accommodate the amount of carbohydrates they ingest per meal. An ICR of 1:10 implies that the patient administers 1 unit of insulin for every 10 grams of carbohydrates ingested. For example, if a patient with an ICR of 1:10 concludes that his meal contains a total of 60 grams of carbohydrates, then he would administer 6 units of insulin prior to this meal to address the anticipated post-meal hyperglycemia.

In order to use the ICR regimen, a patient would need to be able to accurately determine the nutritional content of his meals (starch, protein, carbohydrates, and fat) and calculate the appropriate insulin dosage. For successful diabetes management, it is essential to evaluate the patient’s skills in these areas before starting an ICR regimen, and to routinely assess hypoglycemic episodes at follow-up visits.

An ICR approach is usually reserved for patients who require tighter glucose control than that obtained from fixed prandial insulin doses, such as patients with type 1 diabetes, those with variable meal schedules and content, those with a malabsorption syndrome that requires consuming meals with a specific amount of carbohydrates, athletes on a structured diet with specific carbohydrate content, and patients who want flexibility with carbohydrate intake with meals.

The risk of hypoglycemia is a major barrier to initiating basal-bolus insulin therapy. Hypoglycemia is classified as a blood glucose level of <70 mg/dL, and severe hypoglycemia as <50 mg/dL, regardless of whether the patient develops symptoms.²² Symptoms of hypoglycemia include dizziness, difficulty speaking, anxiety, confusion, and lethargy. Hypoglycemia can result in loss of consciousness or even death.²²

A patient who has frequent hypoglycemic episodes may lose the protective physiologic response and may not recognize that he is experiencing a hypoglycemic episode (“hypoglycemia unawareness”). This is why it is crucial to ask patients if they have had symptoms of hypoglycemia, and to correlate the timing of these symptoms with blood glucose logs. For example, it is possible for a patient to experience hypoglycemic symptoms for blood glucose readings in the 100 to 200 mg/dL range if his or her average blood glucose has been in the 250 to 300 mg/dL range. Such patient may not realize he is experiencing hypoglycemia until he develops severe symptoms, such as loss of consciousness.

Adding a rapid-acting insulin prior to one meal a day is a reasonable starting point for intensifying insulin therapy.

Hypoglycemia unawareness must be addressed immediately by reducing insulin dosing to prevent all hypoglycemic episodes for 2 to 3 weeks. This has been shown to “reset” the normal physiologic response to hypoglycemia, regardless of how long the patient has had diabetes.^23,24 Even if your patient is aware of the warning signs of a hypoglycemic episode, it is important to routinely ask about hypoglycemia at all diabetes visits because patients may reduce insulin doses, skip doses, or eat defensively to prevent hypoglycemia.

Other than the risk of hypoglycemia, insulin typically has fewer adverse effects than oral medications used to treat diabetes. Most common concerns include weight gain, hypoglycemia, injection site reactions and, rarely, allergy to insulin or its vehicle.¹⁶

CORRESPONDENCE
Jay Shubrook, DO, FAAF P, FACOF P, BC-ADM, Touro University College of Osteopathic Medicine, 1310 Club Drive, Vallejo, CA 94592; [email protected]

Diabetes mellitus is a complex, progressive disease that affects every family physician’s practice. Major diabetes organizations recommend that treatment be ongoing and progressive in order to control the disease. The American Diabetes Association (ADA), the European Association for the Study of Diabetes (EASD), and the American Association of Clinical Endocrinologists recommend that patients be assessed every 2 to 3 months after diagnosis and that treatment should be intensified if the patient is not meeting treatment goals.^1,2 Using this approach, all people with type 2 diabetes could be on insulin one year after diagnosis.^1,2 

While many family physicians have become comfortable with using once-daily basal insulin such as glargine or detemir, what to do after basal insulin is much more complex. This review builds upon an earlier article in this journal, “Insulin for type 2 diabetes: How and when to get started,”³ by explaining 3 strategies to consider when basal insulin alone isn't enough.

3 main strategies for intensifying treatment

Basal insulin is indicated for patients who have glucose toxicity and persistently elevated hemoglobin A1c (HbA1c) despite using 2 or more oral agents, or for those who have not achieved glucose goals one year into treatment.^3,4 ADA/EASD recommends initiating a weight-based approach for basal insulin therapy based on initial HbA1c levels >7% or >8%.⁴ Instructing and encouraging patients to titrate their own insulin dose based on fasting glucose readings provides greater and faster glucose control.^1,2

Despite these attempts, some patients will not reach their glucose goals with basal insulin. When intensifying treatment beyond basal insulin therapy, patient preference, cost-effectiveness, safety, tolerability, glycemic efficacy, risk of hypoglycemia, effects on cardiovascular risk factors, and other non-glycemic effects should be considered in the shared decision-making process. There are 3 main strategies for intensifying treatment:
1. Basal plus incretin therapy. Add a newer injectable agent such as a glucagon-like peptide 1 receptor agonist (GLP-1RA).
2. Basal plus one strategy. Add prandial insulin prior to the largest meal of the day.
3. Basal-bolus combination. Add insulin prior to all meals.

TABLE 1^5-8 provides details of several studies that have documented the efficacy of these 3 strategies.

CLICK IMAGE TO ENLARGE

Monitoring blood glucose to guide the way
Blood glucose monitoring using either a 7-point glucose monitoring technique or staggered glucose checks should guide insulin intensification. A 7-point glucose profile includes pre-meal and post-meal readings for 3 meals a day and an additional bedtime reading.⁹ This is typically performed for 3 to 7 days prior to an appointment and provides an estimate of a typical full day’s glucose pattern.

Staggered monitoring includes a pair of glucose checks taken immediately before and typically 90 minutes after a meal. This is assigned to a different meal each day in order to obtain the same information as is achieved with 7-point monitoring, but with fewer checks on any given day. It may take up to 2 to 3 weeks to gather the necessary information using the staggered monitoring technique.

In order to optimize insulin strategies for tighter glycemic control, it is important to review blood glucose logs at each office visit with either of the above techniques.

Basal plus incretin therapy

GLP-1RAs are subcutaneously administered injectable incretin agents. They mimic the action of endogenous GLP-1 hormones, which are normally secreted in response to meals by the cells of the small intestine.¹⁰ GLP-1 stimulates glucose-dependent insulin secretion, suppresses postprandial glucagon release from pancreatic alpha cells, signals satiety, and slows gastric emptying.¹⁰ In other words, GLP-1 appears to be a physiologic regulator of appetite and food intake. GLP-1 is rapidly metabolized and inactivated by dipeptidyl peptidase-4 (DPP-4) enzymes.¹⁰ The amplification of insulin secretion elicited by hormones secreted from the gastrointestinal (GI) tract is called the “incretin effect.”¹⁰ Obesity, insulin resistance, and type 2 diabetes greatly reduce the incretin effect.¹⁰

Insulin intensification should be guided by glucose monitoring using the 7-point technique or staggered glucose checks.

GLP-1RAs mimic the incretin effect and are not degraded by endogenous DPP-4 enzymes.¹⁰ They provide a pharmacologic level of GLP-1 activity, including beneficial glucose effects (via insulin secretion and glucagon suppression), but they also increase GI adverse effects, such as nausea and vomiting.^11-15 Further, they can suppress appetite and contribute to weight loss.^11-15

GLP-1RAs can be considered as an add-on therapy for patients whose HbA1c exceeds 7% and whose fasting blood glucose ranges from 80 to 130 mg/dL, or for patients with a basal insulin dose >0.5 unit/kg/d. The 5 currently available GLP-1RAs (exenatide, exenatide extended-release, liraglutide, albiglutide, and dulaglutide) are compared in TABLE 2.^11-15

Dosing varies with each agent and includes twice daily before meals for exenatide, once daily (independent of meals) for liraglutide, and once weekly for exenatide extended-release, albiglutide, and dulaglutide. These agents should not be used for patients with a history of pancreatitis or a personal or family history of medullary thyroid cancer or multiple endocrine neoplasia type 2. Because exenatide is cleared through the kidneys, its use is contraindicated in patients with a creatinine clearance <30 mL/min or end-stage renal disease. Caution is advised for its use in patients with a creatinine clearance of 30 to 50 mL/min.¹¹

Basal plus one strategy

To best utilize prandial insulin, it is important to know what the patient’s glucose readings are before and after meals as assessed by the 7-point or staggered blood glucose monitoring techniques described earlier. Once you have clarified which meal(s) are raising the patient’s glucose levels, selecting appropriate treatment becomes easier. To reduce the glucose-monitoring burden for the patient, it may be acceptable to allow the patient to omit the fasting glucose measurement (if stable).

The first major decision is whether to treat one meal per day (basal plus one) or all meals (basal-bolus). Adding a rapid-acting insulin prior to one meal a day (usually the largest meal) is a reasonable starting point.¹⁶

The meal that produces the highest postprandial glucose readings can be considered the meal of greatest glycemic impact. The “delta” value—the difference between pre-meal glucose and 2-hour postprandial glucose readings—also helps to determine the largest meal of the day.¹⁷ The average physiologic delta is ≤50 mg/dL.¹⁷ If the delta for a meal is >75 mg/dL, consider initiating prandial insulin prior to that meal and titrating the dose to achieve a target glucose level of <130 mg/dL before the next meal.

A GLP-1RA can be considered for patients with an HbA1c >7% and a fasting glucose ranging from 80 to 130 mg/dL.

Using 4 to 6 units of a rapid-acting insulin per meal is a good initial regimen for a basal plus one (as well as for a basal-bolus) approach.¹⁶ If the patient experiences significantly increased insulin demands as indicated by glucose patterns where the post-meal glucose is still consistently above 180 mg/dL, the initial regimen may be modified to 0.1 unit per kg per meal,^17-19 and then titrated up to a maximum of 50% of the total daily insulin dose (TDD) for basal plus one¹⁶ (or 10%-20% of TDD per meal for basal-bolus).

Consider the timing of administration. Rapid-acting insulin analogs exhibit peak pharmacodynamic activity 60 minutes after injection (TABLE 3).²⁰

Peak carbohydrate absorption following a meal occurs approximately 75 to 90 minutes after eating begins.^17,21 Thus, to synchronize the action of insulin with carbohydrate digestion, the analog should be injected 15 minutes before meals. This can be increased by titrating prandial insulin by 1 unit/d to a goal of either a 90-minute to 2-hour postprandial glucose of <140 to 180 mg/dL or the next preprandial glucose of <130 mg/dL.¹⁶ The goal is to obtain a near-normal physiologic delta of <50 mg/dL. The drop in delta noted with every unit of insulin added to the current dose can provide a rough approximation of how many additional insulin titrations will be needed to achieve a delta of <50 mg/dL.

Basal-bolus combination

A gradual increase from one injection before a single meal each day to as-needed multiple daily injections (MDIs) is the next step in hyperglycemia management. Starting slow and building up to insulin therapy prior to each meal offers structure, simplicity, and physician-patient confidence in diabetes management. The slow progression from basal plus one to basal-bolus combination allows the patient ease into a complex, labor-intensive regimen of MDIs. Additionally, the stepwise reduction of postprandial hyperglycemia with this slow approach often reduces the incidence of hypoglycemia (more on this in a moment).⁸

Advanced insulin users can calculate an “insulin-to-carbohydrate ratio” (ICR) to estimate the amount of insulin they need to accommodate the amount of carbohydrates they ingest per meal. An ICR of 1:10 implies that the patient administers 1 unit of insulin for every 10 grams of carbohydrates ingested. For example, if a patient with an ICR of 1:10 concludes that his meal contains a total of 60 grams of carbohydrates, then he would administer 6 units of insulin prior to this meal to address the anticipated post-meal hyperglycemia.

In order to use the ICR regimen, a patient would need to be able to accurately determine the nutritional content of his meals (starch, protein, carbohydrates, and fat) and calculate the appropriate insulin dosage. For successful diabetes management, it is essential to evaluate the patient’s skills in these areas before starting an ICR regimen, and to routinely assess hypoglycemic episodes at follow-up visits.

An ICR approach is usually reserved for patients who require tighter glucose control than that obtained from fixed prandial insulin doses, such as patients with type 1 diabetes, those with variable meal schedules and content, those with a malabsorption syndrome that requires consuming meals with a specific amount of carbohydrates, athletes on a structured diet with specific carbohydrate content, and patients who want flexibility with carbohydrate intake with meals.

The risk of hypoglycemia is a major barrier to initiating basal-bolus insulin therapy. Hypoglycemia is classified as a blood glucose level of <70 mg/dL, and severe hypoglycemia as <50 mg/dL, regardless of whether the patient develops symptoms.²² Symptoms of hypoglycemia include dizziness, difficulty speaking, anxiety, confusion, and lethargy. Hypoglycemia can result in loss of consciousness or even death.²²

A patient who has frequent hypoglycemic episodes may lose the protective physiologic response and may not recognize that he is experiencing a hypoglycemic episode (“hypoglycemia unawareness”). This is why it is crucial to ask patients if they have had symptoms of hypoglycemia, and to correlate the timing of these symptoms with blood glucose logs. For example, it is possible for a patient to experience hypoglycemic symptoms for blood glucose readings in the 100 to 200 mg/dL range if his or her average blood glucose has been in the 250 to 300 mg/dL range. Such patient may not realize he is experiencing hypoglycemia until he develops severe symptoms, such as loss of consciousness.

Adding a rapid-acting insulin prior to one meal a day is a reasonable starting point for intensifying insulin therapy.

Hypoglycemia unawareness must be addressed immediately by reducing insulin dosing to prevent all hypoglycemic episodes for 2 to 3 weeks. This has been shown to “reset” the normal physiologic response to hypoglycemia, regardless of how long the patient has had diabetes.^23,24 Even if your patient is aware of the warning signs of a hypoglycemic episode, it is important to routinely ask about hypoglycemia at all diabetes visits because patients may reduce insulin doses, skip doses, or eat defensively to prevent hypoglycemia.

Other than the risk of hypoglycemia, insulin typically has fewer adverse effects than oral medications used to treat diabetes. Most common concerns include weight gain, hypoglycemia, injection site reactions and, rarely, allergy to insulin or its vehicle.¹⁶

CORRESPONDENCE
Jay Shubrook, DO, FAAF P, FACOF P, BC-ADM, Touro University College of Osteopathic Medicine, 1310 Club Drive, Vallejo, CA 94592; [email protected]

Diabetes mellitus is a complex, progressive disease that affects every family physician’s practice. Major diabetes organizations recommend that treatment be ongoing and progressive in order to control the disease. The American Diabetes Association (ADA), the European Association for the Study of Diabetes (EASD), and the American Association of Clinical Endocrinologists recommend that patients be assessed every 2 to 3 months after diagnosis and that treatment should be intensified if the patient is not meeting treatment goals.^1,2 Using this approach, all people with type 2 diabetes could be on insulin one year after diagnosis.^1,2 

While many family physicians have become comfortable with using once-daily basal insulin such as glargine or detemir, what to do after basal insulin is much more complex. This review builds upon an earlier article in this journal, “Insulin for type 2 diabetes: How and when to get started,”³ by explaining 3 strategies to consider when basal insulin alone isn't enough.

3 main strategies for intensifying treatment

Basal insulin is indicated for patients who have glucose toxicity and persistently elevated hemoglobin A1c (HbA1c) despite using 2 or more oral agents, or for those who have not achieved glucose goals one year into treatment.^3,4 ADA/EASD recommends initiating a weight-based approach for basal insulin therapy based on initial HbA1c levels >7% or >8%.⁴ Instructing and encouraging patients to titrate their own insulin dose based on fasting glucose readings provides greater and faster glucose control.^1,2

Despite these attempts, some patients will not reach their glucose goals with basal insulin. When intensifying treatment beyond basal insulin therapy, patient preference, cost-effectiveness, safety, tolerability, glycemic efficacy, risk of hypoglycemia, effects on cardiovascular risk factors, and other non-glycemic effects should be considered in the shared decision-making process. There are 3 main strategies for intensifying treatment:
1. Basal plus incretin therapy. Add a newer injectable agent such as a glucagon-like peptide 1 receptor agonist (GLP-1RA).
2. Basal plus one strategy. Add prandial insulin prior to the largest meal of the day.
3. Basal-bolus combination. Add insulin prior to all meals.

TABLE 1^5-8 provides details of several studies that have documented the efficacy of these 3 strategies.

CLICK IMAGE TO ENLARGE

Monitoring blood glucose to guide the way
Blood glucose monitoring using either a 7-point glucose monitoring technique or staggered glucose checks should guide insulin intensification. A 7-point glucose profile includes pre-meal and post-meal readings for 3 meals a day and an additional bedtime reading.⁹ This is typically performed for 3 to 7 days prior to an appointment and provides an estimate of a typical full day’s glucose pattern.

Staggered monitoring includes a pair of glucose checks taken immediately before and typically 90 minutes after a meal. This is assigned to a different meal each day in order to obtain the same information as is achieved with 7-point monitoring, but with fewer checks on any given day. It may take up to 2 to 3 weeks to gather the necessary information using the staggered monitoring technique.

In order to optimize insulin strategies for tighter glycemic control, it is important to review blood glucose logs at each office visit with either of the above techniques.

Basal plus incretin therapy

GLP-1RAs are subcutaneously administered injectable incretin agents. They mimic the action of endogenous GLP-1 hormones, which are normally secreted in response to meals by the cells of the small intestine.¹⁰ GLP-1 stimulates glucose-dependent insulin secretion, suppresses postprandial glucagon release from pancreatic alpha cells, signals satiety, and slows gastric emptying.¹⁰ In other words, GLP-1 appears to be a physiologic regulator of appetite and food intake. GLP-1 is rapidly metabolized and inactivated by dipeptidyl peptidase-4 (DPP-4) enzymes.¹⁰ The amplification of insulin secretion elicited by hormones secreted from the gastrointestinal (GI) tract is called the “incretin effect.”¹⁰ Obesity, insulin resistance, and type 2 diabetes greatly reduce the incretin effect.¹⁰

Insulin intensification should be guided by glucose monitoring using the 7-point technique or staggered glucose checks.

GLP-1RAs mimic the incretin effect and are not degraded by endogenous DPP-4 enzymes.¹⁰ They provide a pharmacologic level of GLP-1 activity, including beneficial glucose effects (via insulin secretion and glucagon suppression), but they also increase GI adverse effects, such as nausea and vomiting.^11-15 Further, they can suppress appetite and contribute to weight loss.^11-15

GLP-1RAs can be considered as an add-on therapy for patients whose HbA1c exceeds 7% and whose fasting blood glucose ranges from 80 to 130 mg/dL, or for patients with a basal insulin dose >0.5 unit/kg/d. The 5 currently available GLP-1RAs (exenatide, exenatide extended-release, liraglutide, albiglutide, and dulaglutide) are compared in TABLE 2.^11-15

Dosing varies with each agent and includes twice daily before meals for exenatide, once daily (independent of meals) for liraglutide, and once weekly for exenatide extended-release, albiglutide, and dulaglutide. These agents should not be used for patients with a history of pancreatitis or a personal or family history of medullary thyroid cancer or multiple endocrine neoplasia type 2. Because exenatide is cleared through the kidneys, its use is contraindicated in patients with a creatinine clearance <30 mL/min or end-stage renal disease. Caution is advised for its use in patients with a creatinine clearance of 30 to 50 mL/min.¹¹

Basal plus one strategy

To best utilize prandial insulin, it is important to know what the patient’s glucose readings are before and after meals as assessed by the 7-point or staggered blood glucose monitoring techniques described earlier. Once you have clarified which meal(s) are raising the patient’s glucose levels, selecting appropriate treatment becomes easier. To reduce the glucose-monitoring burden for the patient, it may be acceptable to allow the patient to omit the fasting glucose measurement (if stable).

The first major decision is whether to treat one meal per day (basal plus one) or all meals (basal-bolus). Adding a rapid-acting insulin prior to one meal a day (usually the largest meal) is a reasonable starting point.¹⁶

The meal that produces the highest postprandial glucose readings can be considered the meal of greatest glycemic impact. The “delta” value—the difference between pre-meal glucose and 2-hour postprandial glucose readings—also helps to determine the largest meal of the day.¹⁷ The average physiologic delta is ≤50 mg/dL.¹⁷ If the delta for a meal is >75 mg/dL, consider initiating prandial insulin prior to that meal and titrating the dose to achieve a target glucose level of <130 mg/dL before the next meal.

A GLP-1RA can be considered for patients with an HbA1c >7% and a fasting glucose ranging from 80 to 130 mg/dL.

Using 4 to 6 units of a rapid-acting insulin per meal is a good initial regimen for a basal plus one (as well as for a basal-bolus) approach.¹⁶ If the patient experiences significantly increased insulin demands as indicated by glucose patterns where the post-meal glucose is still consistently above 180 mg/dL, the initial regimen may be modified to 0.1 unit per kg per meal,^17-19 and then titrated up to a maximum of 50% of the total daily insulin dose (TDD) for basal plus one¹⁶ (or 10%-20% of TDD per meal for basal-bolus).

Consider the timing of administration. Rapid-acting insulin analogs exhibit peak pharmacodynamic activity 60 minutes after injection (TABLE 3).²⁰

Peak carbohydrate absorption following a meal occurs approximately 75 to 90 minutes after eating begins.^17,21 Thus, to synchronize the action of insulin with carbohydrate digestion, the analog should be injected 15 minutes before meals. This can be increased by titrating prandial insulin by 1 unit/d to a goal of either a 90-minute to 2-hour postprandial glucose of <140 to 180 mg/dL or the next preprandial glucose of <130 mg/dL.¹⁶ The goal is to obtain a near-normal physiologic delta of <50 mg/dL. The drop in delta noted with every unit of insulin added to the current dose can provide a rough approximation of how many additional insulin titrations will be needed to achieve a delta of <50 mg/dL.

Basal-bolus combination

A gradual increase from one injection before a single meal each day to as-needed multiple daily injections (MDIs) is the next step in hyperglycemia management. Starting slow and building up to insulin therapy prior to each meal offers structure, simplicity, and physician-patient confidence in diabetes management. The slow progression from basal plus one to basal-bolus combination allows the patient ease into a complex, labor-intensive regimen of MDIs. Additionally, the stepwise reduction of postprandial hyperglycemia with this slow approach often reduces the incidence of hypoglycemia (more on this in a moment).⁸

Advanced insulin users can calculate an “insulin-to-carbohydrate ratio” (ICR) to estimate the amount of insulin they need to accommodate the amount of carbohydrates they ingest per meal. An ICR of 1:10 implies that the patient administers 1 unit of insulin for every 10 grams of carbohydrates ingested. For example, if a patient with an ICR of 1:10 concludes that his meal contains a total of 60 grams of carbohydrates, then he would administer 6 units of insulin prior to this meal to address the anticipated post-meal hyperglycemia.

In order to use the ICR regimen, a patient would need to be able to accurately determine the nutritional content of his meals (starch, protein, carbohydrates, and fat) and calculate the appropriate insulin dosage. For successful diabetes management, it is essential to evaluate the patient’s skills in these areas before starting an ICR regimen, and to routinely assess hypoglycemic episodes at follow-up visits.

An ICR approach is usually reserved for patients who require tighter glucose control than that obtained from fixed prandial insulin doses, such as patients with type 1 diabetes, those with variable meal schedules and content, those with a malabsorption syndrome that requires consuming meals with a specific amount of carbohydrates, athletes on a structured diet with specific carbohydrate content, and patients who want flexibility with carbohydrate intake with meals.

