Case Reports

According to the US Centers for Disease Control and Prevention (CDC), in 2011 there were 25.8 million people in the United States living with diabetes mellitus (DM), 90% to 95% of whom had type 2 DM.1 Seven million of those with DM are undiagnosed. In addition, an estimated 79 million adults in the United States have prediabetes, a condition in which glycated hemoglobin (HbA1_c) levels are above normal but do not yet meet the diagnostic threshold for DM (≥6.5%).¹ This large population with prediabetes, defined by HbA1_c levels of 5.7% to 6.4%, has a 5-year risk of progression to diabetes ranging from 9% to 50%; the higher risk is associated with higher HbA1_c levels as the disease represents a continuum.²

The benefits of diagnosis and treatment of diabetes—preventing the progression of microvascular complications—are well established. Through medication, exercise, and dietary interventions, the progression of both prediabetes and diabetes can be halted and in many cases reversed.

Given the prevalence of DM in the United States and the clear benefits of treatment, the ability to diagnose DM is important for any physician who may encounter patients with undiagnosed diabetes or prediabetes. This article reviews the management of hyperglycemia relevant to the emergency physician (EP) and discusses the criteria used in diagnosing DM.

The American College of Emergency Physicians (ACEP) currently does not have a clinical policy for the initial treatment of patients with DM that is newly diagnosed in the ED. A 2007 survey of 152 EPs in the United States found that 52% would leave initiation of outpatient treatment of diabetes to a patient’s primary care physician (PCP), regardless of the degree of hyperglycemia noted in the ED.³ Almost paradoxically, among these same EPs, the most common reason for not screening for diabetes (cited by 69%) was the inability to secure follow-up for newly diagnosed patients. Given the fragmentary care many emergency patients receive and the difficulty in assuring close ED follow-up, reviewing the American Diabetes Association (ADA) recommendations for initiation of diabetes management is worthwhile.

ADA Treatment Guidelines

Metformin
Metformin, if tolerated and without contraindications, is the preferred initial pharmacologic agent for all patients with newly diagnosed DM. This drug typically lowers HbA1_c by 1.5%, and does not cause weight gain or hypoglycemia associated with some other oral agents (eg, sulfonylureas). Metformin therapy, however, is contraindicated in patients with chronic kidney disease (CKD) (Cr >1.5 in males and >1.4 in females), liver disease, a history of lactic acidosis, and a low-perfusion state. Metformin also should not be used for 48 hours following an intravenous (IV) contrast-enhanced study.

When starting metformin or increasing the dose, patients frequently experience abdominal symptoms, including cramping, nausea, vomiting, and diarrhea. These side effects are usually mild and self-limiting, and they usually resolve within a week. However, patients should be warned of these complications and advised to take metformin with meals to limit side effects. They also should be encouraged to continue the medication through the initial symptoms, if possible.

For a patient with newly diagnosed type 2 DM, a reasonable ED discharge plan is to start metformin at 500 mg daily, taken with a meal. If the patient develops abdominal symptoms, the dose can be reduced to 250 mg daily. Patients should additionally be advised to follow up with their PCP as soon as feasible. Patients can be instructed to increase the dose of metformin on their own. Because the development of gastrointestinal symptoms is the main dose-limiting effect of metformin therapy, it is generally safe to give patients instructions on weekly dose escalation by 500 mg/day to the optimal dose of 1,000 mg twice daily. Although metformin therapy is associated with rare cases of lactic acidosis in patients with CKD, it does not, on its own, cause acute kidney injury; therefore, intense monitoring is not required while doses are increased. Metformin therapy can decrease vitamin B12 absorption, though it rarely results in megaloblastic anemia, and it takes 5 to 10 years for neurological symptoms to manifest.⁵

Other Treatment Options

The Figure highlights the ADA guidelines on additional pharmacologic agents beyond metformin. When these additional agents are used, the patient should follow-up with a PCP every 3 months to determine how far HbA1_c has been lowered. For most patients with DM, the goal for HbA1_c is less than 7%. This goal may be raised for elderly patients and those with a history of cardiovascular disease, as these patients are at higher risk for cardiovascular mortality from an episode of hypoglycemia.

The secondary medication options include sulfonylureas, thiazolidinediones, dipeptidyl peptidase-IV inhibitors, glucagon-like peptide-1 (GLP-1) agonists, sodium-glucose cotransporter-2 inhibitors, and insulin. Insulin therapy is the final medication step for all patients with DM who are willing and able to comply with daily injections and who have failed to meet goals through alternative agents.

Before using additional agents beyond metformin, the PCP or endocrinologist should discuss in detail factors such as side effects (ie, weight gain, risk of hypoglycemia), medication delivery (GLP-1 agonists and insulin require injections), and cost with the patient. It is generally advisable not to try to accomplish this medication adjustment in the ED without consultation with a PCP or endocrinologist.

Diagnosis

The clinical presentation of new-onset DM can range from the relatively benign (eg, polyuria, polydipsia, frequent skin infections) to the life-threatening (eg, diabetic ketoacidosis) (Table 1). The classic symptoms of hyperglycemia include polyuria, polydipsia, unexplained weight loss, easy fatigability, dizziness, and blurred vision. Any one of these symptoms should prompt consideration of DM in the differential diagnosis.

Four abnormal laboratory results comprise the ADA’s diagnostic criteria for DM⁶:

• HbA1_c ≥6.5%;
• 8-hour fasting plasma glucose ≥126 mg/dL;
• 2-hour oral glucose tolerance test (OGTT) ≥200 mg/dL (Patient is given 75 g sugar orally; then plasma glucose is drawn 2 hours later); and
• A random plasma glucose of ≥200 mg/dL in association with “classic symptoms” of DM, defined as polyuria, polydipsia, and unexplained weight loss.

In the absence of an unequivocal indicator, such as the patient presenting in diabetic ketoacidosis, the formal diagnosis of DM should not be made unless at least 2 abnormal test results are obtained simultaneously (from a 2-hour OGTT test and an HbA1_c panel, for example, or when the same test is repeated one month later and both test results indicate DM).

Limitations of Glucometers
When interpreting test results, keep in mind that the US Food and Drug Administration allows point-of-care (POC) glucometers to have an accuracy of +/- 20%. The POC glucometers measure whole-blood glucose rather than plasma glucose; generally the amount of glucose in plasma is about 10% to 15% higher than the amount in whole blood.7 Therefore, the formal diagnosis of DM relies on laboratory plasma glucose results; POC glucometer results may guide the EP in initiating a workup for DM but should not be among the diagnostic criteria.

HbA1c Testing
The HbA1_c percentage represents the preceding 3-month average plasma glucose level. This is due to nonenzymatic glycosylation of Hb, a reaction driven by higher concentrations of circulating plasma glucose. The HbA1_c can be falsely low in states of high red blood cell (RBC) turnover, such as ongoing blood loss or the RBC fragility seen in hemoglobinopathies. Patients who have had RBC transfusions in the preceding months will have inaccurate HbA1_c test results as well.

Keeping these test characteristics in mind, the HbA1_c test may be more useful than glucose-based testing strategies in diagnosing diabetes in the ED. Hyperglycemia can be observed in nondiabetic emergency patients undergoing a stress response, since gluconeogenesis and glycogenolysis are normal hepatic stress responses to circulating epinephrine and cortisol. Unless a patient is on chronic steroids, the HbA1c is not similarly affected.

A recent study of ED patients using HbA1_c as the sole screening modality found that test characteristics for ED patients were similar to those for patients in typical outpatient settings. HbA1_c of 6.5 or greater yielded a sensitivity of 54% and a specificity of 96% in diagnosing diabetes.8 The availability of test results at the time of the emergency visit depends upon the individual hospital laboratory, but most labs run the test itself in less than an hour. Emergency physicians working in hospitals that perform the test will often have the test results prior to patient discharge.

Inpatient Management

Managing a patient with newly diagnosed DM that requires admission to the hospital is different from managing a patient who can be treated as an outpatient. The ADA-recommended treatment regimen for DM in the inpatient setting focuses on the use of insulin. While type 1 DM, by definition, requires insulin treatment, many patients with type 2 DM are treated with oral agents only as outpatients. However, because of complications associated with the use of oral agents in low renal perfusion states (such as during surgery, during IV contrast-enhanced studies, or when the admitting condition causes shock), the ADA generally does not prefer oral agents for inpatient glycemic control. Metformin particularly should not be used in the inpatient setting, due to increased risk of lactic acidosis.

The preferred insulin regimen for noncritically ill diabetic inpatients consists of scheduled subcutaneous therapy with basal, nutritional, and correctional components.⁹ Because absorption from subcutaneous tissue varies among critically ill patients, the ADA recommends an IV insulin therapy protocol for ICU patients with diabetes. The NICE SUGAR trial found that the overall risk of death was 2.6% higher among patients treated with an intensive insulin therapy regimen (finger stick blood glucose goal 80-108 mg/dL) than it was among patients treated using standard therapy with a goal of less than 180 mg/dL, a statistically significant difference.¹⁰ These results form the basis for the ADA-recommended inpatient glycemic target of 140 to 180 mg/dL among critically ill patients.¹¹ While no definitive evidence exists to guide glycemic control among noncritically ill inpatients, the strategy applied to critically ill patients can be extrapolated to other hospitalized patients because the ADA recommends treatment to keep finger stick blood glucose values below 180 mg/dL.

