The Journal of Family Practice

THE CASE

A 75-year-old woman presented to the primary care clinic with right-side rib pain. The patient said the pain started 1 week earlier, after she ate fried chicken for dinner, and had since been exacerbated by rich meals, lying supine, and taking a deep inspiratory breath. She also said that prior to coming to the clinic that day, the pain had been radiating to her right shoulder.

The patient denied experiencing associated fevers, chills, shortness of breath, chest pain, nausea, vomiting, constipation, diarrhea, or changes in stool color. She had a history of hypertension, for which she was taking lisinopril 20 mg/d, and hypercholesterolemia, for which she was on simvastatin 10 mg/d. She was additionally using timolol ophthalmic solution for her glaucoma.

During the examination, the patient’s vital signs were stable, with a pulse of 80 beats/min, a respiratory rate of 16 breaths/min, and an oxygen saturation of 98% on room air. The patient had no abdominal tenderness upon palpation, and the physical exam revealed no abnormalities. An in-office electrocardiogram was performed, with normal results. Additionally, a comprehensive metabolic panel, lipase test, and d-dimer test were ordered. Lab results showed an isolated elevated d-dimer of 2.66 mcg/mL (normal range, < 0.54 mcg/mL), while all other labs were normal.

THE DIAGNOSIS

Based on the lab results, a stat computed tomography pulmonary angiogram (CTPA) was ordered and showed a right segmental and subsegmental pulmonary embolism (PE; FIGURE 1).

DISCUSSION

PE shares pathophysiologic mechanisms with deep vein thrombosis (DVT), and together these comprise venous thromboembolism (VTE). Risk factors for VTE include hypercoagulable disorders, use of estrogens, active malignancy, and immobilization.¹ Unprovoked VTE occurs in the absence of identifiable risk factors and carries a higher risk of recurrence.^2,3 While PE is classically thought to occur in the setting of a DVT, there is increasing literature describing de novo PE that can occur independent of a DVT.⁴

Common symptoms of PE include tachycardia, tachypnea, and pleuritic chest pain.⁵ Abdominal pain is a rare symptom described in some case reports.^6,7 Thus, a high clinical suspicion is needed for diagnosis of PE.

The Wells criteria is an established model for risk stratifying patients presenting with possible VTE (TABLE).⁸ For patients with low pretest probability, as in this case, a d-dimer is an effective diagnostic work-up, as a negative result will rule out PE. (If the d-dimer had been negative in this case, we would have considered other diagnoses, such as acute coronary syndrome, biliary colic, gastritis, pancreatitis, or musculoskeletal pain.) For high-risk patients, immediate anticoagulation and imaging should be performed, frequently with heparin and CTPA.⁹

Continue to: Length of treatment depends on gender and etiology

The cornerstone treatment for stable patients with VTE is therapeutic anticoagulation. The new oral anticoagulants, which directly inhibit factor Xa or thrombin, have become increasingly popular for management of VTE, in part because they don’t require INR testing and monitoring.²

The duration of anticoagulation, particularly in unprovoked PE, is debatable. As noted earlier, patients with an unprovoked PE are at higher risk of recurrence than those with a reversible cause, so the question becomes whether these patients should have indefinite anticoagulation.^2,3 Studies examining risk stratification of patients with a first, unprovoked VTE have found that men have the highest risk of recurrence, followed by women who were not taking estrogen during the index VTE, and lastly women who were taking estrogen therapy during the index VTE and subsequently discontinued it.^2,3,10

Thus, it is reasonable to give women the option to discontinue anticoagulation in the setting of a negative d-dimer follow-up.³ The 2016 CHEST guidelines recommend extended anticoagulation for a first-time, unprovoked VTE, but acknowledge this recommendation is strongest for men and that women with negative d-dimer assays may consider discontinuation.¹⁰

Our patient was directed to the emergency department for further monitoring following CT confirmation. She was discharged home after being deemed stable and prescribed apixaban 10 mg/d. A venous duplex ultrasound performed 12 days later for knee pain revealed no venous thrombosis. A CT of the abdomen performed 3 months later for other reasons revealed a normal gallbladder with no visible stones.

Apixaban was continued for 3 months and discontinued after discussion of risks and benefits of therapy cessation in the setting of a normal d-dimer and the 2016 CHEST guidelines for anticoagulation in VTE.¹⁰

Continue to: THE TAKEAWAY

THE TAKEAWAY

PE carries a significantly high mortality rate and can manifest with nonspecific and masquerading signs. A high index of suspicion is required to place PE on the differential diagnosis and carry out appropriate testing. Our patient presented with a history consistent with biliary colic but with pleuritic chest pain that warranted consideration of a PE.

CORRESPONDENCE
Alyssa Anderson, MD, 1 Continental Drive, Elizabethtown, PA 17022; [email protected]

THE CASE

A 75-year-old woman presented to the primary care clinic with right-side rib pain. The patient said the pain started 1 week earlier, after she ate fried chicken for dinner, and had since been exacerbated by rich meals, lying supine, and taking a deep inspiratory breath. She also said that prior to coming to the clinic that day, the pain had been radiating to her right shoulder.

The patient denied experiencing associated fevers, chills, shortness of breath, chest pain, nausea, vomiting, constipation, diarrhea, or changes in stool color. She had a history of hypertension, for which she was taking lisinopril 20 mg/d, and hypercholesterolemia, for which she was on simvastatin 10 mg/d. She was additionally using timolol ophthalmic solution for her glaucoma.

During the examination, the patient’s vital signs were stable, with a pulse of 80 beats/min, a respiratory rate of 16 breaths/min, and an oxygen saturation of 98% on room air. The patient had no abdominal tenderness upon palpation, and the physical exam revealed no abnormalities. An in-office electrocardiogram was performed, with normal results. Additionally, a comprehensive metabolic panel, lipase test, and d-dimer test were ordered. Lab results showed an isolated elevated d-dimer of 2.66 mcg/mL (normal range, < 0.54 mcg/mL), while all other labs were normal.

THE DIAGNOSIS

Based on the lab results, a stat computed tomography pulmonary angiogram (CTPA) was ordered and showed a right segmental and subsegmental pulmonary embolism (PE; FIGURE 1).

DISCUSSION

PE shares pathophysiologic mechanisms with deep vein thrombosis (DVT), and together these comprise venous thromboembolism (VTE). Risk factors for VTE include hypercoagulable disorders, use of estrogens, active malignancy, and immobilization.¹ Unprovoked VTE occurs in the absence of identifiable risk factors and carries a higher risk of recurrence.^2,3 While PE is classically thought to occur in the setting of a DVT, there is increasing literature describing de novo PE that can occur independent of a DVT.⁴

Common symptoms of PE include tachycardia, tachypnea, and pleuritic chest pain.⁵ Abdominal pain is a rare symptom described in some case reports.^6,7 Thus, a high clinical suspicion is needed for diagnosis of PE.

The Wells criteria is an established model for risk stratifying patients presenting with possible VTE (TABLE).⁸ For patients with low pretest probability, as in this case, a d-dimer is an effective diagnostic work-up, as a negative result will rule out PE. (If the d-dimer had been negative in this case, we would have considered other diagnoses, such as acute coronary syndrome, biliary colic, gastritis, pancreatitis, or musculoskeletal pain.) For high-risk patients, immediate anticoagulation and imaging should be performed, frequently with heparin and CTPA.⁹

Continue to: Length of treatment depends on gender and etiology

The cornerstone treatment for stable patients with VTE is therapeutic anticoagulation. The new oral anticoagulants, which directly inhibit factor Xa or thrombin, have become increasingly popular for management of VTE, in part because they don’t require INR testing and monitoring.²

The duration of anticoagulation, particularly in unprovoked PE, is debatable. As noted earlier, patients with an unprovoked PE are at higher risk of recurrence than those with a reversible cause, so the question becomes whether these patients should have indefinite anticoagulation.^2,3 Studies examining risk stratification of patients with a first, unprovoked VTE have found that men have the highest risk of recurrence, followed by women who were not taking estrogen during the index VTE, and lastly women who were taking estrogen therapy during the index VTE and subsequently discontinued it.^2,3,10

Thus, it is reasonable to give women the option to discontinue anticoagulation in the setting of a negative d-dimer follow-up.³ The 2016 CHEST guidelines recommend extended anticoagulation for a first-time, unprovoked VTE, but acknowledge this recommendation is strongest for men and that women with negative d-dimer assays may consider discontinuation.¹⁰

Our patient was directed to the emergency department for further monitoring following CT confirmation. She was discharged home after being deemed stable and prescribed apixaban 10 mg/d. A venous duplex ultrasound performed 12 days later for knee pain revealed no venous thrombosis. A CT of the abdomen performed 3 months later for other reasons revealed a normal gallbladder with no visible stones.

Apixaban was continued for 3 months and discontinued after discussion of risks and benefits of therapy cessation in the setting of a normal d-dimer and the 2016 CHEST guidelines for anticoagulation in VTE.¹⁰

Continue to: THE TAKEAWAY

THE TAKEAWAY

PE carries a significantly high mortality rate and can manifest with nonspecific and masquerading signs. A high index of suspicion is required to place PE on the differential diagnosis and carry out appropriate testing. Our patient presented with a history consistent with biliary colic but with pleuritic chest pain that warranted consideration of a PE.

CORRESPONDENCE
Alyssa Anderson, MD, 1 Continental Drive, Elizabethtown, PA 17022; [email protected]

THE CASE

A 75-year-old woman presented to the primary care clinic with right-side rib pain. The patient said the pain started 1 week earlier, after she ate fried chicken for dinner, and had since been exacerbated by rich meals, lying supine, and taking a deep inspiratory breath. She also said that prior to coming to the clinic that day, the pain had been radiating to her right shoulder.

The patient denied experiencing associated fevers, chills, shortness of breath, chest pain, nausea, vomiting, constipation, diarrhea, or changes in stool color. She had a history of hypertension, for which she was taking lisinopril 20 mg/d, and hypercholesterolemia, for which she was on simvastatin 10 mg/d. She was additionally using timolol ophthalmic solution for her glaucoma.

During the examination, the patient’s vital signs were stable, with a pulse of 80 beats/min, a respiratory rate of 16 breaths/min, and an oxygen saturation of 98% on room air. The patient had no abdominal tenderness upon palpation, and the physical exam revealed no abnormalities. An in-office electrocardiogram was performed, with normal results. Additionally, a comprehensive metabolic panel, lipase test, and d-dimer test were ordered. Lab results showed an isolated elevated d-dimer of 2.66 mcg/mL (normal range, < 0.54 mcg/mL), while all other labs were normal.

THE DIAGNOSIS

Based on the lab results, a stat computed tomography pulmonary angiogram (CTPA) was ordered and showed a right segmental and subsegmental pulmonary embolism (PE; FIGURE 1).

DISCUSSION

PE shares pathophysiologic mechanisms with deep vein thrombosis (DVT), and together these comprise venous thromboembolism (VTE). Risk factors for VTE include hypercoagulable disorders, use of estrogens, active malignancy, and immobilization.¹ Unprovoked VTE occurs in the absence of identifiable risk factors and carries a higher risk of recurrence.^2,3 While PE is classically thought to occur in the setting of a DVT, there is increasing literature describing de novo PE that can occur independent of a DVT.⁴

Common symptoms of PE include tachycardia, tachypnea, and pleuritic chest pain.⁵ Abdominal pain is a rare symptom described in some case reports.^6,7 Thus, a high clinical suspicion is needed for diagnosis of PE.