The risk of hypoglycemia is a major barrier to initiating basal-bolus insulin therapy. Hypoglycemia is classified as a blood glucose level of <70 mg/dL, and severe hypoglycemia as <50 mg/dL, regardless of whether the patient develops symptoms.²² Symptoms of hypoglycemia include dizziness, difficulty speaking, anxiety, confusion, and lethargy. Hypoglycemia can result in loss of consciousness or even death.²²

A patient who has frequent hypoglycemic episodes may lose the protective physiologic response and may not recognize that he is experiencing a hypoglycemic episode (“hypoglycemia unawareness”). This is why it is crucial to ask patients if they have had symptoms of hypoglycemia, and to correlate the timing of these symptoms with blood glucose logs. For example, it is possible for a patient to experience hypoglycemic symptoms for blood glucose readings in the 100 to 200 mg/dL range if his or her average blood glucose has been in the 250 to 300 mg/dL range. Such patient may not realize he is experiencing hypoglycemia until he develops severe symptoms, such as loss of consciousness.

Adding a rapid-acting insulin prior to one meal a day is a reasonable starting point for intensifying insulin therapy.

Hypoglycemia unawareness must be addressed immediately by reducing insulin dosing to prevent all hypoglycemic episodes for 2 to 3 weeks. This has been shown to “reset” the normal physiologic response to hypoglycemia, regardless of how long the patient has had diabetes.^23,24 Even if your patient is aware of the warning signs of a hypoglycemic episode, it is important to routinely ask about hypoglycemia at all diabetes visits because patients may reduce insulin doses, skip doses, or eat defensively to prevent hypoglycemia.

Other than the risk of hypoglycemia, insulin typically has fewer adverse effects than oral medications used to treat diabetes. Most common concerns include weight gain, hypoglycemia, injection site reactions and, rarely, allergy to insulin or its vehicle.¹⁶

CORRESPONDENCE
Jay Shubrook, DO, FAAF P, FACOF P, BC-ADM, Touro University College of Osteopathic Medicine, 1310 Club Drive, Vallejo, CA 94592; [email protected]

A young mother in your practice arrives with her 2-year-old son for a well-child visit. She remarks that, although her son uses a few single words to indicate hunger and other needs, her sister’s child at the same age had begun using multiple words to ask questions and express her wishes. She’s concerned about whether her son’s behavior is normal. As you start to engage the child, you note that he responds only after you repeat his name a few times. Are these observations indicative of a typical delay in development, or are they clues to a serious medical issue or communication disability? Given the absence of any known medical problem or evident physical or intellectual disability, how would you proceed in this case and in counseling the mother?

Developmental screening minimizes adverse long-term consequences

Speech, language, and hearing delays and disorders in children can lead to learning and socialization problems that may persist into adulthood. Health care providers who monitor speech, language, and hearing development in children can guide parents, as needed, to appropriate services for further assessment or treatment¹ and direct them to advocacy programs such as the Center for Parent Information and Resources (formerly the National Dissemination Center for Children with Disabilities).²

A useful tool at well-child visits is the Denver II, a quick developmental screening test to help identify a variety of disorders of intelligence, language, mental health, and motor and self-help skills.³

Suspicion of a developmental delay not likely due to a medical issue or congenital abnormality requiring examination by an otorhinolaryngologist could warrant referral of the child for early intervention (EI).

Communication disorders and their manifestations

Communication—the ability to receive, process, comprehend, and transmit information—is essential for a successful life.⁴ Speech, language, and hearing impairments affect a child’s ability to send (speak, write, or gesture) and receive (hear, interpret, or decipher) messages.

Speech impairments

Beginning at birth, we systematically develop speech sounds and an ability to use these sounds to convey meaning by forming words and using language.⁵ Speech and language pathologists make a distinction between speech and language impairments.⁶

Speech disorders may involve problems of articulation, fluency, voice, or resonance. About 8% to 9% of preschool children have speech disorders, and approximately 5% of school-age children have speech or language impairments.⁷

Problems of articulation are heard in such instances as substituting a “w” for an “r” (“wabbit” for “rabbit”) or in distorting or omitting sounds or syllables (“tato” for “potato”). Considering that articulation involves the precise coordination of about 70 muscles (tongue, lips, velum, vocal folds, etc), development of this skill normally goes through phases of inaccurate sound productions. Concerns arise when these phases persist or are atypical.

Speech fluency/stuttering is the uncontrollable blocking of speech, sound prolongation (“wwwwater”), or repetition of a sound, syllable, or word during speaking (“pu-pu-pu-puppy”).

Problems of voice include symptoms such as hoarseness, an exceptionally weak voice or one that is too high or too low, or abnormal resonance (hyper- or hyponasality, which gives the impression the child is talking “through the nose” or is constantly congested).

Using common milestones as reference points. The American Speech-Language-Hearing Association (ASHA)⁸ lists the milestones for speech development (English and Spanish) at http://www.asha.org/public/speech/development/chart.htm. For example, between 12 and 24 months of age, a child should be learning vocabulary (“doggie, nana”), combining 2 words (“mommy car”), asking 2-word questions (“where daddy?”), and producing a variety of speech sounds. These milestones represent an average and some children may not master all the items in a category until they reach the upper limit of the age range (TABLE 1).⁸ Roth et al⁹ found that intervention benefited preschoolers with speech and language disabilities when applied earlier than previously recommended. In other words, avoid the “wait-and-see” option. Busari and Weggelaar,¹⁰ studying referral recommendations for children who are “slow to speak,” concluded that EI may diminish further consequences later in the child’s life.

ASHA launched a campaign to increase awareness of communication disorders across the lifespan and to encourage early identification (http://identifythesigns.org/).¹¹ The site has basic lists of signs of common speech and language disorders and hearing loss in children from birth to 4 years of age. This period in a child’s life is “an important stage in early detection of communication disorders.”

Language impairments

One study noted that primary care physicians had low confidence in providing information to families about early intervention resources. Language impairments affect 3 domains of language: form (grammar/syntax), content, or use. These domains are governed by rules specific to the language spoken in the home. Language impairment can interfere with comprehension and formulation of messages. About 2% to 3% of preschoolers have language disorders.¹²

Language impairments may be observed in one or more of the language components, including phonology (rules of the sounds in the language), lexicon (vocabulary), morphology (word markers—eg, final “s” to make the plural of “cat”), syntax (word order) and pragmatics (socially appropriate speech, gesture, eye contact, and language use).

While there is great variability in typical development, atypical language development can be a secondary characteristic of other physical or developmental problems attributed to other conditions such as autism spectrum disorder (ASD), cerebral palsy, childhood apraxia of speech, dysarthria, intellectual disability, or selective mutism.¹³

Verbal communication difficulties may appear in expressive and receptive language.¹⁴Receptive language is the ability to comprehend language communicated by another person.⁶ Receptive language (processing) skills can be demonstrated as follows. If a child is asked “Do you like cats?” she must first decide to whom the question is directed (to her and not someone else). She must then search her long-term memory for the word/concept “cat” (compared with dog, a similar concept; or with mat, a similar-sounding word), and process the word “like” (compared with “dislike”). Now the child understands the question and can decide on an answer.

Expressive language is a child’s ability to speak; the mental process used to produce speech and communicate a message.⁶ To answer, “Yes, I like cats,” the child retrieves the concept “cat” from memory (cognition and semantics: the meanings of words, their relationships and usage), finds the right words (vocabulary), puts them in the right order (syntax), uses the right verb tense (grammar/morphology), assembles the right sounds in order, initiates the neuromotor acts (phonology/speech production), and communicates that she understands what cats are and that she likes them (pragmatics; socially/contextually appropriate responses).

Resources on language development. Typical language development, the length of which varies among children, must be practiced in a rich linguistic environment. Some children are adventurous with language. They babble, talk, and communicate in a carefree manner. Others are cautious. They may wait until they are sure of their skills before attempting a new word. Usually, concern about a child’s speech and language development arises if there is no speech, if speech is not clear, or if speech or language is different from that of peers.

The Centers for Disease Control and Prevention, under their “Learn the signs. Act early” campaign (www.cdc.gov/actearly), has published checklists of children’s developmental milestones from 2 months to 5 years of age on social, communication, cognitive, and motor skills. Health care providers helping parents determine if their child’s communication is developing normally can find information and materials at http://www.cdc.gov/ncbddd/actearly/hcp/index.html.¹⁵

Hearing impairments

Moeller et al¹⁶ surveyed 1968 primary care physicians on their attitude, practices, and knowledge of universal hearing screenings for newborns. They noted limitations in awareness of EI options for infants with hearing loss: proper times and places for referrals, available communication modalities, cochlear implant candidacy, and professionals in their locale with expertise on hearing loss. These knowledge gaps involved some medical issues, such as hearing loss genetics and later-onset hearing loss in infants and children. They also found low confidence in providing information to families about how to proceed with EI and in discussing intervention needs and resources.

Early management of hearing impairment can improve language, especially for children with a severe or profound hearing loss. Adverse effects of hearing loss. Hearing loss can be unilateral or bilateral, conductive or sensorineural, and can range in severity from mild to profound. According to the National Institute on Deafness and Communication Disorders (NIDCD), one in every 350 infants is born with a significant hearing loss, and others become deaf due to childhood illness or injury.¹⁷ According to ASHA,¹⁸ hearing loss can affect children in 4 major ways:¹⁹ delays in the development of receptive (comprehension) and expressive communication skills; a language deficit that causes learning problems and reduces academic achievement; communication difficulties that often lead to social isolation and poor self-concept; and impact on vocational choices and options.²⁰ NIDCD  provides a checklist to determine a child’s hearing status at http://www.nidcd.nih.gov/health/hearing/silence.asp.¹⁷

Timing of intervention is significant. EI is critical in minimizing the deleterious effects of hearing loss and in optimizing speech and language development. Severity of hearing loss influences EI outcome, and treatment options depend on the hearing loss having occurred either before language development (prelingually) or after (postlingually). Early management of hearing impairment can improve language, especially for children with a severe or profound hearing loss.²¹

Devices and methods that promote communication development for students who are deaf or hard of hearing include the use of hearing aids to amplify residual hearing for oral or auditory-oral approaches; the manual approach stressing sign-language (American Sign Language or Signed English); or Total Communication using both the oral and sign-language methods.

With an infant, suspicion of a hearing problem warrants referral to an otorhinolaryngologist or an audiologist for thorough evaluation. In New Jersey, an EI referral typically triggers a referral to one of these specialists. TABLE 2 lists resources for early identification and intervention in communication disorders.

TABLE 2

Autism spectrum disorder

The fifth edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM-5)²² has revised its diagnostic codes for ASD. An in-depth analysis of the new diagnosis is beyond the purview of this article, but it deserves a few comments.

The revised diagnosis of ASD consolidates the previously separate diagnoses of autistic disorder, Asperger’s disorder, childhood disintegrative disorder, and pervasive developmental disorder/not otherwise specified.^22-24 According to DSM-5, an individual with ASD must have 1) persistent deficits in social communication and social interactions, 2) restricted, repetitive patterns of behavior, interests or activities, 3) symptoms present in early childhood (but may not become fully manifest until social demands exceed limited capacities), and 4) symptoms that limit and impair everyday functioning.^22,24

Children who don’t meet DSM-5 criteria for autism spectrum disorder may be included in a new category called social communication disorder. Onset of the disorder must be obvious before age 3 for a child to be eligible for EI. Children who do not qualify for an ASD diagnosis under the new DSM-5 definition may be included in a new category called social communication disorder (SCD) under “Communication Disorders” in DSM-5. SCD is defined as impairment in pragmatics that impacts development of social relationships and comprehension of social conversations. ASD must be ruled out before a diagnosis of SCD can be made.^23,24

Early intervention services

Individual states offer EI services to families to facilitate detection of developmental delays in children and to provide a comprehensive system of support designed to reduce the effects of disabilities (or to prevent learning and developmental problems later in life).⁵ The rationale for EI is that the earlier interventions are started, the less likely later interventions will be needed.²⁵ EI services are provided to children from birth through their third birthday; services are free of charge to eligible families or on a sliding payment scale determined by a family’s income.

Resources for information and referrals. To access EI services on behalf of a family, contact a local hospital or point them to the CPIR (http://www.parentcenterhub.org/repository/disability-landing/).² For families with disabled children older than 3, consider suggesting that parents contact the local school district (even if the child isn’t enrolled there) to arrange an evaluation under the Individuals with Disabilities Education Act (http://idea.ed.gov/).²⁶ Because individual states’ EI organizations may have slightly different procedures, it is best to consult one’s own state EI site for specific information regarding referrals (TABLE 3).

Efficacy of treatment

The prevalence of specific communication disorders varies widely, as do prognoses, possibly due to the variability of underlying causes (physical/biological/medical or environmental/educational).²⁷ Also, as described earlier, “communication disorder” is an umbrella term inclusive of problems as diverse as resonance (eg, hypernasality due to a submucous palatal cleft), severe language delay (eg, due to Down syndrome), or lisp (misarticulation of the “s” sound).

Is therapy effective? Speech and language pathologists use the National Outcomes Measurement System (NOMS) as an index of the outcomes of treatment on functional communication along 6 scales. The most frequent types of communication problems seen in the prekindergarten children’s NOMS were “articulation” (75% of children), “spoken language production” (61%), and “spoken language comprehension” (42%). Problems in the remaining 3 scales (“pragmatics,” “cognitive orientation,” and “swallowing”) were seen in fewer than 15% of the preschool-age students.²⁸

Articulation therapy yielded improvement in 69.3% of cases, spoken language comprehension therapy in 65.3%, and spoken language production therapy in 65.2%.²⁸ Such outcomes support regular screening of children’s communication development and, as needed, referral for EI.

TABLE 3

CORRESPONDENCE
Christopher Mulrine, EdD; William Paterson University, 1600 Valley Road 3003 Wayne, NJ 07474-0920; [email protected]

A young mother in your practice arrives with her 2-year-old son for a well-child visit. She remarks that, although her son uses a few single words to indicate hunger and other needs, her sister’s child at the same age had begun using multiple words to ask questions and express her wishes. She’s concerned about whether her son’s behavior is normal. As you start to engage the child, you note that he responds only after you repeat his name a few times. Are these observations indicative of a typical delay in development, or are they clues to a serious medical issue or communication disability? Given the absence of any known medical problem or evident physical or intellectual disability, how would you proceed in this case and in counseling the mother?

Developmental screening minimizes adverse long-term consequences

Speech, language, and hearing delays and disorders in children can lead to learning and socialization problems that may persist into adulthood. Health care providers who monitor speech, language, and hearing development in children can guide parents, as needed, to appropriate services for further assessment or treatment¹ and direct them to advocacy programs such as the Center for Parent Information and Resources (formerly the National Dissemination Center for Children with Disabilities).²

A useful tool at well-child visits is the Denver II, a quick developmental screening test to help identify a variety of disorders of intelligence, language, mental health, and motor and self-help skills.³

Suspicion of a developmental delay not likely due to a medical issue or congenital abnormality requiring examination by an otorhinolaryngologist could warrant referral of the child for early intervention (EI).

Communication disorders and their manifestations

Communication—the ability to receive, process, comprehend, and transmit information—is essential for a successful life.⁴ Speech, language, and hearing impairments affect a child’s ability to send (speak, write, or gesture) and receive (hear, interpret, or decipher) messages.

Speech impairments

Beginning at birth, we systematically develop speech sounds and an ability to use these sounds to convey meaning by forming words and using language.⁵ Speech and language pathologists make a distinction between speech and language impairments.⁶

Speech disorders may involve problems of articulation, fluency, voice, or resonance. About 8% to 9% of preschool children have speech disorders, and approximately 5% of school-age children have speech or language impairments.⁷

Problems of articulation are heard in such instances as substituting a “w” for an “r” (“wabbit” for “rabbit”) or in distorting or omitting sounds or syllables (“tato” for “potato”). Considering that articulation involves the precise coordination of about 70 muscles (tongue, lips, velum, vocal folds, etc), development of this skill normally goes through phases of inaccurate sound productions. Concerns arise when these phases persist or are atypical.

Speech fluency/stuttering is the uncontrollable blocking of speech, sound prolongation (“wwwwater”), or repetition of a sound, syllable, or word during speaking (“pu-pu-pu-puppy”).

Problems of voice include symptoms such as hoarseness, an exceptionally weak voice or one that is too high or too low, or abnormal resonance (hyper- or hyponasality, which gives the impression the child is talking “through the nose” or is constantly congested).

Using common milestones as reference points. The American Speech-Language-Hearing Association (ASHA)⁸ lists the milestones for speech development (English and Spanish) at http://www.asha.org/public/speech/development/chart.htm. For example, between 12 and 24 months of age, a child should be learning vocabulary (“doggie, nana”), combining 2 words (“mommy car”), asking 2-word questions (“where daddy?”), and producing a variety of speech sounds. These milestones represent an average and some children may not master all the items in a category until they reach the upper limit of the age range (TABLE 1).⁸ Roth et al⁹ found that intervention benefited preschoolers with speech and language disabilities when applied earlier than previously recommended. In other words, avoid the “wait-and-see” option. Busari and Weggelaar,¹⁰ studying referral recommendations for children who are “slow to speak,” concluded that EI may diminish further consequences later in the child’s life.

ASHA launched a campaign to increase awareness of communication disorders across the lifespan and to encourage early identification (http://identifythesigns.org/).¹¹ The site has basic lists of signs of common speech and language disorders and hearing loss in children from birth to 4 years of age. This period in a child’s life is “an important stage in early detection of communication disorders.”

Language impairments

One study noted that primary care physicians had low confidence in providing information to families about early intervention resources. Language impairments affect 3 domains of language: form (grammar/syntax), content, or use. These domains are governed by rules specific to the language spoken in the home. Language impairment can interfere with comprehension and formulation of messages. About 2% to 3% of preschoolers have language disorders.¹²

Language impairments may be observed in one or more of the language components, including phonology (rules of the sounds in the language), lexicon (vocabulary), morphology (word markers—eg, final “s” to make the plural of “cat”), syntax (word order) and pragmatics (socially appropriate speech, gesture, eye contact, and language use).

While there is great variability in typical development, atypical language development can be a secondary characteristic of other physical or developmental problems attributed to other conditions such as autism spectrum disorder (ASD), cerebral palsy, childhood apraxia of speech, dysarthria, intellectual disability, or selective mutism.¹³

Verbal communication difficulties may appear in expressive and receptive language.¹⁴Receptive language is the ability to comprehend language communicated by another person.⁶ Receptive language (processing) skills can be demonstrated as follows. If a child is asked “Do you like cats?” she must first decide to whom the question is directed (to her and not someone else). She must then search her long-term memory for the word/concept “cat” (compared with dog, a similar concept; or with mat, a similar-sounding word), and process the word “like” (compared with “dislike”). Now the child understands the question and can decide on an answer.

Expressive language is a child’s ability to speak; the mental process used to produce speech and communicate a message.⁶ To answer, “Yes, I like cats,” the child retrieves the concept “cat” from memory (cognition and semantics: the meanings of words, their relationships and usage), finds the right words (vocabulary), puts them in the right order (syntax), uses the right verb tense (grammar/morphology), assembles the right sounds in order, initiates the neuromotor acts (phonology/speech production), and communicates that she understands what cats are and that she likes them (pragmatics; socially/contextually appropriate responses).

Resources on language development. Typical language development, the length of which varies among children, must be practiced in a rich linguistic environment. Some children are adventurous with language. They babble, talk, and communicate in a carefree manner. Others are cautious. They may wait until they are sure of their skills before attempting a new word. Usually, concern about a child’s speech and language development arises if there is no speech, if speech is not clear, or if speech or language is different from that of peers.

The Centers for Disease Control and Prevention, under their “Learn the signs. Act early” campaign (www.cdc.gov/actearly), has published checklists of children’s developmental milestones from 2 months to 5 years of age on social, communication, cognitive, and motor skills. Health care providers helping parents determine if their child’s communication is developing normally can find information and materials at http://www.cdc.gov/ncbddd/actearly/hcp/index.html.¹⁵

Hearing impairments

Moeller et al¹⁶ surveyed 1968 primary care physicians on their attitude, practices, and knowledge of universal hearing screenings for newborns. They noted limitations in awareness of EI options for infants with hearing loss: proper times and places for referrals, available communication modalities, cochlear implant candidacy, and professionals in their locale with expertise on hearing loss. These knowledge gaps involved some medical issues, such as hearing loss genetics and later-onset hearing loss in infants and children. They also found low confidence in providing information to families about how to proceed with EI and in discussing intervention needs and resources.

Early management of hearing impairment can improve language, especially for children with a severe or profound hearing loss. Adverse effects of hearing loss. Hearing loss can be unilateral or bilateral, conductive or sensorineural, and can range in severity from mild to profound. According to the National Institute on Deafness and Communication Disorders (NIDCD), one in every 350 infants is born with a significant hearing loss, and others become deaf due to childhood illness or injury.¹⁷ According to ASHA,¹⁸ hearing loss can affect children in 4 major ways:¹⁹ delays in the development of receptive (comprehension) and expressive communication skills; a language deficit that causes learning problems and reduces academic achievement; communication difficulties that often lead to social isolation and poor self-concept; and impact on vocational choices and options.²⁰ NIDCD  provides a checklist to determine a child’s hearing status at http://www.nidcd.nih.gov/health/hearing/silence.asp.¹⁷

Timing of intervention is significant. EI is critical in minimizing the deleterious effects of hearing loss and in optimizing speech and language development. Severity of hearing loss influences EI outcome, and treatment options depend on the hearing loss having occurred either before language development (prelingually) or after (postlingually). Early management of hearing impairment can improve language, especially for children with a severe or profound hearing loss.²¹

Devices and methods that promote communication development for students who are deaf or hard of hearing include the use of hearing aids to amplify residual hearing for oral or auditory-oral approaches; the manual approach stressing sign-language (American Sign Language or Signed English); or Total Communication using both the oral and sign-language methods.

With an infant, suspicion of a hearing problem warrants referral to an otorhinolaryngologist or an audiologist for thorough evaluation. In New Jersey, an EI referral typically triggers a referral to one of these specialists. TABLE 2 lists resources for early identification and intervention in communication disorders.

TABLE 2

Autism spectrum disorder

The fifth edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM-5)²² has revised its diagnostic codes for ASD. An in-depth analysis of the new diagnosis is beyond the purview of this article, but it deserves a few comments.

The revised diagnosis of ASD consolidates the previously separate diagnoses of autistic disorder, Asperger’s disorder, childhood disintegrative disorder, and pervasive developmental disorder/not otherwise specified.^22-24 According to DSM-5, an individual with ASD must have 1) persistent deficits in social communication and social interactions, 2) restricted, repetitive patterns of behavior, interests or activities, 3) symptoms present in early childhood (but may not become fully manifest until social demands exceed limited capacities), and 4) symptoms that limit and impair everyday functioning.^22,24

Children who don’t meet DSM-5 criteria for autism spectrum disorder may be included in a new category called social communication disorder. Onset of the disorder must be obvious before age 3 for a child to be eligible for EI. Children who do not qualify for an ASD diagnosis under the new DSM-5 definition may be included in a new category called social communication disorder (SCD) under “Communication Disorders” in DSM-5. SCD is defined as impairment in pragmatics that impacts development of social relationships and comprehension of social conversations. ASD must be ruled out before a diagnosis of SCD can be made.^23,24

Early intervention services

Individual states offer EI services to families to facilitate detection of developmental delays in children and to provide a comprehensive system of support designed to reduce the effects of disabilities (or to prevent learning and developmental problems later in life).⁵ The rationale for EI is that the earlier interventions are started, the less likely later interventions will be needed.²⁵ EI services are provided to children from birth through their third birthday; services are free of charge to eligible families or on a sliding payment scale determined by a family’s income.