Hypoglycemia

Often a hypoglycemic event can clearly be correlated to a missed meal. This may only require re-education of the patient about the need to be consistent in the timing of his or her dietary regimen. However, often a hypoglycemic event is not clearly related to a missed meal. When a patient has recently taken a long-acting insulin or sulfonylurea or has new renal failure, prolonged observation or admission may be required. In most other cases of medication-induced hypoglycemia, the patients are expediently sent home. As many such patient encounters occur in the early morning hours, EPs need to know how to manage insulin regimens if these patients are to be safely discharged home. Additionally, worsening renal function is a common cause of potentiation of both oral agents and insulin; it is generally worthwhile to check renal function on patients presenting with significant hypoglycemia.

After a significant hypoglycemic event, many guidelines recommend decreasing the daily insulin dose by 10% to 20%. However, with a careful medication history and knowledge of each insulin dose’s activity profile, a physician can often discern which individual insulin dose is responsible for a given hypoglycemic event. That individual daily dose can then be decreased by 20% to better target the cause of the hypoglycemic event.

Conclusion

Diabetes mellitus is one of the most common chronic medical conditions in the United States, and more than 7 million Americans may be living with undiagnosed DM. In the ED setting, physicians regularly encounter patients with undiagnosed type 2 DM. Since treatment is known to prevent the microvascular complications associated with DM, EPs should know the diagnostic criteria and understand the ADA’s inpatient and outpatient treatment recommendations.

According to the US Centers for Disease Control and Prevention (CDC), in 2011 there were 25.8 million people in the United States living with diabetes mellitus (DM), 90% to 95% of whom had type 2 DM.1 Seven million of those with DM are undiagnosed. In addition, an estimated 79 million adults in the United States have prediabetes, a condition in which glycated hemoglobin (HbA1_c) levels are above normal but do not yet meet the diagnostic threshold for DM (≥6.5%).¹ This large population with prediabetes, defined by HbA1_c levels of 5.7% to 6.4%, has a 5-year risk of progression to diabetes ranging from 9% to 50%; the higher risk is associated with higher HbA1_c levels as the disease represents a continuum.²

The benefits of diagnosis and treatment of diabetes—preventing the progression of microvascular complications—are well established. Through medication, exercise, and dietary interventions, the progression of both prediabetes and diabetes can be halted and in many cases reversed.

Given the prevalence of DM in the United States and the clear benefits of treatment, the ability to diagnose DM is important for any physician who may encounter patients with undiagnosed diabetes or prediabetes. This article reviews the management of hyperglycemia relevant to the emergency physician (EP) and discusses the criteria used in diagnosing DM.

The American College of Emergency Physicians (ACEP) currently does not have a clinical policy for the initial treatment of patients with DM that is newly diagnosed in the ED. A 2007 survey of 152 EPs in the United States found that 52% would leave initiation of outpatient treatment of diabetes to a patient’s primary care physician (PCP), regardless of the degree of hyperglycemia noted in the ED.³ Almost paradoxically, among these same EPs, the most common reason for not screening for diabetes (cited by 69%) was the inability to secure follow-up for newly diagnosed patients. Given the fragmentary care many emergency patients receive and the difficulty in assuring close ED follow-up, reviewing the American Diabetes Association (ADA) recommendations for initiation of diabetes management is worthwhile.

ADA Treatment Guidelines

Metformin
Metformin, if tolerated and without contraindications, is the preferred initial pharmacologic agent for all patients with newly diagnosed DM. This drug typically lowers HbA1_c by 1.5%, and does not cause weight gain or hypoglycemia associated with some other oral agents (eg, sulfonylureas). Metformin therapy, however, is contraindicated in patients with chronic kidney disease (CKD) (Cr >1.5 in males and >1.4 in females), liver disease, a history of lactic acidosis, and a low-perfusion state. Metformin also should not be used for 48 hours following an intravenous (IV) contrast-enhanced study.

When starting metformin or increasing the dose, patients frequently experience abdominal symptoms, including cramping, nausea, vomiting, and diarrhea. These side effects are usually mild and self-limiting, and they usually resolve within a week. However, patients should be warned of these complications and advised to take metformin with meals to limit side effects. They also should be encouraged to continue the medication through the initial symptoms, if possible.

For a patient with newly diagnosed type 2 DM, a reasonable ED discharge plan is to start metformin at 500 mg daily, taken with a meal. If the patient develops abdominal symptoms, the dose can be reduced to 250 mg daily. Patients should additionally be advised to follow up with their PCP as soon as feasible. Patients can be instructed to increase the dose of metformin on their own. Because the development of gastrointestinal symptoms is the main dose-limiting effect of metformin therapy, it is generally safe to give patients instructions on weekly dose escalation by 500 mg/day to the optimal dose of 1,000 mg twice daily. Although metformin therapy is associated with rare cases of lactic acidosis in patients with CKD, it does not, on its own, cause acute kidney injury; therefore, intense monitoring is not required while doses are increased. Metformin therapy can decrease vitamin B12 absorption, though it rarely results in megaloblastic anemia, and it takes 5 to 10 years for neurological symptoms to manifest.⁵

Other Treatment Options

The Figure highlights the ADA guidelines on additional pharmacologic agents beyond metformin. When these additional agents are used, the patient should follow-up with a PCP every 3 months to determine how far HbA1_c has been lowered. For most patients with DM, the goal for HbA1_c is less than 7%. This goal may be raised for elderly patients and those with a history of cardiovascular disease, as these patients are at higher risk for cardiovascular mortality from an episode of hypoglycemia.

The secondary medication options include sulfonylureas, thiazolidinediones, dipeptidyl peptidase-IV inhibitors, glucagon-like peptide-1 (GLP-1) agonists, sodium-glucose cotransporter-2 inhibitors, and insulin. Insulin therapy is the final medication step for all patients with DM who are willing and able to comply with daily injections and who have failed to meet goals through alternative agents.

Before using additional agents beyond metformin, the PCP or endocrinologist should discuss in detail factors such as side effects (ie, weight gain, risk of hypoglycemia), medication delivery (GLP-1 agonists and insulin require injections), and cost with the patient. It is generally advisable not to try to accomplish this medication adjustment in the ED without consultation with a PCP or endocrinologist.

Diagnosis

The clinical presentation of new-onset DM can range from the relatively benign (eg, polyuria, polydipsia, frequent skin infections) to the life-threatening (eg, diabetic ketoacidosis) (Table 1). The classic symptoms of hyperglycemia include polyuria, polydipsia, unexplained weight loss, easy fatigability, dizziness, and blurred vision. Any one of these symptoms should prompt consideration of DM in the differential diagnosis.

Four abnormal laboratory results comprise the ADA’s diagnostic criteria for DM⁶:

• HbA1_c ≥6.5%;
• 8-hour fasting plasma glucose ≥126 mg/dL;
• 2-hour oral glucose tolerance test (OGTT) ≥200 mg/dL (Patient is given 75 g sugar orally; then plasma glucose is drawn 2 hours later); and
• A random plasma glucose of ≥200 mg/dL in association with “classic symptoms” of DM, defined as polyuria, polydipsia, and unexplained weight loss.

In the absence of an unequivocal indicator, such as the patient presenting in diabetic ketoacidosis, the formal diagnosis of DM should not be made unless at least 2 abnormal test results are obtained simultaneously (from a 2-hour OGTT test and an HbA1_c panel, for example, or when the same test is repeated one month later and both test results indicate DM).

Limitations of Glucometers
When interpreting test results, keep in mind that the US Food and Drug Administration allows point-of-care (POC) glucometers to have an accuracy of +/- 20%. The POC glucometers measure whole-blood glucose rather than plasma glucose; generally the amount of glucose in plasma is about 10% to 15% higher than the amount in whole blood.7 Therefore, the formal diagnosis of DM relies on laboratory plasma glucose results; POC glucometer results may guide the EP in initiating a workup for DM but should not be among the diagnostic criteria.

HbA1c Testing
The HbA1_c percentage represents the preceding 3-month average plasma glucose level. This is due to nonenzymatic glycosylation of Hb, a reaction driven by higher concentrations of circulating plasma glucose. The HbA1_c can be falsely low in states of high red blood cell (RBC) turnover, such as ongoing blood loss or the RBC fragility seen in hemoglobinopathies. Patients who have had RBC transfusions in the preceding months will have inaccurate HbA1_c test results as well.

Keeping these test characteristics in mind, the HbA1_c test may be more useful than glucose-based testing strategies in diagnosing diabetes in the ED. Hyperglycemia can be observed in nondiabetic emergency patients undergoing a stress response, since gluconeogenesis and glycogenolysis are normal hepatic stress responses to circulating epinephrine and cortisol. Unless a patient is on chronic steroids, the HbA1c is not similarly affected.

A recent study of ED patients using HbA1_c as the sole screening modality found that test characteristics for ED patients were similar to those for patients in typical outpatient settings. HbA1_c of 6.5 or greater yielded a sensitivity of 54% and a specificity of 96% in diagnosing diabetes.8 The availability of test results at the time of the emergency visit depends upon the individual hospital laboratory, but most labs run the test itself in less than an hour. Emergency physicians working in hospitals that perform the test will often have the test results prior to patient discharge.

Inpatient Management

Managing a patient with newly diagnosed DM that requires admission to the hospital is different from managing a patient who can be treated as an outpatient. The ADA-recommended treatment regimen for DM in the inpatient setting focuses on the use of insulin. While type 1 DM, by definition, requires insulin treatment, many patients with type 2 DM are treated with oral agents only as outpatients. However, because of complications associated with the use of oral agents in low renal perfusion states (such as during surgery, during IV contrast-enhanced studies, or when the admitting condition causes shock), the ADA generally does not prefer oral agents for inpatient glycemic control. Metformin particularly should not be used in the inpatient setting, due to increased risk of lactic acidosis.