The Wells criteria is an established model for risk stratifying patients presenting with possible VTE (TABLE).⁸ For patients with low pretest probability, as in this case, a d-dimer is an effective diagnostic work-up, as a negative result will rule out PE. (If the d-dimer had been negative in this case, we would have considered other diagnoses, such as acute coronary syndrome, biliary colic, gastritis, pancreatitis, or musculoskeletal pain.) For high-risk patients, immediate anticoagulation and imaging should be performed, frequently with heparin and CTPA.⁹

Continue to: Length of treatment depends on gender and etiology

The cornerstone treatment for stable patients with VTE is therapeutic anticoagulation. The new oral anticoagulants, which directly inhibit factor Xa or thrombin, have become increasingly popular for management of VTE, in part because they don’t require INR testing and monitoring.²

The duration of anticoagulation, particularly in unprovoked PE, is debatable. As noted earlier, patients with an unprovoked PE are at higher risk of recurrence than those with a reversible cause, so the question becomes whether these patients should have indefinite anticoagulation.^2,3 Studies examining risk stratification of patients with a first, unprovoked VTE have found that men have the highest risk of recurrence, followed by women who were not taking estrogen during the index VTE, and lastly women who were taking estrogen therapy during the index VTE and subsequently discontinued it.^2,3,10

Thus, it is reasonable to give women the option to discontinue anticoagulation in the setting of a negative d-dimer follow-up.³ The 2016 CHEST guidelines recommend extended anticoagulation for a first-time, unprovoked VTE, but acknowledge this recommendation is strongest for men and that women with negative d-dimer assays may consider discontinuation.¹⁰

Our patient was directed to the emergency department for further monitoring following CT confirmation. She was discharged home after being deemed stable and prescribed apixaban 10 mg/d. A venous duplex ultrasound performed 12 days later for knee pain revealed no venous thrombosis. A CT of the abdomen performed 3 months later for other reasons revealed a normal gallbladder with no visible stones.

Apixaban was continued for 3 months and discontinued after discussion of risks and benefits of therapy cessation in the setting of a normal d-dimer and the 2016 CHEST guidelines for anticoagulation in VTE.¹⁰

Continue to: THE TAKEAWAY

THE TAKEAWAY

PE carries a significantly high mortality rate and can manifest with nonspecific and masquerading signs. A high index of suspicion is required to place PE on the differential diagnosis and carry out appropriate testing. Our patient presented with a history consistent with biliary colic but with pleuritic chest pain that warranted consideration of a PE.

CORRESPONDENCE
Alyssa Anderson, MD, 1 Continental Drive, Elizabethtown, PA 17022; [email protected]

ILLUSTRATIVE CASE

A 67-year-old man with a history of coronary artery stenting 7 years prior and nonvalvular AF that is well controlled with a beta-blocker comes in for a routine health maintenance visit. You note that the patient takes warfarin, metoprolol, and aspirin. The patient has not had any thrombotic or bleeding events in his lifetime. Does this patient need to take both warfarin and aspirin? Do the antithrombotic benefits of dual therapy outweigh the risk of bleeding?

Antiplatelet agents have long been recommended for secondary prevention of cardiovascular (CV) events in patients with IHD. The goal is to reduce the risk of coronary artery thrombosis.² Many patients with IHD also develop AF and are treated with OACs such as warfarin or direct oral anticoagulants (DOACs) to prevent thromboembolic events.

There has been a paucity of data to determine the risks and benefits of OAC monotherapy compared to OAC plus single antiplatelet therapy (SAPT). Given research that shows increased risks of bleeding and all-cause mortality when aspirin is used for primary prevention of CV disease,^3,4 it is prudent to examine if the harms of aspirin outweigh its benefits for the secondary prevention of acute coronary events in patients already taking antithrombotic agents.

STUDY SUMMARY

Reduced bleeding risk, with no difference in major adverse cardiovascular events

This study by Lee and colleagues¹ was a meta-analysis of 8855 patients with nonvalvular AF and stable coronary artery disease (CAD), from 6 trials comparing OAC monotherapy vs OAC plus SAPT. The meta-analysis involved 3 studies using patient registries, 2 cohort studies, and an open-label randomized trial that together spanned the period from 2002 to 2016. The longest study period was 9 years (1 study) and the shortest, 1 year (2 studies). Oral anticoagulation consisted of either vitamin K antagonist (VKA) therapy (the majority of the patients studied) or DOAC therapy (8.6% of the patients studied). SAPT was either aspirin or clopidogrel.

The primary outcome measure was major adverse CV events (MACE). Secondary outcome measures included major bleeding, stroke, all-cause mortality, and net adverse events. The definitions used by the studies for major bleeding were deemed “largely consistent” with the International Society on Thrombosis and Haemostasis major bleeding criteria, ie, fatal bleeding, symptomatic bleeding in a critical area or organ (intracranial, intraspinal, intraocular, retroperitoneal, intra-articular, pericardial, or intramuscular causing compartment syndrome), or a drop in hemoglobin (≥ 2 g/dL or requiring transfusion of ≥ 2 units of whole blood or red cells).⁵

This study strongly suggests that there is a large subgroup of patients with stable CAD for whom single antiplatelet therapy should not be prescribed as a preventive medication.

There was no difference in MACE between the monotherapy and OAC plus SAPT groups (hazard ratio [HR] = 1.09; 95% CI, 0.92-1.29). Similarly, there were no differences in stroke and all-cause mortality between the groups. However, there was a significant association of higher risk of major bleeding (HR = 1.61; 95% CI, 1.38-1.87) and net adverse events (HR = 1.21; 95% CI, 1.02-1.43) in the OAC plus SAPT group compared with the OAC monotherapy group.

This study’s limitations included its low percentage of patients taking a DOAC. Also, due to variations in methods of reporting CHA₂DS₂-VASc and HAS-BLED scores among the studies (for risk of stroke in patients with nonrheumatic AF and for risk of bleeding in AF patients taking anticoagulants), this meta-analysis could not determine if different outcomes might be found in patients with different CHA₂DS₂-VASc and HAS-BLED scores.

Continue to: WHAT'S NEW

OAC monotherapy benefit for patients with nonvalvular AF

This study strongly suggests that there is a large subgroup of patients with stable CAD for whom SAPT should not be prescribed as a preventive medication: patients with nonvalvular AF who are receiving OAC therapy. This study concurs with the results of the 2019 AFIRE (Atrial Fibrillation and Ischemic Events with Rivaroxaban in Patients with Stable Coronary Artery Disease) trial in Japan, in which 2236 patients with stable IHD (coronary artery bypass grafting, stenting, or cardiac catheterization > 1 year earlier) were randomized to receive rivaroxaban either alone or with an antiplatelet agent. All-cause mortality and major bleeding were lower in the monotherapy group.⁶

This meta-analysis calls into question the baseline recommendation from the 2012 American College of Cardiology Foundation/American Heart Association (ACCF/AHA) guideline to prescribe aspirin indefinitely for patients with stable CAD unless there is a contraindication (oral anticoagulation is not listed as a contraindication).² The 2020 ACC Expert Consensus Decision Pathway⁷ published in February 2021 stated that for patients requiring long-term anticoagulation therapy who have completed 12 months of SAPT after percutaneous coronary intervention, anticoagulation therapy alone “could be used long-term”; however, the 2019 study by Lee was not listed among their references. Inclusion of the Lee study might have contributed to a stronger recommendation.

Also, the new guidelines include clinical situations in which dual therapy could still be continued: “… if perceived thrombotic risk is high (eg, prior myocardial infarction, complex lesions, presence of select traditional cardiovascular risk factors, or extensive [atherosclerotic cardiovascular disease]), and the patient is at low bleeding risk.” The guidelines state that in this situation, “… it is reasonable to continue SAPT beyond 12 months (in line with prior ACC/AHA recommendations).”⁷ However, the cited study compared dual therapy (dabigatran plus APT) to warfarin triple therapy. Single OAC therapy was not studied.⁸

CAVEATS

DOAC patient populationwas not well represented

The study had a low percentage of patients taking a DOAC. Also, because there were variations in how the studies reported CHA₂DS₂-VASc and HAS-BLED scores, this meta-analysis was unable to determine if different scores might have produced different outcomes. However, the studies involving registries had the advantage of looking at the data for this population over long periods of time and included a wide variety of patients, making the recommendation likely valid.

CHALLENGES TO IMPLEMENTATION

Primary care approach may not sync with specialist practice

We see no challenges to implementation except for potential differences between primary care physicians and specialists regarding the use of antiplatelet agents in this patient population.

ACKNOWLEDGEMENT

The PURLs Surveillance System was supported in part by Grant Number UL1RR024999 from the National Center for Research Resources, a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.

ILLUSTRATIVE CASE

A 67-year-old man with a history of coronary artery stenting 7 years prior and nonvalvular AF that is well controlled with a beta-blocker comes in for a routine health maintenance visit. You note that the patient takes warfarin, metoprolol, and aspirin. The patient has not had any thrombotic or bleeding events in his lifetime. Does this patient need to take both warfarin and aspirin? Do the antithrombotic benefits of dual therapy outweigh the risk of bleeding?

Antiplatelet agents have long been recommended for secondary prevention of cardiovascular (CV) events in patients with IHD. The goal is to reduce the risk of coronary artery thrombosis.² Many patients with IHD also develop AF and are treated with OACs such as warfarin or direct oral anticoagulants (DOACs) to prevent thromboembolic events.

There has been a paucity of data to determine the risks and benefits of OAC monotherapy compared to OAC plus single antiplatelet therapy (SAPT). Given research that shows increased risks of bleeding and all-cause mortality when aspirin is used for primary prevention of CV disease,^3,4 it is prudent to examine if the harms of aspirin outweigh its benefits for the secondary prevention of acute coronary events in patients already taking antithrombotic agents.

STUDY SUMMARY

Reduced bleeding risk, with no difference in major adverse cardiovascular events

This study by Lee and colleagues¹ was a meta-analysis of 8855 patients with nonvalvular AF and stable coronary artery disease (CAD), from 6 trials comparing OAC monotherapy vs OAC plus SAPT. The meta-analysis involved 3 studies using patient registries, 2 cohort studies, and an open-label randomized trial that together spanned the period from 2002 to 2016. The longest study period was 9 years (1 study) and the shortest, 1 year (2 studies). Oral anticoagulation consisted of either vitamin K antagonist (VKA) therapy (the majority of the patients studied) or DOAC therapy (8.6% of the patients studied). SAPT was either aspirin or clopidogrel.

The primary outcome measure was major adverse CV events (MACE). Secondary outcome measures included major bleeding, stroke, all-cause mortality, and net adverse events. The definitions used by the studies for major bleeding were deemed “largely consistent” with the International Society on Thrombosis and Haemostasis major bleeding criteria, ie, fatal bleeding, symptomatic bleeding in a critical area or organ (intracranial, intraspinal, intraocular, retroperitoneal, intra-articular, pericardial, or intramuscular causing compartment syndrome), or a drop in hemoglobin (≥ 2 g/dL or requiring transfusion of ≥ 2 units of whole blood or red cells).⁵

This study strongly suggests that there is a large subgroup of patients with stable CAD for whom single antiplatelet therapy should not be prescribed as a preventive medication.

There was no difference in MACE between the monotherapy and OAC plus SAPT groups (hazard ratio [HR] = 1.09; 95% CI, 0.92-1.29). Similarly, there were no differences in stroke and all-cause mortality between the groups. However, there was a significant association of higher risk of major bleeding (HR = 1.61; 95% CI, 1.38-1.87) and net adverse events (HR = 1.21; 95% CI, 1.02-1.43) in the OAC plus SAPT group compared with the OAC monotherapy group.

This study’s limitations included its low percentage of patients taking a DOAC. Also, due to variations in methods of reporting CHA₂DS₂-VASc and HAS-BLED scores among the studies (for risk of stroke in patients with nonrheumatic AF and for risk of bleeding in AF patients taking anticoagulants), this meta-analysis could not determine if different outcomes might be found in patients with different CHA₂DS₂-VASc and HAS-BLED scores.

