Curtis Olson, MD

The new joint position statement from the American Diabetes Association and the European Association for the Study of Diabetes emphasizes an individualized, patient-centered approach that is less prescriptive and algorithmic than previous guidelines.

Goals

The guidelines recommend lowering HbA_1c to less than 7% in most patients. More important, however, is establishing an individual goal for each patient, taking into account the benefit and risk of tight glycemic control, and the patient’s ability to comply with recommended regimens.

For example, a patient who is newly diagnosed with type 2 diabetes, is highly motivated, is at low risk of hypoglycemic complications, and has a long life expectancy with no significant comorbidities may be encouraged to achieve an HbA_1c of less than 6.5%.

Conversely, a patient with longstanding disease, at high risk of hypoglycemic complications, with a short life span, and having established vascular complications should have a goal of 7.5%-8.0%, or even higher. The guidelines discourage using the percentage of patients with an HbA_1c of less than 7% as a quality indicator, because it undermines individualization of treatment goals.

Therapeutic Options

Lifestyle intervention forms the foundation of treatment for all patients with type 2 diabetes. Weight loss, or at least weight maintenance, should be encouraged through dietary, medical, or surgical means. Weight loss of 5%-10% can contribute to glycemic control. The physical activity goal is at least 150 minutes of moderate activity per week, or as much as possible – noting that any increase in activity level is beneficial.

Metformin is recommended as the usual first-line medical therapy for diabetes. It decreases hepatic gluconeogenesis, is generally weight neutral, and has a low risk of hypoglycemia. Sulfonylureas stimulate insulin release from the pancreas and carry a risk of weight gain and hypoglycemia. The meglinitides, also secretagogues, have less risk of hypoglycemia. Thiazolidinediones (TZDs) improve insulin sensitivity in muscle tissue and decrease hepatic gluconeogenesis. Side effects include weight gain, edema and heart failure, increased risk of bone fractures, and a possible increased risk of bladder cancer.

The injectable GLP-1 (glucagon-like peptide 1) receptor agonists increase insulin and decrease glucagon secretion, slow gastric emptying, and decrease appetite. They stimulate weight loss but also can cause nausea and vomiting. The DPP-4 (dipeptidyl peptidase 4) inhibitors decrease breakdown of GLP-1 and similarly lead to increased insulin and decreased glucagon secretion. Insulin, having many forms and regimens, becomes an option for patients with more severe disease; but it carries the risk of hypoglycemia and weight gain.

Implementation

Newly diagnosed patients near their HbA_1c goal who are highly motivated can be offered a 3- to 6-month period of lifestyle-only intervention before starting pharmacologic therapy. Lifestyle intervention will not be enough for most patients to achieve target HbA_1c goals, however, and oral, noninsulin injectable therapies and insulin therapies should then be added.

Initial drug monotherapy with metformin is appropriate for most patients. If target goals have not been reached after 3 months of titrating up from a low dose, then a second agent from a different class should be added, followed after another 3 months by a third agent.

Because of a relative lack of comparative long-term data about other oral and noninsulin injectable therapies, the next agent to be added will be determined based on the characteristics of a given class of medication, along with how those characteristics match a patient’s needs.

To help choose an individualized regimen, the guidelines outline five factors for each of the other drug categories: efficacy, risk of hypoglycemia, weight change, major side effects, and cost.

For example, sulfonylureas have high efficacy, a moderate risk of hypoglycemia, and low cost, while the TZDs have high efficacy, side effects of heart failure and edema, and high cost.

Adding a second agent will decrease HbA_1c an average of 1%. If a medication has been added as a second- or third-line agent, and there has been no noticeable decrease in glycemic measures after 3 months, the patient should be considered a nonresponder. A drug from a different class should then be substituted.

If a patient has a high initial HbA_1c (9.0%-10%), dual drug therapy may be appropriate from the start. If hyperglycemia is severe (fasting plasma glucose is greater than 300 mg/dL or HbA_1c is greater than 10%), strong consideration should be given to starting therapy with insulin.

Transitioning to insulin often will be required, as progressive beta-cell destruction is a hallmark of the disease. The simplest approach is adding once-daily basal insulin, with patients taught to uptitrate their own insulin in a small incremental manner once or twice weekly based on the results of their daily fasting blood glucose values.