The preferred insulin regimen for noncritically ill diabetic inpatients consists of scheduled subcutaneous therapy with basal, nutritional, and correctional components.⁹ Because absorption from subcutaneous tissue varies among critically ill patients, the ADA recommends an IV insulin therapy protocol for ICU patients with diabetes. The NICE SUGAR trial found that the overall risk of death was 2.6% higher among patients treated with an intensive insulin therapy regimen (finger stick blood glucose goal 80-108 mg/dL) than it was among patients treated using standard therapy with a goal of less than 180 mg/dL, a statistically significant difference.¹⁰ These results form the basis for the ADA-recommended inpatient glycemic target of 140 to 180 mg/dL among critically ill patients.¹¹ While no definitive evidence exists to guide glycemic control among noncritically ill inpatients, the strategy applied to critically ill patients can be extrapolated to other hospitalized patients because the ADA recommends treatment to keep finger stick blood glucose values below 180 mg/dL.

Hypoglycemia

Often a hypoglycemic event can clearly be correlated to a missed meal. This may only require re-education of the patient about the need to be consistent in the timing of his or her dietary regimen. However, often a hypoglycemic event is not clearly related to a missed meal. When a patient has recently taken a long-acting insulin or sulfonylurea or has new renal failure, prolonged observation or admission may be required. In most other cases of medication-induced hypoglycemia, the patients are expediently sent home. As many such patient encounters occur in the early morning hours, EPs need to know how to manage insulin regimens if these patients are to be safely discharged home. Additionally, worsening renal function is a common cause of potentiation of both oral agents and insulin; it is generally worthwhile to check renal function on patients presenting with significant hypoglycemia.

After a significant hypoglycemic event, many guidelines recommend decreasing the daily insulin dose by 10% to 20%. However, with a careful medication history and knowledge of each insulin dose’s activity profile, a physician can often discern which individual insulin dose is responsible for a given hypoglycemic event. That individual daily dose can then be decreased by 20% to better target the cause of the hypoglycemic event.

Conclusion

Diabetes mellitus is one of the most common chronic medical conditions in the United States, and more than 7 million Americans may be living with undiagnosed DM. In the ED setting, physicians regularly encounter patients with undiagnosed type 2 DM. Since treatment is known to prevent the microvascular complications associated with DM, EPs should know the diagnostic criteria and understand the ADA’s inpatient and outpatient treatment recommendations.

According to the US Centers for Disease Control and Prevention (CDC), in 2011 there were 25.8 million people in the United States living with diabetes mellitus (DM), 90% to 95% of whom had type 2 DM.1 Seven million of those with DM are undiagnosed. In addition, an estimated 79 million adults in the United States have prediabetes, a condition in which glycated hemoglobin (HbA1_c) levels are above normal but do not yet meet the diagnostic threshold for DM (≥6.5%).¹ This large population with prediabetes, defined by HbA1_c levels of 5.7% to 6.4%, has a 5-year risk of progression to diabetes ranging from 9% to 50%; the higher risk is associated with higher HbA1_c levels as the disease represents a continuum.²

The benefits of diagnosis and treatment of diabetes—preventing the progression of microvascular complications—are well established. Through medication, exercise, and dietary interventions, the progression of both prediabetes and diabetes can be halted and in many cases reversed.

Given the prevalence of DM in the United States and the clear benefits of treatment, the ability to diagnose DM is important for any physician who may encounter patients with undiagnosed diabetes or prediabetes. This article reviews the management of hyperglycemia relevant to the emergency physician (EP) and discusses the criteria used in diagnosing DM.

The American College of Emergency Physicians (ACEP) currently does not have a clinical policy for the initial treatment of patients with DM that is newly diagnosed in the ED. A 2007 survey of 152 EPs in the United States found that 52% would leave initiation of outpatient treatment of diabetes to a patient’s primary care physician (PCP), regardless of the degree of hyperglycemia noted in the ED.³ Almost paradoxically, among these same EPs, the most common reason for not screening for diabetes (cited by 69%) was the inability to secure follow-up for newly diagnosed patients. Given the fragmentary care many emergency patients receive and the difficulty in assuring close ED follow-up, reviewing the American Diabetes Association (ADA) recommendations for initiation of diabetes management is worthwhile.

ADA Treatment Guidelines

Metformin
Metformin, if tolerated and without contraindications, is the preferred initial pharmacologic agent for all patients with newly diagnosed DM. This drug typically lowers HbA1_c by 1.5%, and does not cause weight gain or hypoglycemia associated with some other oral agents (eg, sulfonylureas). Metformin therapy, however, is contraindicated in patients with chronic kidney disease (CKD) (Cr >1.5 in males and >1.4 in females), liver disease, a history of lactic acidosis, and a low-perfusion state. Metformin also should not be used for 48 hours following an intravenous (IV) contrast-enhanced study.

When starting metformin or increasing the dose, patients frequently experience abdominal symptoms, including cramping, nausea, vomiting, and diarrhea. These side effects are usually mild and self-limiting, and they usually resolve within a week. However, patients should be warned of these complications and advised to take metformin with meals to limit side effects. They also should be encouraged to continue the medication through the initial symptoms, if possible.

For a patient with newly diagnosed type 2 DM, a reasonable ED discharge plan is to start metformin at 500 mg daily, taken with a meal. If the patient develops abdominal symptoms, the dose can be reduced to 250 mg daily. Patients should additionally be advised to follow up with their PCP as soon as feasible. Patients can be instructed to increase the dose of metformin on their own. Because the development of gastrointestinal symptoms is the main dose-limiting effect of metformin therapy, it is generally safe to give patients instructions on weekly dose escalation by 500 mg/day to the optimal dose of 1,000 mg twice daily. Although metformin therapy is associated with rare cases of lactic acidosis in patients with CKD, it does not, on its own, cause acute kidney injury; therefore, intense monitoring is not required while doses are increased. Metformin therapy can decrease vitamin B12 absorption, though it rarely results in megaloblastic anemia, and it takes 5 to 10 years for neurological symptoms to manifest.⁵

Other Treatment Options

The Figure highlights the ADA guidelines on additional pharmacologic agents beyond metformin. When these additional agents are used, the patient should follow-up with a PCP every 3 months to determine how far HbA1_c has been lowered. For most patients with DM, the goal for HbA1_c is less than 7%. This goal may be raised for elderly patients and those with a history of cardiovascular disease, as these patients are at higher risk for cardiovascular mortality from an episode of hypoglycemia.

The secondary medication options include sulfonylureas, thiazolidinediones, dipeptidyl peptidase-IV inhibitors, glucagon-like peptide-1 (GLP-1) agonists, sodium-glucose cotransporter-2 inhibitors, and insulin. Insulin therapy is the final medication step for all patients with DM who are willing and able to comply with daily injections and who have failed to meet goals through alternative agents.

Before using additional agents beyond metformin, the PCP or endocrinologist should discuss in detail factors such as side effects (ie, weight gain, risk of hypoglycemia), medication delivery (GLP-1 agonists and insulin require injections), and cost with the patient. It is generally advisable not to try to accomplish this medication adjustment in the ED without consultation with a PCP or endocrinologist.

Diagnosis

The clinical presentation of new-onset DM can range from the relatively benign (eg, polyuria, polydipsia, frequent skin infections) to the life-threatening (eg, diabetic ketoacidosis) (Table 1). The classic symptoms of hyperglycemia include polyuria, polydipsia, unexplained weight loss, easy fatigability, dizziness, and blurred vision. Any one of these symptoms should prompt consideration of DM in the differential diagnosis.

Four abnormal laboratory results comprise the ADA’s diagnostic criteria for DM⁶:

• HbA1_c ≥6.5%;
• 8-hour fasting plasma glucose ≥126 mg/dL;
• 2-hour oral glucose tolerance test (OGTT) ≥200 mg/dL (Patient is given 75 g sugar orally; then plasma glucose is drawn 2 hours later); and
• A random plasma glucose of ≥200 mg/dL in association with “classic symptoms” of DM, defined as polyuria, polydipsia, and unexplained weight loss.

In the absence of an unequivocal indicator, such as the patient presenting in diabetic ketoacidosis, the formal diagnosis of DM should not be made unless at least 2 abnormal test results are obtained simultaneously (from a 2-hour OGTT test and an HbA1_c panel, for example, or when the same test is repeated one month later and both test results indicate DM).

Limitations of Glucometers
When interpreting test results, keep in mind that the US Food and Drug Administration allows point-of-care (POC) glucometers to have an accuracy of +/- 20%. The POC glucometers measure whole-blood glucose rather than plasma glucose; generally the amount of glucose in plasma is about 10% to 15% higher than the amount in whole blood.7 Therefore, the formal diagnosis of DM relies on laboratory plasma glucose results; POC glucometer results may guide the EP in initiating a workup for DM but should not be among the diagnostic criteria.

HbA1c Testing
The HbA1_c percentage represents the preceding 3-month average plasma glucose level. This is due to nonenzymatic glycosylation of Hb, a reaction driven by higher concentrations of circulating plasma glucose. The HbA1_c can be falsely low in states of high red blood cell (RBC) turnover, such as ongoing blood loss or the RBC fragility seen in hemoglobinopathies. Patients who have had RBC transfusions in the preceding months will have inaccurate HbA1_c test results as well.

Keeping these test characteristics in mind, the HbA1_c test may be more useful than glucose-based testing strategies in diagnosing diabetes in the ED. Hyperglycemia can be observed in nondiabetic emergency patients undergoing a stress response, since gluconeogenesis and glycogenolysis are normal hepatic stress responses to circulating epinephrine and cortisol. Unless a patient is on chronic steroids, the HbA1c is not similarly affected.