Continue to: WHAT'S NEW

OAC monotherapy benefit for patients with nonvalvular AF

This study strongly suggests that there is a large subgroup of patients with stable CAD for whom SAPT should not be prescribed as a preventive medication: patients with nonvalvular AF who are receiving OAC therapy. This study concurs with the results of the 2019 AFIRE (Atrial Fibrillation and Ischemic Events with Rivaroxaban in Patients with Stable Coronary Artery Disease) trial in Japan, in which 2236 patients with stable IHD (coronary artery bypass grafting, stenting, or cardiac catheterization > 1 year earlier) were randomized to receive rivaroxaban either alone or with an antiplatelet agent. All-cause mortality and major bleeding were lower in the monotherapy group.⁶

This meta-analysis calls into question the baseline recommendation from the 2012 American College of Cardiology Foundation/American Heart Association (ACCF/AHA) guideline to prescribe aspirin indefinitely for patients with stable CAD unless there is a contraindication (oral anticoagulation is not listed as a contraindication).² The 2020 ACC Expert Consensus Decision Pathway⁷ published in February 2021 stated that for patients requiring long-term anticoagulation therapy who have completed 12 months of SAPT after percutaneous coronary intervention, anticoagulation therapy alone “could be used long-term”; however, the 2019 study by Lee was not listed among their references. Inclusion of the Lee study might have contributed to a stronger recommendation.

Also, the new guidelines include clinical situations in which dual therapy could still be continued: “… if perceived thrombotic risk is high (eg, prior myocardial infarction, complex lesions, presence of select traditional cardiovascular risk factors, or extensive [atherosclerotic cardiovascular disease]), and the patient is at low bleeding risk.” The guidelines state that in this situation, “… it is reasonable to continue SAPT beyond 12 months (in line with prior ACC/AHA recommendations).”⁷ However, the cited study compared dual therapy (dabigatran plus APT) to warfarin triple therapy. Single OAC therapy was not studied.⁸

CAVEATS

DOAC patient populationwas not well represented

The study had a low percentage of patients taking a DOAC. Also, because there were variations in how the studies reported CHA₂DS₂-VASc and HAS-BLED scores, this meta-analysis was unable to determine if different scores might have produced different outcomes. However, the studies involving registries had the advantage of looking at the data for this population over long periods of time and included a wide variety of patients, making the recommendation likely valid.

CHALLENGES TO IMPLEMENTATION

Primary care approach may not sync with specialist practice

We see no challenges to implementation except for potential differences between primary care physicians and specialists regarding the use of antiplatelet agents in this patient population.

ACKNOWLEDGEMENT

The PURLs Surveillance System was supported in part by Grant Number UL1RR024999 from the National Center for Research Resources, a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.

ILLUSTRATIVE CASE

A 67-year-old man with a history of coronary artery stenting 7 years prior and nonvalvular AF that is well controlled with a beta-blocker comes in for a routine health maintenance visit. You note that the patient takes warfarin, metoprolol, and aspirin. The patient has not had any thrombotic or bleeding events in his lifetime. Does this patient need to take both warfarin and aspirin? Do the antithrombotic benefits of dual therapy outweigh the risk of bleeding?

Antiplatelet agents have long been recommended for secondary prevention of cardiovascular (CV) events in patients with IHD. The goal is to reduce the risk of coronary artery thrombosis.² Many patients with IHD also develop AF and are treated with OACs such as warfarin or direct oral anticoagulants (DOACs) to prevent thromboembolic events.

There has been a paucity of data to determine the risks and benefits of OAC monotherapy compared to OAC plus single antiplatelet therapy (SAPT). Given research that shows increased risks of bleeding and all-cause mortality when aspirin is used for primary prevention of CV disease,^3,4 it is prudent to examine if the harms of aspirin outweigh its benefits for the secondary prevention of acute coronary events in patients already taking antithrombotic agents.

STUDY SUMMARY

Reduced bleeding risk, with no difference in major adverse cardiovascular events

This study by Lee and colleagues¹ was a meta-analysis of 8855 patients with nonvalvular AF and stable coronary artery disease (CAD), from 6 trials comparing OAC monotherapy vs OAC plus SAPT. The meta-analysis involved 3 studies using patient registries, 2 cohort studies, and an open-label randomized trial that together spanned the period from 2002 to 2016. The longest study period was 9 years (1 study) and the shortest, 1 year (2 studies). Oral anticoagulation consisted of either vitamin K antagonist (VKA) therapy (the majority of the patients studied) or DOAC therapy (8.6% of the patients studied). SAPT was either aspirin or clopidogrel.

The primary outcome measure was major adverse CV events (MACE). Secondary outcome measures included major bleeding, stroke, all-cause mortality, and net adverse events. The definitions used by the studies for major bleeding were deemed “largely consistent” with the International Society on Thrombosis and Haemostasis major bleeding criteria, ie, fatal bleeding, symptomatic bleeding in a critical area or organ (intracranial, intraspinal, intraocular, retroperitoneal, intra-articular, pericardial, or intramuscular causing compartment syndrome), or a drop in hemoglobin (≥ 2 g/dL or requiring transfusion of ≥ 2 units of whole blood or red cells).⁵

This study strongly suggests that there is a large subgroup of patients with stable CAD for whom single antiplatelet therapy should not be prescribed as a preventive medication.

There was no difference in MACE between the monotherapy and OAC plus SAPT groups (hazard ratio [HR] = 1.09; 95% CI, 0.92-1.29). Similarly, there were no differences in stroke and all-cause mortality between the groups. However, there was a significant association of higher risk of major bleeding (HR = 1.61; 95% CI, 1.38-1.87) and net adverse events (HR = 1.21; 95% CI, 1.02-1.43) in the OAC plus SAPT group compared with the OAC monotherapy group.

This study’s limitations included its low percentage of patients taking a DOAC. Also, due to variations in methods of reporting CHA₂DS₂-VASc and HAS-BLED scores among the studies (for risk of stroke in patients with nonrheumatic AF and for risk of bleeding in AF patients taking anticoagulants), this meta-analysis could not determine if different outcomes might be found in patients with different CHA₂DS₂-VASc and HAS-BLED scores.

Continue to: WHAT'S NEW

OAC monotherapy benefit for patients with nonvalvular AF

This study strongly suggests that there is a large subgroup of patients with stable CAD for whom SAPT should not be prescribed as a preventive medication: patients with nonvalvular AF who are receiving OAC therapy. This study concurs with the results of the 2019 AFIRE (Atrial Fibrillation and Ischemic Events with Rivaroxaban in Patients with Stable Coronary Artery Disease) trial in Japan, in which 2236 patients with stable IHD (coronary artery bypass grafting, stenting, or cardiac catheterization > 1 year earlier) were randomized to receive rivaroxaban either alone or with an antiplatelet agent. All-cause mortality and major bleeding were lower in the monotherapy group.⁶

This meta-analysis calls into question the baseline recommendation from the 2012 American College of Cardiology Foundation/American Heart Association (ACCF/AHA) guideline to prescribe aspirin indefinitely for patients with stable CAD unless there is a contraindication (oral anticoagulation is not listed as a contraindication).² The 2020 ACC Expert Consensus Decision Pathway⁷ published in February 2021 stated that for patients requiring long-term anticoagulation therapy who have completed 12 months of SAPT after percutaneous coronary intervention, anticoagulation therapy alone “could be used long-term”; however, the 2019 study by Lee was not listed among their references. Inclusion of the Lee study might have contributed to a stronger recommendation.

Also, the new guidelines include clinical situations in which dual therapy could still be continued: “… if perceived thrombotic risk is high (eg, prior myocardial infarction, complex lesions, presence of select traditional cardiovascular risk factors, or extensive [atherosclerotic cardiovascular disease]), and the patient is at low bleeding risk.” The guidelines state that in this situation, “… it is reasonable to continue SAPT beyond 12 months (in line with prior ACC/AHA recommendations).”⁷ However, the cited study compared dual therapy (dabigatran plus APT) to warfarin triple therapy. Single OAC therapy was not studied.⁸

CAVEATS

DOAC patient populationwas not well represented

The study had a low percentage of patients taking a DOAC. Also, because there were variations in how the studies reported CHA₂DS₂-VASc and HAS-BLED scores, this meta-analysis was unable to determine if different scores might have produced different outcomes. However, the studies involving registries had the advantage of looking at the data for this population over long periods of time and included a wide variety of patients, making the recommendation likely valid.

CHALLENGES TO IMPLEMENTATION

Primary care approach may not sync with specialist practice

We see no challenges to implementation except for potential differences between primary care physicians and specialists regarding the use of antiplatelet agents in this patient population.

ACKNOWLEDGEMENT

The PURLs Surveillance System was supported in part by Grant Number UL1RR024999 from the National Center for Research Resources, a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.

Since their discovery in the early 1900s as the treatment for life-threatening deficiency syndromes, vitamins have been touted as panaceas for numerous ailments. While observational data have suggested potential correlations between vitamin status and every imaginable disease, randomized controlled trials (RCTs) have generally failed to find benefits from supplementation. Despite this lack of proven efficacy, more than half of older adults reported taking vitamins regularly.¹

While most clinicians consider vitamins to be, at worst, expensive placebos, the potential for harm and dangerous interactions exists. Unlike pharmaceuticals, vitamins are generally unregulated, and the true content of many dietary supplements is often difficult to elucidate. Understanding the physiologic role, foundational evidence, and specific indications for the various vitamins is key to providing the best recommendations to patients.

Vitamins are essential organic nutrients, required in small quantities for normal metabolism. Since they are not synthesized endogenously, they must be ingested via food intake. In the developed world, vitamin deficiency syndromes are rare, thanks to sufficiently balanced diets and availability of fortified foods. The focus of this article will be on vitamin supplementation in healthy patients with well-balanced diets. TABLE W1² (available at mdedge.com/familymedicine) lists the 13 recognized vitamins, their recommended dietary allowances, and any known toxicity risks. TABLE 2² outlines elements of the history to consider when evaluating for deficiency. A summary of the most clinically significant evidence for vitamin supplementation follows; a more comprehensive review can be found in TABLE 3.^3-96

B COMPLEX VITAMINS

Vitamin B1

Vitamers: Thiamine (thiamin)

Physiologic role: Critical in carbohydrate and amino-acid catabolism and energy metabolism

Dietary sources: Whole grains, meat, fish, fortified cereals, and breads

Thiamine serves as an essential cofactor in energy metabolism.² Thiamine deficiency is responsible for beriberi syndrome (rare in the developed world) and Wernicke-Korsakoff syndrome, the latter of which is a relatively common complication of chronic alcohol dependence. Although thiamine’s administration in these conditions can be curative, evidence is lacking to support its use preventively in patients with alcoholism.³ Thiamine has additionally been theorized to play a role in cardiac and cognitive function, but RCT data has not shown consistent patient-oriented benefits.^4,5

The takeaway: Given the lack of evidence, supplementation in the general population is not recommended.

Vitamin B2

Vitamers: Riboflavin

Physiologic role: Essential component of cellular function and growth, energy production, and metabolism of fats and drugs

Dietary sources: Eggs, organ meats, lean meats, milk, green vegetables, fortified cereals and grains

Riboflavin is essential to energy production, cellular growth, and metabolism.²

The takeaway: Its use as migraine prophylaxis has limited data,⁹⁷ but there is otherwise no evidence to support health benefits of riboflavin supplementation.

Vitamin B3

Vitamers: Nicotinic acid (niacin); nicotinamide (niacinamide); nicotinamide riboside

Physiologic role: Converted to nicotinamide adenine dinucleotide (NAD), which is widely required in most cellular metabolic redox processes. Crucial to the synthesis and metabolism of carbohydrates, fatty acids, and proteins

Dietary sources: Poultry, beef, fish, nuts, legumes, grains. (Tryptophan can also be converted to NAD.)