When fasting glucose levels are controlled but HbA_1c is not at goal, it is likely due to postprandial glucose excursions, and prandial insulin is indicated. Short-acting bolus insulin may be started before the evening meal, which usually has the largest carbohydrate content. Then, if needed, add it to other meals. Another option is twice-daily premixed insulin, given before morning and evening meals. This regimen lowers HbA_1c more, but it also has a greater risk of hypoglycemia and weight gain.

Metformin may be continued with basal insulin, and may allow less weight gain than insulin alone. Continuing secretagogues initially when insulin is started minimizes initial deterioration of control. But it offers little advantage after that, and thus should then be stopped. It also needs to be discontinued if prandial insulin therapy is added. TZDs should generally be stopped when starting insulin, while the GLP-1 receptor agonists may be helpful with insulin.

Summary

The clear message of these guidelines is to individualize therapy to maximize adherence and benefit. Establish a target HbA_1c goal and start with lifestyle modification with or without metformin. If HbA_1c is not at goal after 3 months, add a second and then possibly a third oral or noninsulin injectable agent from different classes based on efficacy, side effects, weight change, and cost.

Reference

• Management of Hyperglycemia in Type 2 Diabetes: A Patient-Centered Approach. Diabetes Care 2012;35:1364-79.

To listen to an interview with Dr. Skolnik and the chair of the guidelines committee, Dr. Silvio Inzucchi, going over the details of the guidelines, click here, or search for "Diabetes Core Update" in iTunes for the special June 1, 2012, issue.

Dr. Olson is a third-year resident and chief resident in the family medicine residency program at Abington (Pa.) Memorial Hospital. Dr. Skolnik is an associate director of the family medicine residency program at Abington Memorial Hospital.

The new joint position statement from the American Diabetes Association and the European Association for the Study of Diabetes emphasizes an individualized, patient-centered approach that is less prescriptive and algorithmic than previous guidelines.

Goals

The guidelines recommend lowering HbA_1c to less than 7% in most patients. More important, however, is establishing an individual goal for each patient, taking into account the benefit and risk of tight glycemic control, and the patient’s ability to comply with recommended regimens.

For example, a patient who is newly diagnosed with type 2 diabetes, is highly motivated, is at low risk of hypoglycemic complications, and has a long life expectancy with no significant comorbidities may be encouraged to achieve an HbA_1c of less than 6.5%.

Conversely, a patient with longstanding disease, at high risk of hypoglycemic complications, with a short life span, and having established vascular complications should have a goal of 7.5%-8.0%, or even higher. The guidelines discourage using the percentage of patients with an HbA_1c of less than 7% as a quality indicator, because it undermines individualization of treatment goals.

Therapeutic Options

Lifestyle intervention forms the foundation of treatment for all patients with type 2 diabetes. Weight loss, or at least weight maintenance, should be encouraged through dietary, medical, or surgical means. Weight loss of 5%-10% can contribute to glycemic control. The physical activity goal is at least 150 minutes of moderate activity per week, or as much as possible – noting that any increase in activity level is beneficial.

Metformin is recommended as the usual first-line medical therapy for diabetes. It decreases hepatic gluconeogenesis, is generally weight neutral, and has a low risk of hypoglycemia. Sulfonylureas stimulate insulin release from the pancreas and carry a risk of weight gain and hypoglycemia. The meglinitides, also secretagogues, have less risk of hypoglycemia. Thiazolidinediones (TZDs) improve insulin sensitivity in muscle tissue and decrease hepatic gluconeogenesis. Side effects include weight gain, edema and heart failure, increased risk of bone fractures, and a possible increased risk of bladder cancer.

The injectable GLP-1 (glucagon-like peptide 1) receptor agonists increase insulin and decrease glucagon secretion, slow gastric emptying, and decrease appetite. They stimulate weight loss but also can cause nausea and vomiting. The DPP-4 (dipeptidyl peptidase 4) inhibitors decrease breakdown of GLP-1 and similarly lead to increased insulin and decreased glucagon secretion. Insulin, having many forms and regimens, becomes an option for patients with more severe disease; but it carries the risk of hypoglycemia and weight gain.

Implementation

Newly diagnosed patients near their HbA_1c goal who are highly motivated can be offered a 3- to 6-month period of lifestyle-only intervention before starting pharmacologic therapy. Lifestyle intervention will not be enough for most patients to achieve target HbA_1c goals, however, and oral, noninsulin injectable therapies and insulin therapies should then be added.