A recent study of ED patients using HbA1_c as the sole screening modality found that test characteristics for ED patients were similar to those for patients in typical outpatient settings. HbA1_c of 6.5 or greater yielded a sensitivity of 54% and a specificity of 96% in diagnosing diabetes.8 The availability of test results at the time of the emergency visit depends upon the individual hospital laboratory, but most labs run the test itself in less than an hour. Emergency physicians working in hospitals that perform the test will often have the test results prior to patient discharge.

Inpatient Management

Managing a patient with newly diagnosed DM that requires admission to the hospital is different from managing a patient who can be treated as an outpatient. The ADA-recommended treatment regimen for DM in the inpatient setting focuses on the use of insulin. While type 1 DM, by definition, requires insulin treatment, many patients with type 2 DM are treated with oral agents only as outpatients. However, because of complications associated with the use of oral agents in low renal perfusion states (such as during surgery, during IV contrast-enhanced studies, or when the admitting condition causes shock), the ADA generally does not prefer oral agents for inpatient glycemic control. Metformin particularly should not be used in the inpatient setting, due to increased risk of lactic acidosis.

The preferred insulin regimen for noncritically ill diabetic inpatients consists of scheduled subcutaneous therapy with basal, nutritional, and correctional components.⁹ Because absorption from subcutaneous tissue varies among critically ill patients, the ADA recommends an IV insulin therapy protocol for ICU patients with diabetes. The NICE SUGAR trial found that the overall risk of death was 2.6% higher among patients treated with an intensive insulin therapy regimen (finger stick blood glucose goal 80-108 mg/dL) than it was among patients treated using standard therapy with a goal of less than 180 mg/dL, a statistically significant difference.¹⁰ These results form the basis for the ADA-recommended inpatient glycemic target of 140 to 180 mg/dL among critically ill patients.¹¹ While no definitive evidence exists to guide glycemic control among noncritically ill inpatients, the strategy applied to critically ill patients can be extrapolated to other hospitalized patients because the ADA recommends treatment to keep finger stick blood glucose values below 180 mg/dL.

Hypoglycemia

Often a hypoglycemic event can clearly be correlated to a missed meal. This may only require re-education of the patient about the need to be consistent in the timing of his or her dietary regimen. However, often a hypoglycemic event is not clearly related to a missed meal. When a patient has recently taken a long-acting insulin or sulfonylurea or has new renal failure, prolonged observation or admission may be required. In most other cases of medication-induced hypoglycemia, the patients are expediently sent home. As many such patient encounters occur in the early morning hours, EPs need to know how to manage insulin regimens if these patients are to be safely discharged home. Additionally, worsening renal function is a common cause of potentiation of both oral agents and insulin; it is generally worthwhile to check renal function on patients presenting with significant hypoglycemia.

After a significant hypoglycemic event, many guidelines recommend decreasing the daily insulin dose by 10% to 20%. However, with a careful medication history and knowledge of each insulin dose’s activity profile, a physician can often discern which individual insulin dose is responsible for a given hypoglycemic event. That individual daily dose can then be decreased by 20% to better target the cause of the hypoglycemic event.

Conclusion

Diabetes mellitus is one of the most common chronic medical conditions in the United States, and more than 7 million Americans may be living with undiagnosed DM. In the ED setting, physicians regularly encounter patients with undiagnosed type 2 DM. Since treatment is known to prevent the microvascular complications associated with DM, EPs should know the diagnostic criteria and understand the ADA’s inpatient and outpatient treatment recommendations.

Rapid advances in technology have brought improvements in prosthetic components. In particular, prosthetic knees and ankle/foot complexes have made substantial advancements with the incorporation of computer technology. For example, microprocessor knees are relatively new; the X2 knee from Ottobock (Minneapolis, Minnesota) represents one of the latest and most advanced units and has just been upgraded.

Until recently, there have been no similarly functioning ankle/foot components except for the Proprio Foot from Össur (Foothill Ranch, California), which also provides powered dorsiflexion.

Also, recently BiOM introduced the BiOM T2 foot and ankle system with the added technology of powered plantarflexion to further normalize amputee prosthetic gait. Active patients who have successfully used a microprocessor knee, such as the X2, have generally paired that technology with a variety of foot/ankle components, ranging from passive-elastic units to advanced-energy storing units.

To normalize gait and improve biomechanics even further in select above-knee amputees, experts in the field have suggested combining a microprocessor knee with a powered foot/ankle complex. One potential obstacle to this combination, however, concerns the possible conflict between the active components of the individual units, such as over- or underengagement of component sensors. This situation, theoretically, could compromise patient safety. BiOM, however, provides training to prosthetic providers to address possible component integration issues, including microprocessor conflict and methods to safely use the components together. Once the prosthetist received this training, the patient in this study was fitted with the T2 foot and the X2 knee with excellent results and no perceived disadvantages.

Case Presentation

The patient was a 32-year-old man with a right transfemoral amputation due to trauma from a blast injury, which occurred during Marine service in Iraq. He also had a gunshot wound to his left leg, which resulted in severe injury, but this limb was salvaged and now has good residual function. Before the trauma, the patient was very athletic and involved in long-distance running and bicycling. Once he recovered from his acute injuries, the patient expressed a desire to return to his previous high level of activity and sport participation.

The experiences of these limitations pushed him to look for other prosthetic options that would offer better performance in these situations. Ultimately, he received the T2 ankle/foot with the X2 microprocessor knee after using a different combination for 2 years. He felt substantial improvements in all the aforementioned limitations and has been using the X2 and T2 combination ever since. The prosthetist provided training in both instances. For distance running, the patient uses the Flex-Run (Össur) Foot.

The Trinity Amputation and Prosthesis Experience Scale (TAPES) and the Locomotor Capabilities Index in Amputees (LCI) were used to assess his adjustment to the prosthetic and performance, respectively, before and after use of the aforementioned combination.

The LCI is a validated measure of lower-extremity amputees’ ability to perform activities with a prosthesis.¹ The patient scored the maximum of 7 for all parameters of the LCI (a total of 28 parameters) while using his baseline prosthetic configuration of the X2 knee and the Triton foot (Ottobock). This score did not change when he used the X2/T2 combination (Figure 1; Table).

The TAPES Index is a validated measure of psychological adjustment to prosthetic integration.² The measure consists of 12 items, rated 1 to 3 (1 = limited a lot; 2 = limited a little; and 3 = not limited at all). His total score was 25 using the X2 alone without the T2 but with the Triton foot. The patient reported that he was “limited a lot” on 2 activity measures (climbing several flights of stairs and running to catch a bus). This measure was reapplied after the patient used the T2 ankle/foot and X2 knee for several weeks. His new sum score was 36, the highest possible for this measure, indicating no functional, social, or athletic restrictions.

Furthermore, the patient reported other improvements, including an almost complete elimination of long-standing back pain, present since amputation. He reported he was able to climb hills with increased speed and less fatigue. The patient also reported he could stand more comfortably and don his shoes more easily, because the T2 would “bend.” Other subjective activity improvements included the ability to easily pick an object off the floor, step up curbs, walk on uneven ground, perform a mountain-climber exercise, and go through small spaces. He reported he was able to do all these activities previously, but the X2/T2 combination made these tasks easier than before to accomplish (Figures 2A and 2B).

Discussion

The subject of this case report is a physically active traumatic transfemoral amputee who had previous experience with several prosthetic components with the ultimate preference and use of the X2 microprocessor knee. Because of the patient’s desire for the most natural and energy-sparing gait he could achieve, a T2 foot and ankle system was added. Though objective measures of locomotion (LCI) did not change, he reported significant improvement in subjective measures of function and prosthetic acceptance (TAPES).

Reported objective advantages favoring the use of microprocessor prosthetic components most often refer to the decrease in energy consumption during locomotion. Several small studies have compared powered with nonpowered, energy-storing, or passive-elastic components and demonstrated at least modest energy savings. In a study of transtibial amputees, researchers compared oxygen consumption during locomotion in patients fitted with a passive-elastic ankle/foot with patients fitted with the powered T2.³ The researchers reported an average decrease in overall energy consumption of 8.4%. Plantarflexion and p

eak ankle-power production at push-off were both increased. The authors of this study conclude that the T2 ankle/foot allows achievement of greater biological realism.

A 2010 review by Highsmith and colleagues concluded that the microprocessor knee C-Leg demonstrated increased efficacy in safety and energy efficiency compared with other prosthetic knees for transfemoral amputees.⁴

Subjectively, the study patient reported less fatigue when using the X2/T2 combination than when using the X2 knee without the T2 ankle/foot. It is currently unknown whether the combination provided additive energy savings, and this area would be a good course for future investigation.

The study patient reported several subjective improvements, including reduced back pain, a more natural gait, and improved mobility. Hammarlund and colleagues found a significant prevalence of postamputation lower-extremity back pain compared with preamputation symptoms.⁵ This pain resulted in at least moderate disability in all subjects during prosthetic use. Morgenroth and colleagues went on to speculate that abnormal lumbar spinal kinematics could be a contributing factor for back pain in transfemoral amputees.⁶

Though not specifically causative, the study found that those transfemoral amputees with increased lumbar spine transverse plane motion experienced significantly more back pain than did similar amputees without lumbar spine transverse plane motion. An abnormal gait would promote more transverse plane motion than that seen in a normal gait. Normalizing prosthetic gait to best simulate the patient’s preamputation biomechanical baseline could reduce transverse lumbar spine motion, reduce back and other mechanical pain, and ultimately, reduce overall disability.

Similarly, the patient in this study also reported increased ease with hills and stairs. Many studies exist that attest to the advantages of microprocessor knees in providing improvements such as decreased stumbles, increased ability to multitask, increased satisfaction with the prosthesis, and improved stair and stance functions, such as with the Genium (Ottobock).^7,8 Whether the combination of a microprocessor knee with a powered ankle/foot would further improve these aspects is yet to be objectively investigated. The report of this study patient who used the combination suggests these types of advantages but certainly as a single case report does not provide definitive answers.