Niacin is readily converted to NAD, an essential coenzyme for multiple catalytic processes in the body. While niacin at doses more than 100 times the recommended dietary allowance (RDA; 1-3 g/d) has been extensively studied for its role in dyslipidemias,² pharmacologic dosing is beyond the scope of this article.

The takeaway: There is no evidence supporting a clinical benefit from niacin supplementation.

Vitamin B5

Vitamers: Pantothenic acid; pantethine

Physiologic role: Required for synthesis of coenzyme A (CoA) and acyl carrier protein, both essential in fatty acid and other anabolic/catabolic processes

Dietary sources: Almost all plant/animal-based foods. Richest sources include beef, chicken, organ meats, whole grains, and some vegetables

Pantothenic acid is essential to multiple metabolic processes and readily available in sufficient amounts in most foods.² Although limited RCT data suggest pantethine may improve lipid measures,^12,98,99 pantothenic acid itself does not seem to share this effect.

The takeaway: There is no data that supplementation of any form of vitamin B5 has any patient-oriented clinical benefits.

Continue to: Vitamin B6

Vitamers: Pyridoxine; pyridoxamine; pyridoxal

Physiologic role: Widely involved coenzyme for cognitive development, neurotransmitter biosynthesis, homocysteine and glucose metabolism, immune function, and hemoglobin formation

Dietary sources: Fish, organ meats, potatoes/starchy vegetables, fruit (other than citrus), and fortified cereals

Pyridoxine is required for numerous enzymatic processes in the body, including biosynthesis of neurotransmitters and homeostasis of the amino acid homocysteine.² While overt deficiency is rare, marginal insufficiency may become clinically apparent and has been associated with malabsorption, malignancies, pregnancy, heart disease, alcoholism, and use of drugs such as isoniazid, hydralazine, and levodopa/carbidopa.² Vitamin B6 supplementation is known to decrease plasma homocysteine levels, a theorized intermediary for cardiovascular disease; however, studies have failed to consistently demonstrate patient-­oriented benefits.^100-102 While observational data has suggested a correlation between vitamin B6 status and cancer risk, RCTs have not supported benefit from supplementation.^14-16 Potential effects of vitamin B6 supplementation on cognitive function have also been studied without observed benefit.^17,18

The takeaway: Vitamin B6 is recommended as a potential treatment option for nausea in pregnancy.¹⁹ Otherwise, vitamin B6 is readily available in food, deficiency is rare, and no patient-oriented evidence supports supplementation in the general population.

Vitamin B7

Vitamers: Biotin

Physiologic role: Cofactor in the metabolism of fatty acids, glucose, and amino acids. Also plays key role in histone modifications, gene regulation, and cell signaling

Dietary sources: Widely available; most prevalent in organ meats, fish, meat, seeds, nuts, and vegetables (eg, sweet potatoes). Whole cooked eggs are a major source, but raw eggs contain avidin, which blocks absorption

Biotin serves a key role in metabolism, gene regulation, and cell signaling.² Biotin is known to interfere with laboratory assays— including cardiac enzymes, thyroid studies, and hormone studies—at normal supplementation doses, resulting in both false-positive and false-negative results.¹⁰³

The takeaway: No evidence supports the health benefits of biotin supplementation.

Vitamin B9

Vitamers: Folates; folic acid

Physiologic role: Functions as a coenzyme in the synthesis of DNA/RNA and metabolism of amino acids

Dietary sources: Highest content in spinach, liver, asparagus, and brussels sprouts. Generally found in green leafy vegetables, fruits, nuts, beans, peas, seafood, eggs, dairy, meat, poultry, grains, and fortified cereals.

Continue to: Vitamin B12

Vitamers: Cyanocobalamin; hydroxocobalamin; methylcobalamin; adenosylcobalamin

Physiologic role: Required for red blood cell formation, neurologic function, and DNA synthesis

Dietary sources: Only in animal products: fish, poultry, meat, eggs, and milk/dairy products. Not present in plant foods. Fortified cereals, nutritional yeast are sources for vegans/vegetarians.

Given their linked physiologic roles, vitamins B9 and B12 are frequently studied together. Folate and cobalamins play key roles in nucleic acid synthesis and amino acid metabolism, with their most clinically significant role in hematopoiesis. Vitamin B12 is also essential to normal neurologic function.²

The US Preventive Services Task Force (USPSTF) recommends preconceptual folate supplementation of 0.4-0.8 mg/d in women of childbearing age to decrease the risk of fetal neural tube defects (grade A).²¹ This is supported by high-quality RCT evidence demonstrating a protective effect of daily folate supplementation in preventing neural tube defects.²² Folate supplementation’s effect on other fetal birth defects has been investigated, but no benefit has been demonstrated. While observational studies have suggested an inverse relationship with folate status and fetal autism spectrum disorder,^23-25 the RCT data is mixed.²⁶

A potential role for folate in cancer prevention has been extensively investigated. An expert panel of the National Toxicology Program (NTP) concluded that folate supplementation does not reduce cancer risk in people with adequate baseline folate status based on high-quality meta-analysis data.^27,104 Conversely, long-term follow-up from RCTs demonstrated an increased risk of colorectal adenomas and cancers,^28,29 leading the NTP panel to conclude there is sufficient concern for adverse effects of folate on cancer growth to justify further research.¹⁰⁴

While observational studies have found a correlation of increased risk for disease with lower antioxidant serum levels, RCTs have not demonstrated a reduction in disease risk with supplementation.

Given folate and vitamin B12’s ­homocysteine-reducing effects, it has been theorized that supplementation may protect from cardiovascular disease. However, despite extensive research, there remains no consistent patient-oriented outcomes data to support such a benefit.^31,32,105

The evidence is mixed but generally has found no benefit of folate or vitamin B12 supplementation on cognitive function.^18,33-35 Finally, RCT data has failed to demonstrate a reduction in fracture risk with supplementation.^36,106

The takeaway: High-quality RCT evidence demonstrates a protective effect of preconceptual daily folate supplementation in preventing neural tube defects.²² The USPSTF recommends preconceptual folate supplementation of 0.4-0.8 mg/d in women of childbearing age to decrease the risk of fetal neural tube defects.

Continue to: ANTIOXIDANTS

In addition to their individual roles, vitamins A, E, and C are antioxidants, functioning to protect cells from oxidative damage by free radical species.² Due to this shared role, these vitamins are commonly studied together. Antioxidants are hypothesized to protect from various diseases, including cancer, cardiovascular disease, dementia, autoimmune disorders, depression, cataracts, and age-related vision decline.^2,37,107-112

Though observational studies have found a correlation of increased risk for disease with lower antioxidant serum levels, RCTs have not demonstrated a reduction in disease risk with supplementation and, in some cases, have found an increased risk of mortality. While several studies have found potential benefit of antioxidant use in reducing colon and breast cancer risk,^38,113-115 vitamins A and E have been associated with increased risk of lung and prostate cancer, respectively.^47,110 Cardiovascular disease and antioxidant vitamin supplementation has similar inconsistent data, ranging from slight benefit to harm.^2,116 After a large Cochrane review in 2012 found a significant increase in all-cause mortality associated with vitamin E and ­beta-carotene,¹¹⁷ the USPSTF made a specific recommendation against supplementation of these vitamins for the prevention of cardiovascular disease or cancer (grade D).¹¹⁸ Given its limited risk for harm, vitamin C was excluded from this recommendation.

Vitamin A

Vitamers: Retinol; retinal; retinyl esters; provitamin A carotenoids (beta-carotene, alpha-carotene, beta-cryptoxanthin)

Physiologic role: Essential for vision and corneal development. Also involved in general cell differentiation and immune function

Dietary sources: Liver, fish oil, dairy, and fortified cereals. Provitamin A sources: leafy green vegetables, orange/yellow vegetables, tomato products, fruits, and vegetable oils

Retinoids and their precursors, carotenoids, serve a critical function in vision, as well as regulating cell differentiation and proliferation throughout the body.² While evidence suggests mortality benefit of supplementation in populations at risk of deficiency,⁴⁵ wide-ranging studies have found either inconsistent benefit or outright harms in the developed world.

The takeaway: Given the USPSTF grade “D” recommendation and concern for potential harms, supplementation is not recommended in healthy patients without risk factors for deficiency.²

Vitamin E

Vitamers: Tocopherols (alpha-, beta-, ­gamma-, delta-); tocotrienol (alpha-, beta-, gamma-, delta-)

Physiologic role: Antioxidant; protects polyunsaturated fats from free radical oxidative damage. Involved in immune function, cell signaling, and regulation of gene expression

Dietary sources: Nuts, seeds, vegetable oil, green leafy vegetables, and fortified cereals

Vitamin E is the collective name of 8 compounds; alpha-tocopherol is the physiologically active form. Vitamin E is involved with cell proliferation as well as endothelial and platelet function.²

The takeaway: Vitamin E supplementation’s effects on cancer, cardiovascular disease, ophthalmologic disorders, and cognition have been investigated; data is either lacking to support a benefit or demonstrates harms as outlined above. Given this and the USPSTF grade “D” recommendation, supplementation is not recommended in healthy patients.²

Vitamin C

Vitamers: Ascorbic acid

Physiologic role: Required for synthesis of collagen, L-carnitine, and some neurotransmitters. Also involved in protein metabolism

Dietary sources: Primarily in fruits and vegetables: citrus, tomato, potatoes, red/green peppers, kiwi fruit, broccoli, strawberries, brussels sprouts, cantaloupe, and fortified cereals

Vitamin C supplementation at the onset of illness does not seem to have benefit.

Ascorbic acid is a required cofactor for biosynthesis of collagen, neurotransmitters, and protein metabolism.² In addition to the shared hypothesized benefits of antioxidants, vitamin C supplementation has undergone extensive research into its potential role in augmenting the immune system and preventing the common cold. Systematic reviews have found daily vitamin C supplementation of at least 200 mg did not affect the incidence of the common cold in healthy adults but may shorten duration and could be of benefit in those exposed to extreme physical exercise or cold.⁴⁸ Vitamin C supplementation at the onset of illness does not seem to have benefit.⁴⁸ Data is insufficient to draw conclusions about a potential effect on pneumonia incidence or severity.^119,120

The takeaway: Overall, data remain inconclusive as to potential benefits of vitamin C supplementation, although risks of potential harms are likely low.

Continue to: Vitamin D

Vitamers: Cholecalciferol (D3); ergocalciferol (D2)

Physiologic role: Hydroxylation in liver and kidney required to activate. Promotes dietary calcium absorption, enables normal bone mineralization. Also involved in modulation of cell growth, and neuromuscular and immune function

Dietary sources: Few natural dietary sources, which include fatty fish, fish liver oils; small amount in beef liver, cheese, egg yolks. Primary sources include fortified milk and endogenous synthesis in skin with UV exposure

Calciferol is a fat-soluble vitamin required for calcium and bone homeostasis. It is not naturally available in many foods but is primarily produced endogenously in the skin with ultraviolet light exposure.²

The AAP recommends supplementing exclusively breastfed infants with 400 IU/d of vitamin D to prevent rickets.