Resources for information and referrals. To access EI services on behalf of a family, contact a local hospital or point them to the CPIR (http://www.parentcenterhub.org/repository/disability-landing/).² For families with disabled children older than 3, consider suggesting that parents contact the local school district (even if the child isn’t enrolled there) to arrange an evaluation under the Individuals with Disabilities Education Act (http://idea.ed.gov/).²⁶ Because individual states’ EI organizations may have slightly different procedures, it is best to consult one’s own state EI site for specific information regarding referrals (TABLE 3).

Efficacy of treatment

The prevalence of specific communication disorders varies widely, as do prognoses, possibly due to the variability of underlying causes (physical/biological/medical or environmental/educational).²⁷ Also, as described earlier, “communication disorder” is an umbrella term inclusive of problems as diverse as resonance (eg, hypernasality due to a submucous palatal cleft), severe language delay (eg, due to Down syndrome), or lisp (misarticulation of the “s” sound).

Is therapy effective? Speech and language pathologists use the National Outcomes Measurement System (NOMS) as an index of the outcomes of treatment on functional communication along 6 scales. The most frequent types of communication problems seen in the prekindergarten children’s NOMS were “articulation” (75% of children), “spoken language production” (61%), and “spoken language comprehension” (42%). Problems in the remaining 3 scales (“pragmatics,” “cognitive orientation,” and “swallowing”) were seen in fewer than 15% of the preschool-age students.²⁸

Articulation therapy yielded improvement in 69.3% of cases, spoken language comprehension therapy in 65.3%, and spoken language production therapy in 65.2%.²⁸ Such outcomes support regular screening of children’s communication development and, as needed, referral for EI.

TABLE 3

CORRESPONDENCE
Christopher Mulrine, EdD; William Paterson University, 1600 Valley Road 3003 Wayne, NJ 07474-0920; [email protected]

A young mother in your practice arrives with her 2-year-old son for a well-child visit. She remarks that, although her son uses a few single words to indicate hunger and other needs, her sister’s child at the same age had begun using multiple words to ask questions and express her wishes. She’s concerned about whether her son’s behavior is normal. As you start to engage the child, you note that he responds only after you repeat his name a few times. Are these observations indicative of a typical delay in development, or are they clues to a serious medical issue or communication disability? Given the absence of any known medical problem or evident physical or intellectual disability, how would you proceed in this case and in counseling the mother?

Developmental screening minimizes adverse long-term consequences

Speech, language, and hearing delays and disorders in children can lead to learning and socialization problems that may persist into adulthood. Health care providers who monitor speech, language, and hearing development in children can guide parents, as needed, to appropriate services for further assessment or treatment¹ and direct them to advocacy programs such as the Center for Parent Information and Resources (formerly the National Dissemination Center for Children with Disabilities).²

A useful tool at well-child visits is the Denver II, a quick developmental screening test to help identify a variety of disorders of intelligence, language, mental health, and motor and self-help skills.³

Suspicion of a developmental delay not likely due to a medical issue or congenital abnormality requiring examination by an otorhinolaryngologist could warrant referral of the child for early intervention (EI).

Communication disorders and their manifestations

Communication—the ability to receive, process, comprehend, and transmit information—is essential for a successful life.⁴ Speech, language, and hearing impairments affect a child’s ability to send (speak, write, or gesture) and receive (hear, interpret, or decipher) messages.

Speech impairments

Beginning at birth, we systematically develop speech sounds and an ability to use these sounds to convey meaning by forming words and using language.⁵ Speech and language pathologists make a distinction between speech and language impairments.⁶

Speech disorders may involve problems of articulation, fluency, voice, or resonance. About 8% to 9% of preschool children have speech disorders, and approximately 5% of school-age children have speech or language impairments.⁷

Problems of articulation are heard in such instances as substituting a “w” for an “r” (“wabbit” for “rabbit”) or in distorting or omitting sounds or syllables (“tato” for “potato”). Considering that articulation involves the precise coordination of about 70 muscles (tongue, lips, velum, vocal folds, etc), development of this skill normally goes through phases of inaccurate sound productions. Concerns arise when these phases persist or are atypical.

Speech fluency/stuttering is the uncontrollable blocking of speech, sound prolongation (“wwwwater”), or repetition of a sound, syllable, or word during speaking (“pu-pu-pu-puppy”).

Problems of voice include symptoms such as hoarseness, an exceptionally weak voice or one that is too high or too low, or abnormal resonance (hyper- or hyponasality, which gives the impression the child is talking “through the nose” or is constantly congested).

Using common milestones as reference points. The American Speech-Language-Hearing Association (ASHA)⁸ lists the milestones for speech development (English and Spanish) at http://www.asha.org/public/speech/development/chart.htm. For example, between 12 and 24 months of age, a child should be learning vocabulary (“doggie, nana”), combining 2 words (“mommy car”), asking 2-word questions (“where daddy?”), and producing a variety of speech sounds. These milestones represent an average and some children may not master all the items in a category until they reach the upper limit of the age range (TABLE 1).⁸ Roth et al⁹ found that intervention benefited preschoolers with speech and language disabilities when applied earlier than previously recommended. In other words, avoid the “wait-and-see” option. Busari and Weggelaar,¹⁰ studying referral recommendations for children who are “slow to speak,” concluded that EI may diminish further consequences later in the child’s life.

ASHA launched a campaign to increase awareness of communication disorders across the lifespan and to encourage early identification (http://identifythesigns.org/).¹¹ The site has basic lists of signs of common speech and language disorders and hearing loss in children from birth to 4 years of age. This period in a child’s life is “an important stage in early detection of communication disorders.”

Language impairments

One study noted that primary care physicians had low confidence in providing information to families about early intervention resources. Language impairments affect 3 domains of language: form (grammar/syntax), content, or use. These domains are governed by rules specific to the language spoken in the home. Language impairment can interfere with comprehension and formulation of messages. About 2% to 3% of preschoolers have language disorders.¹²

Language impairments may be observed in one or more of the language components, including phonology (rules of the sounds in the language), lexicon (vocabulary), morphology (word markers—eg, final “s” to make the plural of “cat”), syntax (word order) and pragmatics (socially appropriate speech, gesture, eye contact, and language use).

While there is great variability in typical development, atypical language development can be a secondary characteristic of other physical or developmental problems attributed to other conditions such as autism spectrum disorder (ASD), cerebral palsy, childhood apraxia of speech, dysarthria, intellectual disability, or selective mutism.¹³

Verbal communication difficulties may appear in expressive and receptive language.¹⁴Receptive language is the ability to comprehend language communicated by another person.⁶ Receptive language (processing) skills can be demonstrated as follows. If a child is asked “Do you like cats?” she must first decide to whom the question is directed (to her and not someone else). She must then search her long-term memory for the word/concept “cat” (compared with dog, a similar concept; or with mat, a similar-sounding word), and process the word “like” (compared with “dislike”). Now the child understands the question and can decide on an answer.

Expressive language is a child’s ability to speak; the mental process used to produce speech and communicate a message.⁶ To answer, “Yes, I like cats,” the child retrieves the concept “cat” from memory (cognition and semantics: the meanings of words, their relationships and usage), finds the right words (vocabulary), puts them in the right order (syntax), uses the right verb tense (grammar/morphology), assembles the right sounds in order, initiates the neuromotor acts (phonology/speech production), and communicates that she understands what cats are and that she likes them (pragmatics; socially/contextually appropriate responses).

Resources on language development. Typical language development, the length of which varies among children, must be practiced in a rich linguistic environment. Some children are adventurous with language. They babble, talk, and communicate in a carefree manner. Others are cautious. They may wait until they are sure of their skills before attempting a new word. Usually, concern about a child’s speech and language development arises if there is no speech, if speech is not clear, or if speech or language is different from that of peers.

The Centers for Disease Control and Prevention, under their “Learn the signs. Act early” campaign (www.cdc.gov/actearly), has published checklists of children’s developmental milestones from 2 months to 5 years of age on social, communication, cognitive, and motor skills. Health care providers helping parents determine if their child’s communication is developing normally can find information and materials at http://www.cdc.gov/ncbddd/actearly/hcp/index.html.¹⁵

Hearing impairments

Moeller et al¹⁶ surveyed 1968 primary care physicians on their attitude, practices, and knowledge of universal hearing screenings for newborns. They noted limitations in awareness of EI options for infants with hearing loss: proper times and places for referrals, available communication modalities, cochlear implant candidacy, and professionals in their locale with expertise on hearing loss. These knowledge gaps involved some medical issues, such as hearing loss genetics and later-onset hearing loss in infants and children. They also found low confidence in providing information to families about how to proceed with EI and in discussing intervention needs and resources.

Early management of hearing impairment can improve language, especially for children with a severe or profound hearing loss. Adverse effects of hearing loss. Hearing loss can be unilateral or bilateral, conductive or sensorineural, and can range in severity from mild to profound. According to the National Institute on Deafness and Communication Disorders (NIDCD), one in every 350 infants is born with a significant hearing loss, and others become deaf due to childhood illness or injury.¹⁷ According to ASHA,¹⁸ hearing loss can affect children in 4 major ways:¹⁹ delays in the development of receptive (comprehension) and expressive communication skills; a language deficit that causes learning problems and reduces academic achievement; communication difficulties that often lead to social isolation and poor self-concept; and impact on vocational choices and options.²⁰ NIDCD  provides a checklist to determine a child’s hearing status at http://www.nidcd.nih.gov/health/hearing/silence.asp.¹⁷

Timing of intervention is significant. EI is critical in minimizing the deleterious effects of hearing loss and in optimizing speech and language development. Severity of hearing loss influences EI outcome, and treatment options depend on the hearing loss having occurred either before language development (prelingually) or after (postlingually). Early management of hearing impairment can improve language, especially for children with a severe or profound hearing loss.²¹

Devices and methods that promote communication development for students who are deaf or hard of hearing include the use of hearing aids to amplify residual hearing for oral or auditory-oral approaches; the manual approach stressing sign-language (American Sign Language or Signed English); or Total Communication using both the oral and sign-language methods.

With an infant, suspicion of a hearing problem warrants referral to an otorhinolaryngologist or an audiologist for thorough evaluation. In New Jersey, an EI referral typically triggers a referral to one of these specialists. TABLE 2 lists resources for early identification and intervention in communication disorders.

TABLE 2

Autism spectrum disorder

The fifth edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM-5)²² has revised its diagnostic codes for ASD. An in-depth analysis of the new diagnosis is beyond the purview of this article, but it deserves a few comments.

The revised diagnosis of ASD consolidates the previously separate diagnoses of autistic disorder, Asperger’s disorder, childhood disintegrative disorder, and pervasive developmental disorder/not otherwise specified.^22-24 According to DSM-5, an individual with ASD must have 1) persistent deficits in social communication and social interactions, 2) restricted, repetitive patterns of behavior, interests or activities, 3) symptoms present in early childhood (but may not become fully manifest until social demands exceed limited capacities), and 4) symptoms that limit and impair everyday functioning.^22,24

Children who don’t meet DSM-5 criteria for autism spectrum disorder may be included in a new category called social communication disorder. Onset of the disorder must be obvious before age 3 for a child to be eligible for EI. Children who do not qualify for an ASD diagnosis under the new DSM-5 definition may be included in a new category called social communication disorder (SCD) under “Communication Disorders” in DSM-5. SCD is defined as impairment in pragmatics that impacts development of social relationships and comprehension of social conversations. ASD must be ruled out before a diagnosis of SCD can be made.^23,24

Early intervention services

Individual states offer EI services to families to facilitate detection of developmental delays in children and to provide a comprehensive system of support designed to reduce the effects of disabilities (or to prevent learning and developmental problems later in life).⁵ The rationale for EI is that the earlier interventions are started, the less likely later interventions will be needed.²⁵ EI services are provided to children from birth through their third birthday; services are free of charge to eligible families or on a sliding payment scale determined by a family’s income.

Resources for information and referrals. To access EI services on behalf of a family, contact a local hospital or point them to the CPIR (http://www.parentcenterhub.org/repository/disability-landing/).² For families with disabled children older than 3, consider suggesting that parents contact the local school district (even if the child isn’t enrolled there) to arrange an evaluation under the Individuals with Disabilities Education Act (http://idea.ed.gov/).²⁶ Because individual states’ EI organizations may have slightly different procedures, it is best to consult one’s own state EI site for specific information regarding referrals (TABLE 3).

Efficacy of treatment

The prevalence of specific communication disorders varies widely, as do prognoses, possibly due to the variability of underlying causes (physical/biological/medical or environmental/educational).²⁷ Also, as described earlier, “communication disorder” is an umbrella term inclusive of problems as diverse as resonance (eg, hypernasality due to a submucous palatal cleft), severe language delay (eg, due to Down syndrome), or lisp (misarticulation of the “s” sound).

Is therapy effective? Speech and language pathologists use the National Outcomes Measurement System (NOMS) as an index of the outcomes of treatment on functional communication along 6 scales. The most frequent types of communication problems seen in the prekindergarten children’s NOMS were “articulation” (75% of children), “spoken language production” (61%), and “spoken language comprehension” (42%). Problems in the remaining 3 scales (“pragmatics,” “cognitive orientation,” and “swallowing”) were seen in fewer than 15% of the preschool-age students.²⁸

Articulation therapy yielded improvement in 69.3% of cases, spoken language comprehension therapy in 65.3%, and spoken language production therapy in 65.2%.²⁸ Such outcomes support regular screening of children’s communication development and, as needed, referral for EI.

TABLE 3

CORRESPONDENCE
Christopher Mulrine, EdD; William Paterson University, 1600 Valley Road 3003 Wayne, NJ 07474-0920; [email protected]

A switch to hydromorphone 20 mg/d—the physician used the 5:1 morphine-to-hydromorphone conversion ratio, then decreased the dose by 50% to account for incomplete cross-tolerance—left Ms. G lethargic. In addition, her pain score rose to 5, and she began having difficulty swallowing the medication. Prior to the drug rotation, she was able to perform light tasks and was alert enough to interact with her family.

If Ms. G were your patient, what would be your next step?

By bypassing 
first-pass metabolism, intrathecal drug delivery provides greater pain relief at lower dosages and with fewer adverse effects than systemic delivery. Continuous intrathecal (IT) drug delivery systems have been in use for more than 30 years.¹ And, while IT administration of analgesia has become increasingly useful for patients with refractory chronic pain and spasticity, it remains an underutilized resource.² Delivered directly into the pre- and post-synaptic opioid receptors in the dorsal horn of the spinal cord, IT analgesia bypasses first-pass metabolism. The result: a higher rate of efficacy, with smaller dosages and fewer adverse effects than systemic delivery.¹

The drugs are delivered via a small battery-powered programmable pump that is implanted under the subcutaneous tissue of the abdomen and connected to a catheter tunneled to the site of spinal entry. The device must be refilled periodically—typically every one to 3 months—but this is not a difficult process. It can be done in an office setting or in the patient’s home by a specially trained visiting nurse.³

There is ample reason to consider this approach when systemic analgesics or antispasmodics fail to control pain or cause unacceptable adverse effects. So why isn’t it used more frequently? One factor may be that many primary care physicians—often the first practitioners called upon to manage these complicated cases—know too little about it.

Who is a potential candidate for IT analgesia? What medications can be administered via this route? What is the role of a family physician (FP) in coordinating and overseeing the care of a patient being treated with IT therapy? Our goals in writing this review are to address these questions.

Patient selection: Not just for cancer pain

Because of the invasive nature and high cost of implantation, intrathecal pumps are typically reserved for patients with a life expectancy of more than 6 months. FPs interested in referring patients for IT therapy have many factors to consider before consulting a pain specialist. Foremost among them are the different criteria for individuals with cancer-related pain and those with chronic nonmalignant pain.

IT analgesia for cancer pain has been shown to improve patients’ quality of life and potentially increase long-term survival due to a decrease in systemic toxicity.^4-6 An appropriate candidate is an individual who, like Ms. G, was initially responsive to systemic opioids but later developed refractory symptoms or intolerance.⁷ Because of the invasive nature and high cost of implantation, subcutaneous IT pumps are typically reserved for patients with a life expectancy of more than 6 months.⁷ But implantation may be considered for those with a shorter life expectancy if they have severe pain or cannot tolerate the adverse effects of systemic analgesia.

Noncancer pain is more complex

The use of IT analgesia in patients with chronic nonmalignant pain, such as failed back surgery syndrome, spasticity associated with multiple sclerosis, or diabetic neuropathy, is both more controversial and more complex. It is important for FPs to recognize the multidimensional nature of this type of pain, which may be complicated by physical, psychological, and behavioral factors, including the possibility of addiction.^8-11

Although IT analgesia is less subject to abuse and diversion than systemic opioids, the dependent relationship associated with a continuous delivery system makes risk stratification a necessity.¹² Psychological testing is commonly used to evaluate potential candidates for long-term IT analgesia.

Prior to placement, patients must have had a failed course of conservative pain management and have no surgical options, no medical contraindications (eg, spinal pathology or susceptibility to infection), and no evidence of active addiction.¹² A medication history is crucial, too, to identify use of anticoagulation therapy—a relative contraindication—as well as drug allergies and potential drug-drug interactions to guard against.³

An IT trial may be required

Psychological testing is commonly used to evaluate potential candidates for long-term intrathecal analgesia. It is common practice for patients to undergo an IT analgesia trial prior to implantation of a subcutaneous pump. This involves using an external pump to infuse the selected medication intrathecally and slowly titrating it according to symptoms for 2 to 3 days. During this time frame, the patient records his or her response; a reduction by more than half in VAS pain score is considered a success, indicating that the patient is an appropriate candidate for placement of the device.^3,13

Drug choices—a look at the evidence

The US Food and Drug Administration (FDA) has approved 3 medications for continuous IT delivery: morphine, ziconotide, and baclofen. But it is common practice to use alternative agents, such as other opioids, local anesthetics, or alpha 2-adrenergic agonists (TABLE).^3,14-34 

CASE  › Ms. G’s primary care physician referred her to a pain specialist, who thought she would benefit from IT analgesia. After a successful single-shot IT trial with 0.5 mg morphine, the patient underwent implantation.  The specialist chose morphine as the IT agent because of Ms. G’s history of successful pain relief with it, and because such a low dose was unlikely to be a problem for a patient with renal failure.

A month later, when she returned to the specialist to have the pump refilled, Ms. G reported a pain score of 3.

Opioids such as morphine exhibit a wider spread of analgesia when administered intrathecally, resulting in fewer adverse effects than systemic opioids.^13,35,36 The mu-opioid receptors in the dorsal horn of the spinal cord are the primary target of IT opioids.

In a multicenter randomized trial involving 200 cancer patients on opioids, Smith et al⁴ compared implantable IT drug delivery systems with comprehensive medical management. The mean VAS pain score in the IT group fell 52% vs a decline of 39% in the medical management group. 

The evidence supporting IT opioids for nonmalignant pain is not as strong. This may be due to inherent differences in pain mechanisms. In cancer pain, between 75% and 90% of pain is either nociceptive or mixed nociceptive-neuropathic; the etiology of noncancer pain is more variable.^37-39 

Although IT opioid therapy is associated with a lower incidence of adverse effects than systemic therapy, this route is not devoid of adverse effects. Opioids delivered intrathecally may still be associated with respiratory depression, constipation, urinary retention, nausea/vomiting, sweating, and hyperalgesia.³⁹ In addition, chronic opioid use suppresses the hypothalamic-pituitary-gonadal axis and the hypothalamic-pituitary-adrenal axis^14,40,41—a risk with long-term IT as well as systemic administration.¹⁴  Respiratory depression most commonly results from accidental overdosing, and patients must be monitored during initiation and dose escalation of IT opioid therapy.¹⁵

Local anesthetics. Numerous studies have documented the favorable outcomes of combining local anesthetics with opioids for patients with cancer^16-20 and noncancer pain.^21,22 Local anesthetics work via the blockade of voltage-gated sodium channels, interfering with neuron depolarization.¹⁷

Numerous studies have documented the favorable outcomes of combining local anesthetics with opioids for cancer and noncancer pain. A retrospective study in which patients with malignant pain and those with failed back surgery syndrome had bupivacaine added to their IT opioid solution found that the combination led to lower pain scores and a 23% reduction in opioid dosage.²⁰ In another retrospective review, researchers demonstrated that the coadministration of IT bupivacaine and an opioid decreased the rate of opioid dose escalation by 65% over the first year in patients with noncancer pain.²³

However, a double-blind randomized, crossover multicenter study found that in patients with chronic nonmalignant pain, the addition of bupivacaine to IT opioids failed to produce significant improvement in pain control compared with opioid use alone. Quality of life scores did improve, however, in the group receiving combination therapy.²⁴

Adverse effects of local anesthetics delivered intrathecally include numbness, paresthesias, weakness, bowel/bladder dysfunction, and neurotoxicity.^17,19,25  

Calcium channel blockers. Found in venom produced by the marine snail Conus magus, ziconotide blocks presynaptic N-type channels. It is the only calcium channel blocker used to manage chronic pain.²⁶ Several trials in patients with malignant and nonmalignant pain have shown a significant decrease in VAS pain scores compared with placebo.^25,26 In addition, a multicenter, double-blind placebo-controlled crossover study evaluating IT ziconotide for the treatment of refractory pain in 111 patients with cancer and AIDS found that the treatment group obtained significantly better pain relief than the controls (53% vs 17.5% using a VAS pain intensity score).²⁵ However, 31% of those in the treatment group experienced adverse effects, the most common of which were confusion, somnolence, and urinary retention.

Ziconotide has FDA approval only as monotherapy. But because of its high cost and adverse effect profile, it is mainly used in combination with other IT drugs.²⁷ Ziconotide increases the risk of suicide in patients with a history of depression.²⁸ The prevalence of adverse effects correlates with a higher dose, faster titration rate, and older age.^26,28

Alpha-2 adrenergic agonists. Clonidine is the only alpha-2 agonist with FDA approval for epidural use, with several studies supporting its off-label use in combination with IT therapy.^22,29 In a prospective open-label study evaluating combination IT therapy in patients with failed back surgery syndrome, 73% reported subjective ratings of good or excellent at 2-year follow-up.²² The most common adverse effects were sedation, hypotension, nausea, and dry mouth.