The patient achieved the highest possible score on the LCI before using the X2/T2 combination and thus demonstrated a ceiling effect that has been discussed in several studies.⁹ Furthermore, Larsson and colleagues noted that because of the ceiling effect, the LCI was more useful for amputees of low to moderate activity levels.¹⁰ The TAPES, however, showed an improvement in before and after measurements, and assessment with it was not hindered by a ceiling effect.

Conclusion

The patient in this case report noted substantial subjective functional improvements when using the X2 compared with prior mechanical prosthetic knees paired with the T2 foot/ankle. The functional gains were further verified by significant improvement in the TAPES Index score, a validated measure of prosthetic integration. Specific subjective advantages included energy savings, almost complete resolution of back pain, and improved facility with hills, stairs, and crawl spaces. No perceived disadvantages were reported. 

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Rapid advances in technology have brought improvements in prosthetic components. In particular, prosthetic knees and ankle/foot complexes have made substantial advancements with the incorporation of computer technology. For example, microprocessor knees are relatively new; the X2 knee from Ottobock (Minneapolis, Minnesota) represents one of the latest and most advanced units and has just been upgraded.

Until recently, there have been no similarly functioning ankle/foot components except for the Proprio Foot from Össur (Foothill Ranch, California), which also provides powered dorsiflexion.

Also, recently BiOM introduced the BiOM T2 foot and ankle system with the added technology of powered plantarflexion to further normalize amputee prosthetic gait. Active patients who have successfully used a microprocessor knee, such as the X2, have generally paired that technology with a variety of foot/ankle components, ranging from passive-elastic units to advanced-energy storing units.

To normalize gait and improve biomechanics even further in select above-knee amputees, experts in the field have suggested combining a microprocessor knee with a powered foot/ankle complex. One potential obstacle to this combination, however, concerns the possible conflict between the active components of the individual units, such as over- or underengagement of component sensors. This situation, theoretically, could compromise patient safety. BiOM, however, provides training to prosthetic providers to address possible component integration issues, including microprocessor conflict and methods to safely use the components together. Once the prosthetist received this training, the patient in this study was fitted with the T2 foot and the X2 knee with excellent results and no perceived disadvantages.

Case Presentation

The patient was a 32-year-old man with a right transfemoral amputation due to trauma from a blast injury, which occurred during Marine service in Iraq. He also had a gunshot wound to his left leg, which resulted in severe injury, but this limb was salvaged and now has good residual function. Before the trauma, the patient was very athletic and involved in long-distance running and bicycling. Once he recovered from his acute injuries, the patient expressed a desire to return to his previous high level of activity and sport participation.

The experiences of these limitations pushed him to look for other prosthetic options that would offer better performance in these situations. Ultimately, he received the T2 ankle/foot with the X2 microprocessor knee after using a different combination for 2 years. He felt substantial improvements in all the aforementioned limitations and has been using the X2 and T2 combination ever since. The prosthetist provided training in both instances. For distance running, the patient uses the Flex-Run (Össur) Foot.

The Trinity Amputation and Prosthesis Experience Scale (TAPES) and the Locomotor Capabilities Index in Amputees (LCI) were used to assess his adjustment to the prosthetic and performance, respectively, before and after use of the aforementioned combination.

The LCI is a validated measure of lower-extremity amputees’ ability to perform activities with a prosthesis.¹ The patient scored the maximum of 7 for all parameters of the LCI (a total of 28 parameters) while using his baseline prosthetic configuration of the X2 knee and the Triton foot (Ottobock). This score did not change when he used the X2/T2 combination (Figure 1; Table).

The TAPES Index is a validated measure of psychological adjustment to prosthetic integration.² The measure consists of 12 items, rated 1 to 3 (1 = limited a lot; 2 = limited a little; and 3 = not limited at all). His total score was 25 using the X2 alone without the T2 but with the Triton foot. The patient reported that he was “limited a lot” on 2 activity measures (climbing several flights of stairs and running to catch a bus). This measure was reapplied after the patient used the T2 ankle/foot and X2 knee for several weeks. His new sum score was 36, the highest possible for this measure, indicating no functional, social, or athletic restrictions.

Furthermore, the patient reported other improvements, including an almost complete elimination of long-standing back pain, present since amputation. He reported he was able to climb hills with increased speed and less fatigue. The patient also reported he could stand more comfortably and don his shoes more easily, because the T2 would “bend.” Other subjective activity improvements included the ability to easily pick an object off the floor, step up curbs, walk on uneven ground, perform a mountain-climber exercise, and go through small spaces. He reported he was able to do all these activities previously, but the X2/T2 combination made these tasks easier than before to accomplish (Figures 2A and 2B).

Discussion

The subject of this case report is a physically active traumatic transfemoral amputee who had previous experience with several prosthetic components with the ultimate preference and use of the X2 microprocessor knee. Because of the patient’s desire for the most natural and energy-sparing gait he could achieve, a T2 foot and ankle system was added. Though objective measures of locomotion (LCI) did not change, he reported significant improvement in subjective measures of function and prosthetic acceptance (TAPES).

Reported objective advantages favoring the use of microprocessor prosthetic components most often refer to the decrease in energy consumption during locomotion. Several small studies have compared powered with nonpowered, energy-storing, or passive-elastic components and demonstrated at least modest energy savings. In a study of transtibial amputees, researchers compared oxygen consumption during locomotion in patients fitted with a passive-elastic ankle/foot with patients fitted with the powered T2.³ The researchers reported an average decrease in overall energy consumption of 8.4%. Plantarflexion and p

eak ankle-power production at push-off were both increased. The authors of this study conclude that the T2 ankle/foot allows achievement of greater biological realism.

A 2010 review by Highsmith and colleagues concluded that the microprocessor knee C-Leg demonstrated increased efficacy in safety and energy efficiency compared with other prosthetic knees for transfemoral amputees.⁴

Subjectively, the study patient reported less fatigue when using the X2/T2 combination than when using the X2 knee without the T2 ankle/foot. It is currently unknown whether the combination provided additive energy savings, and this area would be a good course for future investigation.

The study patient reported several subjective improvements, including reduced back pain, a more natural gait, and improved mobility. Hammarlund and colleagues found a significant prevalence of postamputation lower-extremity back pain compared with preamputation symptoms.⁵ This pain resulted in at least moderate disability in all subjects during prosthetic use. Morgenroth and colleagues went on to speculate that abnormal lumbar spinal kinematics could be a contributing factor for back pain in transfemoral amputees.⁶

Though not specifically causative, the study found that those transfemoral amputees with increased lumbar spine transverse plane motion experienced significantly more back pain than did similar amputees without lumbar spine transverse plane motion. An abnormal gait would promote more transverse plane motion than that seen in a normal gait. Normalizing prosthetic gait to best simulate the patient’s preamputation biomechanical baseline could reduce transverse lumbar spine motion, reduce back and other mechanical pain, and ultimately, reduce overall disability.

Similarly, the patient in this study also reported increased ease with hills and stairs. Many studies exist that attest to the advantages of microprocessor knees in providing improvements such as decreased stumbles, increased ability to multitask, increased satisfaction with the prosthesis, and improved stair and stance functions, such as with the Genium (Ottobock).^7,8 Whether the combination of a microprocessor knee with a powered ankle/foot would further improve these aspects is yet to be objectively investigated. The report of this study patient who used the combination suggests these types of advantages but certainly as a single case report does not provide definitive answers.

The patient achieved the highest possible score on the LCI before using the X2/T2 combination and thus demonstrated a ceiling effect that has been discussed in several studies.⁹ Furthermore, Larsson and colleagues noted that because of the ceiling effect, the LCI was more useful for amputees of low to moderate activity levels.¹⁰ The TAPES, however, showed an improvement in before and after measurements, and assessment with it was not hindered by a ceiling effect.

Conclusion

The patient in this case report noted substantial subjective functional improvements when using the X2 compared with prior mechanical prosthetic knees paired with the T2 foot/ankle. The functional gains were further verified by significant improvement in the TAPES Index score, a validated measure of prosthetic integration. Specific subjective advantages included energy savings, almost complete resolution of back pain, and improved facility with hills, stairs, and crawl spaces. No perceived disadvantages were reported. 

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Rapid advances in technology have brought improvements in prosthetic components. In particular, prosthetic knees and ankle/foot complexes have made substantial advancements with the incorporation of computer technology. For example, microprocessor knees are relatively new; the X2 knee from Ottobock (Minneapolis, Minnesota) represents one of the latest and most advanced units and has just been upgraded.

Until recently, there have been no similarly functioning ankle/foot components except for the Proprio Foot from Össur (Foothill Ranch, California), which also provides powered dorsiflexion.

Also, recently BiOM introduced the BiOM T2 foot and ankle system with the added technology of powered plantarflexion to further normalize amputee prosthetic gait. Active patients who have successfully used a microprocessor knee, such as the X2, have generally paired that technology with a variety of foot/ankle components, ranging from passive-elastic units to advanced-energy storing units.

To normalize gait and improve biomechanics even further in select above-knee amputees, experts in the field have suggested combining a microprocessor knee with a powered foot/ankle complex. One potential obstacle to this combination, however, concerns the possible conflict between the active components of the individual units, such as over- or underengagement of component sensors. This situation, theoretically, could compromise patient safety. BiOM, however, provides training to prosthetic providers to address possible component integration issues, including microprocessor conflict and methods to safely use the components together. Once the prosthetist received this training, the patient in this study was fitted with the T2 foot and the X2 knee with excellent results and no perceived disadvantages.