Initial drug monotherapy with metformin is appropriate for most patients. If target goals have not been reached after 3 months of titrating up from a low dose, then a second agent from a different class should be added, followed after another 3 months by a third agent.

Because of a relative lack of comparative long-term data about other oral and noninsulin injectable therapies, the next agent to be added will be determined based on the characteristics of a given class of medication, along with how those characteristics match a patient’s needs.

To help choose an individualized regimen, the guidelines outline five factors for each of the other drug categories: efficacy, risk of hypoglycemia, weight change, major side effects, and cost.

For example, sulfonylureas have high efficacy, a moderate risk of hypoglycemia, and low cost, while the TZDs have high efficacy, side effects of heart failure and edema, and high cost.

Adding a second agent will decrease HbA_1c an average of 1%. If a medication has been added as a second- or third-line agent, and there has been no noticeable decrease in glycemic measures after 3 months, the patient should be considered a nonresponder. A drug from a different class should then be substituted.

If a patient has a high initial HbA_1c (9.0%-10%), dual drug therapy may be appropriate from the start. If hyperglycemia is severe (fasting plasma glucose is greater than 300 mg/dL or HbA_1c is greater than 10%), strong consideration should be given to starting therapy with insulin.

Transitioning to insulin often will be required, as progressive beta-cell destruction is a hallmark of the disease. The simplest approach is adding once-daily basal insulin, with patients taught to uptitrate their own insulin in a small incremental manner once or twice weekly based on the results of their daily fasting blood glucose values.

When fasting glucose levels are controlled but HbA_1c is not at goal, it is likely due to postprandial glucose excursions, and prandial insulin is indicated. Short-acting bolus insulin may be started before the evening meal, which usually has the largest carbohydrate content. Then, if needed, add it to other meals. Another option is twice-daily premixed insulin, given before morning and evening meals. This regimen lowers HbA_1c more, but it also has a greater risk of hypoglycemia and weight gain.

Metformin may be continued with basal insulin, and may allow less weight gain than insulin alone. Continuing secretagogues initially when insulin is started minimizes initial deterioration of control. But it offers little advantage after that, and thus should then be stopped. It also needs to be discontinued if prandial insulin therapy is added. TZDs should generally be stopped when starting insulin, while the GLP-1 receptor agonists may be helpful with insulin.

Summary

The clear message of these guidelines is to individualize therapy to maximize adherence and benefit. Establish a target HbA_1c goal and start with lifestyle modification with or without metformin. If HbA_1c is not at goal after 3 months, add a second and then possibly a third oral or noninsulin injectable agent from different classes based on efficacy, side effects, weight change, and cost.

Reference

• Management of Hyperglycemia in Type 2 Diabetes: A Patient-Centered Approach. Diabetes Care 2012;35:1364-79.

To listen to an interview with Dr. Skolnik and the chair of the guidelines committee, Dr. Silvio Inzucchi, going over the details of the guidelines, click here, or search for "Diabetes Core Update" in iTunes for the special June 1, 2012, issue.

Dr. Olson is a third-year resident and chief resident in the family medicine residency program at Abington (Pa.) Memorial Hospital. Dr. Skolnik is an associate director of the family medicine residency program at Abington Memorial Hospital.

The new joint position statement from the American Diabetes Association and the European Association for the Study of Diabetes emphasizes an individualized, patient-centered approach that is less prescriptive and algorithmic than previous guidelines.

Goals

The guidelines recommend lowering HbA_1c to less than 7% in most patients. More important, however, is establishing an individual goal for each patient, taking into account the benefit and risk of tight glycemic control, and the patient’s ability to comply with recommended regimens.

For example, a patient who is newly diagnosed with type 2 diabetes, is highly motivated, is at low risk of hypoglycemic complications, and has a long life expectancy with no significant comorbidities may be encouraged to achieve an HbA_1c of less than 6.5%.

Conversely, a patient with longstanding disease, at high risk of hypoglycemic complications, with a short life span, and having established vascular complications should have a goal of 7.5%-8.0%, or even higher. The guidelines discourage using the percentage of patients with an HbA_1c of less than 7% as a quality indicator, because it undermines individualization of treatment goals.

Therapeutic Options

Lifestyle intervention forms the foundation of treatment for all patients with type 2 diabetes. Weight loss, or at least weight maintenance, should be encouraged through dietary, medical, or surgical means. Weight loss of 5%-10% can contribute to glycemic control. The physical activity goal is at least 150 minutes of moderate activity per week, or as much as possible – noting that any increase in activity level is beneficial.