Gamma-aminobutyric acid (GABA) agonists. Baclofen, a GABA agonist with FDA approval for the treatment of spasticity, has been used intrathecally since the mid-1980s.³² Several studies have supported its effectiveness for this purpose.^30,42  Clinical studies have also found IT baclofen to be effective in treating conditions such as complex regional pain syndrome, central pain, and neuropathic pain secondary to failed back surgery syndrome.^31,32 In one randomized double-blind crossover trial, 7 women with complex regional pain syndrome were given bolus injections of baclofen or saline. Those treated with baclofen experienced a reduction in pain and regained function.³¹

In another trial—a double-blind placebo-controlled study of patients with multiple sclerosis and spinal cord injury comparing baclofen with placebo—those treated with baclofen showed significant reductions in dysesthetic and spasm-related pain.³² The most common adverse effects of baclofen are drowsiness, cognitive impairment, weakness, gastrointestinal complaints, and sexual dysfunction.³¹

Which patients and which drugs? An expert consensus

Due to the potential for inconsistent patient management and the use of therapies with anecdotal evidence, the Polyanalgesic Consensus Conference (PACC)—a panel of experts in IT therapy—convened in 2000, 2003, 2007, and 2011 to develop recommendations for IT therapy and an algorithm for drug selection. PACC’s list of chronic conditions for which IT should be considered includes axial low back pain, postherpetic neuralgia, spinal cord injury, spinal stenosis, pancreatitis, osteoporosis, compression fracture, and phantom limb pain, among others.

The algorithm contains separate arms for neuropathic, nociceptive, and mixed pain states. First-line agents for neuropathic pain include morphine, alone or combined with bupivacaine, and ziconotide. For nociceptive pain, morphine, hydromorphone, fentanyl, and ziconotide are all first-line agents; for mixed pain states, the appropriate choice should be based on the clinical scenario.³³

Overseeing IT pain management in primary care

Referring potential candidates for IT therapy to specialists in pain management is just the beginning. While patients typically return to the specialist for pump refills, it is important that they see their primary care physician regularly, as well. Vigilance is required of both the FP and the patient. Any sudden worsening in pain level or acute change in neurologic function must be reported to the pain specialist immediately. 

Adverse effects of medications are the most common complications

Kamran and Wright⁴³ performed a retrospective review of their practice’s Intrathecal Drug Delivery Systems database of 122 patients and found that adverse medication effects were most common, accounting for 77% of complications.

Catheter malfunctions were next, at 16%, followed by infections, at 5%.⁴³ In other studies, catheter-related complications were found to have an incidence of 15% to 25%.^44,45 Problems include kinking, breaking, leaking, and migration of the catheter. Advise patients to immediately contact their pain specialist for evaluation if they experience a sudden loss of, or change in, pain control.

Any sudden worsening in pain level or acute change in neurologic function must be reported immediately. Infectious complications, which occur infrequently, are usually limited to superficial wounds, although epidural abscesses and meningitis are possible.⁴⁶ Standard perioperative antibiotic administration helps to minimize the risk of infection. If a patient presents with signs and symptoms of an epidural abscess—back pain, fever, and variable neurologic deficits—emergent initiation of intravenous antibiotics is needed. Magnetic resonance imaging (MRI) with and without gadolinium should be obtained, as well.²²

Spinal damage. Although IT catheters are placed under fluoroscopic guidance, there is a risk of direct injury to the spinal cord; this is more common if the catheter is placed above the level of the conus medullaris. Damage to the spinal cord or exiting spinal nerves will manifest as pain, sensory loss, and/or weakness over a dermatomal distribution.⁴³

Neurologic sequelae, ranging from mild symptoms to paraplegia, can  result from the formation of a granuloma at the tip of the spinal catheter. A sudden increase in pain usually occurs prior to neurologic deterioration, thereby allowing for early detection and intervention.⁴⁷ Development of a granuloma appears to be related to the long-term infusion of high-concentration opioids.³⁴ The diagnosis is confirmed by MRI, but physical exam and history are imperative in making the initial diagnosis.

In cases of mild neurologic symptoms, a transition to saline infusion through the pump may allow the granuloma to absorb; more severe cases may require neurosurgical intervention.⁴⁷

Is your patient scheduled for an IT drug trial?

If a patient of yours is scheduled for an IT drug trial, ideally followed by pump implantation, microdosing—the practice of weaning the individual from oral opioids prior to the procedure so that very low doses of IT opioids will suffice—may play a role.^48,49 While this approach appears promising, however, there is little in the way of definitive evidence of efficacy.

CASE › Over time, Ms. G’s maintenance IT dose of morphine had to be slowly increased from 0.5 mg to 1 mg/d. At bimonthly visits with her FP, she consistently reports pain scores of 3 on a scale of 1 to 10. The patient’s function has returned to baseline, and she has minimal adverse effects. 

CORRESPONDENCE
Jessica Tsukanov, DO, Montefiore Medical Center, 3347 Steuben Avenue, Bronx, NY 10467; [email protected]

A switch to hydromorphone 20 mg/d—the physician used the 5:1 morphine-to-hydromorphone conversion ratio, then decreased the dose by 50% to account for incomplete cross-tolerance—left Ms. G lethargic. In addition, her pain score rose to 5, and she began having difficulty swallowing the medication. Prior to the drug rotation, she was able to perform light tasks and was alert enough to interact with her family.

If Ms. G were your patient, what would be your next step?

By bypassing 
first-pass metabolism, intrathecal drug delivery provides greater pain relief at lower dosages and with fewer adverse effects than systemic delivery. Continuous intrathecal (IT) drug delivery systems have been in use for more than 30 years.¹ And, while IT administration of analgesia has become increasingly useful for patients with refractory chronic pain and spasticity, it remains an underutilized resource.² Delivered directly into the pre- and post-synaptic opioid receptors in the dorsal horn of the spinal cord, IT analgesia bypasses first-pass metabolism. The result: a higher rate of efficacy, with smaller dosages and fewer adverse effects than systemic delivery.¹

The drugs are delivered via a small battery-powered programmable pump that is implanted under the subcutaneous tissue of the abdomen and connected to a catheter tunneled to the site of spinal entry. The device must be refilled periodically—typically every one to 3 months—but this is not a difficult process. It can be done in an office setting or in the patient’s home by a specially trained visiting nurse.³

There is ample reason to consider this approach when systemic analgesics or antispasmodics fail to control pain or cause unacceptable adverse effects. So why isn’t it used more frequently? One factor may be that many primary care physicians—often the first practitioners called upon to manage these complicated cases—know too little about it.

Who is a potential candidate for IT analgesia? What medications can be administered via this route? What is the role of a family physician (FP) in coordinating and overseeing the care of a patient being treated with IT therapy? Our goals in writing this review are to address these questions.

Patient selection: Not just for cancer pain

Because of the invasive nature and high cost of implantation, intrathecal pumps are typically reserved for patients with a life expectancy of more than 6 months. FPs interested in referring patients for IT therapy have many factors to consider before consulting a pain specialist. Foremost among them are the different criteria for individuals with cancer-related pain and those with chronic nonmalignant pain.

IT analgesia for cancer pain has been shown to improve patients’ quality of life and potentially increase long-term survival due to a decrease in systemic toxicity.^4-6 An appropriate candidate is an individual who, like Ms. G, was initially responsive to systemic opioids but later developed refractory symptoms or intolerance.⁷ Because of the invasive nature and high cost of implantation, subcutaneous IT pumps are typically reserved for patients with a life expectancy of more than 6 months.⁷ But implantation may be considered for those with a shorter life expectancy if they have severe pain or cannot tolerate the adverse effects of systemic analgesia.

Noncancer pain is more complex

The use of IT analgesia in patients with chronic nonmalignant pain, such as failed back surgery syndrome, spasticity associated with multiple sclerosis, or diabetic neuropathy, is both more controversial and more complex. It is important for FPs to recognize the multidimensional nature of this type of pain, which may be complicated by physical, psychological, and behavioral factors, including the possibility of addiction.^8-11

Although IT analgesia is less subject to abuse and diversion than systemic opioids, the dependent relationship associated with a continuous delivery system makes risk stratification a necessity.¹² Psychological testing is commonly used to evaluate potential candidates for long-term IT analgesia.

Prior to placement, patients must have had a failed course of conservative pain management and have no surgical options, no medical contraindications (eg, spinal pathology or susceptibility to infection), and no evidence of active addiction.¹² A medication history is crucial, too, to identify use of anticoagulation therapy—a relative contraindication—as well as drug allergies and potential drug-drug interactions to guard against.³

An IT trial may be required

Psychological testing is commonly used to evaluate potential candidates for long-term intrathecal analgesia. It is common practice for patients to undergo an IT analgesia trial prior to implantation of a subcutaneous pump. This involves using an external pump to infuse the selected medication intrathecally and slowly titrating it according to symptoms for 2 to 3 days. During this time frame, the patient records his or her response; a reduction by more than half in VAS pain score is considered a success, indicating that the patient is an appropriate candidate for placement of the device.^3,13

Drug choices—a look at the evidence

The US Food and Drug Administration (FDA) has approved 3 medications for continuous IT delivery: morphine, ziconotide, and baclofen. But it is common practice to use alternative agents, such as other opioids, local anesthetics, or alpha 2-adrenergic agonists (TABLE).^3,14-34 

CASE  › Ms. G’s primary care physician referred her to a pain specialist, who thought she would benefit from IT analgesia. After a successful single-shot IT trial with 0.5 mg morphine, the patient underwent implantation.  The specialist chose morphine as the IT agent because of Ms. G’s history of successful pain relief with it, and because such a low dose was unlikely to be a problem for a patient with renal failure.

A month later, when she returned to the specialist to have the pump refilled, Ms. G reported a pain score of 3.

Opioids such as morphine exhibit a wider spread of analgesia when administered intrathecally, resulting in fewer adverse effects than systemic opioids.^13,35,36 The mu-opioid receptors in the dorsal horn of the spinal cord are the primary target of IT opioids.

In a multicenter randomized trial involving 200 cancer patients on opioids, Smith et al⁴ compared implantable IT drug delivery systems with comprehensive medical management. The mean VAS pain score in the IT group fell 52% vs a decline of 39% in the medical management group. 

The evidence supporting IT opioids for nonmalignant pain is not as strong. This may be due to inherent differences in pain mechanisms. In cancer pain, between 75% and 90% of pain is either nociceptive or mixed nociceptive-neuropathic; the etiology of noncancer pain is more variable.^37-39 

Although IT opioid therapy is associated with a lower incidence of adverse effects than systemic therapy, this route is not devoid of adverse effects. Opioids delivered intrathecally may still be associated with respiratory depression, constipation, urinary retention, nausea/vomiting, sweating, and hyperalgesia.³⁹ In addition, chronic opioid use suppresses the hypothalamic-pituitary-gonadal axis and the hypothalamic-pituitary-adrenal axis^14,40,41—a risk with long-term IT as well as systemic administration.¹⁴  Respiratory depression most commonly results from accidental overdosing, and patients must be monitored during initiation and dose escalation of IT opioid therapy.¹⁵

Local anesthetics. Numerous studies have documented the favorable outcomes of combining local anesthetics with opioids for patients with cancer^16-20 and noncancer pain.^21,22 Local anesthetics work via the blockade of voltage-gated sodium channels, interfering with neuron depolarization.¹⁷

Numerous studies have documented the favorable outcomes of combining local anesthetics with opioids for cancer and noncancer pain. A retrospective study in which patients with malignant pain and those with failed back surgery syndrome had bupivacaine added to their IT opioid solution found that the combination led to lower pain scores and a 23% reduction in opioid dosage.²⁰ In another retrospective review, researchers demonstrated that the coadministration of IT bupivacaine and an opioid decreased the rate of opioid dose escalation by 65% over the first year in patients with noncancer pain.²³

However, a double-blind randomized, crossover multicenter study found that in patients with chronic nonmalignant pain, the addition of bupivacaine to IT opioids failed to produce significant improvement in pain control compared with opioid use alone. Quality of life scores did improve, however, in the group receiving combination therapy.²⁴

Adverse effects of local anesthetics delivered intrathecally include numbness, paresthesias, weakness, bowel/bladder dysfunction, and neurotoxicity.^17,19,25  

Calcium channel blockers. Found in venom produced by the marine snail Conus magus, ziconotide blocks presynaptic N-type channels. It is the only calcium channel blocker used to manage chronic pain.²⁶ Several trials in patients with malignant and nonmalignant pain have shown a significant decrease in VAS pain scores compared with placebo.^25,26 In addition, a multicenter, double-blind placebo-controlled crossover study evaluating IT ziconotide for the treatment of refractory pain in 111 patients with cancer and AIDS found that the treatment group obtained significantly better pain relief than the controls (53% vs 17.5% using a VAS pain intensity score).²⁵ However, 31% of those in the treatment group experienced adverse effects, the most common of which were confusion, somnolence, and urinary retention.

Ziconotide has FDA approval only as monotherapy. But because of its high cost and adverse effect profile, it is mainly used in combination with other IT drugs.²⁷ Ziconotide increases the risk of suicide in patients with a history of depression.²⁸ The prevalence of adverse effects correlates with a higher dose, faster titration rate, and older age.^26,28

Alpha-2 adrenergic agonists. Clonidine is the only alpha-2 agonist with FDA approval for epidural use, with several studies supporting its off-label use in combination with IT therapy.^22,29 In a prospective open-label study evaluating combination IT therapy in patients with failed back surgery syndrome, 73% reported subjective ratings of good or excellent at 2-year follow-up.²² The most common adverse effects were sedation, hypotension, nausea, and dry mouth.

Gamma-aminobutyric acid (GABA) agonists. Baclofen, a GABA agonist with FDA approval for the treatment of spasticity, has been used intrathecally since the mid-1980s.³² Several studies have supported its effectiveness for this purpose.^30,42  Clinical studies have also found IT baclofen to be effective in treating conditions such as complex regional pain syndrome, central pain, and neuropathic pain secondary to failed back surgery syndrome.^31,32 In one randomized double-blind crossover trial, 7 women with complex regional pain syndrome were given bolus injections of baclofen or saline. Those treated with baclofen experienced a reduction in pain and regained function.³¹

In another trial—a double-blind placebo-controlled study of patients with multiple sclerosis and spinal cord injury comparing baclofen with placebo—those treated with baclofen showed significant reductions in dysesthetic and spasm-related pain.³² The most common adverse effects of baclofen are drowsiness, cognitive impairment, weakness, gastrointestinal complaints, and sexual dysfunction.³¹

Which patients and which drugs? An expert consensus

Due to the potential for inconsistent patient management and the use of therapies with anecdotal evidence, the Polyanalgesic Consensus Conference (PACC)—a panel of experts in IT therapy—convened in 2000, 2003, 2007, and 2011 to develop recommendations for IT therapy and an algorithm for drug selection. PACC’s list of chronic conditions for which IT should be considered includes axial low back pain, postherpetic neuralgia, spinal cord injury, spinal stenosis, pancreatitis, osteoporosis, compression fracture, and phantom limb pain, among others.

The algorithm contains separate arms for neuropathic, nociceptive, and mixed pain states. First-line agents for neuropathic pain include morphine, alone or combined with bupivacaine, and ziconotide. For nociceptive pain, morphine, hydromorphone, fentanyl, and ziconotide are all first-line agents; for mixed pain states, the appropriate choice should be based on the clinical scenario.³³

Overseeing IT pain management in primary care

Referring potential candidates for IT therapy to specialists in pain management is just the beginning. While patients typically return to the specialist for pump refills, it is important that they see their primary care physician regularly, as well. Vigilance is required of both the FP and the patient. Any sudden worsening in pain level or acute change in neurologic function must be reported to the pain specialist immediately. 

Adverse effects of medications are the most common complications

Kamran and Wright⁴³ performed a retrospective review of their practice’s Intrathecal Drug Delivery Systems database of 122 patients and found that adverse medication effects were most common, accounting for 77% of complications.

Catheter malfunctions were next, at 16%, followed by infections, at 5%.⁴³ In other studies, catheter-related complications were found to have an incidence of 15% to 25%.^44,45 Problems include kinking, breaking, leaking, and migration of the catheter. Advise patients to immediately contact their pain specialist for evaluation if they experience a sudden loss of, or change in, pain control.

Any sudden worsening in pain level or acute change in neurologic function must be reported immediately. Infectious complications, which occur infrequently, are usually limited to superficial wounds, although epidural abscesses and meningitis are possible.⁴⁶ Standard perioperative antibiotic administration helps to minimize the risk of infection. If a patient presents with signs and symptoms of an epidural abscess—back pain, fever, and variable neurologic deficits—emergent initiation of intravenous antibiotics is needed. Magnetic resonance imaging (MRI) with and without gadolinium should be obtained, as well.²²

Spinal damage. Although IT catheters are placed under fluoroscopic guidance, there is a risk of direct injury to the spinal cord; this is more common if the catheter is placed above the level of the conus medullaris. Damage to the spinal cord or exiting spinal nerves will manifest as pain, sensory loss, and/or weakness over a dermatomal distribution.⁴³

Neurologic sequelae, ranging from mild symptoms to paraplegia, can  result from the formation of a granuloma at the tip of the spinal catheter. A sudden increase in pain usually occurs prior to neurologic deterioration, thereby allowing for early detection and intervention.⁴⁷ Development of a granuloma appears to be related to the long-term infusion of high-concentration opioids.³⁴ The diagnosis is confirmed by MRI, but physical exam and history are imperative in making the initial diagnosis.

In cases of mild neurologic symptoms, a transition to saline infusion through the pump may allow the granuloma to absorb; more severe cases may require neurosurgical intervention.⁴⁷

Is your patient scheduled for an IT drug trial?

If a patient of yours is scheduled for an IT drug trial, ideally followed by pump implantation, microdosing—the practice of weaning the individual from oral opioids prior to the procedure so that very low doses of IT opioids will suffice—may play a role.^48,49 While this approach appears promising, however, there is little in the way of definitive evidence of efficacy.

CASE › Over time, Ms. G’s maintenance IT dose of morphine had to be slowly increased from 0.5 mg to 1 mg/d. At bimonthly visits with her FP, she consistently reports pain scores of 3 on a scale of 1 to 10. The patient’s function has returned to baseline, and she has minimal adverse effects. 

CORRESPONDENCE
Jessica Tsukanov, DO, Montefiore Medical Center, 3347 Steuben Avenue, Bronx, NY 10467; [email protected]

A switch to hydromorphone 20 mg/d—the physician used the 5:1 morphine-to-hydromorphone conversion ratio, then decreased the dose by 50% to account for incomplete cross-tolerance—left Ms. G lethargic. In addition, her pain score rose to 5, and she began having difficulty swallowing the medication. Prior to the drug rotation, she was able to perform light tasks and was alert enough to interact with her family.

If Ms. G were your patient, what would be your next step?

By bypassing 
first-pass metabolism, intrathecal drug delivery provides greater pain relief at lower dosages and with fewer adverse effects than systemic delivery. Continuous intrathecal (IT) drug delivery systems have been in use for more than 30 years.¹ And, while IT administration of analgesia has become increasingly useful for patients with refractory chronic pain and spasticity, it remains an underutilized resource.² Delivered directly into the pre- and post-synaptic opioid receptors in the dorsal horn of the spinal cord, IT analgesia bypasses first-pass metabolism. The result: a higher rate of efficacy, with smaller dosages and fewer adverse effects than systemic delivery.¹

The drugs are delivered via a small battery-powered programmable pump that is implanted under the subcutaneous tissue of the abdomen and connected to a catheter tunneled to the site of spinal entry. The device must be refilled periodically—typically every one to 3 months—but this is not a difficult process. It can be done in an office setting or in the patient’s home by a specially trained visiting nurse.³

There is ample reason to consider this approach when systemic analgesics or antispasmodics fail to control pain or cause unacceptable adverse effects. So why isn’t it used more frequently? One factor may be that many primary care physicians—often the first practitioners called upon to manage these complicated cases—know too little about it.

Who is a potential candidate for IT analgesia? What medications can be administered via this route? What is the role of a family physician (FP) in coordinating and overseeing the care of a patient being treated with IT therapy? Our goals in writing this review are to address these questions.

Patient selection: Not just for cancer pain

Because of the invasive nature and high cost of implantation, intrathecal pumps are typically reserved for patients with a life expectancy of more than 6 months. FPs interested in referring patients for IT therapy have many factors to consider before consulting a pain specialist. Foremost among them are the different criteria for individuals with cancer-related pain and those with chronic nonmalignant pain.

IT analgesia for cancer pain has been shown to improve patients’ quality of life and potentially increase long-term survival due to a decrease in systemic toxicity.^4-6 An appropriate candidate is an individual who, like Ms. G, was initially responsive to systemic opioids but later developed refractory symptoms or intolerance.⁷ Because of the invasive nature and high cost of implantation, subcutaneous IT pumps are typically reserved for patients with a life expectancy of more than 6 months.⁷ But implantation may be considered for those with a shorter life expectancy if they have severe pain or cannot tolerate the adverse effects of systemic analgesia.

Noncancer pain is more complex

The use of IT analgesia in patients with chronic nonmalignant pain, such as failed back surgery syndrome, spasticity associated with multiple sclerosis, or diabetic neuropathy, is both more controversial and more complex. It is important for FPs to recognize the multidimensional nature of this type of pain, which may be complicated by physical, psychological, and behavioral factors, including the possibility of addiction.^8-11

Although IT analgesia is less subject to abuse and diversion than systemic opioids, the dependent relationship associated with a continuous delivery system makes risk stratification a necessity.¹² Psychological testing is commonly used to evaluate potential candidates for long-term IT analgesia.

Prior to placement, patients must have had a failed course of conservative pain management and have no surgical options, no medical contraindications (eg, spinal pathology or susceptibility to infection), and no evidence of active addiction.¹² A medication history is crucial, too, to identify use of anticoagulation therapy—a relative contraindication—as well as drug allergies and potential drug-drug interactions to guard against.³

An IT trial may be required

Psychological testing is commonly used to evaluate potential candidates for long-term intrathecal analgesia. It is common practice for patients to undergo an IT analgesia trial prior to implantation of a subcutaneous pump. This involves using an external pump to infuse the selected medication intrathecally and slowly titrating it according to symptoms for 2 to 3 days. During this time frame, the patient records his or her response; a reduction by more than half in VAS pain score is considered a success, indicating that the patient is an appropriate candidate for placement of the device.^3,13

Drug choices—a look at the evidence

The US Food and Drug Administration (FDA) has approved 3 medications for continuous IT delivery: morphine, ziconotide, and baclofen. But it is common practice to use alternative agents, such as other opioids, local anesthetics, or alpha 2-adrenergic agonists (TABLE).^3,14-34 

CASE  › Ms. G’s primary care physician referred her to a pain specialist, who thought she would benefit from IT analgesia. After a successful single-shot IT trial with 0.5 mg morphine, the patient underwent implantation.  The specialist chose morphine as the IT agent because of Ms. G’s history of successful pain relief with it, and because such a low dose was unlikely to be a problem for a patient with renal failure.

A month later, when she returned to the specialist to have the pump refilled, Ms. G reported a pain score of 3.

Opioids such as morphine exhibit a wider spread of analgesia when administered intrathecally, resulting in fewer adverse effects than systemic opioids.^13,35,36 The mu-opioid receptors in the dorsal horn of the spinal cord are the primary target of IT opioids.

In a multicenter randomized trial involving 200 cancer patients on opioids, Smith et al⁴ compared implantable IT drug delivery systems with comprehensive medical management. The mean VAS pain score in the IT group fell 52% vs a decline of 39% in the medical management group. 