Case Presentation

The patient was a 32-year-old man with a right transfemoral amputation due to trauma from a blast injury, which occurred during Marine service in Iraq. He also had a gunshot wound to his left leg, which resulted in severe injury, but this limb was salvaged and now has good residual function. Before the trauma, the patient was very athletic and involved in long-distance running and bicycling. Once he recovered from his acute injuries, the patient expressed a desire to return to his previous high level of activity and sport participation.

The experiences of these limitations pushed him to look for other prosthetic options that would offer better performance in these situations. Ultimately, he received the T2 ankle/foot with the X2 microprocessor knee after using a different combination for 2 years. He felt substantial improvements in all the aforementioned limitations and has been using the X2 and T2 combination ever since. The prosthetist provided training in both instances. For distance running, the patient uses the Flex-Run (Össur) Foot.

The Trinity Amputation and Prosthesis Experience Scale (TAPES) and the Locomotor Capabilities Index in Amputees (LCI) were used to assess his adjustment to the prosthetic and performance, respectively, before and after use of the aforementioned combination.

The LCI is a validated measure of lower-extremity amputees’ ability to perform activities with a prosthesis.¹ The patient scored the maximum of 7 for all parameters of the LCI (a total of 28 parameters) while using his baseline prosthetic configuration of the X2 knee and the Triton foot (Ottobock). This score did not change when he used the X2/T2 combination (Figure 1; Table).

The TAPES Index is a validated measure of psychological adjustment to prosthetic integration.² The measure consists of 12 items, rated 1 to 3 (1 = limited a lot; 2 = limited a little; and 3 = not limited at all). His total score was 25 using the X2 alone without the T2 but with the Triton foot. The patient reported that he was “limited a lot” on 2 activity measures (climbing several flights of stairs and running to catch a bus). This measure was reapplied after the patient used the T2 ankle/foot and X2 knee for several weeks. His new sum score was 36, the highest possible for this measure, indicating no functional, social, or athletic restrictions.

Furthermore, the patient reported other improvements, including an almost complete elimination of long-standing back pain, present since amputation. He reported he was able to climb hills with increased speed and less fatigue. The patient also reported he could stand more comfortably and don his shoes more easily, because the T2 would “bend.” Other subjective activity improvements included the ability to easily pick an object off the floor, step up curbs, walk on uneven ground, perform a mountain-climber exercise, and go through small spaces. He reported he was able to do all these activities previously, but the X2/T2 combination made these tasks easier than before to accomplish (Figures 2A and 2B).

Discussion

The subject of this case report is a physically active traumatic transfemoral amputee who had previous experience with several prosthetic components with the ultimate preference and use of the X2 microprocessor knee. Because of the patient’s desire for the most natural and energy-sparing gait he could achieve, a T2 foot and ankle system was added. Though objective measures of locomotion (LCI) did not change, he reported significant improvement in subjective measures of function and prosthetic acceptance (TAPES).

Reported objective advantages favoring the use of microprocessor prosthetic components most often refer to the decrease in energy consumption during locomotion. Several small studies have compared powered with nonpowered, energy-storing, or passive-elastic components and demonstrated at least modest energy savings. In a study of transtibial amputees, researchers compared oxygen consumption during locomotion in patients fitted with a passive-elastic ankle/foot with patients fitted with the powered T2.³ The researchers reported an average decrease in overall energy consumption of 8.4%. Plantarflexion and p

eak ankle-power production at push-off were both increased. The authors of this study conclude that the T2 ankle/foot allows achievement of greater biological realism.

A 2010 review by Highsmith and colleagues concluded that the microprocessor knee C-Leg demonstrated increased efficacy in safety and energy efficiency compared with other prosthetic knees for transfemoral amputees.⁴

Subjectively, the study patient reported less fatigue when using the X2/T2 combination than when using the X2 knee without the T2 ankle/foot. It is currently unknown whether the combination provided additive energy savings, and this area would be a good course for future investigation.

The study patient reported several subjective improvements, including reduced back pain, a more natural gait, and improved mobility. Hammarlund and colleagues found a significant prevalence of postamputation lower-extremity back pain compared with preamputation symptoms.⁵ This pain resulted in at least moderate disability in all subjects during prosthetic use. Morgenroth and colleagues went on to speculate that abnormal lumbar spinal kinematics could be a contributing factor for back pain in transfemoral amputees.⁶

Though not specifically causative, the study found that those transfemoral amputees with increased lumbar spine transverse plane motion experienced significantly more back pain than did similar amputees without lumbar spine transverse plane motion. An abnormal gait would promote more transverse plane motion than that seen in a normal gait. Normalizing prosthetic gait to best simulate the patient’s preamputation biomechanical baseline could reduce transverse lumbar spine motion, reduce back and other mechanical pain, and ultimately, reduce overall disability.

Similarly, the patient in this study also reported increased ease with hills and stairs. Many studies exist that attest to the advantages of microprocessor knees in providing improvements such as decreased stumbles, increased ability to multitask, increased satisfaction with the prosthesis, and improved stair and stance functions, such as with the Genium (Ottobock).^7,8 Whether the combination of a microprocessor knee with a powered ankle/foot would further improve these aspects is yet to be objectively investigated. The report of this study patient who used the combination suggests these types of advantages but certainly as a single case report does not provide definitive answers.

The patient achieved the highest possible score on the LCI before using the X2/T2 combination and thus demonstrated a ceiling effect that has been discussed in several studies.⁹ Furthermore, Larsson and colleagues noted that because of the ceiling effect, the LCI was more useful for amputees of low to moderate activity levels.¹⁰ The TAPES, however, showed an improvement in before and after measurements, and assessment with it was not hindered by a ceiling effect.

Conclusion

The patient in this case report noted substantial subjective functional improvements when using the X2 compared with prior mechanical prosthetic knees paired with the T2 foot/ankle. The functional gains were further verified by significant improvement in the TAPES Index score, a validated measure of prosthetic integration. Specific subjective advantages included energy savings, almost complete resolution of back pain, and improved facility with hills, stairs, and crawl spaces. No perceived disadvantages were reported. 

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

The most common microbes causing postoperative wound infection are Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, Staphylococcus epidermidis, and Enterococcus faecalis.1 Pasteurella multocida (P multocida) is an uncommon organism causing surgical-site infections.2 This article reports a case of a patient who developed a postoperative wound infection due to P multocida complicated by a bloodstream infection.

CASE REPORT

A 54-year-old man presented with pain and discharge surrounding his recent surgical site and a 1-day history of fever and chills. Four weeks prior to the presentation, he underwent nodal dissection and reconstructive surgery with a muscle flap on his right leg for a superficial spreading melanoma. The patient noticed wound dehiscence the previous 2 to 3 days and a white-yellowish discharge from the wound.

Six years previously, the patient was diagnosed with a superficial spreading melanoma on his right thigh. At that time, he underwent wide excision for the localized disease. The patient was doing well until 3 months prior to the admission when he noticed a nodule in his right groin area, which had enlarged to a lump. Fine needle aspiration of the lump and immunohistochemistry showed the presence of malignant cells, positive for S100 and MART-1.These markers are useful for a diagnosis of metastatic melanoma.^3,4

A positron emission tomography (PET) scan revealed localized disease with increased uptake only in the groin region. Superficial nodal dissection and sartorius muscle flap placement were performed, and the patient was discharged to his home with drains in place.

Two weeks after his node dissection, the patient was seen in the surgery clinic, and his drains were removed. Three days after his clinic visit, he noticed erythema around the suture line and discharge from the wound, followed by increasing pain, high fever, and chills. The patient reported no history of nausea, vomiting, diarrhea, painful urination, cough, shortness of breath, chest pain, or abdominal pain. His medical history was also significant for hypertension and multiple lipomas on his back and extremities, which were removed surgically 15 years previously. He had been a chronic smoker and occasionally used alcohol but did not use illicit drugs. The patient was unmarried and lived alone with his domestic cats.

The patient was febrile on admission (38.9°C); heart rate 110 bpm. His blood pressure and respiratory rate were normal. An examination of the right groin revealed an incision mark of about 20 cm in length from his right anterior superior iliac spine travelling inferomedially into his groin with breakdown at the infero-medial aspect, extending one-half to one-third the wound distance. The wound had clean borders with minimal amount of purulent drainage. Induration was noted, extending down the thigh to above the knee level. Sutures remained in place along the superior aspect of the incision site. No other abnormalities were found.

Laboratory studies revealed leucocytosis (white blood cell count, 12,200/mm³). Blood glucose, blood urea nitrogen, serum creatinine, and electrolytes were within normal range. Computerized tomography (CT) scans of the pelvis and thighs with IV contrast revealed normal right common femoral artery and vein with postsurgical changes and thickening of the skin and small air lucencies in the right groin. The patient was started on empiric treatment with IV vancomycin and piperacillin/tazobactam.

Due to initial worsening of erythema, piperacillin/tazobactam were stopped and ertapenem was initiated. Blood cultures sets and the culture of the surgical wound were positive for P multocida. Vancomycin was discontinued, and the patient rapidly improved. He was discharged on oral moxifloxacin for 2 weeks. After 10 months of follow-up, the patient was doing well and the surgical site wound had healed.