Metformin is recommended as the usual first-line medical therapy for diabetes. It decreases hepatic gluconeogenesis, is generally weight neutral, and has a low risk of hypoglycemia. Sulfonylureas stimulate insulin release from the pancreas and carry a risk of weight gain and hypoglycemia. The meglinitides, also secretagogues, have less risk of hypoglycemia. Thiazolidinediones (TZDs) improve insulin sensitivity in muscle tissue and decrease hepatic gluconeogenesis. Side effects include weight gain, edema and heart failure, increased risk of bone fractures, and a possible increased risk of bladder cancer.

The injectable GLP-1 (glucagon-like peptide 1) receptor agonists increase insulin and decrease glucagon secretion, slow gastric emptying, and decrease appetite. They stimulate weight loss but also can cause nausea and vomiting. The DPP-4 (dipeptidyl peptidase 4) inhibitors decrease breakdown of GLP-1 and similarly lead to increased insulin and decreased glucagon secretion. Insulin, having many forms and regimens, becomes an option for patients with more severe disease; but it carries the risk of hypoglycemia and weight gain.

Implementation

Newly diagnosed patients near their HbA_1c goal who are highly motivated can be offered a 3- to 6-month period of lifestyle-only intervention before starting pharmacologic therapy. Lifestyle intervention will not be enough for most patients to achieve target HbA_1c goals, however, and oral, noninsulin injectable therapies and insulin therapies should then be added.

Initial drug monotherapy with metformin is appropriate for most patients. If target goals have not been reached after 3 months of titrating up from a low dose, then a second agent from a different class should be added, followed after another 3 months by a third agent.

Because of a relative lack of comparative long-term data about other oral and noninsulin injectable therapies, the next agent to be added will be determined based on the characteristics of a given class of medication, along with how those characteristics match a patient’s needs.

To help choose an individualized regimen, the guidelines outline five factors for each of the other drug categories: efficacy, risk of hypoglycemia, weight change, major side effects, and cost.

For example, sulfonylureas have high efficacy, a moderate risk of hypoglycemia, and low cost, while the TZDs have high efficacy, side effects of heart failure and edema, and high cost.

Adding a second agent will decrease HbA_1c an average of 1%. If a medication has been added as a second- or third-line agent, and there has been no noticeable decrease in glycemic measures after 3 months, the patient should be considered a nonresponder. A drug from a different class should then be substituted.

If a patient has a high initial HbA_1c (9.0%-10%), dual drug therapy may be appropriate from the start. If hyperglycemia is severe (fasting plasma glucose is greater than 300 mg/dL or HbA_1c is greater than 10%), strong consideration should be given to starting therapy with insulin.

Transitioning to insulin often will be required, as progressive beta-cell destruction is a hallmark of the disease. The simplest approach is adding once-daily basal insulin, with patients taught to uptitrate their own insulin in a small incremental manner once or twice weekly based on the results of their daily fasting blood glucose values.

When fasting glucose levels are controlled but HbA_1c is not at goal, it is likely due to postprandial glucose excursions, and prandial insulin is indicated. Short-acting bolus insulin may be started before the evening meal, which usually has the largest carbohydrate content. Then, if needed, add it to other meals. Another option is twice-daily premixed insulin, given before morning and evening meals. This regimen lowers HbA_1c more, but it also has a greater risk of hypoglycemia and weight gain.

Metformin may be continued with basal insulin, and may allow less weight gain than insulin alone. Continuing secretagogues initially when insulin is started minimizes initial deterioration of control. But it offers little advantage after that, and thus should then be stopped. It also needs to be discontinued if prandial insulin therapy is added. TZDs should generally be stopped when starting insulin, while the GLP-1 receptor agonists may be helpful with insulin.

Summary

The clear message of these guidelines is to individualize therapy to maximize adherence and benefit. Establish a target HbA_1c goal and start with lifestyle modification with or without metformin. If HbA_1c is not at goal after 3 months, add a second and then possibly a third oral or noninsulin injectable agent from different classes based on efficacy, side effects, weight change, and cost.

Reference

• Management of Hyperglycemia in Type 2 Diabetes: A Patient-Centered Approach. Diabetes Care 2012;35:1364-79.