Bone density and fracture risk reduction are the most often cited benefits of vitamin D supplementation, but this has not been demonstrated consistently in RCTs. Multiple systematic reviews showing inconsistent benefit of vitamin D (with or without calcium) on fracture risk led the USPSTF to conclude that there is insufficient evidence (grade I) to issue a recommendation on vitamin D and calcium supplementation for primary prevention of fractures in postmenopausal women.^49-51 Despite some initial evidence suggesting a benefit of vitamin D supplementation on falls reduction, 3 recent systematic reviews did not demonstrate this in community-dwelling elders,^54-56 although a separate Cochrane review did suggest a reduction in rate of falls among institutionalized elders.⁵⁷

The takeaway: Given these findings, the USPSTF has recommended against (grade D) vitamin D supplementation to prevent falls in community-dwelling elders.⁵⁵

Beyond falls. While the vitamin D receptor is expressed throughout the body and observational studies have suggested a correlation between vitamin D status and many outcomes, extensive RCT data has generally failed to demonstrate extraskeletal benefits from supplementation. Meta-analysis data have demonstrated potential reductions in acute respiratory infection rates and asthma exacerbations with vitamin D supplementation. There is also limited evidence suggesting a reduction in preeclampsia and low-birthweight infant risk with vitamin D supplementation in pregnancy. However, several large meta-analyses and systematic reviews have investigated vitamin D supplementation’s effect on all-cause mortality and found no consistent data to support an association.^41,58-62

Multiple systematic reviews have investigated and found high-quality evidence demonstrating no association between vitamin D supplementation and cancer^41,63-66,121 or cardiovascular disease risk.^41,70,71 There is high-quality data showing no benefit of vitamin D supplementation for multiple additional diseases, including diabetes, cognitive decline, depression, pain, obesity, and liver disease.^{43,72-75,85-90,122}

The takeaway: Due to poor availability in breastmilk, the American Academy of Pediatrics (AAP) recommends supplementing exclusively breastfed infants with 400 IU/d of vitamin D to prevent rickets.¹²³ RCT data support high-dose supplementation of lactating women (6400 IU/d) as an alternative strategy to supplementation of the infant.¹²⁴ The AAP recommends that all nonbreastfeeding infants and older children ingesting < 1000 mL/d of vitamin D–fortified formula or milk should also be supplemented with 400 IU/d of vitamin D.¹²³ Despite these universal recommendations for supplementation, evidence is mixed on the effect of vitamin D supplementation on bone health in children.^52,53

Although concerns about vitamin D supplementation and increased risk of urolithiasis and hypercalcemia have been raised,^51,62,121 systematic reviews have not demonstrated significant, clinically relevant risks, even with high-dose supplementation (> 2800 IU/d).^125,126

Vitamin K

Vitamers: Phylloquinone (K1); menaquinones (K2)

Physiologic role: Coenzyme for synthesis of proteins involved in hemostasis and bone metabolism

Dietary sources: Phylloquinone is found in green leafy vegetables, vegetable oils, some fruits, meat, dairy, and eggs. Menaquinone is produced by gut bacteria and present in fermented foods

Vitamin K includes 2 groups of similar compounds: phylloquinone and menaquinones. Unlike other fat-soluble vitamins, vitamin K is rapidly metabolized and has low tissue storage.²

Children taking multivitamins were often found to have excess levels of potentially harmful nutrients, such as retinol, zinc, and folic acid.

Administration of vitamin K 0.5 to 1 mg intramuscularly (IM) to newborns is standard of care for the prevention of vitamin K deficiency bleeding (VKDB). This is supported by RCT data demonstrating a reduction in classic VKDB (occurring within 7 days)⁹¹ and epidemiologic data from various countries showing a reduction in late-onset VKDB with vitamin K prophylaxis programs.¹²⁷ Oral dosing appears to reduce the risk of VKDB in the setting of parental refusal but is less effective than IM dosing.^128,129

Vitamin K’s effects on bone density and fracture risk have also been investigated. Systematic reviews have demonstrated a reduction in fracture risk with vitamin K supplementation,^92,93 and European and Asian regulatory bodies have recognized a potential benefit on bone health.² The FDA considers the evidence insufficient at this time to support such a claim.² Higher dietary vitamin K consumption has been associated with lower risk of cardiovascular disease in observational studies⁹⁴ and supplementation was associated with improved disease measures,¹³⁰ but no patient-oriented outcomes have been demonstrated.¹³¹

The takeaway: The administration of vitamin K 0.5 to 1 mg intramuscularly (IM) to newborns is standard of care for the prevention of VKDB. Vitamin K may lead to a reduction in fracture risk, but the FDA considers the evidence insufficient. Vitamin K’s potential link to a lowered risk of cardiovascular disease has not been demonstrated with patient-­oriented outcomes. Vitamin K has low potential for toxicity, although its interaction with vitamin K antagonists (ie, warfarin) is clinically relevant.

Continue to: MULTIVITAMINS

Multivitamins are often defined as a supplement containing 3 or more vitamins and minerals but without herbs, hormones, or drugs.¹³² Many multivitamins do contain additional substances, and some include levels of vitamins that exceed the RDA or even the established tolerable upper intake level.¹³³

Safe medication storage should be practiced, as multivitamins with iron are a leading cause of poisoning in children.

A 2013 systematic review found limited evidence to support any benefit from multivitamin supplementation.⁴¹ Two included RCTs demonstrated a narrowly significant decrease in cancer rates among men, but saw no effect in women or the combined population.^134,135 This benefit appears to disappear at 5 years of follow-up.¹³⁶ RCT data have shown no benefit of multivitamin use on cognitive function,⁹⁵ and high-quality data suggest there is no effect on all-cause mortality.¹³⁷ Given this lack of supporting evidence, the USPSTF has concluded that there is insufficient evidence (grade I) to recommend vitamin supplementation in general to prevent cardiovascular disease or cancer.⁴¹

The use of prenatal multivitamins is generally recommended in the pregnancy and preconception period and has been associated with reduced risk of autism spectrum disorders, pediatric cancer rates, small-for-gestational-age infants, and multiple birth defects in offspring; however, studies have not examined if this benefit exceeds that of folate supplementation alone.^138-140 AAP does not recommend multivitamins for children with a well-balanced diet.¹⁴¹ Of concern, children taking multivitamins were often found to have excess levels of potentially harmful nutrients such as retinol, zinc, and folic acid.¹⁴²

The takeaway: There is limited evidence to support any benefit from multivitamin supplementation. Prenatal multivitamins are generally recommended in the pregnancy and preconception period. Overall, the risks of multivitamins are minimal, although that risk is dependent on the multivitamin’s constituent components.¹⁴³ Components such as vitamin K may interact with a patient’s medications, and multivitamins have been shown to reduce the circulating levels of antiretrovirals.¹⁴⁴ Specifically, multivitamins with iron should be avoided in men and postmenopausal women, and safe medication storage should be practiced as multivitamins with iron are a leading cause of poisoning in children.²

SUMMARY

Vitamin supplementation in the developed world remains common despite a paucity of RCT data supporting it. Supplementation of folate in women planning to conceive, vitamin D in breastfeeding infants, and vitamin K in newborns are well supported by clinical evidence. Otherwise, there is limited evidence supporting clinically significant benefit from supplementation in healthy patients with well-balanced diets—and in the case of vitamins A and E, there may be outright harms.

CORRESPONDENCE
Joel Herness, MD, 4700 North Las Vegas Boulevard, Nellis AFB, NV 89191; [email protected]

Since their discovery in the early 1900s as the treatment for life-threatening deficiency syndromes, vitamins have been touted as panaceas for numerous ailments. While observational data have suggested potential correlations between vitamin status and every imaginable disease, randomized controlled trials (RCTs) have generally failed to find benefits from supplementation. Despite this lack of proven efficacy, more than half of older adults reported taking vitamins regularly.¹

While most clinicians consider vitamins to be, at worst, expensive placebos, the potential for harm and dangerous interactions exists. Unlike pharmaceuticals, vitamins are generally unregulated, and the true content of many dietary supplements is often difficult to elucidate. Understanding the physiologic role, foundational evidence, and specific indications for the various vitamins is key to providing the best recommendations to patients.

Vitamins are essential organic nutrients, required in small quantities for normal metabolism. Since they are not synthesized endogenously, they must be ingested via food intake. In the developed world, vitamin deficiency syndromes are rare, thanks to sufficiently balanced diets and availability of fortified foods. The focus of this article will be on vitamin supplementation in healthy patients with well-balanced diets. TABLE W1² (available at mdedge.com/familymedicine) lists the 13 recognized vitamins, their recommended dietary allowances, and any known toxicity risks. TABLE 2² outlines elements of the history to consider when evaluating for deficiency. A summary of the most clinically significant evidence for vitamin supplementation follows; a more comprehensive review can be found in TABLE 3.^3-96

B COMPLEX VITAMINS

Vitamin B1

Vitamers: Thiamine (thiamin)

Physiologic role: Critical in carbohydrate and amino-acid catabolism and energy metabolism

Dietary sources: Whole grains, meat, fish, fortified cereals, and breads

Thiamine serves as an essential cofactor in energy metabolism.² Thiamine deficiency is responsible for beriberi syndrome (rare in the developed world) and Wernicke-Korsakoff syndrome, the latter of which is a relatively common complication of chronic alcohol dependence. Although thiamine’s administration in these conditions can be curative, evidence is lacking to support its use preventively in patients with alcoholism.³ Thiamine has additionally been theorized to play a role in cardiac and cognitive function, but RCT data has not shown consistent patient-oriented benefits.^4,5

The takeaway: Given the lack of evidence, supplementation in the general population is not recommended.

Vitamin B2

Vitamers: Riboflavin

Physiologic role: Essential component of cellular function and growth, energy production, and metabolism of fats and drugs

Dietary sources: Eggs, organ meats, lean meats, milk, green vegetables, fortified cereals and grains

Riboflavin is essential to energy production, cellular growth, and metabolism.²

The takeaway: Its use as migraine prophylaxis has limited data,⁹⁷ but there is otherwise no evidence to support health benefits of riboflavin supplementation.

Vitamin B3

Vitamers: Nicotinic acid (niacin); nicotinamide (niacinamide); nicotinamide riboside

Physiologic role: Converted to nicotinamide adenine dinucleotide (NAD), which is widely required in most cellular metabolic redox processes. Crucial to the synthesis and metabolism of carbohydrates, fatty acids, and proteins

Dietary sources: Poultry, beef, fish, nuts, legumes, grains. (Tryptophan can also be converted to NAD.)

Niacin is readily converted to NAD, an essential coenzyme for multiple catalytic processes in the body. While niacin at doses more than 100 times the recommended dietary allowance (RDA; 1-3 g/d) has been extensively studied for its role in dyslipidemias,² pharmacologic dosing is beyond the scope of this article.

The takeaway: There is no evidence supporting a clinical benefit from niacin supplementation.

Vitamin B5

Vitamers: Pantothenic acid; pantethine

Physiologic role: Required for synthesis of coenzyme A (CoA) and acyl carrier protein, both essential in fatty acid and other anabolic/catabolic processes

Dietary sources: Almost all plant/animal-based foods. Richest sources include beef, chicken, organ meats, whole grains, and some vegetables

Pantothenic acid is essential to multiple metabolic processes and readily available in sufficient amounts in most foods.² Although limited RCT data suggest pantethine may improve lipid measures,^12,98,99 pantothenic acid itself does not seem to share this effect.

The takeaway: There is no data that supplementation of any form of vitamin B5 has any patient-oriented clinical benefits.