DISCUSSION

Pasteurella multocida is a small, Gram-negative, nonmotile, non–spore-forming coccobacillus with bipolar staining features.⁵ It often exists as a commensal in the upper respiratory tracts of domestic pet species, especially cats and dogs, and may cause hemorrhagic septicemia in cattle, fowl cholera in chickens, and atrophic rhinitis in pigs.^6,7Pasteurella multocida infection in humans is often associated with an animal bite, scratch, or lick, but infection without epidemiologic evidence of animal contact may occur.⁸

The majority of animal bites involve dogs (85%-90%), followed by cats (5%-10%). Infectious complications occur in about 15% to 20% of dog-related bites and > 50% of cat-related bites. The sharp and long teeth of cats can easily penetrate human skin, create a deep puncture wound, and even inoculate the periosteal component of bones. Cat-related wounds more commonly progress to serious and deeper tissue infections, including osteomyelitis and meningitis.⁹ Human-to-human transmission is rare, but there are reported cases of respiratory transmission and maternal-to-fetal transmission.^10,11

The most common presentation of P multocida infection in humans is soft-tissue infection, appearing as purulent wounds (48%), cellulitis (36%), or abscesses (16%).¹² The infection is characterized by the development of soft-tissue inflammation at the site of contact, which may progress to diffuse, localized cellulitis. High leukocyte and neutrophil counts are typically observed.¹³ Complications of localized infection include rapidly progressive cellulitis, abscesses, tenosynovitis, osteomyelitis, and septic arthritis. The localized infection can also lead to septicemia, which carries a high mortality rate (31%).^12,14

Pasteurella multocida septicemia commonly occurs in patients with an immune-compromised status, but septicemia in healthy individuals has also been reported.^15,16 Apart from local skin and soft-tissue infections, P multocida can cause upper respiratory tract infections, lower respiratory tract infections leading to pneumonia, trachea-bronchitis, lung abscess, and empyema, usually in individuals with underlying pulmonary disease. Pasteurella multocida meningitis has been associated with cat licks and bites occurring on the face in both the young and the elderly.^17-19

Gram stains of purulent material or other fluid specimens, including blood, sputum, and cerebrospinal fluid, may show small, Gram-negative, nonmotile, non–spore-forming pleomorphic coccobacilli. Wright, Giemsa, and Wayson stains enhance bipolar staining. The quickest and most accurate method for confirming an active P multocida infection is molecular detection using polymerase chain reaction.²⁰ Evaluations of tenosynovitis, septic arthritis, osteomyelitis, and meningeal enhancement, when appropriate, should be done with CT scans or MRIs.

In < 10% of cases, localized infection by P multocida may lead to bacteremia. The most common predisposing factors associated with bacteremia identified in a review of cases over 20 years at an urban medical center included old age and chronic medical conditions.²¹ Chronic medical conditions involved most commonly were diabetes mellitus, hypertension, and congestive heart failure. Liver dysfunction has also been reported as a significant risk factor in cases of P multocida bacteremia. The most common antibiotics used to treat the patients with bacteremia were ampicillin/sulbactam, cephalosporins, and fluoroquinolones.

A brief review of the literature of 21 cases of bacteremia showed that the most common antibiotics used for treatment were penicillins (ampicillin, amoxicillin, and piperacillin) in 11 of 21 cases. After penicillin, ciprofloxacin was the most commonly used antibiotic to treat bacteremia secondary to P multocida. The review also identified chronic medical conditions, such as diabetes mellitus, hypertension, congestive heart failure, and chronic obstructive pulmonary disease, as the most common risk factors associated in these cases with bacteremia.

In 17 of 20 cases, patients reported being present in an environment with pets, mostly cats and dogs. Twelve out of 17 patients reported having contact with pet cats and dogs, mostly in the form of bites or pets licking their wounds, and 3 of these were postoperative patients with external wounds.^22-24

This patient reported that after discharge from the hospital following surgery for his spreading melanoma, domestic cats at his home repeatedly licked the postoperative wound. This almost certainly was the source of the infection and bacteremia in this patient. These findings stress the importance of educating patients about proper postoperative wound care and precautions needed if there is potential exposure to domestic animals, such as cats and dogs.

CONCLUSIONS

Pasteurella multocida is an unusual cause of postoperative wound infection. The most common method of acquiring a P multocida infection is through contact with pet animals, mostly cats and dogs. Infection can occur not only with animal bites and scratches, but also with licking of open wounds. The rate of infections can be decreased significantly by educating patients about the mode of transmission of infection, its complications, and safety measures needed if they have pets at home. 

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

The most common microbes causing postoperative wound infection are Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, Staphylococcus epidermidis, and Enterococcus faecalis.1 Pasteurella multocida (P multocida) is an uncommon organism causing surgical-site infections.2 This article reports a case of a patient who developed a postoperative wound infection due to P multocida complicated by a bloodstream infection.

CASE REPORT

A 54-year-old man presented with pain and discharge surrounding his recent surgical site and a 1-day history of fever and chills. Four weeks prior to the presentation, he underwent nodal dissection and reconstructive surgery with a muscle flap on his right leg for a superficial spreading melanoma. The patient noticed wound dehiscence the previous 2 to 3 days and a white-yellowish discharge from the wound.

Six years previously, the patient was diagnosed with a superficial spreading melanoma on his right thigh. At that time, he underwent wide excision for the localized disease. The patient was doing well until 3 months prior to the admission when he noticed a nodule in his right groin area, which had enlarged to a lump. Fine needle aspiration of the lump and immunohistochemistry showed the presence of malignant cells, positive for S100 and MART-1.These markers are useful for a diagnosis of metastatic melanoma.^3,4

A positron emission tomography (PET) scan revealed localized disease with increased uptake only in the groin region. Superficial nodal dissection and sartorius muscle flap placement were performed, and the patient was discharged to his home with drains in place.

Two weeks after his node dissection, the patient was seen in the surgery clinic, and his drains were removed. Three days after his clinic visit, he noticed erythema around the suture line and discharge from the wound, followed by increasing pain, high fever, and chills. The patient reported no history of nausea, vomiting, diarrhea, painful urination, cough, shortness of breath, chest pain, or abdominal pain. His medical history was also significant for hypertension and multiple lipomas on his back and extremities, which were removed surgically 15 years previously. He had been a chronic smoker and occasionally used alcohol but did not use illicit drugs. The patient was unmarried and lived alone with his domestic cats.

The patient was febrile on admission (38.9°C); heart rate 110 bpm. His blood pressure and respiratory rate were normal. An examination of the right groin revealed an incision mark of about 20 cm in length from his right anterior superior iliac spine travelling inferomedially into his groin with breakdown at the infero-medial aspect, extending one-half to one-third the wound distance. The wound had clean borders with minimal amount of purulent drainage. Induration was noted, extending down the thigh to above the knee level. Sutures remained in place along the superior aspect of the incision site. No other abnormalities were found.

Laboratory studies revealed leucocytosis (white blood cell count, 12,200/mm³). Blood glucose, blood urea nitrogen, serum creatinine, and electrolytes were within normal range. Computerized tomography (CT) scans of the pelvis and thighs with IV contrast revealed normal right common femoral artery and vein with postsurgical changes and thickening of the skin and small air lucencies in the right groin. The patient was started on empiric treatment with IV vancomycin and piperacillin/tazobactam.

Due to initial worsening of erythema, piperacillin/tazobactam were stopped and ertapenem was initiated. Blood cultures sets and the culture of the surgical wound were positive for P multocida. Vancomycin was discontinued, and the patient rapidly improved. He was discharged on oral moxifloxacin for 2 weeks. After 10 months of follow-up, the patient was doing well and the surgical site wound had healed.

DISCUSSION

Pasteurella multocida is a small, Gram-negative, nonmotile, non–spore-forming coccobacillus with bipolar staining features.⁵ It often exists as a commensal in the upper respiratory tracts of domestic pet species, especially cats and dogs, and may cause hemorrhagic septicemia in cattle, fowl cholera in chickens, and atrophic rhinitis in pigs.^6,7Pasteurella multocida infection in humans is often associated with an animal bite, scratch, or lick, but infection without epidemiologic evidence of animal contact may occur.⁸

The majority of animal bites involve dogs (85%-90%), followed by cats (5%-10%). Infectious complications occur in about 15% to 20% of dog-related bites and > 50% of cat-related bites. The sharp and long teeth of cats can easily penetrate human skin, create a deep puncture wound, and even inoculate the periosteal component of bones. Cat-related wounds more commonly progress to serious and deeper tissue infections, including osteomyelitis and meningitis.⁹ Human-to-human transmission is rare, but there are reported cases of respiratory transmission and maternal-to-fetal transmission.^10,11

The most common presentation of P multocida infection in humans is soft-tissue infection, appearing as purulent wounds (48%), cellulitis (36%), or abscesses (16%).¹² The infection is characterized by the development of soft-tissue inflammation at the site of contact, which may progress to diffuse, localized cellulitis. High leukocyte and neutrophil counts are typically observed.¹³ Complications of localized infection include rapidly progressive cellulitis, abscesses, tenosynovitis, osteomyelitis, and septic arthritis. The localized infection can also lead to septicemia, which carries a high mortality rate (31%).^12,14

Pasteurella multocida septicemia commonly occurs in patients with an immune-compromised status, but septicemia in healthy individuals has also been reported.^15,16 Apart from local skin and soft-tissue infections, P multocida can cause upper respiratory tract infections, lower respiratory tract infections leading to pneumonia, trachea-bronchitis, lung abscess, and empyema, usually in individuals with underlying pulmonary disease. Pasteurella multocida meningitis has been associated with cat licks and bites occurring on the face in both the young and the elderly.^17-19

Gram stains of purulent material or other fluid specimens, including blood, sputum, and cerebrospinal fluid, may show small, Gram-negative, nonmotile, non–spore-forming pleomorphic coccobacilli. Wright, Giemsa, and Wayson stains enhance bipolar staining. The quickest and most accurate method for confirming an active P multocida infection is molecular detection using polymerase chain reaction.²⁰ Evaluations of tenosynovitis, septic arthritis, osteomyelitis, and meningeal enhancement, when appropriate, should be done with CT scans or MRIs.