The evidence supporting IT opioids for nonmalignant pain is not as strong. This may be due to inherent differences in pain mechanisms. In cancer pain, between 75% and 90% of pain is either nociceptive or mixed nociceptive-neuropathic; the etiology of noncancer pain is more variable.^37-39 

Although IT opioid therapy is associated with a lower incidence of adverse effects than systemic therapy, this route is not devoid of adverse effects. Opioids delivered intrathecally may still be associated with respiratory depression, constipation, urinary retention, nausea/vomiting, sweating, and hyperalgesia.³⁹ In addition, chronic opioid use suppresses the hypothalamic-pituitary-gonadal axis and the hypothalamic-pituitary-adrenal axis^14,40,41—a risk with long-term IT as well as systemic administration.¹⁴  Respiratory depression most commonly results from accidental overdosing, and patients must be monitored during initiation and dose escalation of IT opioid therapy.¹⁵

Local anesthetics. Numerous studies have documented the favorable outcomes of combining local anesthetics with opioids for patients with cancer^16-20 and noncancer pain.^21,22 Local anesthetics work via the blockade of voltage-gated sodium channels, interfering with neuron depolarization.¹⁷

Numerous studies have documented the favorable outcomes of combining local anesthetics with opioids for cancer and noncancer pain. A retrospective study in which patients with malignant pain and those with failed back surgery syndrome had bupivacaine added to their IT opioid solution found that the combination led to lower pain scores and a 23% reduction in opioid dosage.²⁰ In another retrospective review, researchers demonstrated that the coadministration of IT bupivacaine and an opioid decreased the rate of opioid dose escalation by 65% over the first year in patients with noncancer pain.²³

However, a double-blind randomized, crossover multicenter study found that in patients with chronic nonmalignant pain, the addition of bupivacaine to IT opioids failed to produce significant improvement in pain control compared with opioid use alone. Quality of life scores did improve, however, in the group receiving combination therapy.²⁴

Adverse effects of local anesthetics delivered intrathecally include numbness, paresthesias, weakness, bowel/bladder dysfunction, and neurotoxicity.^17,19,25  

Calcium channel blockers. Found in venom produced by the marine snail Conus magus, ziconotide blocks presynaptic N-type channels. It is the only calcium channel blocker used to manage chronic pain.²⁶ Several trials in patients with malignant and nonmalignant pain have shown a significant decrease in VAS pain scores compared with placebo.^25,26 In addition, a multicenter, double-blind placebo-controlled crossover study evaluating IT ziconotide for the treatment of refractory pain in 111 patients with cancer and AIDS found that the treatment group obtained significantly better pain relief than the controls (53% vs 17.5% using a VAS pain intensity score).²⁵ However, 31% of those in the treatment group experienced adverse effects, the most common of which were confusion, somnolence, and urinary retention.

Ziconotide has FDA approval only as monotherapy. But because of its high cost and adverse effect profile, it is mainly used in combination with other IT drugs.²⁷ Ziconotide increases the risk of suicide in patients with a history of depression.²⁸ The prevalence of adverse effects correlates with a higher dose, faster titration rate, and older age.^26,28

Alpha-2 adrenergic agonists. Clonidine is the only alpha-2 agonist with FDA approval for epidural use, with several studies supporting its off-label use in combination with IT therapy.^22,29 In a prospective open-label study evaluating combination IT therapy in patients with failed back surgery syndrome, 73% reported subjective ratings of good or excellent at 2-year follow-up.²² The most common adverse effects were sedation, hypotension, nausea, and dry mouth.

Gamma-aminobutyric acid (GABA) agonists. Baclofen, a GABA agonist with FDA approval for the treatment of spasticity, has been used intrathecally since the mid-1980s.³² Several studies have supported its effectiveness for this purpose.^30,42  Clinical studies have also found IT baclofen to be effective in treating conditions such as complex regional pain syndrome, central pain, and neuropathic pain secondary to failed back surgery syndrome.^31,32 In one randomized double-blind crossover trial, 7 women with complex regional pain syndrome were given bolus injections of baclofen or saline. Those treated with baclofen experienced a reduction in pain and regained function.³¹

In another trial—a double-blind placebo-controlled study of patients with multiple sclerosis and spinal cord injury comparing baclofen with placebo—those treated with baclofen showed significant reductions in dysesthetic and spasm-related pain.³² The most common adverse effects of baclofen are drowsiness, cognitive impairment, weakness, gastrointestinal complaints, and sexual dysfunction.³¹

Which patients and which drugs? An expert consensus

Due to the potential for inconsistent patient management and the use of therapies with anecdotal evidence, the Polyanalgesic Consensus Conference (PACC)—a panel of experts in IT therapy—convened in 2000, 2003, 2007, and 2011 to develop recommendations for IT therapy and an algorithm for drug selection. PACC’s list of chronic conditions for which IT should be considered includes axial low back pain, postherpetic neuralgia, spinal cord injury, spinal stenosis, pancreatitis, osteoporosis, compression fracture, and phantom limb pain, among others.

The algorithm contains separate arms for neuropathic, nociceptive, and mixed pain states. First-line agents for neuropathic pain include morphine, alone or combined with bupivacaine, and ziconotide. For nociceptive pain, morphine, hydromorphone, fentanyl, and ziconotide are all first-line agents; for mixed pain states, the appropriate choice should be based on the clinical scenario.³³

Overseeing IT pain management in primary care

Referring potential candidates for IT therapy to specialists in pain management is just the beginning. While patients typically return to the specialist for pump refills, it is important that they see their primary care physician regularly, as well. Vigilance is required of both the FP and the patient. Any sudden worsening in pain level or acute change in neurologic function must be reported to the pain specialist immediately. 

Adverse effects of medications are the most common complications

Kamran and Wright⁴³ performed a retrospective review of their practice’s Intrathecal Drug Delivery Systems database of 122 patients and found that adverse medication effects were most common, accounting for 77% of complications.

Catheter malfunctions were next, at 16%, followed by infections, at 5%.⁴³ In other studies, catheter-related complications were found to have an incidence of 15% to 25%.^44,45 Problems include kinking, breaking, leaking, and migration of the catheter. Advise patients to immediately contact their pain specialist for evaluation if they experience a sudden loss of, or change in, pain control.

Any sudden worsening in pain level or acute change in neurologic function must be reported immediately. Infectious complications, which occur infrequently, are usually limited to superficial wounds, although epidural abscesses and meningitis are possible.⁴⁶ Standard perioperative antibiotic administration helps to minimize the risk of infection. If a patient presents with signs and symptoms of an epidural abscess—back pain, fever, and variable neurologic deficits—emergent initiation of intravenous antibiotics is needed. Magnetic resonance imaging (MRI) with and without gadolinium should be obtained, as well.²²

Spinal damage. Although IT catheters are placed under fluoroscopic guidance, there is a risk of direct injury to the spinal cord; this is more common if the catheter is placed above the level of the conus medullaris. Damage to the spinal cord or exiting spinal nerves will manifest as pain, sensory loss, and/or weakness over a dermatomal distribution.⁴³

Neurologic sequelae, ranging from mild symptoms to paraplegia, can  result from the formation of a granuloma at the tip of the spinal catheter. A sudden increase in pain usually occurs prior to neurologic deterioration, thereby allowing for early detection and intervention.⁴⁷ Development of a granuloma appears to be related to the long-term infusion of high-concentration opioids.³⁴ The diagnosis is confirmed by MRI, but physical exam and history are imperative in making the initial diagnosis.

In cases of mild neurologic symptoms, a transition to saline infusion through the pump may allow the granuloma to absorb; more severe cases may require neurosurgical intervention.⁴⁷

Is your patient scheduled for an IT drug trial?

If a patient of yours is scheduled for an IT drug trial, ideally followed by pump implantation, microdosing—the practice of weaning the individual from oral opioids prior to the procedure so that very low doses of IT opioids will suffice—may play a role.^48,49 While this approach appears promising, however, there is little in the way of definitive evidence of efficacy.

CASE › Over time, Ms. G’s maintenance IT dose of morphine had to be slowly increased from 0.5 mg to 1 mg/d. At bimonthly visits with her FP, she consistently reports pain scores of 3 on a scale of 1 to 10. The patient’s function has returned to baseline, and she has minimal adverse effects. 

CORRESPONDENCE
Jessica Tsukanov, DO, Montefiore Medical Center, 3347 Steuben Avenue, Bronx, NY 10467; [email protected]

Ms. C says she sleeps well, getting more than 8 hours of sleep per night on weekends but fewer than 7 hours per night during the week. But no matter how long she sleeps, she never awakens feeling refreshed. Ms. C reports that she doesn’t smoke, has no more than 4 alcoholic drinks per month, and adheres to an “average” diet. She is too tired to exercise.

Ms. C is single, with no children. Although she says she has a strong network of family and friends, she increasingly finds she has no energy for socializing. If Ms. C were your patient, what would you do?

Patients with an organ-based medical illness tend to associate their fatigue with activities that they are unable to complete, while those with fatigue that is not organ-based typically say that they're tired all the time.Fatigue is a common presenting symptom in primary care, accounting for about 5% of adult visits.¹ Defined as a generalized lack of energy, fatigue that persists despite adequate rest or is severe enough to disrupt an individual’s ability to participate in key social and/or occupational activities warrants a thorough investigation.

Because fatigue is a nonspecific symptom that may be linked to a number of medical and psychiatric illnesses or medications used to treat them, determining the cause can be difficult. In about half of all cases, no specific etiology is found.² This review, which includes the elements of a work-up and management strategies for patients presenting with ongoing fatigue, will help you arrive at the appropriate diagnosis and provide optimal treatment.

Chronic fatigue: Defining the terms

A definition of chronic fatigue syndrome (CFS) was initially published in 1988.³ In subsequent years, the term myalgic encephalomyelitis (ME) became popular. Although the terms are sometimes used interchangeably, ME often refers to patients whose condition is thought to have an infectious cause and for whom postexertional malaise is a hallmark symptom.⁴

CDC criteria. While several sets of diagnostic criteria for CFS have been developed, the most widely used is that of the Centers for Disease Control and Prevention (CDC), published in 1994 (TABLE 1).^5,6 A diagnosis of CFS is made on the basis of exclusion, subjective clinical interpretation, and patient self-report.

When the first 2 criteria—fatigue not due to ongoing exertion or other medical conditions that has lasted ≥6 months and is severe enough to interfere with daily activities—but fewer than 4 of the CDC’s 8 concurrent symptoms (eg, headache, unrefreshing sleep, and postexertion malaise lasting >24 hours) are present, idiopathic fatigue, rather than CFS, is diagnosed.⁶

International Consensus Criteria (ICC). In 2011, the ICC for ME were proposed in an effort to provide more specific diagnostic criteria (TABLE 2).⁷ The ICC emphasize fatigability, or what the authors identify as “post-exertional neuroimmune exhaustion.”

The ICC have not yet been broadly researched.  But an Australian study of patients with chronic fatigue found that those who met the ICC definition were sicker and more homogeneous, with significantly lower scores for physical and social functioning and bodily pain compared with those who fulfilled the CDC criteria alone.⁸

Chronic fatigue & neuropsychiatric conditions: Common threads

Recent research has made it clear that depression, somatization, and CFS share some biological underpinnings. These include biomarkers for inflammation, cell-mediated immune activation—which may be related to the symptoms of fatigue—autonomic dysfunction, and hyperalgesia.⁹ Evidence suggests that up to two-thirds of patients with CFS also meet the criteria for a psychiatric disorder.¹⁰ The most common psychiatric conditions are major depressive disorder (MDD), affecting an estimated 22% to 32% of those with CFS; anxiety disorder, affecting about 20%; and somatization disorder, affecting about 10%—at least double the incidence of the general population.¹⁰

Others point out, however, that up to half of those with CFS do not have a psychiatric disorder.¹¹ A diagnosis of somatization disorder, in particular, depends largely on a subjective interpretation of whether or not the presenting symptoms have a physical cause.¹⁰

CFS and MDD comorbidity. The most widely studied association between CFS and psychiatric disorders involves MDD. Observational studies have found patients with CFS have a lifetime prevalence of MDD of 65%,^12,13 which is higher than that of patients with other chronic diseases. Overlapping symptoms include fatigue, sleep disturbance, poor concentration, and memory problems. However, those with CFS have fewer symptoms related to anhedonia, poor self-esteem, guilt, and suicidal ideation compared with individuals with MDD.^12,13

There are several possible explanations for CFS and MDD comorbidity, which are not necessarily mutually exclusive.¹⁰ One theory is that CFS is an atypical form of depression; another holds that the disability associated with CFS leads to depression, as is the case with many other chronic illnesses; and a third points to overlapping pathophysiology.¹⁰

An emerging body of evidence suggests that CFS and MDD have some common oxidative and nitrosative biochemical pathways. Activated by infection, psychological stress, and immune disorders, they are believed to have damaging free radical and nitric oxide effects at the cellular level.¹⁴ The cellular effects can result in fatigue, muscle pain, and flu-like malaise.

Cortisol response differs

The majority of patients with CFS report symptoms of cognitive decline, with the acquisition of new verbal learning and information-processing speed particularly likely to be impaired.CFS and MDD might be distinguishable by another pathway—the hypothalamic-pituitary-adrenal (HPA) axis. MDD is classically associated with activation and raised cortisol levels, while CFS is consistently associated with impaired HPA axis functioning and reduced cortisol levels.¹⁰ The majority of patients with CFS report symptoms of cognitive decline, with the acquisition of new verbal learning and information-processing speed particularly likely to be impaired.¹⁵

A meta-analysis of 50 studies of patients with CFS showed deficits in attention, memory, and reaction time, but not in fine motor speed, vocabulary, or reasoning.¹⁶ Autonomic dysfunction has also been observed, including disordered sympathetic activity. The most frequently observed abnormalities on autonomic testing are postural hypotension, tachycardia syndrome, neurally mediated hypotension, and heart rate variability during tilt table testing.¹⁶

The link between infection and CFS

Several infectious agents have been associated with ME/CFS, including the Epstein-Barr virus (EBV), herpes simplex virus 6, parvovirus, Q fever, and Lyme disease.^8,17,18 Most agents that have been linked to ME/CFS are associated with persistent infection and thus incitement of the immune system.

Numerous observational studies^17,18 have documented postinfectious fatigue syndromes after acute viral and bacterial infections and symptoms suggestive of infection, such as fever, myalgias, and respiratory and gastrointestinal distress. In one prospective Australian study,¹⁹ investigators identified 253 cases of acute EBV, Ross River virus, and Q fever. Of those 253 patients, 12% went on to develop CFS, with a higher likelihood among those with more severe acute symptoms. No correlation with preexisting psychiatric disorders was found.

Several infectious agents have been associated with ME/CFS, including the Epstein-Barr virus, herpes simplex virus 6, parvovirus, Q fever, and Lyme disease.Muscle mitochondria studies have demonstrated what appear to be acquired abnormalities in those with CFS.^20,21 Signs of increased oxidative stress have been found in both blood and muscle samples from patients with CFS, and longitudinal studies suggest that oxidative stress is greatest during periods of clinical exacerbation.²² Increased lactate levels suggest increased anaerobic metabolism in the central nervous system consistent with mitochondrial dysfunction. Several studies have demonstrated that exercise can precipitate oxidative stress in patients with CFS, in contrast with healthy controls and controls with other chronic illnesses, suggesting a physiologic basis for their postexertional symptoms.¹⁷

Autoinflammatory syndrome induced by adjuvants, a rare syndrome associated with vaccine administration, has been linked to postvaccination adverse events, exposure to silicone implants, Gulf War syndrome (related to multiple vaccinations), and macrophagic myofasciitis. All involve exposure to immune adjuvants and have similar clinical manifestations. The corresponding exposures appear to trigger an autoimmune response in susceptible individuals.  The hepatitis B vaccine is most often associated with CFS, with symptoms occurring within 90 days of administration.²³

The clinical work-up: Putting knowledge into practice

Familiarity with potential causes of and connections with ME/CFS will help you ensure that patients who say they’re always tired receive a thorough work-up. Start with a medical history, inquiring directly about medical and psychiatric disorders that may contribute to fatigue (TABLE 3).^5,24,25 Include a medication history, as well, to help determine whether the fatigue is drug-related (TABLE 4).^5,25 

What to ask. Determine the onset, course, duration, daily pattern, and impact of fatigue on the patient’s daily life. Inquire, too, about related symptoms of daytime sleepiness, dyspnea on exertion, generalized weakness, and depressed mood. The prominence of any of these rather than fatigue, per se, point to a diagnosis of a chronic illness other than ME/CFS.

Keep in mind, too, that patients with an organ-based medical illness tend to associate their fatigue with activities that they are unable to complete, such as shopping or light housework. In contrast, those with fatigue that is not organ-based typically say that they’re tired all the time. Their fatigue is not necessarily related to exertion, nor does it improve with rest.²⁶

To address this distinction, take a sleep history, assessing both the quality and quantity of the patient’s sleep to determine how it affects symptoms.²⁷ Consider using a questionnaire designed to help distinguish between sleepiness—and a primary sleep disorder—and fatigue,²⁸ such as the Fatigue Severity Scale of Sleep Disorders (http://www.healthywomen.org/sites/default/files/FatigueSeverityScale.pdf).

What to rule out. In addition to a medical history, the physical examination should be oriented toward ruling out secondary causes of fatigue. In addition to a system-by-system approach to the differential diagnosis, carefully observe the patient’s general appearance, with attention to his or her level of alertness, grooming, and psychomotor agitation or retardation as possible signs of a psychiatric disorder.  To rule out neurologic causes, evaluate muscle bulk, tone, and strength; deep tendon reflexes; and sensory and cranial nerves, as well.²⁹

Lab tests to consider. In most cases of ME/CFS, basic studies—complete blood count with differential, erythrocyte sedimentation rate, blood chemistry, and thyroid-stimulating hormone (TSH) levels—are sufficient. When no medical or psychiatric cause has been found, additional tests may be ordered on a case-by-case basis, although laboratory analysis affects the management of fatigue in less than 5% of such patients.³⁰ Tests to consider include:

• Creatine kinase (for patients who report pain or muscle weakness)
• pregnancy test (for women of childbearing age)
• ferritin testing (for young women who might benefit from iron supplementation for levels <50 ng/mL even if anemia is not present)³¹
• hepatitis C screening (recommended by the US Preventive Services Task Force for those born between 1945 and 1965)³²
• human immunodeficiency virus screening and the purified protein derivative test for tuberculosis (based on patient history).

Forego routine testing for other infections

Routine testing for infectious diseases and conditions associated with fatigue, such as Epstein-Barr virus or Lyme disease, immune deficiency, and inflammatory disease, is unlikely to be helpful. Routine testing for infectious diseases and conditions associated with fatigue, such as EBV or Lyme disease, immune deficiency (eg, immunoglobulins), inflammatory disease (eg, antinuclear antibodies, rheumatoid factor), celiac disease, vitamin D deficiency, vitamin B₁₂ deficiency, or heavy metal toxicity, is unlikely to be helpful.²⁹ Additional testing simply to reassure a worried patient usually does not accomplish that objective.³³

Additional studies, referrals to consider. If you suspect that a patient has a sleep disorder, a referral to a sleep clinic to rule out idiopathic sleep disorders, obstructive sleep apnea, or movement disorders that interfere with sleep may be in order. Spirometry and echocardiography may be helpful for some patients. If you suspect peripheral muscle fatigue, a referral for neuromuscular testing is indicated.

CASE › Ms. C’s medical history reveals that she also suffers from irritable bowel syndrome, which she manages with diet and over-the-counter medication, as needed, for constipation or diarrhea. She denies having any other chronic conditions. Her only other symptoms, she reports, are mild upper back pain after spending long hours at the computer, and arthralgias in her left knee and both hands. She admits to being “somewhat depressed” in the last few months, but denies the presence of anhedonia.

The patient’s physical examination is normal, and her depression screen does not meet the criteria for MDD. Her metabolic chemistry panel, complete blood count, TSH, and sedimentation rate are all normal, as well.

Symptom management and coping strategies

When no specific cause of chronic fatigue is found, the focus shifts from diagnosis to symptom management and coping strategies.

Evidence of the effectiveness of specific therapies for ME/CFS is limited; however, the best-studied approaches are cognitive behavioral therapy (CBT) and graded exercise therapy.This requires engagement with the patient. It is important to acknowledge the existence of his or her symptoms and to reassure the patient that further investigation may be warranted later, should new symptoms emerge. Advise the patient, too, that periods of remission and relapse are likely.

Strategies designed to motivate patient self-management, as well as the formulation of patient-centered treatment plans, have been shown to reduce symptom scores.³⁴ Participation in a support group, as well as frequent follow-up visits with a primary care physician, a behavioral therapist, or both, may help to provide needed psychological support.

Evidence of the effectiveness of specific therapies for ME/CFS is limited; however, the best-studied approaches are cognitive behavioral therapy (CBT) and graded exercise therapy.³⁵ Exercise should be low intensity, such as walking or cycling for 30 minutes 3 times a week, with a gradual increase in duration and frequency over a period of weeks to months. Patients with cancer-related fatigue may benefit from yoga, group therapy, and stress management.³⁶

Associated mood and pain symptoms should be treated, as well. Bupropion, which is somewhat stimulating, may be considered as an initial treatment for patients with depression and clinically significant fatigue.

Other potentially beneficial approaches include a healthy diet, avoidance of more than nominal amounts of alcohol, relative avoidance of caffeine (no more than one cup of a caffeinated beverage in the morning), and stress reduction techniques.

Attention to good sleep hygiene may be especially beneficial, including a regular bedtime routine and sleep schedule, and elimination of bedroom light and noise. Pharmacologic treatments for insomnia should be used with caution, if at all.

Bupropion, which is somewhat stimulating, may be considered as an initial treatment for patients with depression and clinically significant fatigue.CASE › Ms. C receives a referral for CBT and is scheduled for a return visit in 4 weeks.

At the advice of both her primary care physician and the behavioral therapist, Ms. C gradually makes several lifestyle changes. She begins going to bed earlier on weeknights to ensure that she sleeps for at least 7 hours. She improves her diet, with increasing emphasis on vegetables, fruits, and whole grains. She also starts a walking program, increasing gradually to a total of 3 hours per week. After 4 months she adds a weekly trip to a gym, where she practices resistance training for about 40 minutes.

Ms. C also increases her social activities on weekends, and recently accepted an invitation to join a book club. Six months from her initial visit, Ms. C notes that although she is still more easily fatigued than most people, she has made significant improvement.

CORRESPONDENCE
Linda Speer, MD; University of Toledo, 3000 Arlington Avenue, MS 1179, Toledo, OH 43614; [email protected]

Ms. C says she sleeps well, getting more than 8 hours of sleep per night on weekends but fewer than 7 hours per night during the week. But no matter how long she sleeps, she never awakens feeling refreshed. Ms. C reports that she doesn’t smoke, has no more than 4 alcoholic drinks per month, and adheres to an “average” diet. She is too tired to exercise.

Ms. C is single, with no children. Although she says she has a strong network of family and friends, she increasingly finds she has no energy for socializing. If Ms. C were your patient, what would you do?

Patients with an organ-based medical illness tend to associate their fatigue with activities that they are unable to complete, while those with fatigue that is not organ-based typically say that they're tired all the time.Fatigue is a common presenting symptom in primary care, accounting for about 5% of adult visits.¹ Defined as a generalized lack of energy, fatigue that persists despite adequate rest or is severe enough to disrupt an individual’s ability to participate in key social and/or occupational activities warrants a thorough investigation.

Because fatigue is a nonspecific symptom that may be linked to a number of medical and psychiatric illnesses or medications used to treat them, determining the cause can be difficult. In about half of all cases, no specific etiology is found.² This review, which includes the elements of a work-up and management strategies for patients presenting with ongoing fatigue, will help you arrive at the appropriate diagnosis and provide optimal treatment.