Continue to: Vitamin B6

Vitamers: Pyridoxine; pyridoxamine; pyridoxal

Physiologic role: Widely involved coenzyme for cognitive development, neurotransmitter biosynthesis, homocysteine and glucose metabolism, immune function, and hemoglobin formation

Dietary sources: Fish, organ meats, potatoes/starchy vegetables, fruit (other than citrus), and fortified cereals

Pyridoxine is required for numerous enzymatic processes in the body, including biosynthesis of neurotransmitters and homeostasis of the amino acid homocysteine.² While overt deficiency is rare, marginal insufficiency may become clinically apparent and has been associated with malabsorption, malignancies, pregnancy, heart disease, alcoholism, and use of drugs such as isoniazid, hydralazine, and levodopa/carbidopa.² Vitamin B6 supplementation is known to decrease plasma homocysteine levels, a theorized intermediary for cardiovascular disease; however, studies have failed to consistently demonstrate patient-­oriented benefits.^100-102 While observational data has suggested a correlation between vitamin B6 status and cancer risk, RCTs have not supported benefit from supplementation.^14-16 Potential effects of vitamin B6 supplementation on cognitive function have also been studied without observed benefit.^17,18

The takeaway: Vitamin B6 is recommended as a potential treatment option for nausea in pregnancy.¹⁹ Otherwise, vitamin B6 is readily available in food, deficiency is rare, and no patient-oriented evidence supports supplementation in the general population.

Vitamin B7

Vitamers: Biotin

Physiologic role: Cofactor in the metabolism of fatty acids, glucose, and amino acids. Also plays key role in histone modifications, gene regulation, and cell signaling

Dietary sources: Widely available; most prevalent in organ meats, fish, meat, seeds, nuts, and vegetables (eg, sweet potatoes). Whole cooked eggs are a major source, but raw eggs contain avidin, which blocks absorption

Biotin serves a key role in metabolism, gene regulation, and cell signaling.² Biotin is known to interfere with laboratory assays— including cardiac enzymes, thyroid studies, and hormone studies—at normal supplementation doses, resulting in both false-positive and false-negative results.¹⁰³

The takeaway: No evidence supports the health benefits of biotin supplementation.

Vitamin B9

Vitamers: Folates; folic acid

Physiologic role: Functions as a coenzyme in the synthesis of DNA/RNA and metabolism of amino acids

Dietary sources: Highest content in spinach, liver, asparagus, and brussels sprouts. Generally found in green leafy vegetables, fruits, nuts, beans, peas, seafood, eggs, dairy, meat, poultry, grains, and fortified cereals.

Continue to: Vitamin B12

Vitamers: Cyanocobalamin; hydroxocobalamin; methylcobalamin; adenosylcobalamin

Physiologic role: Required for red blood cell formation, neurologic function, and DNA synthesis

Dietary sources: Only in animal products: fish, poultry, meat, eggs, and milk/dairy products. Not present in plant foods. Fortified cereals, nutritional yeast are sources for vegans/vegetarians.

Given their linked physiologic roles, vitamins B9 and B12 are frequently studied together. Folate and cobalamins play key roles in nucleic acid synthesis and amino acid metabolism, with their most clinically significant role in hematopoiesis. Vitamin B12 is also essential to normal neurologic function.²

The US Preventive Services Task Force (USPSTF) recommends preconceptual folate supplementation of 0.4-0.8 mg/d in women of childbearing age to decrease the risk of fetal neural tube defects (grade A).²¹ This is supported by high-quality RCT evidence demonstrating a protective effect of daily folate supplementation in preventing neural tube defects.²² Folate supplementation’s effect on other fetal birth defects has been investigated, but no benefit has been demonstrated. While observational studies have suggested an inverse relationship with folate status and fetal autism spectrum disorder,^23-25 the RCT data is mixed.²⁶

A potential role for folate in cancer prevention has been extensively investigated. An expert panel of the National Toxicology Program (NTP) concluded that folate supplementation does not reduce cancer risk in people with adequate baseline folate status based on high-quality meta-analysis data.^27,104 Conversely, long-term follow-up from RCTs demonstrated an increased risk of colorectal adenomas and cancers,^28,29 leading the NTP panel to conclude there is sufficient concern for adverse effects of folate on cancer growth to justify further research.¹⁰⁴

While observational studies have found a correlation of increased risk for disease with lower antioxidant serum levels, RCTs have not demonstrated a reduction in disease risk with supplementation.

Given folate and vitamin B12’s ­homocysteine-reducing effects, it has been theorized that supplementation may protect from cardiovascular disease. However, despite extensive research, there remains no consistent patient-oriented outcomes data to support such a benefit.^31,32,105

The evidence is mixed but generally has found no benefit of folate or vitamin B12 supplementation on cognitive function.^18,33-35 Finally, RCT data has failed to demonstrate a reduction in fracture risk with supplementation.^36,106

The takeaway: High-quality RCT evidence demonstrates a protective effect of preconceptual daily folate supplementation in preventing neural tube defects.²² The USPSTF recommends preconceptual folate supplementation of 0.4-0.8 mg/d in women of childbearing age to decrease the risk of fetal neural tube defects.

Continue to: ANTIOXIDANTS

In addition to their individual roles, vitamins A, E, and C are antioxidants, functioning to protect cells from oxidative damage by free radical species.² Due to this shared role, these vitamins are commonly studied together. Antioxidants are hypothesized to protect from various diseases, including cancer, cardiovascular disease, dementia, autoimmune disorders, depression, cataracts, and age-related vision decline.^2,37,107-112

Though observational studies have found a correlation of increased risk for disease with lower antioxidant serum levels, RCTs have not demonstrated a reduction in disease risk with supplementation and, in some cases, have found an increased risk of mortality. While several studies have found potential benefit of antioxidant use in reducing colon and breast cancer risk,^38,113-115 vitamins A and E have been associated with increased risk of lung and prostate cancer, respectively.^47,110 Cardiovascular disease and antioxidant vitamin supplementation has similar inconsistent data, ranging from slight benefit to harm.^2,116 After a large Cochrane review in 2012 found a significant increase in all-cause mortality associated with vitamin E and ­beta-carotene,¹¹⁷ the USPSTF made a specific recommendation against supplementation of these vitamins for the prevention of cardiovascular disease or cancer (grade D).¹¹⁸ Given its limited risk for harm, vitamin C was excluded from this recommendation.

Vitamin A

Vitamers: Retinol; retinal; retinyl esters; provitamin A carotenoids (beta-carotene, alpha-carotene, beta-cryptoxanthin)

Physiologic role: Essential for vision and corneal development. Also involved in general cell differentiation and immune function

Dietary sources: Liver, fish oil, dairy, and fortified cereals. Provitamin A sources: leafy green vegetables, orange/yellow vegetables, tomato products, fruits, and vegetable oils

Retinoids and their precursors, carotenoids, serve a critical function in vision, as well as regulating cell differentiation and proliferation throughout the body.² While evidence suggests mortality benefit of supplementation in populations at risk of deficiency,⁴⁵ wide-ranging studies have found either inconsistent benefit or outright harms in the developed world.

The takeaway: Given the USPSTF grade “D” recommendation and concern for potential harms, supplementation is not recommended in healthy patients without risk factors for deficiency.²

Vitamin E

Vitamers: Tocopherols (alpha-, beta-, ­gamma-, delta-); tocotrienol (alpha-, beta-, gamma-, delta-)

Physiologic role: Antioxidant; protects polyunsaturated fats from free radical oxidative damage. Involved in immune function, cell signaling, and regulation of gene expression

Dietary sources: Nuts, seeds, vegetable oil, green leafy vegetables, and fortified cereals

Vitamin E is the collective name of 8 compounds; alpha-tocopherol is the physiologically active form. Vitamin E is involved with cell proliferation as well as endothelial and platelet function.²

The takeaway: Vitamin E supplementation’s effects on cancer, cardiovascular disease, ophthalmologic disorders, and cognition have been investigated; data is either lacking to support a benefit or demonstrates harms as outlined above. Given this and the USPSTF grade “D” recommendation, supplementation is not recommended in healthy patients.²

Vitamin C

Vitamers: Ascorbic acid

Physiologic role: Required for synthesis of collagen, L-carnitine, and some neurotransmitters. Also involved in protein metabolism

Dietary sources: Primarily in fruits and vegetables: citrus, tomato, potatoes, red/green peppers, kiwi fruit, broccoli, strawberries, brussels sprouts, cantaloupe, and fortified cereals

Vitamin C supplementation at the onset of illness does not seem to have benefit.

Ascorbic acid is a required cofactor for biosynthesis of collagen, neurotransmitters, and protein metabolism.² In addition to the shared hypothesized benefits of antioxidants, vitamin C supplementation has undergone extensive research into its potential role in augmenting the immune system and preventing the common cold. Systematic reviews have found daily vitamin C supplementation of at least 200 mg did not affect the incidence of the common cold in healthy adults but may shorten duration and could be of benefit in those exposed to extreme physical exercise or cold.⁴⁸ Vitamin C supplementation at the onset of illness does not seem to have benefit.⁴⁸ Data is insufficient to draw conclusions about a potential effect on pneumonia incidence or severity.^119,120

The takeaway: Overall, data remain inconclusive as to potential benefits of vitamin C supplementation, although risks of potential harms are likely low.

Continue to: Vitamin D

Vitamers: Cholecalciferol (D3); ergocalciferol (D2)

Physiologic role: Hydroxylation in liver and kidney required to activate. Promotes dietary calcium absorption, enables normal bone mineralization. Also involved in modulation of cell growth, and neuromuscular and immune function

Dietary sources: Few natural dietary sources, which include fatty fish, fish liver oils; small amount in beef liver, cheese, egg yolks. Primary sources include fortified milk and endogenous synthesis in skin with UV exposure

Calciferol is a fat-soluble vitamin required for calcium and bone homeostasis. It is not naturally available in many foods but is primarily produced endogenously in the skin with ultraviolet light exposure.²

The AAP recommends supplementing exclusively breastfed infants with 400 IU/d of vitamin D to prevent rickets.

Bone density and fracture risk reduction are the most often cited benefits of vitamin D supplementation, but this has not been demonstrated consistently in RCTs. Multiple systematic reviews showing inconsistent benefit of vitamin D (with or without calcium) on fracture risk led the USPSTF to conclude that there is insufficient evidence (grade I) to issue a recommendation on vitamin D and calcium supplementation for primary prevention of fractures in postmenopausal women.^49-51 Despite some initial evidence suggesting a benefit of vitamin D supplementation on falls reduction, 3 recent systematic reviews did not demonstrate this in community-dwelling elders,^54-56 although a separate Cochrane review did suggest a reduction in rate of falls among institutionalized elders.⁵⁷

The takeaway: Given these findings, the USPSTF has recommended against (grade D) vitamin D supplementation to prevent falls in community-dwelling elders.⁵⁵

Beyond falls. While the vitamin D receptor is expressed throughout the body and observational studies have suggested a correlation between vitamin D status and many outcomes, extensive RCT data has generally failed to demonstrate extraskeletal benefits from supplementation. Meta-analysis data have demonstrated potential reductions in acute respiratory infection rates and asthma exacerbations with vitamin D supplementation. There is also limited evidence suggesting a reduction in preeclampsia and low-birthweight infant risk with vitamin D supplementation in pregnancy. However, several large meta-analyses and systematic reviews have investigated vitamin D supplementation’s effect on all-cause mortality and found no consistent data to support an association.^41,58-62

Multiple systematic reviews have investigated and found high-quality evidence demonstrating no association between vitamin D supplementation and cancer^41,63-66,121 or cardiovascular disease risk.^41,70,71 There is high-quality data showing no benefit of vitamin D supplementation for multiple additional diseases, including diabetes, cognitive decline, depression, pain, obesity, and liver disease.^{43,72-75,85-90,122}

The takeaway: Due to poor availability in breastmilk, the American Academy of Pediatrics (AAP) recommends supplementing exclusively breastfed infants with 400 IU/d of vitamin D to prevent rickets.¹²³ RCT data support high-dose supplementation of lactating women (6400 IU/d) as an alternative strategy to supplementation of the infant.¹²⁴ The AAP recommends that all nonbreastfeeding infants and older children ingesting < 1000 mL/d of vitamin D–fortified formula or milk should also be supplemented with 400 IU/d of vitamin D.¹²³ Despite these universal recommendations for supplementation, evidence is mixed on the effect of vitamin D supplementation on bone health in children.^52,53

Although concerns about vitamin D supplementation and increased risk of urolithiasis and hypercalcemia have been raised,^51,62,121 systematic reviews have not demonstrated significant, clinically relevant risks, even with high-dose supplementation (> 2800 IU/d).^125,126

Vitamin K

Vitamers: Phylloquinone (K1); menaquinones (K2)

Physiologic role: Coenzyme for synthesis of proteins involved in hemostasis and bone metabolism

Dietary sources: Phylloquinone is found in green leafy vegetables, vegetable oils, some fruits, meat, dairy, and eggs. Menaquinone is produced by gut bacteria and present in fermented foods

Vitamin K includes 2 groups of similar compounds: phylloquinone and menaquinones. Unlike other fat-soluble vitamins, vitamin K is rapidly metabolized and has low tissue storage.²

Children taking multivitamins were often found to have excess levels of potentially harmful nutrients, such as retinol, zinc, and folic acid.