To listen to an interview with Dr. Skolnik and the chair of the guidelines committee, Dr. Silvio Inzucchi, going over the details of the guidelines, click here, or search for "Diabetes Core Update" in iTunes for the special June 1, 2012, issue.

Dr. Olson is a third-year resident and chief resident in the family medicine residency program at Abington (Pa.) Memorial Hospital. Dr. Skolnik is an associate director of the family medicine residency program at Abington Memorial Hospital.

The American College of Physicians recently published guidelines on venous thromboembolism prophylaxis in hospitalized medical patients (Ann. Intern. Med. 2011;155:625-32).

Venous thromboembolism (VTE) – including pulmonary embolism (PE) and deep vein thrombosis (DVT) – is a cause of serious morbidity and mortality in hospitalized patients. Between 5% and 10% of all in-hospital deaths are a direct result of PE, and PE accounts for 200,000-300,000 hospitalizations per year.

Two main approaches to VTE prophylaxis exist. The first is heparin or related agents, including low-dose unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), or fondaparinux, all given subcutaneously. The second option is mechanical leg compression, which primarily includes graduated compression stockings and intermittent pneumatic compression.

Studies from 1950 and 2011 were analyzed that compared heparin or related agents vs. no heparin; LMWH vs. UFH; and mechanical prophylaxis vs. no prophylaxis. The primary outcome for the extracted data was total mortality up to 120 days from randomization, whereas secondary outcomes included symptomatic DVT, all PE, fatal PE, all bleedings events, major bleeding events, and – for mechanical prophylaxis – skin damage.

Prophylaxis vs. No Prophylaxis

In nonstroke medical patients (10 trials, n = 20,717), heparin or related agent (UFH, LMWH, or fondaparinux) prophylaxis did not significantly reduce total mortality, compared with no prophylaxis. There was a statistically significant reduced risk of PE and an increased risk of bleeding events with heparin or related agents. There was an increase in major bleeding and a reduction in symptomatic DVT with heparin that was not statistically significant.

Heparin prophylaxis in acute stroke patients (eight trials, n = 15,405) did not show a significant reduction in mortality, PE, or symptomatic DVT, although there was a nonsignificant increased risk for major bleeding events. Most of the evidence for this group came from one large study that also showed a significant increase in 14-day hemorrhagic stroke or serious extracranial hemorrhage, as well as a significant decrease in 14-day recurrent ischemic stroke with heparin prophylaxis.

Analyzing combined data from acute stroke and nonstroke medical patients, heparin prophylaxis decreased mortality to a borderline but not to a statistically significant degree (relative risk, 93%). There was a significant reduction in risk of PE (RR, 0.70) but no significant decrease in symptomatic DVT with heparin prophylaxis. With heparin, there was a significantly increased risk of all bleeding events (RR, 1.28) and major bleeding events (RR, 1.61).

LMWH vs. UFH

In a comparison of LMWH vs. UFH, nonstroke medical patients (nine trials, n = 11,650) and acute stroke patients showed no significant difference in mortality, PE, symptomatic DVT, or major bleeding. Combined data from acute stroke and nonstroke medical patients also showed no significant difference for any outcome.

Mechanical vs. No Prophylaxis

Most data come from one large study of acute stroke patients that compared graduated compression stockings vs. no prophylaxis. There was not a significant difference in mortality, symptomatic DVT, or PE. There was, however, a significant increase in skin damage with compression stockings. There are insufficient data to evaluate benefit and risk of intermittent pneumatic compression in medical patients. Based on data from surgical patients, intermittent pneumatic compression may be used for VTE prophylaxis in medical patients when bleeding risk makes heparin contraindicated for prophylaxis.

Recommendations

• Assess all medical inpatients for the risk for thromboembolism and bleeding prior to initiation of prophylaxis. A thoughtful assessment of benefit to risk should be done for each patient, and then a decision should be made about initiating prophylaxis.

• Pharmacologic prophylaxis with heparin or a related drug for VTE in medical (including stroke) patients is usually beneficial, unless the assessed risk for bleeding outweighs the likely benefits. In medical inpatients, prophylaxis leads to a statistically significant reduction in PE (four events per 1,000 people treated) and an increase in all bleeding events (nine per 1,000 people) and a nonstatistically significant increase in major bleeding events (one per 1,000 people). There is no effect on mortality or DVT. The increased risk for major bleeding is greater in patients with stroke (six per 1,000). For most people, the reduction in PE outweighs the harm of increased bleeding. The evidence in patients with stroke is weaker than for medical inpatients in general.