In < 10% of cases, localized infection by P multocida may lead to bacteremia. The most common predisposing factors associated with bacteremia identified in a review of cases over 20 years at an urban medical center included old age and chronic medical conditions.²¹ Chronic medical conditions involved most commonly were diabetes mellitus, hypertension, and congestive heart failure. Liver dysfunction has also been reported as a significant risk factor in cases of P multocida bacteremia. The most common antibiotics used to treat the patients with bacteremia were ampicillin/sulbactam, cephalosporins, and fluoroquinolones.

A brief review of the literature of 21 cases of bacteremia showed that the most common antibiotics used for treatment were penicillins (ampicillin, amoxicillin, and piperacillin) in 11 of 21 cases. After penicillin, ciprofloxacin was the most commonly used antibiotic to treat bacteremia secondary to P multocida. The review also identified chronic medical conditions, such as diabetes mellitus, hypertension, congestive heart failure, and chronic obstructive pulmonary disease, as the most common risk factors associated in these cases with bacteremia.

In 17 of 20 cases, patients reported being present in an environment with pets, mostly cats and dogs. Twelve out of 17 patients reported having contact with pet cats and dogs, mostly in the form of bites or pets licking their wounds, and 3 of these were postoperative patients with external wounds.^22-24

This patient reported that after discharge from the hospital following surgery for his spreading melanoma, domestic cats at his home repeatedly licked the postoperative wound. This almost certainly was the source of the infection and bacteremia in this patient. These findings stress the importance of educating patients about proper postoperative wound care and precautions needed if there is potential exposure to domestic animals, such as cats and dogs.

CONCLUSIONS

Pasteurella multocida is an unusual cause of postoperative wound infection. The most common method of acquiring a P multocida infection is through contact with pet animals, mostly cats and dogs. Infection can occur not only with animal bites and scratches, but also with licking of open wounds. The rate of infections can be decreased significantly by educating patients about the mode of transmission of infection, its complications, and safety measures needed if they have pets at home. 

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

The most common microbes causing postoperative wound infection are Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, Staphylococcus epidermidis, and Enterococcus faecalis.1 Pasteurella multocida (P multocida) is an uncommon organism causing surgical-site infections.2 This article reports a case of a patient who developed a postoperative wound infection due to P multocida complicated by a bloodstream infection.

CASE REPORT

A 54-year-old man presented with pain and discharge surrounding his recent surgical site and a 1-day history of fever and chills. Four weeks prior to the presentation, he underwent nodal dissection and reconstructive surgery with a muscle flap on his right leg for a superficial spreading melanoma. The patient noticed wound dehiscence the previous 2 to 3 days and a white-yellowish discharge from the wound.

Six years previously, the patient was diagnosed with a superficial spreading melanoma on his right thigh. At that time, he underwent wide excision for the localized disease. The patient was doing well until 3 months prior to the admission when he noticed a nodule in his right groin area, which had enlarged to a lump. Fine needle aspiration of the lump and immunohistochemistry showed the presence of malignant cells, positive for S100 and MART-1.These markers are useful for a diagnosis of metastatic melanoma.^3,4

A positron emission tomography (PET) scan revealed localized disease with increased uptake only in the groin region. Superficial nodal dissection and sartorius muscle flap placement were performed, and the patient was discharged to his home with drains in place.

Two weeks after his node dissection, the patient was seen in the surgery clinic, and his drains were removed. Three days after his clinic visit, he noticed erythema around the suture line and discharge from the wound, followed by increasing pain, high fever, and chills. The patient reported no history of nausea, vomiting, diarrhea, painful urination, cough, shortness of breath, chest pain, or abdominal pain. His medical history was also significant for hypertension and multiple lipomas on his back and extremities, which were removed surgically 15 years previously. He had been a chronic smoker and occasionally used alcohol but did not use illicit drugs. The patient was unmarried and lived alone with his domestic cats.

The patient was febrile on admission (38.9°C); heart rate 110 bpm. His blood pressure and respiratory rate were normal. An examination of the right groin revealed an incision mark of about 20 cm in length from his right anterior superior iliac spine travelling inferomedially into his groin with breakdown at the infero-medial aspect, extending one-half to one-third the wound distance. The wound had clean borders with minimal amount of purulent drainage. Induration was noted, extending down the thigh to above the knee level. Sutures remained in place along the superior aspect of the incision site. No other abnormalities were found.

Laboratory studies revealed leucocytosis (white blood cell count, 12,200/mm³). Blood glucose, blood urea nitrogen, serum creatinine, and electrolytes were within normal range. Computerized tomography (CT) scans of the pelvis and thighs with IV contrast revealed normal right common femoral artery and vein with postsurgical changes and thickening of the skin and small air lucencies in the right groin. The patient was started on empiric treatment with IV vancomycin and piperacillin/tazobactam.

Due to initial worsening of erythema, piperacillin/tazobactam were stopped and ertapenem was initiated. Blood cultures sets and the culture of the surgical wound were positive for P multocida. Vancomycin was discontinued, and the patient rapidly improved. He was discharged on oral moxifloxacin for 2 weeks. After 10 months of follow-up, the patient was doing well and the surgical site wound had healed.

DISCUSSION

Pasteurella multocida is a small, Gram-negative, nonmotile, non–spore-forming coccobacillus with bipolar staining features.⁵ It often exists as a commensal in the upper respiratory tracts of domestic pet species, especially cats and dogs, and may cause hemorrhagic septicemia in cattle, fowl cholera in chickens, and atrophic rhinitis in pigs.^6,7Pasteurella multocida infection in humans is often associated with an animal bite, scratch, or lick, but infection without epidemiologic evidence of animal contact may occur.⁸

The majority of animal bites involve dogs (85%-90%), followed by cats (5%-10%). Infectious complications occur in about 15% to 20% of dog-related bites and > 50% of cat-related bites. The sharp and long teeth of cats can easily penetrate human skin, create a deep puncture wound, and even inoculate the periosteal component of bones. Cat-related wounds more commonly progress to serious and deeper tissue infections, including osteomyelitis and meningitis.⁹ Human-to-human transmission is rare, but there are reported cases of respiratory transmission and maternal-to-fetal transmission.^10,11

The most common presentation of P multocida infection in humans is soft-tissue infection, appearing as purulent wounds (48%), cellulitis (36%), or abscesses (16%).¹² The infection is characterized by the development of soft-tissue inflammation at the site of contact, which may progress to diffuse, localized cellulitis. High leukocyte and neutrophil counts are typically observed.¹³ Complications of localized infection include rapidly progressive cellulitis, abscesses, tenosynovitis, osteomyelitis, and septic arthritis. The localized infection can also lead to septicemia, which carries a high mortality rate (31%).^12,14

Pasteurella multocida septicemia commonly occurs in patients with an immune-compromised status, but septicemia in healthy individuals has also been reported.^15,16 Apart from local skin and soft-tissue infections, P multocida can cause upper respiratory tract infections, lower respiratory tract infections leading to pneumonia, trachea-bronchitis, lung abscess, and empyema, usually in individuals with underlying pulmonary disease. Pasteurella multocida meningitis has been associated with cat licks and bites occurring on the face in both the young and the elderly.^17-19

Gram stains of purulent material or other fluid specimens, including blood, sputum, and cerebrospinal fluid, may show small, Gram-negative, nonmotile, non–spore-forming pleomorphic coccobacilli. Wright, Giemsa, and Wayson stains enhance bipolar staining. The quickest and most accurate method for confirming an active P multocida infection is molecular detection using polymerase chain reaction.²⁰ Evaluations of tenosynovitis, septic arthritis, osteomyelitis, and meningeal enhancement, when appropriate, should be done with CT scans or MRIs.

In < 10% of cases, localized infection by P multocida may lead to bacteremia. The most common predisposing factors associated with bacteremia identified in a review of cases over 20 years at an urban medical center included old age and chronic medical conditions.²¹ Chronic medical conditions involved most commonly were diabetes mellitus, hypertension, and congestive heart failure. Liver dysfunction has also been reported as a significant risk factor in cases of P multocida bacteremia. The most common antibiotics used to treat the patients with bacteremia were ampicillin/sulbactam, cephalosporins, and fluoroquinolones.

A brief review of the literature of 21 cases of bacteremia showed that the most common antibiotics used for treatment were penicillins (ampicillin, amoxicillin, and piperacillin) in 11 of 21 cases. After penicillin, ciprofloxacin was the most commonly used antibiotic to treat bacteremia secondary to P multocida. The review also identified chronic medical conditions, such as diabetes mellitus, hypertension, congestive heart failure, and chronic obstructive pulmonary disease, as the most common risk factors associated in these cases with bacteremia.

In 17 of 20 cases, patients reported being present in an environment with pets, mostly cats and dogs. Twelve out of 17 patients reported having contact with pet cats and dogs, mostly in the form of bites or pets licking their wounds, and 3 of these were postoperative patients with external wounds.^22-24

This patient reported that after discharge from the hospital following surgery for his spreading melanoma, domestic cats at his home repeatedly licked the postoperative wound. This almost certainly was the source of the infection and bacteremia in this patient. These findings stress the importance of educating patients about proper postoperative wound care and precautions needed if there is potential exposure to domestic animals, such as cats and dogs.

CONCLUSIONS

Pasteurella multocida is an unusual cause of postoperative wound infection. The most common method of acquiring a P multocida infection is through contact with pet animals, mostly cats and dogs. Infection can occur not only with animal bites and scratches, but also with licking of open wounds. The rate of infections can be decreased significantly by educating patients about the mode of transmission of infection, its complications, and safety measures needed if they have pets at home. 

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.