Administration of vitamin K 0.5 to 1 mg intramuscularly (IM) to newborns is standard of care for the prevention of vitamin K deficiency bleeding (VKDB). This is supported by RCT data demonstrating a reduction in classic VKDB (occurring within 7 days)⁹¹ and epidemiologic data from various countries showing a reduction in late-onset VKDB with vitamin K prophylaxis programs.¹²⁷ Oral dosing appears to reduce the risk of VKDB in the setting of parental refusal but is less effective than IM dosing.^128,129

Vitamin K’s effects on bone density and fracture risk have also been investigated. Systematic reviews have demonstrated a reduction in fracture risk with vitamin K supplementation,^92,93 and European and Asian regulatory bodies have recognized a potential benefit on bone health.² The FDA considers the evidence insufficient at this time to support such a claim.² Higher dietary vitamin K consumption has been associated with lower risk of cardiovascular disease in observational studies⁹⁴ and supplementation was associated with improved disease measures,¹³⁰ but no patient-oriented outcomes have been demonstrated.¹³¹

The takeaway: The administration of vitamin K 0.5 to 1 mg intramuscularly (IM) to newborns is standard of care for the prevention of VKDB. Vitamin K may lead to a reduction in fracture risk, but the FDA considers the evidence insufficient. Vitamin K’s potential link to a lowered risk of cardiovascular disease has not been demonstrated with patient-­oriented outcomes. Vitamin K has low potential for toxicity, although its interaction with vitamin K antagonists (ie, warfarin) is clinically relevant.

Continue to: MULTIVITAMINS

Multivitamins are often defined as a supplement containing 3 or more vitamins and minerals but without herbs, hormones, or drugs.¹³² Many multivitamins do contain additional substances, and some include levels of vitamins that exceed the RDA or even the established tolerable upper intake level.¹³³

Safe medication storage should be practiced, as multivitamins with iron are a leading cause of poisoning in children.

A 2013 systematic review found limited evidence to support any benefit from multivitamin supplementation.⁴¹ Two included RCTs demonstrated a narrowly significant decrease in cancer rates among men, but saw no effect in women or the combined population.^134,135 This benefit appears to disappear at 5 years of follow-up.¹³⁶ RCT data have shown no benefit of multivitamin use on cognitive function,⁹⁵ and high-quality data suggest there is no effect on all-cause mortality.¹³⁷ Given this lack of supporting evidence, the USPSTF has concluded that there is insufficient evidence (grade I) to recommend vitamin supplementation in general to prevent cardiovascular disease or cancer.⁴¹

The use of prenatal multivitamins is generally recommended in the pregnancy and preconception period and has been associated with reduced risk of autism spectrum disorders, pediatric cancer rates, small-for-gestational-age infants, and multiple birth defects in offspring; however, studies have not examined if this benefit exceeds that of folate supplementation alone.^138-140 AAP does not recommend multivitamins for children with a well-balanced diet.¹⁴¹ Of concern, children taking multivitamins were often found to have excess levels of potentially harmful nutrients such as retinol, zinc, and folic acid.¹⁴²

The takeaway: There is limited evidence to support any benefit from multivitamin supplementation. Prenatal multivitamins are generally recommended in the pregnancy and preconception period. Overall, the risks of multivitamins are minimal, although that risk is dependent on the multivitamin’s constituent components.¹⁴³ Components such as vitamin K may interact with a patient’s medications, and multivitamins have been shown to reduce the circulating levels of antiretrovirals.¹⁴⁴ Specifically, multivitamins with iron should be avoided in men and postmenopausal women, and safe medication storage should be practiced as multivitamins with iron are a leading cause of poisoning in children.²

SUMMARY

Vitamin supplementation in the developed world remains common despite a paucity of RCT data supporting it. Supplementation of folate in women planning to conceive, vitamin D in breastfeeding infants, and vitamin K in newborns are well supported by clinical evidence. Otherwise, there is limited evidence supporting clinically significant benefit from supplementation in healthy patients with well-balanced diets—and in the case of vitamins A and E, there may be outright harms.

CORRESPONDENCE
Joel Herness, MD, 4700 North Las Vegas Boulevard, Nellis AFB, NV 89191; [email protected]

Since their discovery in the early 1900s as the treatment for life-threatening deficiency syndromes, vitamins have been touted as panaceas for numerous ailments. While observational data have suggested potential correlations between vitamin status and every imaginable disease, randomized controlled trials (RCTs) have generally failed to find benefits from supplementation. Despite this lack of proven efficacy, more than half of older adults reported taking vitamins regularly.¹

While most clinicians consider vitamins to be, at worst, expensive placebos, the potential for harm and dangerous interactions exists. Unlike pharmaceuticals, vitamins are generally unregulated, and the true content of many dietary supplements is often difficult to elucidate. Understanding the physiologic role, foundational evidence, and specific indications for the various vitamins is key to providing the best recommendations to patients.

Vitamins are essential organic nutrients, required in small quantities for normal metabolism. Since they are not synthesized endogenously, they must be ingested via food intake. In the developed world, vitamin deficiency syndromes are rare, thanks to sufficiently balanced diets and availability of fortified foods. The focus of this article will be on vitamin supplementation in healthy patients with well-balanced diets. TABLE W1² (available at mdedge.com/familymedicine) lists the 13 recognized vitamins, their recommended dietary allowances, and any known toxicity risks. TABLE 2² outlines elements of the history to consider when evaluating for deficiency. A summary of the most clinically significant evidence for vitamin supplementation follows; a more comprehensive review can be found in TABLE 3.^3-96

B COMPLEX VITAMINS

Vitamin B1

Vitamers: Thiamine (thiamin)

Physiologic role: Critical in carbohydrate and amino-acid catabolism and energy metabolism

Dietary sources: Whole grains, meat, fish, fortified cereals, and breads

Thiamine serves as an essential cofactor in energy metabolism.² Thiamine deficiency is responsible for beriberi syndrome (rare in the developed world) and Wernicke-Korsakoff syndrome, the latter of which is a relatively common complication of chronic alcohol dependence. Although thiamine’s administration in these conditions can be curative, evidence is lacking to support its use preventively in patients with alcoholism.³ Thiamine has additionally been theorized to play a role in cardiac and cognitive function, but RCT data has not shown consistent patient-oriented benefits.^4,5

The takeaway: Given the lack of evidence, supplementation in the general population is not recommended.

Vitamin B2

Vitamers: Riboflavin

Physiologic role: Essential component of cellular function and growth, energy production, and metabolism of fats and drugs

Dietary sources: Eggs, organ meats, lean meats, milk, green vegetables, fortified cereals and grains

Riboflavin is essential to energy production, cellular growth, and metabolism.²

The takeaway: Its use as migraine prophylaxis has limited data,⁹⁷ but there is otherwise no evidence to support health benefits of riboflavin supplementation.

Vitamin B3

Vitamers: Nicotinic acid (niacin); nicotinamide (niacinamide); nicotinamide riboside

Physiologic role: Converted to nicotinamide adenine dinucleotide (NAD), which is widely required in most cellular metabolic redox processes. Crucial to the synthesis and metabolism of carbohydrates, fatty acids, and proteins

Dietary sources: Poultry, beef, fish, nuts, legumes, grains. (Tryptophan can also be converted to NAD.)

Niacin is readily converted to NAD, an essential coenzyme for multiple catalytic processes in the body. While niacin at doses more than 100 times the recommended dietary allowance (RDA; 1-3 g/d) has been extensively studied for its role in dyslipidemias,² pharmacologic dosing is beyond the scope of this article.

The takeaway: There is no evidence supporting a clinical benefit from niacin supplementation.

Vitamin B5

Vitamers: Pantothenic acid; pantethine

Physiologic role: Required for synthesis of coenzyme A (CoA) and acyl carrier protein, both essential in fatty acid and other anabolic/catabolic processes

Dietary sources: Almost all plant/animal-based foods. Richest sources include beef, chicken, organ meats, whole grains, and some vegetables

Pantothenic acid is essential to multiple metabolic processes and readily available in sufficient amounts in most foods.² Although limited RCT data suggest pantethine may improve lipid measures,^12,98,99 pantothenic acid itself does not seem to share this effect.

The takeaway: There is no data that supplementation of any form of vitamin B5 has any patient-oriented clinical benefits.

Continue to: Vitamin B6

Vitamers: Pyridoxine; pyridoxamine; pyridoxal

Physiologic role: Widely involved coenzyme for cognitive development, neurotransmitter biosynthesis, homocysteine and glucose metabolism, immune function, and hemoglobin formation

Dietary sources: Fish, organ meats, potatoes/starchy vegetables, fruit (other than citrus), and fortified cereals

Pyridoxine is required for numerous enzymatic processes in the body, including biosynthesis of neurotransmitters and homeostasis of the amino acid homocysteine.² While overt deficiency is rare, marginal insufficiency may become clinically apparent and has been associated with malabsorption, malignancies, pregnancy, heart disease, alcoholism, and use of drugs such as isoniazid, hydralazine, and levodopa/carbidopa.² Vitamin B6 supplementation is known to decrease plasma homocysteine levels, a theorized intermediary for cardiovascular disease; however, studies have failed to consistently demonstrate patient-­oriented benefits.^100-102 While observational data has suggested a correlation between vitamin B6 status and cancer risk, RCTs have not supported benefit from supplementation.^14-16 Potential effects of vitamin B6 supplementation on cognitive function have also been studied without observed benefit.^17,18

The takeaway: Vitamin B6 is recommended as a potential treatment option for nausea in pregnancy.¹⁹ Otherwise, vitamin B6 is readily available in food, deficiency is rare, and no patient-oriented evidence supports supplementation in the general population.

Vitamin B7

Vitamers: Biotin

Physiologic role: Cofactor in the metabolism of fatty acids, glucose, and amino acids. Also plays key role in histone modifications, gene regulation, and cell signaling

Dietary sources: Widely available; most prevalent in organ meats, fish, meat, seeds, nuts, and vegetables (eg, sweet potatoes). Whole cooked eggs are a major source, but raw eggs contain avidin, which blocks absorption

Biotin serves a key role in metabolism, gene regulation, and cell signaling.² Biotin is known to interfere with laboratory assays— including cardiac enzymes, thyroid studies, and hormone studies—at normal supplementation doses, resulting in both false-positive and false-negative results.¹⁰³

The takeaway: No evidence supports the health benefits of biotin supplementation.

Vitamin B9

Vitamers: Folates; folic acid

Physiologic role: Functions as a coenzyme in the synthesis of DNA/RNA and metabolism of amino acids

Dietary sources: Highest content in spinach, liver, asparagus, and brussels sprouts. Generally found in green leafy vegetables, fruits, nuts, beans, peas, seafood, eggs, dairy, meat, poultry, grains, and fortified cereals.