• Recommend against the use of mechanical prophylaxis with graduated compression stockings for prevention of VTE.

Notably, the guidelines do not support the application in hospitals of quality performance measures that encourage universal VTE prophylaxis, because such measures would encourage increased use of prophylaxis without regard to risk stratification and assessment. The evidence simply does not support routine prophylaxis of all medical inpatients.

Risk factors for thromboembolism include inherited conditions (such as protein C and protein S deficiencies and the factor V Leiden mutation), and acquired-risk factors (such as cancer, immobilization, presence of central venous catheters, heart failure, smoking, chronic kidney disease, and history of thromboembolism). Risk factors for bleeding with anticoagulant therapy include hypertension, cancer, prior stroke, diabetes, a bleeding disorder, and use of NSAIDs, aspirin, or antiplatelet agents. Unfortunately, these conditions exist together in many hospitalized patients, and there are no validated tools to formally assess risk and benefit of VTE prophylaxis.

Bottom Line

Assess risk for venous thromboembolism as well as bleeding risk in all hospitalized medical patients. Heparin or related agent prophylaxis does not decrease total mortality, but it does decrease the risk of PE while increasing risk of bleeding. For most people, the reduction in PE outweighs the harm of increased bleeding. When benefit outweighs risk, use subcutaneous low-dose UFH, LMWH, or fondaparinux for VTE prophylaxis. Graduated compression stockings are not recommended for VTE prophylaxis.

Dr. Olsen is a chief resident in the family medicine residency program at Abington (Pa.) Memorial Hospital. Dr. Skolnik is an associate director of the family medicine residency program at Abington Memorial Hospital.

The American College of Physicians recently published guidelines on venous thromboembolism prophylaxis in hospitalized medical patients (Ann. Intern. Med. 2011;155:625-32).

Venous thromboembolism (VTE) – including pulmonary embolism (PE) and deep vein thrombosis (DVT) – is a cause of serious morbidity and mortality in hospitalized patients. Between 5% and 10% of all in-hospital deaths are a direct result of PE, and PE accounts for 200,000-300,000 hospitalizations per year.

Two main approaches to VTE prophylaxis exist. The first is heparin or related agents, including low-dose unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), or fondaparinux, all given subcutaneously. The second option is mechanical leg compression, which primarily includes graduated compression stockings and intermittent pneumatic compression.

Studies from 1950 and 2011 were analyzed that compared heparin or related agents vs. no heparin; LMWH vs. UFH; and mechanical prophylaxis vs. no prophylaxis. The primary outcome for the extracted data was total mortality up to 120 days from randomization, whereas secondary outcomes included symptomatic DVT, all PE, fatal PE, all bleedings events, major bleeding events, and – for mechanical prophylaxis – skin damage.

Prophylaxis vs. No Prophylaxis

In nonstroke medical patients (10 trials, n = 20,717), heparin or related agent (UFH, LMWH, or fondaparinux) prophylaxis did not significantly reduce total mortality, compared with no prophylaxis. There was a statistically significant reduced risk of PE and an increased risk of bleeding events with heparin or related agents. There was an increase in major bleeding and a reduction in symptomatic DVT with heparin that was not statistically significant.

Heparin prophylaxis in acute stroke patients (eight trials, n = 15,405) did not show a significant reduction in mortality, PE, or symptomatic DVT, although there was a nonsignificant increased risk for major bleeding events. Most of the evidence for this group came from one large study that also showed a significant increase in 14-day hemorrhagic stroke or serious extracranial hemorrhage, as well as a significant decrease in 14-day recurrent ischemic stroke with heparin prophylaxis.

Analyzing combined data from acute stroke and nonstroke medical patients, heparin prophylaxis decreased mortality to a borderline but not to a statistically significant degree (relative risk, 93%). There was a significant reduction in risk of PE (RR, 0.70) but no significant decrease in symptomatic DVT with heparin prophylaxis. With heparin, there was a significantly increased risk of all bleeding events (RR, 1.28) and major bleeding events (RR, 1.61).

LMWH vs. UFH

In a comparison of LMWH vs. UFH, nonstroke medical patients (nine trials, n = 11,650) and acute stroke patients showed no significant difference in mortality, PE, symptomatic DVT, or major bleeding. Combined data from acute stroke and nonstroke medical patients also showed no significant difference for any outcome.