Chronic fatigue: Defining the terms

A definition of chronic fatigue syndrome (CFS) was initially published in 1988.³ In subsequent years, the term myalgic encephalomyelitis (ME) became popular. Although the terms are sometimes used interchangeably, ME often refers to patients whose condition is thought to have an infectious cause and for whom postexertional malaise is a hallmark symptom.⁴

CDC criteria. While several sets of diagnostic criteria for CFS have been developed, the most widely used is that of the Centers for Disease Control and Prevention (CDC), published in 1994 (TABLE 1).^5,6 A diagnosis of CFS is made on the basis of exclusion, subjective clinical interpretation, and patient self-report.

When the first 2 criteria—fatigue not due to ongoing exertion or other medical conditions that has lasted ≥6 months and is severe enough to interfere with daily activities—but fewer than 4 of the CDC’s 8 concurrent symptoms (eg, headache, unrefreshing sleep, and postexertion malaise lasting >24 hours) are present, idiopathic fatigue, rather than CFS, is diagnosed.⁶

International Consensus Criteria (ICC). In 2011, the ICC for ME were proposed in an effort to provide more specific diagnostic criteria (TABLE 2).⁷ The ICC emphasize fatigability, or what the authors identify as “post-exertional neuroimmune exhaustion.”

The ICC have not yet been broadly researched.  But an Australian study of patients with chronic fatigue found that those who met the ICC definition were sicker and more homogeneous, with significantly lower scores for physical and social functioning and bodily pain compared with those who fulfilled the CDC criteria alone.⁸

Chronic fatigue & neuropsychiatric conditions: Common threads

Recent research has made it clear that depression, somatization, and CFS share some biological underpinnings. These include biomarkers for inflammation, cell-mediated immune activation—which may be related to the symptoms of fatigue—autonomic dysfunction, and hyperalgesia.⁹ Evidence suggests that up to two-thirds of patients with CFS also meet the criteria for a psychiatric disorder.¹⁰ The most common psychiatric conditions are major depressive disorder (MDD), affecting an estimated 22% to 32% of those with CFS; anxiety disorder, affecting about 20%; and somatization disorder, affecting about 10%—at least double the incidence of the general population.¹⁰

Others point out, however, that up to half of those with CFS do not have a psychiatric disorder.¹¹ A diagnosis of somatization disorder, in particular, depends largely on a subjective interpretation of whether or not the presenting symptoms have a physical cause.¹⁰

CFS and MDD comorbidity. The most widely studied association between CFS and psychiatric disorders involves MDD. Observational studies have found patients with CFS have a lifetime prevalence of MDD of 65%,^12,13 which is higher than that of patients with other chronic diseases. Overlapping symptoms include fatigue, sleep disturbance, poor concentration, and memory problems. However, those with CFS have fewer symptoms related to anhedonia, poor self-esteem, guilt, and suicidal ideation compared with individuals with MDD.^12,13

There are several possible explanations for CFS and MDD comorbidity, which are not necessarily mutually exclusive.¹⁰ One theory is that CFS is an atypical form of depression; another holds that the disability associated with CFS leads to depression, as is the case with many other chronic illnesses; and a third points to overlapping pathophysiology.¹⁰

An emerging body of evidence suggests that CFS and MDD have some common oxidative and nitrosative biochemical pathways. Activated by infection, psychological stress, and immune disorders, they are believed to have damaging free radical and nitric oxide effects at the cellular level.¹⁴ The cellular effects can result in fatigue, muscle pain, and flu-like malaise.

Cortisol response differs

The majority of patients with CFS report symptoms of cognitive decline, with the acquisition of new verbal learning and information-processing speed particularly likely to be impaired.CFS and MDD might be distinguishable by another pathway—the hypothalamic-pituitary-adrenal (HPA) axis. MDD is classically associated with activation and raised cortisol levels, while CFS is consistently associated with impaired HPA axis functioning and reduced cortisol levels.¹⁰ The majority of patients with CFS report symptoms of cognitive decline, with the acquisition of new verbal learning and information-processing speed particularly likely to be impaired.¹⁵

A meta-analysis of 50 studies of patients with CFS showed deficits in attention, memory, and reaction time, but not in fine motor speed, vocabulary, or reasoning.¹⁶ Autonomic dysfunction has also been observed, including disordered sympathetic activity. The most frequently observed abnormalities on autonomic testing are postural hypotension, tachycardia syndrome, neurally mediated hypotension, and heart rate variability during tilt table testing.¹⁶

The link between infection and CFS

Several infectious agents have been associated with ME/CFS, including the Epstein-Barr virus (EBV), herpes simplex virus 6, parvovirus, Q fever, and Lyme disease.^8,17,18 Most agents that have been linked to ME/CFS are associated with persistent infection and thus incitement of the immune system.

Numerous observational studies^17,18 have documented postinfectious fatigue syndromes after acute viral and bacterial infections and symptoms suggestive of infection, such as fever, myalgias, and respiratory and gastrointestinal distress. In one prospective Australian study,¹⁹ investigators identified 253 cases of acute EBV, Ross River virus, and Q fever. Of those 253 patients, 12% went on to develop CFS, with a higher likelihood among those with more severe acute symptoms. No correlation with preexisting psychiatric disorders was found.

Several infectious agents have been associated with ME/CFS, including the Epstein-Barr virus, herpes simplex virus 6, parvovirus, Q fever, and Lyme disease.Muscle mitochondria studies have demonstrated what appear to be acquired abnormalities in those with CFS.^20,21 Signs of increased oxidative stress have been found in both blood and muscle samples from patients with CFS, and longitudinal studies suggest that oxidative stress is greatest during periods of clinical exacerbation.²² Increased lactate levels suggest increased anaerobic metabolism in the central nervous system consistent with mitochondrial dysfunction. Several studies have demonstrated that exercise can precipitate oxidative stress in patients with CFS, in contrast with healthy controls and controls with other chronic illnesses, suggesting a physiologic basis for their postexertional symptoms.¹⁷

Autoinflammatory syndrome induced by adjuvants, a rare syndrome associated with vaccine administration, has been linked to postvaccination adverse events, exposure to silicone implants, Gulf War syndrome (related to multiple vaccinations), and macrophagic myofasciitis. All involve exposure to immune adjuvants and have similar clinical manifestations. The corresponding exposures appear to trigger an autoimmune response in susceptible individuals.  The hepatitis B vaccine is most often associated with CFS, with symptoms occurring within 90 days of administration.²³

The clinical work-up: Putting knowledge into practice

Familiarity with potential causes of and connections with ME/CFS will help you ensure that patients who say they’re always tired receive a thorough work-up. Start with a medical history, inquiring directly about medical and psychiatric disorders that may contribute to fatigue (TABLE 3).^5,24,25 Include a medication history, as well, to help determine whether the fatigue is drug-related (TABLE 4).^5,25 

What to ask. Determine the onset, course, duration, daily pattern, and impact of fatigue on the patient’s daily life. Inquire, too, about related symptoms of daytime sleepiness, dyspnea on exertion, generalized weakness, and depressed mood. The prominence of any of these rather than fatigue, per se, point to a diagnosis of a chronic illness other than ME/CFS.

Keep in mind, too, that patients with an organ-based medical illness tend to associate their fatigue with activities that they are unable to complete, such as shopping or light housework. In contrast, those with fatigue that is not organ-based typically say that they’re tired all the time. Their fatigue is not necessarily related to exertion, nor does it improve with rest.²⁶

To address this distinction, take a sleep history, assessing both the quality and quantity of the patient’s sleep to determine how it affects symptoms.²⁷ Consider using a questionnaire designed to help distinguish between sleepiness—and a primary sleep disorder—and fatigue,²⁸ such as the Fatigue Severity Scale of Sleep Disorders (http://www.healthywomen.org/sites/default/files/FatigueSeverityScale.pdf).

What to rule out. In addition to a medical history, the physical examination should be oriented toward ruling out secondary causes of fatigue. In addition to a system-by-system approach to the differential diagnosis, carefully observe the patient’s general appearance, with attention to his or her level of alertness, grooming, and psychomotor agitation or retardation as possible signs of a psychiatric disorder.  To rule out neurologic causes, evaluate muscle bulk, tone, and strength; deep tendon reflexes; and sensory and cranial nerves, as well.²⁹

Lab tests to consider. In most cases of ME/CFS, basic studies—complete blood count with differential, erythrocyte sedimentation rate, blood chemistry, and thyroid-stimulating hormone (TSH) levels—are sufficient. When no medical or psychiatric cause has been found, additional tests may be ordered on a case-by-case basis, although laboratory analysis affects the management of fatigue in less than 5% of such patients.³⁰ Tests to consider include:

• Creatine kinase (for patients who report pain or muscle weakness)
• pregnancy test (for women of childbearing age)
• ferritin testing (for young women who might benefit from iron supplementation for levels <50 ng/mL even if anemia is not present)³¹
• hepatitis C screening (recommended by the US Preventive Services Task Force for those born between 1945 and 1965)³²
• human immunodeficiency virus screening and the purified protein derivative test for tuberculosis (based on patient history).

Forego routine testing for other infections

Routine testing for infectious diseases and conditions associated with fatigue, such as Epstein-Barr virus or Lyme disease, immune deficiency, and inflammatory disease, is unlikely to be helpful. Routine testing for infectious diseases and conditions associated with fatigue, such as EBV or Lyme disease, immune deficiency (eg, immunoglobulins), inflammatory disease (eg, antinuclear antibodies, rheumatoid factor), celiac disease, vitamin D deficiency, vitamin B₁₂ deficiency, or heavy metal toxicity, is unlikely to be helpful.²⁹ Additional testing simply to reassure a worried patient usually does not accomplish that objective.³³

Additional studies, referrals to consider. If you suspect that a patient has a sleep disorder, a referral to a sleep clinic to rule out idiopathic sleep disorders, obstructive sleep apnea, or movement disorders that interfere with sleep may be in order. Spirometry and echocardiography may be helpful for some patients. If you suspect peripheral muscle fatigue, a referral for neuromuscular testing is indicated.

CASE › Ms. C’s medical history reveals that she also suffers from irritable bowel syndrome, which she manages with diet and over-the-counter medication, as needed, for constipation or diarrhea. She denies having any other chronic conditions. Her only other symptoms, she reports, are mild upper back pain after spending long hours at the computer, and arthralgias in her left knee and both hands. She admits to being “somewhat depressed” in the last few months, but denies the presence of anhedonia.

The patient’s physical examination is normal, and her depression screen does not meet the criteria for MDD. Her metabolic chemistry panel, complete blood count, TSH, and sedimentation rate are all normal, as well.

Symptom management and coping strategies

When no specific cause of chronic fatigue is found, the focus shifts from diagnosis to symptom management and coping strategies.

Evidence of the effectiveness of specific therapies for ME/CFS is limited; however, the best-studied approaches are cognitive behavioral therapy (CBT) and graded exercise therapy.This requires engagement with the patient. It is important to acknowledge the existence of his or her symptoms and to reassure the patient that further investigation may be warranted later, should new symptoms emerge. Advise the patient, too, that periods of remission and relapse are likely.

Strategies designed to motivate patient self-management, as well as the formulation of patient-centered treatment plans, have been shown to reduce symptom scores.³⁴ Participation in a support group, as well as frequent follow-up visits with a primary care physician, a behavioral therapist, or both, may help to provide needed psychological support.

Evidence of the effectiveness of specific therapies for ME/CFS is limited; however, the best-studied approaches are cognitive behavioral therapy (CBT) and graded exercise therapy.³⁵ Exercise should be low intensity, such as walking or cycling for 30 minutes 3 times a week, with a gradual increase in duration and frequency over a period of weeks to months. Patients with cancer-related fatigue may benefit from yoga, group therapy, and stress management.³⁶

Associated mood and pain symptoms should be treated, as well. Bupropion, which is somewhat stimulating, may be considered as an initial treatment for patients with depression and clinically significant fatigue.

Other potentially beneficial approaches include a healthy diet, avoidance of more than nominal amounts of alcohol, relative avoidance of caffeine (no more than one cup of a caffeinated beverage in the morning), and stress reduction techniques.

Attention to good sleep hygiene may be especially beneficial, including a regular bedtime routine and sleep schedule, and elimination of bedroom light and noise. Pharmacologic treatments for insomnia should be used with caution, if at all.

Bupropion, which is somewhat stimulating, may be considered as an initial treatment for patients with depression and clinically significant fatigue.CASE › Ms. C receives a referral for CBT and is scheduled for a return visit in 4 weeks.

At the advice of both her primary care physician and the behavioral therapist, Ms. C gradually makes several lifestyle changes. She begins going to bed earlier on weeknights to ensure that she sleeps for at least 7 hours. She improves her diet, with increasing emphasis on vegetables, fruits, and whole grains. She also starts a walking program, increasing gradually to a total of 3 hours per week. After 4 months she adds a weekly trip to a gym, where she practices resistance training for about 40 minutes.

Ms. C also increases her social activities on weekends, and recently accepted an invitation to join a book club. Six months from her initial visit, Ms. C notes that although she is still more easily fatigued than most people, she has made significant improvement.

CORRESPONDENCE
Linda Speer, MD; University of Toledo, 3000 Arlington Avenue, MS 1179, Toledo, OH 43614; [email protected]

Ms. C says she sleeps well, getting more than 8 hours of sleep per night on weekends but fewer than 7 hours per night during the week. But no matter how long she sleeps, she never awakens feeling refreshed. Ms. C reports that she doesn’t smoke, has no more than 4 alcoholic drinks per month, and adheres to an “average” diet. She is too tired to exercise.

Ms. C is single, with no children. Although she says she has a strong network of family and friends, she increasingly finds she has no energy for socializing. If Ms. C were your patient, what would you do?

Patients with an organ-based medical illness tend to associate their fatigue with activities that they are unable to complete, while those with fatigue that is not organ-based typically say that they're tired all the time.Fatigue is a common presenting symptom in primary care, accounting for about 5% of adult visits.¹ Defined as a generalized lack of energy, fatigue that persists despite adequate rest or is severe enough to disrupt an individual’s ability to participate in key social and/or occupational activities warrants a thorough investigation.

Because fatigue is a nonspecific symptom that may be linked to a number of medical and psychiatric illnesses or medications used to treat them, determining the cause can be difficult. In about half of all cases, no specific etiology is found.² This review, which includes the elements of a work-up and management strategies for patients presenting with ongoing fatigue, will help you arrive at the appropriate diagnosis and provide optimal treatment.

Chronic fatigue: Defining the terms

A definition of chronic fatigue syndrome (CFS) was initially published in 1988.³ In subsequent years, the term myalgic encephalomyelitis (ME) became popular. Although the terms are sometimes used interchangeably, ME often refers to patients whose condition is thought to have an infectious cause and for whom postexertional malaise is a hallmark symptom.⁴

CDC criteria. While several sets of diagnostic criteria for CFS have been developed, the most widely used is that of the Centers for Disease Control and Prevention (CDC), published in 1994 (TABLE 1).^5,6 A diagnosis of CFS is made on the basis of exclusion, subjective clinical interpretation, and patient self-report.

When the first 2 criteria—fatigue not due to ongoing exertion or other medical conditions that has lasted ≥6 months and is severe enough to interfere with daily activities—but fewer than 4 of the CDC’s 8 concurrent symptoms (eg, headache, unrefreshing sleep, and postexertion malaise lasting >24 hours) are present, idiopathic fatigue, rather than CFS, is diagnosed.⁶

International Consensus Criteria (ICC). In 2011, the ICC for ME were proposed in an effort to provide more specific diagnostic criteria (TABLE 2).⁷ The ICC emphasize fatigability, or what the authors identify as “post-exertional neuroimmune exhaustion.”

The ICC have not yet been broadly researched.  But an Australian study of patients with chronic fatigue found that those who met the ICC definition were sicker and more homogeneous, with significantly lower scores for physical and social functioning and bodily pain compared with those who fulfilled the CDC criteria alone.⁸

Chronic fatigue & neuropsychiatric conditions: Common threads

Recent research has made it clear that depression, somatization, and CFS share some biological underpinnings. These include biomarkers for inflammation, cell-mediated immune activation—which may be related to the symptoms of fatigue—autonomic dysfunction, and hyperalgesia.⁹ Evidence suggests that up to two-thirds of patients with CFS also meet the criteria for a psychiatric disorder.¹⁰ The most common psychiatric conditions are major depressive disorder (MDD), affecting an estimated 22% to 32% of those with CFS; anxiety disorder, affecting about 20%; and somatization disorder, affecting about 10%—at least double the incidence of the general population.¹⁰

Others point out, however, that up to half of those with CFS do not have a psychiatric disorder.¹¹ A diagnosis of somatization disorder, in particular, depends largely on a subjective interpretation of whether or not the presenting symptoms have a physical cause.¹⁰

CFS and MDD comorbidity. The most widely studied association between CFS and psychiatric disorders involves MDD. Observational studies have found patients with CFS have a lifetime prevalence of MDD of 65%,^12,13 which is higher than that of patients with other chronic diseases. Overlapping symptoms include fatigue, sleep disturbance, poor concentration, and memory problems. However, those with CFS have fewer symptoms related to anhedonia, poor self-esteem, guilt, and suicidal ideation compared with individuals with MDD.^12,13

There are several possible explanations for CFS and MDD comorbidity, which are not necessarily mutually exclusive.¹⁰ One theory is that CFS is an atypical form of depression; another holds that the disability associated with CFS leads to depression, as is the case with many other chronic illnesses; and a third points to overlapping pathophysiology.¹⁰

An emerging body of evidence suggests that CFS and MDD have some common oxidative and nitrosative biochemical pathways. Activated by infection, psychological stress, and immune disorders, they are believed to have damaging free radical and nitric oxide effects at the cellular level.¹⁴ The cellular effects can result in fatigue, muscle pain, and flu-like malaise.

Cortisol response differs

The majority of patients with CFS report symptoms of cognitive decline, with the acquisition of new verbal learning and information-processing speed particularly likely to be impaired.CFS and MDD might be distinguishable by another pathway—the hypothalamic-pituitary-adrenal (HPA) axis. MDD is classically associated with activation and raised cortisol levels, while CFS is consistently associated with impaired HPA axis functioning and reduced cortisol levels.¹⁰ The majority of patients with CFS report symptoms of cognitive decline, with the acquisition of new verbal learning and information-processing speed particularly likely to be impaired.¹⁵

A meta-analysis of 50 studies of patients with CFS showed deficits in attention, memory, and reaction time, but not in fine motor speed, vocabulary, or reasoning.¹⁶ Autonomic dysfunction has also been observed, including disordered sympathetic activity. The most frequently observed abnormalities on autonomic testing are postural hypotension, tachycardia syndrome, neurally mediated hypotension, and heart rate variability during tilt table testing.¹⁶

The link between infection and CFS

Several infectious agents have been associated with ME/CFS, including the Epstein-Barr virus (EBV), herpes simplex virus 6, parvovirus, Q fever, and Lyme disease.^8,17,18 Most agents that have been linked to ME/CFS are associated with persistent infection and thus incitement of the immune system.

Numerous observational studies^17,18 have documented postinfectious fatigue syndromes after acute viral and bacterial infections and symptoms suggestive of infection, such as fever, myalgias, and respiratory and gastrointestinal distress. In one prospective Australian study,¹⁹ investigators identified 253 cases of acute EBV, Ross River virus, and Q fever. Of those 253 patients, 12% went on to develop CFS, with a higher likelihood among those with more severe acute symptoms. No correlation with preexisting psychiatric disorders was found.

Several infectious agents have been associated with ME/CFS, including the Epstein-Barr virus, herpes simplex virus 6, parvovirus, Q fever, and Lyme disease.Muscle mitochondria studies have demonstrated what appear to be acquired abnormalities in those with CFS.^20,21 Signs of increased oxidative stress have been found in both blood and muscle samples from patients with CFS, and longitudinal studies suggest that oxidative stress is greatest during periods of clinical exacerbation.²² Increased lactate levels suggest increased anaerobic metabolism in the central nervous system consistent with mitochondrial dysfunction. Several studies have demonstrated that exercise can precipitate oxidative stress in patients with CFS, in contrast with healthy controls and controls with other chronic illnesses, suggesting a physiologic basis for their postexertional symptoms.¹⁷

Autoinflammatory syndrome induced by adjuvants, a rare syndrome associated with vaccine administration, has been linked to postvaccination adverse events, exposure to silicone implants, Gulf War syndrome (related to multiple vaccinations), and macrophagic myofasciitis. All involve exposure to immune adjuvants and have similar clinical manifestations. The corresponding exposures appear to trigger an autoimmune response in susceptible individuals.  The hepatitis B vaccine is most often associated with CFS, with symptoms occurring within 90 days of administration.²³

The clinical work-up: Putting knowledge into practice

Familiarity with potential causes of and connections with ME/CFS will help you ensure that patients who say they’re always tired receive a thorough work-up. Start with a medical history, inquiring directly about medical and psychiatric disorders that may contribute to fatigue (TABLE 3).^5,24,25 Include a medication history, as well, to help determine whether the fatigue is drug-related (TABLE 4).^5,25 

What to ask. Determine the onset, course, duration, daily pattern, and impact of fatigue on the patient’s daily life. Inquire, too, about related symptoms of daytime sleepiness, dyspnea on exertion, generalized weakness, and depressed mood. The prominence of any of these rather than fatigue, per se, point to a diagnosis of a chronic illness other than ME/CFS.

Keep in mind, too, that patients with an organ-based medical illness tend to associate their fatigue with activities that they are unable to complete, such as shopping or light housework. In contrast, those with fatigue that is not organ-based typically say that they’re tired all the time. Their fatigue is not necessarily related to exertion, nor does it improve with rest.²⁶

To address this distinction, take a sleep history, assessing both the quality and quantity of the patient’s sleep to determine how it affects symptoms.²⁷ Consider using a questionnaire designed to help distinguish between sleepiness—and a primary sleep disorder—and fatigue,²⁸ such as the Fatigue Severity Scale of Sleep Disorders (http://www.healthywomen.org/sites/default/files/FatigueSeverityScale.pdf).

What to rule out. In addition to a medical history, the physical examination should be oriented toward ruling out secondary causes of fatigue. In addition to a system-by-system approach to the differential diagnosis, carefully observe the patient’s general appearance, with attention to his or her level of alertness, grooming, and psychomotor agitation or retardation as possible signs of a psychiatric disorder.  To rule out neurologic causes, evaluate muscle bulk, tone, and strength; deep tendon reflexes; and sensory and cranial nerves, as well.²⁹

Lab tests to consider. In most cases of ME/CFS, basic studies—complete blood count with differential, erythrocyte sedimentation rate, blood chemistry, and thyroid-stimulating hormone (TSH) levels—are sufficient. When no medical or psychiatric cause has been found, additional tests may be ordered on a case-by-case basis, although laboratory analysis affects the management of fatigue in less than 5% of such patients.³⁰ Tests to consider include:

• Creatine kinase (for patients who report pain or muscle weakness)
• pregnancy test (for women of childbearing age)
• ferritin testing (for young women who might benefit from iron supplementation for levels <50 ng/mL even if anemia is not present)³¹
• hepatitis C screening (recommended by the US Preventive Services Task Force for those born between 1945 and 1965)³²
• human immunodeficiency virus screening and the purified protein derivative test for tuberculosis (based on patient history).