Mechanical vs. No Prophylaxis

Most data come from one large study of acute stroke patients that compared graduated compression stockings vs. no prophylaxis. There was not a significant difference in mortality, symptomatic DVT, or PE. There was, however, a significant increase in skin damage with compression stockings. There are insufficient data to evaluate benefit and risk of intermittent pneumatic compression in medical patients. Based on data from surgical patients, intermittent pneumatic compression may be used for VTE prophylaxis in medical patients when bleeding risk makes heparin contraindicated for prophylaxis.

Recommendations

• Assess all medical inpatients for the risk for thromboembolism and bleeding prior to initiation of prophylaxis. A thoughtful assessment of benefit to risk should be done for each patient, and then a decision should be made about initiating prophylaxis.

• Pharmacologic prophylaxis with heparin or a related drug for VTE in medical (including stroke) patients is usually beneficial, unless the assessed risk for bleeding outweighs the likely benefits. In medical inpatients, prophylaxis leads to a statistically significant reduction in PE (four events per 1,000 people treated) and an increase in all bleeding events (nine per 1,000 people) and a nonstatistically significant increase in major bleeding events (one per 1,000 people). There is no effect on mortality or DVT. The increased risk for major bleeding is greater in patients with stroke (six per 1,000). For most people, the reduction in PE outweighs the harm of increased bleeding. The evidence in patients with stroke is weaker than for medical inpatients in general.

• Recommend against the use of mechanical prophylaxis with graduated compression stockings for prevention of VTE.

Notably, the guidelines do not support the application in hospitals of quality performance measures that encourage universal VTE prophylaxis, because such measures would encourage increased use of prophylaxis without regard to risk stratification and assessment. The evidence simply does not support routine prophylaxis of all medical inpatients.

Risk factors for thromboembolism include inherited conditions (such as protein C and protein S deficiencies and the factor V Leiden mutation), and acquired-risk factors (such as cancer, immobilization, presence of central venous catheters, heart failure, smoking, chronic kidney disease, and history of thromboembolism). Risk factors for bleeding with anticoagulant therapy include hypertension, cancer, prior stroke, diabetes, a bleeding disorder, and use of NSAIDs, aspirin, or antiplatelet agents. Unfortunately, these conditions exist together in many hospitalized patients, and there are no validated tools to formally assess risk and benefit of VTE prophylaxis.

Bottom Line

Assess risk for venous thromboembolism as well as bleeding risk in all hospitalized medical patients. Heparin or related agent prophylaxis does not decrease total mortality, but it does decrease the risk of PE while increasing risk of bleeding. For most people, the reduction in PE outweighs the harm of increased bleeding. When benefit outweighs risk, use subcutaneous low-dose UFH, LMWH, or fondaparinux for VTE prophylaxis. Graduated compression stockings are not recommended for VTE prophylaxis.

Dr. Olsen is a chief resident in the family medicine residency program at Abington (Pa.) Memorial Hospital. Dr. Skolnik is an associate director of the family medicine residency program at Abington Memorial Hospital.

The American College of Physicians recently published guidelines on venous thromboembolism prophylaxis in hospitalized medical patients (Ann. Intern. Med. 2011;155:625-32).

Venous thromboembolism (VTE) – including pulmonary embolism (PE) and deep vein thrombosis (DVT) – is a cause of serious morbidity and mortality in hospitalized patients. Between 5% and 10% of all in-hospital deaths are a direct result of PE, and PE accounts for 200,000-300,000 hospitalizations per year.

Two main approaches to VTE prophylaxis exist. The first is heparin or related agents, including low-dose unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), or fondaparinux, all given subcutaneously. The second option is mechanical leg compression, which primarily includes graduated compression stockings and intermittent pneumatic compression.

Studies from 1950 and 2011 were analyzed that compared heparin or related agents vs. no heparin; LMWH vs. UFH; and mechanical prophylaxis vs. no prophylaxis. The primary outcome for the extracted data was total mortality up to 120 days from randomization, whereas secondary outcomes included symptomatic DVT, all PE, fatal PE, all bleedings events, major bleeding events, and – for mechanical prophylaxis – skin damage.

Prophylaxis vs. No Prophylaxis

In nonstroke medical patients (10 trials, n = 20,717), heparin or related agent (UFH, LMWH, or fondaparinux) prophylaxis did not significantly reduce total mortality, compared with no prophylaxis. There was a statistically significant reduced risk of PE and an increased risk of bleeding events with heparin or related agents. There was an increase in major bleeding and a reduction in symptomatic DVT with heparin that was not statistically significant.