Forego routine testing for other infections

Routine testing for infectious diseases and conditions associated with fatigue, such as Epstein-Barr virus or Lyme disease, immune deficiency, and inflammatory disease, is unlikely to be helpful. Routine testing for infectious diseases and conditions associated with fatigue, such as EBV or Lyme disease, immune deficiency (eg, immunoglobulins), inflammatory disease (eg, antinuclear antibodies, rheumatoid factor), celiac disease, vitamin D deficiency, vitamin B₁₂ deficiency, or heavy metal toxicity, is unlikely to be helpful.²⁹ Additional testing simply to reassure a worried patient usually does not accomplish that objective.³³

Additional studies, referrals to consider. If you suspect that a patient has a sleep disorder, a referral to a sleep clinic to rule out idiopathic sleep disorders, obstructive sleep apnea, or movement disorders that interfere with sleep may be in order. Spirometry and echocardiography may be helpful for some patients. If you suspect peripheral muscle fatigue, a referral for neuromuscular testing is indicated.

CASE › Ms. C’s medical history reveals that she also suffers from irritable bowel syndrome, which she manages with diet and over-the-counter medication, as needed, for constipation or diarrhea. She denies having any other chronic conditions. Her only other symptoms, she reports, are mild upper back pain after spending long hours at the computer, and arthralgias in her left knee and both hands. She admits to being “somewhat depressed” in the last few months, but denies the presence of anhedonia.

The patient’s physical examination is normal, and her depression screen does not meet the criteria for MDD. Her metabolic chemistry panel, complete blood count, TSH, and sedimentation rate are all normal, as well.

Symptom management and coping strategies

When no specific cause of chronic fatigue is found, the focus shifts from diagnosis to symptom management and coping strategies.

Evidence of the effectiveness of specific therapies for ME/CFS is limited; however, the best-studied approaches are cognitive behavioral therapy (CBT) and graded exercise therapy.This requires engagement with the patient. It is important to acknowledge the existence of his or her symptoms and to reassure the patient that further investigation may be warranted later, should new symptoms emerge. Advise the patient, too, that periods of remission and relapse are likely.

Strategies designed to motivate patient self-management, as well as the formulation of patient-centered treatment plans, have been shown to reduce symptom scores.³⁴ Participation in a support group, as well as frequent follow-up visits with a primary care physician, a behavioral therapist, or both, may help to provide needed psychological support.

Evidence of the effectiveness of specific therapies for ME/CFS is limited; however, the best-studied approaches are cognitive behavioral therapy (CBT) and graded exercise therapy.³⁵ Exercise should be low intensity, such as walking or cycling for 30 minutes 3 times a week, with a gradual increase in duration and frequency over a period of weeks to months. Patients with cancer-related fatigue may benefit from yoga, group therapy, and stress management.³⁶

Associated mood and pain symptoms should be treated, as well. Bupropion, which is somewhat stimulating, may be considered as an initial treatment for patients with depression and clinically significant fatigue.

Other potentially beneficial approaches include a healthy diet, avoidance of more than nominal amounts of alcohol, relative avoidance of caffeine (no more than one cup of a caffeinated beverage in the morning), and stress reduction techniques.

Attention to good sleep hygiene may be especially beneficial, including a regular bedtime routine and sleep schedule, and elimination of bedroom light and noise. Pharmacologic treatments for insomnia should be used with caution, if at all.

Bupropion, which is somewhat stimulating, may be considered as an initial treatment for patients with depression and clinically significant fatigue.CASE › Ms. C receives a referral for CBT and is scheduled for a return visit in 4 weeks.

At the advice of both her primary care physician and the behavioral therapist, Ms. C gradually makes several lifestyle changes. She begins going to bed earlier on weeknights to ensure that she sleeps for at least 7 hours. She improves her diet, with increasing emphasis on vegetables, fruits, and whole grains. She also starts a walking program, increasing gradually to a total of 3 hours per week. After 4 months she adds a weekly trip to a gym, where she practices resistance training for about 40 minutes.

Ms. C also increases her social activities on weekends, and recently accepted an invitation to join a book club. Six months from her initial visit, Ms. C notes that although she is still more easily fatigued than most people, she has made significant improvement.

CORRESPONDENCE
Linda Speer, MD; University of Toledo, 3000 Arlington Avenue, MS 1179, Toledo, OH 43614; [email protected]

A young mother in your practice wants her toddler to begin swimming lessons because her family loves water activities. How would you advise her? In fielding an urgent call about a drowning incident, what priorities would you urge regarding resuscitation at the scene? For a stabilized patient following a drowning episode, when might antibiotics be indicated? This article covers these issues as well as follow-up matters such as assisted ventilation and tiered hypothermia intervention.

Drowning likely occurs more often than is reported

Worldwide, drowning accounts for more than 388,000 deaths annually and is the third leading cause of unintentional injury death. Low- and middle-income countries represent 96% of the yearly total.¹ As reported in the United States, nearly 6000 individuals are hospitalized and nearly 4000 die from drowning events annually.² But these figures likely underestimate the true rate, as many drowning fatalities are officially attributed to floods, boating accidents, or other associated events. Nonfatal drownings often go unreported.

Children under the age of 5 years have the highest drowning mortality worldwide, and drowning is the leading cause of unintentional injury death for this age group in many countries, including the United States.^1,2 Men are nearly 4 times as likely to die from drowning than women.² Predictably, in the United States most drowning happens on the weekend and during summer months. More than half of drownings in children younger than 4 years occur in swimming pools; with increasing age, drowning is more likely to occur in natural bodies of water.² With adults in higher income countries, alcohol is a significant contributor to drowning events during recreational activities.^3-5

Much effort has been made in recent years to standardize the nomenclature and treatment of drowning episodes. The International World Congress on Drowning met in the Netherlands in 2002, and established the definition of drowning as “the process of experiencing respiratory impairment from submersion/immersion in liquid.”⁶ Submersion refers to the complete submergence of the victim under the water, while immersion implies that the victim’s airway remains above the water.

CPR should begin, if possible, the moment the victim is out of the water.The Congress recommended that terms such as “wet-drowning,” “dry-drowning,” and “near-drowning” be discontinued in favor of the outcome classifications “death,” “no morbidity,” and “morbidity.” The “morbidity” subgroup was further characterized as “moderately disabled,” “severely disabled,” “vegetative state/coma,” and “brain death.” This meeting established guidelines on the treatment of drowning victims in addition to outlining points for future research.⁶

Physiologic chain of events in drowning

An unexpected immersion in water, particularly cold water, causes a reflexive inspiratory gasp, and some degree of aspiration occurs in most, if not all, cases of drowning. Aspiration further impairs victims’ ability to hold their breath or breathe normally.^5,7,8 It decreases lung compliance due to surfactant washout or intrapulmonary shunting and thereby leads to hypoxia. Aspiration-induced severe laryngospasm can also lead to hypoxia. Pulmonary edema and acute respiratory distress syndrome (ARDS) can follow.

The cardiovascular effects of drowning mirror those seen in hypoxia. Initially, apnea leads to decreased oxygen saturation and precipitates tachycardia and hypertension. Bradycardia and hypotension follow and blood is shunted to vital organs, such as the brain, heart, and lungs.⁹ This phenomenon is accelerated in cold water and leads to “swimming failure,” the impaired ability of the victim to swim because of decreased perfusion of the extremities.^5,7,8,10

“Autonomic conflict” has been proposed as an additional mechanism for morbidity and mortality from drowning episodes. Breath holding and immersion in cold water each can induce cardiac arrhythmias. When combined, these events may increase the risk of an arrhythmogenic state secondary to opposing chronotropic effects: the diving reflex (bradycardia via parasympathetic activation), and the cold shock response (tachycardia via sympathetic activation). This is thought to be an underreported cause of death in drowning, as arrhythmias are undetectable during autopsy.^8,11

Drowning prevention

The American Academy of Pediatrics (AAP) recommends swimming lessons for most children ages 4 years and older.¹² Previously, swimming lessons were not recommended for children ages 1 to 4 because evidence of benefit was lacking, and there was some concern that it might reduce children’s caution around water and reduce parents’ perceived level of need for supervision. Although data are still conflicting, some reports have since shown benefit in early swimming lessons for
toddlers.^13,14

As of 2010, the AAP acknowledges that training may be beneficial for children in this age group, but cautions that not all children will be ready for swimming by this age.¹² Infant water safety programs for children under the age of 1 are not recommended because evidence of benefit is lacking.¹²

Evidence is growing to support teaching basic water survival skills in low- to middle-income countries where water sources are abundant, particularly in Southeast Asia. Specifically, the SwimSafe survival swimming program has yielded impressive results in Bangladesh.¹⁵ This program targets children starting at age 5, and involves 20 lessons teaching basic water survival and rescue skills.

Results have shown a 93% reduction in drowning rates for children enrolled in the program, compared with those not enrolled.¹⁵ Subsequent analyses have proposed that swimming lessons for children in these parts of the world would be as cost effective as current attempts to prevent diarrheal and respiratory diseases in the same areas.¹⁶

Additional preventive measures that are effective in the United States include 4-sided pool fencing, use of personal flotation devices, and bystander cardiopulmonary
resuscitation (CPR).^2,5,17

On-scene evaluation and treatment

CPR for victims of drowning should always involve rescue breathing in addition to chest compressions.Drowning victims can appear mottled and have minimal or no peripheral pulses despite a heartbeat. Rescuers may assume the victim is dead when, in fact, there is cardiac function. Because initial assessment in this situation is difficult, CPR should begin, if possible, the moment the victim is out of the water. Successful on-scene resuscitation is the surest predictor of survival.^9,18 In fact, delay of CPR until the arrival of emergency personnel lessens the likelihood of survival.¹⁹

CPR applied to drowning. For cardiogenic cardiac arrest, chest compressions alone may be better than compressions with rescue breathing. For victims of drowning, though, coordinated compressions and rescue breathing are recommended.²⁰ The
2010 revision of the American Heart Association Guidelines for CPR and Emergency Cardiovascular Care emphasize “compression first” for CPR in cases of cardiogenic cardiac arrest, but continue to support the traditional Airway-Breathing-Chest Compressions sequence for drowning victims in its Special Situations section.²¹

Ventricular fibrillation (VF) is rare after submersion injury. An external defibrillator should be used when available, but it is unlikely to play a significant role in initial resuscitation.⁹

Don’t attempt to remove water from the victim’s mouth before resuscitation. The volume of fluid in the oral cavity is usually insignificant, and trying to remove it by abdominal thrusts or Heimlich maneuver will delay CPR and may injure the patient.²¹

Cervical spine injury is uncommon in drowning episodes, making cervical spine immobilization unnecessary unless the mechanism of injury is known or if there are clinical signs suggesting such injury. Needless cervical spine immobilization can interfere with adequate ventilation.^22,23 However, concern for head or cervical spine injury is warranted when recovering an unconscious victim from shallow water, where such injuries are more likely to result from falling or diving into the water.²⁴

Unless the mechanism of injury is known, cervical spine immobilization is typically unnecessary; needless c-spine immobilization can interfere with adequate ventilation.Administer oxygen supplementation when available to all spontaneously breathing individuals. Individuals who respond well to initial resuscitation and who don’t require intubation tend to have a very good prognosis overall.²⁵

Total time of submersion and the temperature of the water have bearing on the likelihood of survival. Only in rare cases have victims survived submersion lasting longer than 30 minutes. Ten minutes generally is considered the “point of no return.”^9,26 This is consistent with data suggesting 10 minutes of hypoxic insult causes irreversible neurologic damage, with each additional minute rapidly leading to coma.²⁰ However, to complicate matters, unlike cardiac arrest victims, drowning victims can lose cerebral blood flow slowly after respiratory impairment, which makes duration of submersion a potentially unreliable predictor of neurologic outcome.^20,26

Does hypothermia have a protective effect? Hypothermia can occur in water 85°F (30°C) or cooler.¹⁰ It has been hypothesized that resuscitation can be achieved after longer periods of submersion in cold water. However, the considerable debate on this topic has been based on little more than case reports.

For hypothermia to have a protective effect on neurologic function, cooling must take place rapidly and, ideally, before any hypoxic insult. The water would have to be exceptionally cold, likely less than 50°F (10°C).²⁷ The greater surface-to-volume ratio in children enables more rapid cooling and quicker onset of hypothermia, which may explain why they seem to have better neurologic outcomes than adults after prolonged submersion.²⁸

Hypothermia can also be protective if the victim is breathing when cooling begins, such as while floating or swimming in cold water before drowning.²⁹ This was the likely scenario in a reported case of a Norwegian kayaker who called for help after capsizing in 38°F (3.3°C) seawater. Despite having been in cardiac arrest for over 3 hours, the individual experienced a spontaneous return of circulation and was discharged after 32 days with no neurologic deficits.²⁹

Correcting hypothermia after rescue. Conscious patients with no cardiovascular or respiratory compromise should have wet clothing removed at the scene in exchange for blankets, towels, or warm dry clothing. Advise rescuers to attempt no further rewarming at the scene. With unconscious patients, take only simple measures to prevent further heat loss, and focus on transport and resuscitative efforts.²⁴

Hospital management

Attempts have been made to create a prognostic or predictive scoring system for drowning victims presenting to the emergency department. Factors thought to have bearing on mortality include duration of submersion, victim’s age, Glasgow Coma Scale (GCS) score, pupillary reactivity, and the Acute Physiology and Chronic Health Evaluation II (APACHE II) score.^23,30,31 Other measurements, such as core temperature, blood pH, and response to painful stimuli, correlate poorly with mortality.³¹ Hyperkalemia is repeatedly mentioned as a predictor of a poor neurologic outcome, as it is thought to indicate hypoxia before the onset of cooling.^8,26,32 The best predictor of a good outcome is consciousness at the time of arrival at the emergency department.⁴

Continued ventilation assistance is critical. As with prehospital resuscitation efforts, ventilation is critical to in-hospital management. For patients who are breathing spontaneously, continuous positive airway pressure or bi-level positive airway pressure can reduce hypoxia in pulmonary edema. Standard indications of the need for intubation include decreased level of consciousness or concern for ability to protect the airway, hypoxia despite a high fraction of inspired oxygen (FiO₂), or persistent hypercapnia even with adequate noninvasive ventilatory support.³³

Victims may swallow large amounts of liquid during drowning; for intubated patients, advise orogastric tube placement to prevent aspiration of gastric contents. The reliability of pulse oximetry has been called into question in this setting and may be less accurate for victims who are hypothermic or who have been submerged in cold water.^23,34

Tiered intervention for hypothermia. In the hospital setting, passive rewarming is indicated for individuals with a body core temperature of 89.6°F to 95°F (32°C-35°C). Remove wet clothing; cover the victim with warm, dry towels or blankets; and give warm oral fluids and urge movement.

Individuals with core temperatures between 82.4°F to 89.6°F (28°C-32°C) require active external rewarming by applying heat directly to the skin via hot packs, warm blankets, and insulation. These patients should remain in a horizontal position with little movement, if possible, to avoid cold peripheral blood rapidly shifting to the core and precipitating an arrhythmia. Warmed intravenous fluids are appropriate as well.

With a core temperature less than 82.4°F (28°C), aggressive rewarming with extracorporeal membrane oxygenation (ECMO) or cardiopulmonary bypass (CPB) may be warranted as this individual will likely have unstable or absent vital signs (TABLE).^35,36 Some authorities advise that drowning victims with severe hypothermia and cardiac arrest should be resuscitated at facilities with CPB capability.^27,28,37

Inducing hypothermia therapeutically is still unproven. Although not tested specifically on drowning victims, therapeutic hypothermia would seem to have theoretical benefit for their resuscitation. In cardiogenic cardiac arrest, good evidence exists for improved neurologic outcomes with therapeutic hypothermia,³⁸ and this benefit might extend to drowning victims given their similar neurologic injury. No specific pharmacologic therapies have shown benefit in preventing loss of cerebral function.⁹

Monitor for clinical worsening. The concept of “secondary drowning,” a term now abandoned, referred to the phenomenon of clinical worsening hours after the initial drowning episode and resuscitation. This occurrence is now thought to be due to laryngospasm or to the progressive development of pulmonary edema from the aspiration of small amounts of water. Evidence supports monitoring asymptomatic patients; however, the period of suggested monitoring varies between 4 and 24 hours after the incident.^8,22,24

Imaging, if delayed, may be useful. Imaging immediately after a drowning episode is an unreliable predictor of outcome and should be sought only if trauma or symptoms dictate. Cranial computed tomography (CT) has yielded normal findings in drowning victims with a GCS score as low as 4.

If CT is performed, any abnormality detected within 2 to 3 days of injury is a strong predictor of a poor neurologic outcome.³⁹ Magnetic resonance imaging can be beneficial when performed more than 24 hours after resuscitation, preferably within a 4- to 7-day window.²⁰ Lung ultrasound has been used as a bedside tool to monitor progression of pulmonary edema, and could serve the same purpose in drowning recovery.⁴⁰

Anticipate respiratory complications. Since only a small amount of water is usually aspirated during a drowning event, the salinity of the aspirate is unlikely to cause significant disruption in hemodynamic or electrolyte balance.^17,20 However, even a small amount of aspirated water, particularly fresh water, can disrupt gas exchange by washing out surfactant. This can rapidly precipitate ARDS. Not surprisingly, the use of exogenous surfactant has been studied in limited case reports and has had positive results.^41-43 However, large trials have not yet been conducted, mostly because of the significant cost associated with surfactant therapy.

Antibiotics are rarely indicated prophylactically. Pneumonia after a drowning event is potentially fatal. It is more common in patients who have been intubated, and is therefore thought to be a hospital-acquired infection rather than a direct result of the drowning event.

The best predictor of a good outcome is consciousness at the time of arrival at the emergency department.Frequently, pneumonia after drowning is caused by pathogens native to the upper airway, when a victim is unable to protect his or her upper airway.⁴⁴ In these cases, start broad spectrum antibiotics, with particular concern for organisms of the upper oropharynx. Also take into consideration species native to the body of water in which the victim was immersed.⁴⁴

Routine prophylaxis with antibiotics, although common, is not recommended. Exceptions may be victims of drowning in known contaminated water or victims with high volumes of water aspiration.²⁵ Some experts recommend blood cultures for victims who have aspirated, regardless of the presence or absence of infection.²⁴ However, this recommendation seems to be based on opinion.

CORRESPONDENCE
Sean C. Engel, MD, 6600 Excelsior Boulevard Suite 100, St. Louis Park, MN 55426; [email protected]

A young mother in your practice wants her toddler to begin swimming lessons because her family loves water activities. How would you advise her? In fielding an urgent call about a drowning incident, what priorities would you urge regarding resuscitation at the scene? For a stabilized patient following a drowning episode, when might antibiotics be indicated? This article covers these issues as well as follow-up matters such as assisted ventilation and tiered hypothermia intervention.

Drowning likely occurs more often than is reported

Worldwide, drowning accounts for more than 388,000 deaths annually and is the third leading cause of unintentional injury death. Low- and middle-income countries represent 96% of the yearly total.¹ As reported in the United States, nearly 6000 individuals are hospitalized and nearly 4000 die from drowning events annually.² But these figures likely underestimate the true rate, as many drowning fatalities are officially attributed to floods, boating accidents, or other associated events. Nonfatal drownings often go unreported.

Children under the age of 5 years have the highest drowning mortality worldwide, and drowning is the leading cause of unintentional injury death for this age group in many countries, including the United States.^1,2 Men are nearly 4 times as likely to die from drowning than women.² Predictably, in the United States most drowning happens on the weekend and during summer months. More than half of drownings in children younger than 4 years occur in swimming pools; with increasing age, drowning is more likely to occur in natural bodies of water.² With adults in higher income countries, alcohol is a significant contributor to drowning events during recreational activities.^3-5

Much effort has been made in recent years to standardize the nomenclature and treatment of drowning episodes. The International World Congress on Drowning met in the Netherlands in 2002, and established the definition of drowning as “the process of experiencing respiratory impairment from submersion/immersion in liquid.”⁶ Submersion refers to the complete submergence of the victim under the water, while immersion implies that the victim’s airway remains above the water.

CPR should begin, if possible, the moment the victim is out of the water.The Congress recommended that terms such as “wet-drowning,” “dry-drowning,” and “near-drowning” be discontinued in favor of the outcome classifications “death,” “no morbidity,” and “morbidity.” The “morbidity” subgroup was further characterized as “moderately disabled,” “severely disabled,” “vegetative state/coma,” and “brain death.” This meeting established guidelines on the treatment of drowning victims in addition to outlining points for future research.⁶

Physiologic chain of events in drowning

An unexpected immersion in water, particularly cold water, causes a reflexive inspiratory gasp, and some degree of aspiration occurs in most, if not all, cases of drowning. Aspiration further impairs victims’ ability to hold their breath or breathe normally.^5,7,8 It decreases lung compliance due to surfactant washout or intrapulmonary shunting and thereby leads to hypoxia. Aspiration-induced severe laryngospasm can also lead to hypoxia. Pulmonary edema and acute respiratory distress syndrome (ARDS) can follow.

The cardiovascular effects of drowning mirror those seen in hypoxia. Initially, apnea leads to decreased oxygen saturation and precipitates tachycardia and hypertension. Bradycardia and hypotension follow and blood is shunted to vital organs, such as the brain, heart, and lungs.⁹ This phenomenon is accelerated in cold water and leads to “swimming failure,” the impaired ability of the victim to swim because of decreased perfusion of the extremities.^5,7,8,10

“Autonomic conflict” has been proposed as an additional mechanism for morbidity and mortality from drowning episodes. Breath holding and immersion in cold water each can induce cardiac arrhythmias. When combined, these events may increase the risk of an arrhythmogenic state secondary to opposing chronotropic effects: the diving reflex (bradycardia via parasympathetic activation), and the cold shock response (tachycardia via sympathetic activation). This is thought to be an underreported cause of death in drowning, as arrhythmias are undetectable during autopsy.^8,11

Drowning prevention

The American Academy of Pediatrics (AAP) recommends swimming lessons for most children ages 4 years and older.¹² Previously, swimming lessons were not recommended for children ages 1 to 4 because evidence of benefit was lacking, and there was some concern that it might reduce children’s caution around water and reduce parents’ perceived level of need for supervision. Although data are still conflicting, some reports have since shown benefit in early swimming lessons for
toddlers.^13,14

As of 2010, the AAP acknowledges that training may be beneficial for children in this age group, but cautions that not all children will be ready for swimming by this age.¹² Infant water safety programs for children under the age of 1 are not recommended because evidence of benefit is lacking.¹²

Evidence is growing to support teaching basic water survival skills in low- to middle-income countries where water sources are abundant, particularly in Southeast Asia. Specifically, the SwimSafe survival swimming program has yielded impressive results in Bangladesh.¹⁵ This program targets children starting at age 5, and involves 20 lessons teaching basic water survival and rescue skills.

Results have shown a 93% reduction in drowning rates for children enrolled in the program, compared with those not enrolled.¹⁵ Subsequent analyses have proposed that swimming lessons for children in these parts of the world would be as cost effective as current attempts to prevent diarrheal and respiratory diseases in the same areas.¹⁶

Additional preventive measures that are effective in the United States include 4-sided pool fencing, use of personal flotation devices, and bystander cardiopulmonary
resuscitation (CPR).^2,5,17

On-scene evaluation and treatment

CPR for victims of drowning should always involve rescue breathing in addition to chest compressions.Drowning victims can appear mottled and have minimal or no peripheral pulses despite a heartbeat. Rescuers may assume the victim is dead when, in fact, there is cardiac function. Because initial assessment in this situation is difficult, CPR should begin, if possible, the moment the victim is out of the water. Successful on-scene resuscitation is the surest predictor of survival.^9,18 In fact, delay of CPR until the arrival of emergency personnel lessens the likelihood of survival.¹⁹

CPR applied to drowning. For cardiogenic cardiac arrest, chest compressions alone may be better than compressions with rescue breathing. For victims of drowning, though, coordinated compressions and rescue breathing are recommended.²⁰ The
2010 revision of the American Heart Association Guidelines for CPR and Emergency Cardiovascular Care emphasize “compression first” for CPR in cases of cardiogenic cardiac arrest, but continue to support the traditional Airway-Breathing-Chest Compressions sequence for drowning victims in its Special Situations section.²¹

Ventricular fibrillation (VF) is rare after submersion injury. An external defibrillator should be used when available, but it is unlikely to play a significant role in initial resuscitation.⁹

Don’t attempt to remove water from the victim’s mouth before resuscitation. The volume of fluid in the oral cavity is usually insignificant, and trying to remove it by abdominal thrusts or Heimlich maneuver will delay CPR and may injure the patient.²¹

Cervical spine injury is uncommon in drowning episodes, making cervical spine immobilization unnecessary unless the mechanism of injury is known or if there are clinical signs suggesting such injury. Needless cervical spine immobilization can interfere with adequate ventilation.^22,23 However, concern for head or cervical spine injury is warranted when recovering an unconscious victim from shallow water, where such injuries are more likely to result from falling or diving into the water.²⁴

Unless the mechanism of injury is known, cervical spine immobilization is typically unnecessary; needless c-spine immobilization can interfere with adequate ventilation.Administer oxygen supplementation when available to all spontaneously breathing individuals. Individuals who respond well to initial resuscitation and who don’t require intubation tend to have a very good prognosis overall.²⁵

Total time of submersion and the temperature of the water have bearing on the likelihood of survival. Only in rare cases have victims survived submersion lasting longer than 30 minutes. Ten minutes generally is considered the “point of no return.”^9,26 This is consistent with data suggesting 10 minutes of hypoxic insult causes irreversible neurologic damage, with each additional minute rapidly leading to coma.²⁰ However, to complicate matters, unlike cardiac arrest victims, drowning victims can lose cerebral blood flow slowly after respiratory impairment, which makes duration of submersion a potentially unreliable predictor of neurologic outcome.^20,26

Does hypothermia have a protective effect? Hypothermia can occur in water 85°F (30°C) or cooler.¹⁰ It has been hypothesized that resuscitation can be achieved after longer periods of submersion in cold water. However, the considerable debate on this topic has been based on little more than case reports.

For hypothermia to have a protective effect on neurologic function, cooling must take place rapidly and, ideally, before any hypoxic insult. The water would have to be exceptionally cold, likely less than 50°F (10°C).²⁷ The greater surface-to-volume ratio in children enables more rapid cooling and quicker onset of hypothermia, which may explain why they seem to have better neurologic outcomes than adults after prolonged submersion.²⁸

Hypothermia can also be protective if the victim is breathing when cooling begins, such as while floating or swimming in cold water before drowning.²⁹ This was the likely scenario in a reported case of a Norwegian kayaker who called for help after capsizing in 38°F (3.3°C) seawater. Despite having been in cardiac arrest for over 3 hours, the individual experienced a spontaneous return of circulation and was discharged after 32 days with no neurologic deficits.²⁹

Correcting hypothermia after rescue. Conscious patients with no cardiovascular or respiratory compromise should have wet clothing removed at the scene in exchange for blankets, towels, or warm dry clothing. Advise rescuers to attempt no further rewarming at the scene. With unconscious patients, take only simple measures to prevent further heat loss, and focus on transport and resuscitative efforts.²⁴

Hospital management

Attempts have been made to create a prognostic or predictive scoring system for drowning victims presenting to the emergency department. Factors thought to have bearing on mortality include duration of submersion, victim’s age, Glasgow Coma Scale (GCS) score, pupillary reactivity, and the Acute Physiology and Chronic Health Evaluation II (APACHE II) score.^23,30,31 Other measurements, such as core temperature, blood pH, and response to painful stimuli, correlate poorly with mortality.³¹ Hyperkalemia is repeatedly mentioned as a predictor of a poor neurologic outcome, as it is thought to indicate hypoxia before the onset of cooling.^8,26,32 The best predictor of a good outcome is consciousness at the time of arrival at the emergency department.⁴

Continued ventilation assistance is critical. As with prehospital resuscitation efforts, ventilation is critical to in-hospital management. For patients who are breathing spontaneously, continuous positive airway pressure or bi-level positive airway pressure can reduce hypoxia in pulmonary edema. Standard indications of the need for intubation include decreased level of consciousness or concern for ability to protect the airway, hypoxia despite a high fraction of inspired oxygen (FiO₂), or persistent hypercapnia even with adequate noninvasive ventilatory support.³³

Victims may swallow large amounts of liquid during drowning; for intubated patients, advise orogastric tube placement to prevent aspiration of gastric contents. The reliability of pulse oximetry has been called into question in this setting and may be less accurate for victims who are hypothermic or who have been submerged in cold water.^23,34

Tiered intervention for hypothermia. In the hospital setting, passive rewarming is indicated for individuals with a body core temperature of 89.6°F to 95°F (32°C-35°C). Remove wet clothing; cover the victim with warm, dry towels or blankets; and give warm oral fluids and urge movement.

Individuals with core temperatures between 82.4°F to 89.6°F (28°C-32°C) require active external rewarming by applying heat directly to the skin via hot packs, warm blankets, and insulation. These patients should remain in a horizontal position with little movement, if possible, to avoid cold peripheral blood rapidly shifting to the core and precipitating an arrhythmia. Warmed intravenous fluids are appropriate as well.

With a core temperature less than 82.4°F (28°C), aggressive rewarming with extracorporeal membrane oxygenation (ECMO) or cardiopulmonary bypass (CPB) may be warranted as this individual will likely have unstable or absent vital signs (TABLE).^35,36 Some authorities advise that drowning victims with severe hypothermia and cardiac arrest should be resuscitated at facilities with CPB capability.^27,28,37

Inducing hypothermia therapeutically is still unproven. Although not tested specifically on drowning victims, therapeutic hypothermia would seem to have theoretical benefit for their resuscitation. In cardiogenic cardiac arrest, good evidence exists for improved neurologic outcomes with therapeutic hypothermia,³⁸ and this benefit might extend to drowning victims given their similar neurologic injury. No specific pharmacologic therapies have shown benefit in preventing loss of cerebral function.⁹

Monitor for clinical worsening. The concept of “secondary drowning,” a term now abandoned, referred to the phenomenon of clinical worsening hours after the initial drowning episode and resuscitation. This occurrence is now thought to be due to laryngospasm or to the progressive development of pulmonary edema from the aspiration of small amounts of water. Evidence supports monitoring asymptomatic patients; however, the period of suggested monitoring varies between 4 and 24 hours after the incident.^8,22,24

Imaging, if delayed, may be useful. Imaging immediately after a drowning episode is an unreliable predictor of outcome and should be sought only if trauma or symptoms dictate. Cranial computed tomography (CT) has yielded normal findings in drowning victims with a GCS score as low as 4.

If CT is performed, any abnormality detected within 2 to 3 days of injury is a strong predictor of a poor neurologic outcome.³⁹ Magnetic resonance imaging can be beneficial when performed more than 24 hours after resuscitation, preferably within a 4- to 7-day window.²⁰ Lung ultrasound has been used as a bedside tool to monitor progression of pulmonary edema, and could serve the same purpose in drowning recovery.⁴⁰

Anticipate respiratory complications. Since only a small amount of water is usually aspirated during a drowning event, the salinity of the aspirate is unlikely to cause significant disruption in hemodynamic or electrolyte balance.^17,20 However, even a small amount of aspirated water, particularly fresh water, can disrupt gas exchange by washing out surfactant. This can rapidly precipitate ARDS. Not surprisingly, the use of exogenous surfactant has been studied in limited case reports and has had positive results.^41-43 However, large trials have not yet been conducted, mostly because of the significant cost associated with surfactant therapy.

Antibiotics are rarely indicated prophylactically. Pneumonia after a drowning event is potentially fatal. It is more common in patients who have been intubated, and is therefore thought to be a hospital-acquired infection rather than a direct result of the drowning event.

The best predictor of a good outcome is consciousness at the time of arrival at the emergency department.Frequently, pneumonia after drowning is caused by pathogens native to the upper airway, when a victim is unable to protect his or her upper airway.⁴⁴ In these cases, start broad spectrum antibiotics, with particular concern for organisms of the upper oropharynx. Also take into consideration species native to the body of water in which the victim was immersed.⁴⁴

Routine prophylaxis with antibiotics, although common, is not recommended. Exceptions may be victims of drowning in known contaminated water or victims with high volumes of water aspiration.²⁵ Some experts recommend blood cultures for victims who have aspirated, regardless of the presence or absence of infection.²⁴ However, this recommendation seems to be based on opinion.

CORRESPONDENCE
Sean C. Engel, MD, 6600 Excelsior Boulevard Suite 100, St. Louis Park, MN 55426; [email protected]

A young mother in your practice wants her toddler to begin swimming lessons because her family loves water activities. How would you advise her? In fielding an urgent call about a drowning incident, what priorities would you urge regarding resuscitation at the scene? For a stabilized patient following a drowning episode, when might antibiotics be indicated? This article covers these issues as well as follow-up matters such as assisted ventilation and tiered hypothermia intervention.

Drowning likely occurs more often than is reported

Worldwide, drowning accounts for more than 388,000 deaths annually and is the third leading cause of unintentional injury death. Low- and middle-income countries represent 96% of the yearly total.¹ As reported in the United States, nearly 6000 individuals are hospitalized and nearly 4000 die from drowning events annually.² But these figures likely underestimate the true rate, as many drowning fatalities are officially attributed to floods, boating accidents, or other associated events. Nonfatal drownings often go unreported.

Children under the age of 5 years have the highest drowning mortality worldwide, and drowning is the leading cause of unintentional injury death for this age group in many countries, including the United States.^1,2 Men are nearly 4 times as likely to die from drowning than women.² Predictably, in the United States most drowning happens on the weekend and during summer months. More than half of drownings in children younger than 4 years occur in swimming pools; with increasing age, drowning is more likely to occur in natural bodies of water.² With adults in higher income countries, alcohol is a significant contributor to drowning events during recreational activities.^3-5

Much effort has been made in recent years to standardize the nomenclature and treatment of drowning episodes. The International World Congress on Drowning met in the Netherlands in 2002, and established the definition of drowning as “the process of experiencing respiratory impairment from submersion/immersion in liquid.”⁶ Submersion refers to the complete submergence of the victim under the water, while immersion implies that the victim’s airway remains above the water.

CPR should begin, if possible, the moment the victim is out of the water.The Congress recommended that terms such as “wet-drowning,” “dry-drowning,” and “near-drowning” be discontinued in favor of the outcome classifications “death,” “no morbidity,” and “morbidity.” The “morbidity” subgroup was further characterized as “moderately disabled,” “severely disabled,” “vegetative state/coma,” and “brain death.” This meeting established guidelines on the treatment of drowning victims in addition to outlining points for future research.⁶

Physiologic chain of events in drowning

An unexpected immersion in water, particularly cold water, causes a reflexive inspiratory gasp, and some degree of aspiration occurs in most, if not all, cases of drowning. Aspiration further impairs victims’ ability to hold their breath or breathe normally.^5,7,8 It decreases lung compliance due to surfactant washout or intrapulmonary shunting and thereby leads to hypoxia. Aspiration-induced severe laryngospasm can also lead to hypoxia. Pulmonary edema and acute respiratory distress syndrome (ARDS) can follow.

The cardiovascular effects of drowning mirror those seen in hypoxia. Initially, apnea leads to decreased oxygen saturation and precipitates tachycardia and hypertension. Bradycardia and hypotension follow and blood is shunted to vital organs, such as the brain, heart, and lungs.⁹ This phenomenon is accelerated in cold water and leads to “swimming failure,” the impaired ability of the victim to swim because of decreased perfusion of the extremities.^5,7,8,10

“Autonomic conflict” has been proposed as an additional mechanism for morbidity and mortality from drowning episodes. Breath holding and immersion in cold water each can induce cardiac arrhythmias. When combined, these events may increase the risk of an arrhythmogenic state secondary to opposing chronotropic effects: the diving reflex (bradycardia via parasympathetic activation), and the cold shock response (tachycardia via sympathetic activation). This is thought to be an underreported cause of death in drowning, as arrhythmias are undetectable during autopsy.^8,11

Drowning prevention

The American Academy of Pediatrics (AAP) recommends swimming lessons for most children ages 4 years and older.¹² Previously, swimming lessons were not recommended for children ages 1 to 4 because evidence of benefit was lacking, and there was some concern that it might reduce children’s caution around water and reduce parents’ perceived level of need for supervision. Although data are still conflicting, some reports have since shown benefit in early swimming lessons for
toddlers.^13,14

As of 2010, the AAP acknowledges that training may be beneficial for children in this age group, but cautions that not all children will be ready for swimming by this age.¹² Infant water safety programs for children under the age of 1 are not recommended because evidence of benefit is lacking.¹²

Evidence is growing to support teaching basic water survival skills in low- to middle-income countries where water sources are abundant, particularly in Southeast Asia. Specifically, the SwimSafe survival swimming program has yielded impressive results in Bangladesh.¹⁵ This program targets children starting at age 5, and involves 20 lessons teaching basic water survival and rescue skills.

Results have shown a 93% reduction in drowning rates for children enrolled in the program, compared with those not enrolled.¹⁵ Subsequent analyses have proposed that swimming lessons for children in these parts of the world would be as cost effective as current attempts to prevent diarrheal and respiratory diseases in the same areas.¹⁶

Additional preventive measures that are effective in the United States include 4-sided pool fencing, use of personal flotation devices, and bystander cardiopulmonary
resuscitation (CPR).^2,5,17

On-scene evaluation and treatment

CPR for victims of drowning should always involve rescue breathing in addition to chest compressions.Drowning victims can appear mottled and have minimal or no peripheral pulses despite a heartbeat. Rescuers may assume the victim is dead when, in fact, there is cardiac function. Because initial assessment in this situation is difficult, CPR should begin, if possible, the moment the victim is out of the water. Successful on-scene resuscitation is the surest predictor of survival.^9,18 In fact, delay of CPR until the arrival of emergency personnel lessens the likelihood of survival.¹⁹

CPR applied to drowning. For cardiogenic cardiac arrest, chest compressions alone may be better than compressions with rescue breathing. For victims of drowning, though, coordinated compressions and rescue breathing are recommended.²⁰ The
2010 revision of the American Heart Association Guidelines for CPR and Emergency Cardiovascular Care emphasize “compression first” for CPR in cases of cardiogenic cardiac arrest, but continue to support the traditional Airway-Breathing-Chest Compressions sequence for drowning victims in its Special Situations section.²¹

Ventricular fibrillation (VF) is rare after submersion injury. An external defibrillator should be used when available, but it is unlikely to play a significant role in initial resuscitation.⁹

Don’t attempt to remove water from the victim’s mouth before resuscitation. The volume of fluid in the oral cavity is usually insignificant, and trying to remove it by abdominal thrusts or Heimlich maneuver will delay CPR and may injure the patient.²¹

Cervical spine injury is uncommon in drowning episodes, making cervical spine immobilization unnecessary unless the mechanism of injury is known or if there are clinical signs suggesting such injury. Needless cervical spine immobilization can interfere with adequate ventilation.^22,23 However, concern for head or cervical spine injury is warranted when recovering an unconscious victim from shallow water, where such injuries are more likely to result from falling or diving into the water.²⁴

Unless the mechanism of injury is known, cervical spine immobilization is typically unnecessary; needless c-spine immobilization can interfere with adequate ventilation.Administer oxygen supplementation when available to all spontaneously breathing individuals. Individuals who respond well to initial resuscitation and who don’t require intubation tend to have a very good prognosis overall.²⁵

Total time of submersion and the temperature of the water have bearing on the likelihood of survival. Only in rare cases have victims survived submersion lasting longer than 30 minutes. Ten minutes generally is considered the “point of no return.”^9,26 This is consistent with data suggesting 10 minutes of hypoxic insult causes irreversible neurologic damage, with each additional minute rapidly leading to coma.²⁰ However, to complicate matters, unlike cardiac arrest victims, drowning victims can lose cerebral blood flow slowly after respiratory impairment, which makes duration of submersion a potentially unreliable predictor of neurologic outcome.^20,26

Does hypothermia have a protective effect? Hypothermia can occur in water 85°F (30°C) or cooler.¹⁰ It has been hypothesized that resuscitation can be achieved after longer periods of submersion in cold water. However, the considerable debate on this topic has been based on little more than case reports.

For hypothermia to have a protective effect on neurologic function, cooling must take place rapidly and, ideally, before any hypoxic insult. The water would have to be exceptionally cold, likely less than 50°F (10°C).²⁷ The greater surface-to-volume ratio in children enables more rapid cooling and quicker onset of hypothermia, which may explain why they seem to have better neurologic outcomes than adults after prolonged submersion.²⁸

Hypothermia can also be protective if the victim is breathing when cooling begins, such as while floating or swimming in cold water before drowning.²⁹ This was the likely scenario in a reported case of a Norwegian kayaker who called for help after capsizing in 38°F (3.3°C) seawater. Despite having been in cardiac arrest for over 3 hours, the individual experienced a spontaneous return of circulation and was discharged after 32 days with no neurologic deficits.²⁹

Correcting hypothermia after rescue. Conscious patients with no cardiovascular or respiratory compromise should have wet clothing removed at the scene in exchange for blankets, towels, or warm dry clothing. Advise rescuers to attempt no further rewarming at the scene. With unconscious patients, take only simple measures to prevent further heat loss, and focus on transport and resuscitative efforts.²⁴

Hospital management

Attempts have been made to create a prognostic or predictive scoring system for drowning victims presenting to the emergency department. Factors thought to have bearing on mortality include duration of submersion, victim’s age, Glasgow Coma Scale (GCS) score, pupillary reactivity, and the Acute Physiology and Chronic Health Evaluation II (APACHE II) score.^23,30,31 Other measurements, such as core temperature, blood pH, and response to painful stimuli, correlate poorly with mortality.³¹ Hyperkalemia is repeatedly mentioned as a predictor of a poor neurologic outcome, as it is thought to indicate hypoxia before the onset of cooling.^8,26,32 The best predictor of a good outcome is consciousness at the time of arrival at the emergency department.⁴

Continued ventilation assistance is critical. As with prehospital resuscitation efforts, ventilation is critical to in-hospital management. For patients who are breathing spontaneously, continuous positive airway pressure or bi-level positive airway pressure can reduce hypoxia in pulmonary edema. Standard indications of the need for intubation include decreased level of consciousness or concern for ability to protect the airway, hypoxia despite a high fraction of inspired oxygen (FiO₂), or persistent hypercapnia even with adequate noninvasive ventilatory support.³³

Victims may swallow large amounts of liquid during drowning; for intubated patients, advise orogastric tube placement to prevent aspiration of gastric contents. The reliability of pulse oximetry has been called into question in this setting and may be less accurate for victims who are hypothermic or who have been submerged in cold water.^23,34

Tiered intervention for hypothermia. In the hospital setting, passive rewarming is indicated for individuals with a body core temperature of 89.6°F to 95°F (32°C-35°C). Remove wet clothing; cover the victim with warm, dry towels or blankets; and give warm oral fluids and urge movement.

Individuals with core temperatures between 82.4°F to 89.6°F (28°C-32°C) require active external rewarming by applying heat directly to the skin via hot packs, warm blankets, and insulation. These patients should remain in a horizontal position with little movement, if possible, to avoid cold peripheral blood rapidly shifting to the core and precipitating an arrhythmia. Warmed intravenous fluids are appropriate as well.

With a core temperature less than 82.4°F (28°C), aggressive rewarming with extracorporeal membrane oxygenation (ECMO) or cardiopulmonary bypass (CPB) may be warranted as this individual will likely have unstable or absent vital signs (TABLE).^35,36 Some authorities advise that drowning victims with severe hypothermia and cardiac arrest should be resuscitated at facilities with CPB capability.^27,28,37

Inducing hypothermia therapeutically is still unproven. Although not tested specifically on drowning victims, therapeutic hypothermia would seem to have theoretical benefit for their resuscitation. In cardiogenic cardiac arrest, good evidence exists for improved neurologic outcomes with therapeutic hypothermia,³⁸ and this benefit might extend to drowning victims given their similar neurologic injury. No specific pharmacologic therapies have shown benefit in preventing loss of cerebral function.⁹

Monitor for clinical worsening. The concept of “secondary drowning,” a term now abandoned, referred to the phenomenon of clinical worsening hours after the initial drowning episode and resuscitation. This occurrence is now thought to be due to laryngospasm or to the progressive development of pulmonary edema from the aspiration of small amounts of water. Evidence supports monitoring asymptomatic patients; however, the period of suggested monitoring varies between 4 and 24 hours after the incident.^8,22,24

Imaging, if delayed, may be useful. Imaging immediately after a drowning episode is an unreliable predictor of outcome and should be sought only if trauma or symptoms dictate. Cranial computed tomography (CT) has yielded normal findings in drowning victims with a GCS score as low as 4.

If CT is performed, any abnormality detected within 2 to 3 days of injury is a strong predictor of a poor neurologic outcome.³⁹ Magnetic resonance imaging can be beneficial when performed more than 24 hours after resuscitation, preferably within a 4- to 7-day window.²⁰ Lung ultrasound has been used as a bedside tool to monitor progression of pulmonary edema, and could serve the same purpose in drowning recovery.⁴⁰

Anticipate respiratory complications. Since only a small amount of water is usually aspirated during a drowning event, the salinity of the aspirate is unlikely to cause significant disruption in hemodynamic or electrolyte balance.^17,20 However, even a small amount of aspirated water, particularly fresh water, can disrupt gas exchange by washing out surfactant. This can rapidly precipitate ARDS. Not surprisingly, the use of exogenous surfactant has been studied in limited case reports and has had positive results.^41-43 However, large trials have not yet been conducted, mostly because of the significant cost associated with surfactant therapy.

Antibiotics are rarely indicated prophylactically. Pneumonia after a drowning event is potentially fatal. It is more common in patients who have been intubated, and is therefore thought to be a hospital-acquired infection rather than a direct result of the drowning event.

The best predictor of a good outcome is consciousness at the time of arrival at the emergency department.Frequently, pneumonia after drowning is caused by pathogens native to the upper airway, when a victim is unable to protect his or her upper airway.⁴⁴ In these cases, start broad spectrum antibiotics, with particular concern for organisms of the upper oropharynx. Also take into consideration species native to the body of water in which the victim was immersed.⁴⁴

Routine prophylaxis with antibiotics, although common, is not recommended. Exceptions may be victims of drowning in known contaminated water or victims with high volumes of water aspiration.²⁵ Some experts recommend blood cultures for victims who have aspirated, regardless of the presence or absence of infection.²⁴ However, this recommendation seems to be based on opinion.

CORRESPONDENCE
Sean C. Engel, MD, 6600 Excelsior Boulevard Suite 100, St. Louis Park, MN 55426; [email protected]