Heparin prophylaxis in acute stroke patients (eight trials, n = 15,405) did not show a significant reduction in mortality, PE, or symptomatic DVT, although there was a nonsignificant increased risk for major bleeding events. Most of the evidence for this group came from one large study that also showed a significant increase in 14-day hemorrhagic stroke or serious extracranial hemorrhage, as well as a significant decrease in 14-day recurrent ischemic stroke with heparin prophylaxis.

Analyzing combined data from acute stroke and nonstroke medical patients, heparin prophylaxis decreased mortality to a borderline but not to a statistically significant degree (relative risk, 93%). There was a significant reduction in risk of PE (RR, 0.70) but no significant decrease in symptomatic DVT with heparin prophylaxis. With heparin, there was a significantly increased risk of all bleeding events (RR, 1.28) and major bleeding events (RR, 1.61).

LMWH vs. UFH

In a comparison of LMWH vs. UFH, nonstroke medical patients (nine trials, n = 11,650) and acute stroke patients showed no significant difference in mortality, PE, symptomatic DVT, or major bleeding. Combined data from acute stroke and nonstroke medical patients also showed no significant difference for any outcome.

Mechanical vs. No Prophylaxis

Most data come from one large study of acute stroke patients that compared graduated compression stockings vs. no prophylaxis. There was not a significant difference in mortality, symptomatic DVT, or PE. There was, however, a significant increase in skin damage with compression stockings. There are insufficient data to evaluate benefit and risk of intermittent pneumatic compression in medical patients. Based on data from surgical patients, intermittent pneumatic compression may be used for VTE prophylaxis in medical patients when bleeding risk makes heparin contraindicated for prophylaxis.

Recommendations

• Assess all medical inpatients for the risk for thromboembolism and bleeding prior to initiation of prophylaxis. A thoughtful assessment of benefit to risk should be done for each patient, and then a decision should be made about initiating prophylaxis.

• Pharmacologic prophylaxis with heparin or a related drug for VTE in medical (including stroke) patients is usually beneficial, unless the assessed risk for bleeding outweighs the likely benefits. In medical inpatients, prophylaxis leads to a statistically significant reduction in PE (four events per 1,000 people treated) and an increase in all bleeding events (nine per 1,000 people) and a nonstatistically significant increase in major bleeding events (one per 1,000 people). There is no effect on mortality or DVT. The increased risk for major bleeding is greater in patients with stroke (six per 1,000). For most people, the reduction in PE outweighs the harm of increased bleeding. The evidence in patients with stroke is weaker than for medical inpatients in general.

• Recommend against the use of mechanical prophylaxis with graduated compression stockings for prevention of VTE.

Notably, the guidelines do not support the application in hospitals of quality performance measures that encourage universal VTE prophylaxis, because such measures would encourage increased use of prophylaxis without regard to risk stratification and assessment. The evidence simply does not support routine prophylaxis of all medical inpatients.

Risk factors for thromboembolism include inherited conditions (such as protein C and protein S deficiencies and the factor V Leiden mutation), and acquired-risk factors (such as cancer, immobilization, presence of central venous catheters, heart failure, smoking, chronic kidney disease, and history of thromboembolism). Risk factors for bleeding with anticoagulant therapy include hypertension, cancer, prior stroke, diabetes, a bleeding disorder, and use of NSAIDs, aspirin, or antiplatelet agents. Unfortunately, these conditions exist together in many hospitalized patients, and there are no validated tools to formally assess risk and benefit of VTE prophylaxis.

Bottom Line

Assess risk for venous thromboembolism as well as bleeding risk in all hospitalized medical patients. Heparin or related agent prophylaxis does not decrease total mortality, but it does decrease the risk of PE while increasing risk of bleeding. For most people, the reduction in PE outweighs the harm of increased bleeding. When benefit outweighs risk, use subcutaneous low-dose UFH, LMWH, or fondaparinux for VTE prophylaxis. Graduated compression stockings are not recommended for VTE prophylaxis.

Dr. Olsen is a chief resident in the family medicine residency program at Abington (Pa.) Memorial Hospital. Dr. Skolnik is an associate director of the family medicine residency program at Abington Memorial Hospital.