Continue to: Vitamin B12

Vitamers: Cyanocobalamin; hydroxocobalamin; methylcobalamin; adenosylcobalamin

Physiologic role: Required for red blood cell formation, neurologic function, and DNA synthesis

Dietary sources: Only in animal products: fish, poultry, meat, eggs, and milk/dairy products. Not present in plant foods. Fortified cereals, nutritional yeast are sources for vegans/vegetarians.

Given their linked physiologic roles, vitamins B9 and B12 are frequently studied together. Folate and cobalamins play key roles in nucleic acid synthesis and amino acid metabolism, with their most clinically significant role in hematopoiesis. Vitamin B12 is also essential to normal neurologic function.²

The US Preventive Services Task Force (USPSTF) recommends preconceptual folate supplementation of 0.4-0.8 mg/d in women of childbearing age to decrease the risk of fetal neural tube defects (grade A).²¹ This is supported by high-quality RCT evidence demonstrating a protective effect of daily folate supplementation in preventing neural tube defects.²² Folate supplementation’s effect on other fetal birth defects has been investigated, but no benefit has been demonstrated. While observational studies have suggested an inverse relationship with folate status and fetal autism spectrum disorder,^23-25 the RCT data is mixed.²⁶

A potential role for folate in cancer prevention has been extensively investigated. An expert panel of the National Toxicology Program (NTP) concluded that folate supplementation does not reduce cancer risk in people with adequate baseline folate status based on high-quality meta-analysis data.^27,104 Conversely, long-term follow-up from RCTs demonstrated an increased risk of colorectal adenomas and cancers,^28,29 leading the NTP panel to conclude there is sufficient concern for adverse effects of folate on cancer growth to justify further research.¹⁰⁴

While observational studies have found a correlation of increased risk for disease with lower antioxidant serum levels, RCTs have not demonstrated a reduction in disease risk with supplementation.

Given folate and vitamin B12’s ­homocysteine-reducing effects, it has been theorized that supplementation may protect from cardiovascular disease. However, despite extensive research, there remains no consistent patient-oriented outcomes data to support such a benefit.^31,32,105

The evidence is mixed but generally has found no benefit of folate or vitamin B12 supplementation on cognitive function.^18,33-35 Finally, RCT data has failed to demonstrate a reduction in fracture risk with supplementation.^36,106

The takeaway: High-quality RCT evidence demonstrates a protective effect of preconceptual daily folate supplementation in preventing neural tube defects.²² The USPSTF recommends preconceptual folate supplementation of 0.4-0.8 mg/d in women of childbearing age to decrease the risk of fetal neural tube defects.

Continue to: ANTIOXIDANTS

In addition to their individual roles, vitamins A, E, and C are antioxidants, functioning to protect cells from oxidative damage by free radical species.² Due to this shared role, these vitamins are commonly studied together. Antioxidants are hypothesized to protect from various diseases, including cancer, cardiovascular disease, dementia, autoimmune disorders, depression, cataracts, and age-related vision decline.^2,37,107-112

Though observational studies have found a correlation of increased risk for disease with lower antioxidant serum levels, RCTs have not demonstrated a reduction in disease risk with supplementation and, in some cases, have found an increased risk of mortality. While several studies have found potential benefit of antioxidant use in reducing colon and breast cancer risk,^38,113-115 vitamins A and E have been associated with increased risk of lung and prostate cancer, respectively.^47,110 Cardiovascular disease and antioxidant vitamin supplementation has similar inconsistent data, ranging from slight benefit to harm.^2,116 After a large Cochrane review in 2012 found a significant increase in all-cause mortality associated with vitamin E and ­beta-carotene,¹¹⁷ the USPSTF made a specific recommendation against supplementation of these vitamins for the prevention of cardiovascular disease or cancer (grade D).¹¹⁸ Given its limited risk for harm, vitamin C was excluded from this recommendation.

Vitamin A

Vitamers: Retinol; retinal; retinyl esters; provitamin A carotenoids (beta-carotene, alpha-carotene, beta-cryptoxanthin)

Physiologic role: Essential for vision and corneal development. Also involved in general cell differentiation and immune function

Dietary sources: Liver, fish oil, dairy, and fortified cereals. Provitamin A sources: leafy green vegetables, orange/yellow vegetables, tomato products, fruits, and vegetable oils

Retinoids and their precursors, carotenoids, serve a critical function in vision, as well as regulating cell differentiation and proliferation throughout the body.² While evidence suggests mortality benefit of supplementation in populations at risk of deficiency,⁴⁵ wide-ranging studies have found either inconsistent benefit or outright harms in the developed world.

The takeaway: Given the USPSTF grade “D” recommendation and concern for potential harms, supplementation is not recommended in healthy patients without risk factors for deficiency.²

Vitamin E

Vitamers: Tocopherols (alpha-, beta-, ­gamma-, delta-); tocotrienol (alpha-, beta-, gamma-, delta-)

Physiologic role: Antioxidant; protects polyunsaturated fats from free radical oxidative damage. Involved in immune function, cell signaling, and regulation of gene expression

Dietary sources: Nuts, seeds, vegetable oil, green leafy vegetables, and fortified cereals

Vitamin E is the collective name of 8 compounds; alpha-tocopherol is the physiologically active form. Vitamin E is involved with cell proliferation as well as endothelial and platelet function.²

The takeaway: Vitamin E supplementation’s effects on cancer, cardiovascular disease, ophthalmologic disorders, and cognition have been investigated; data is either lacking to support a benefit or demonstrates harms as outlined above. Given this and the USPSTF grade “D” recommendation, supplementation is not recommended in healthy patients.²

Vitamin C

Vitamers: Ascorbic acid

Physiologic role: Required for synthesis of collagen, L-carnitine, and some neurotransmitters. Also involved in protein metabolism

Dietary sources: Primarily in fruits and vegetables: citrus, tomato, potatoes, red/green peppers, kiwi fruit, broccoli, strawberries, brussels sprouts, cantaloupe, and fortified cereals

Vitamin C supplementation at the onset of illness does not seem to have benefit.

Ascorbic acid is a required cofactor for biosynthesis of collagen, neurotransmitters, and protein metabolism.² In addition to the shared hypothesized benefits of antioxidants, vitamin C supplementation has undergone extensive research into its potential role in augmenting the immune system and preventing the common cold. Systematic reviews have found daily vitamin C supplementation of at least 200 mg did not affect the incidence of the common cold in healthy adults but may shorten duration and could be of benefit in those exposed to extreme physical exercise or cold.⁴⁸ Vitamin C supplementation at the onset of illness does not seem to have benefit.⁴⁸ Data is insufficient to draw conclusions about a potential effect on pneumonia incidence or severity.^119,120

The takeaway: Overall, data remain inconclusive as to potential benefits of vitamin C supplementation, although risks of potential harms are likely low.

Continue to: Vitamin D

Vitamers: Cholecalciferol (D3); ergocalciferol (D2)

Physiologic role: Hydroxylation in liver and kidney required to activate. Promotes dietary calcium absorption, enables normal bone mineralization. Also involved in modulation of cell growth, and neuromuscular and immune function

Dietary sources: Few natural dietary sources, which include fatty fish, fish liver oils; small amount in beef liver, cheese, egg yolks. Primary sources include fortified milk and endogenous synthesis in skin with UV exposure

Calciferol is a fat-soluble vitamin required for calcium and bone homeostasis. It is not naturally available in many foods but is primarily produced endogenously in the skin with ultraviolet light exposure.²

The AAP recommends supplementing exclusively breastfed infants with 400 IU/d of vitamin D to prevent rickets.

Bone density and fracture risk reduction are the most often cited benefits of vitamin D supplementation, but this has not been demonstrated consistently in RCTs. Multiple systematic reviews showing inconsistent benefit of vitamin D (with or without calcium) on fracture risk led the USPSTF to conclude that there is insufficient evidence (grade I) to issue a recommendation on vitamin D and calcium supplementation for primary prevention of fractures in postmenopausal women.^49-51 Despite some initial evidence suggesting a benefit of vitamin D supplementation on falls reduction, 3 recent systematic reviews did not demonstrate this in community-dwelling elders,^54-56 although a separate Cochrane review did suggest a reduction in rate of falls among institutionalized elders.⁵⁷

The takeaway: Given these findings, the USPSTF has recommended against (grade D) vitamin D supplementation to prevent falls in community-dwelling elders.⁵⁵

Beyond falls. While the vitamin D receptor is expressed throughout the body and observational studies have suggested a correlation between vitamin D status and many outcomes, extensive RCT data has generally failed to demonstrate extraskeletal benefits from supplementation. Meta-analysis data have demonstrated potential reductions in acute respiratory infection rates and asthma exacerbations with vitamin D supplementation. There is also limited evidence suggesting a reduction in preeclampsia and low-birthweight infant risk with vitamin D supplementation in pregnancy. However, several large meta-analyses and systematic reviews have investigated vitamin D supplementation’s effect on all-cause mortality and found no consistent data to support an association.^41,58-62

Multiple systematic reviews have investigated and found high-quality evidence demonstrating no association between vitamin D supplementation and cancer^41,63-66,121 or cardiovascular disease risk.^41,70,71 There is high-quality data showing no benefit of vitamin D supplementation for multiple additional diseases, including diabetes, cognitive decline, depression, pain, obesity, and liver disease.^{43,72-75,85-90,122}

The takeaway: Due to poor availability in breastmilk, the American Academy of Pediatrics (AAP) recommends supplementing exclusively breastfed infants with 400 IU/d of vitamin D to prevent rickets.¹²³ RCT data support high-dose supplementation of lactating women (6400 IU/d) as an alternative strategy to supplementation of the infant.¹²⁴ The AAP recommends that all nonbreastfeeding infants and older children ingesting < 1000 mL/d of vitamin D–fortified formula or milk should also be supplemented with 400 IU/d of vitamin D.¹²³ Despite these universal recommendations for supplementation, evidence is mixed on the effect of vitamin D supplementation on bone health in children.^52,53

Although concerns about vitamin D supplementation and increased risk of urolithiasis and hypercalcemia have been raised,^51,62,121 systematic reviews have not demonstrated significant, clinically relevant risks, even with high-dose supplementation (> 2800 IU/d).^125,126

Vitamin K

Vitamers: Phylloquinone (K1); menaquinones (K2)

Physiologic role: Coenzyme for synthesis of proteins involved in hemostasis and bone metabolism

Dietary sources: Phylloquinone is found in green leafy vegetables, vegetable oils, some fruits, meat, dairy, and eggs. Menaquinone is produced by gut bacteria and present in fermented foods

Vitamin K includes 2 groups of similar compounds: phylloquinone and menaquinones. Unlike other fat-soluble vitamins, vitamin K is rapidly metabolized and has low tissue storage.²

Children taking multivitamins were often found to have excess levels of potentially harmful nutrients, such as retinol, zinc, and folic acid.

Administration of vitamin K 0.5 to 1 mg intramuscularly (IM) to newborns is standard of care for the prevention of vitamin K deficiency bleeding (VKDB). This is supported by RCT data demonstrating a reduction in classic VKDB (occurring within 7 days)⁹¹ and epidemiologic data from various countries showing a reduction in late-onset VKDB with vitamin K prophylaxis programs.¹²⁷ Oral dosing appears to reduce the risk of VKDB in the setting of parental refusal but is less effective than IM dosing.^128,129

Vitamin K’s effects on bone density and fracture risk have also been investigated. Systematic reviews have demonstrated a reduction in fracture risk with vitamin K supplementation,^92,93 and European and Asian regulatory bodies have recognized a potential benefit on bone health.² The FDA considers the evidence insufficient at this time to support such a claim.² Higher dietary vitamin K consumption has been associated with lower risk of cardiovascular disease in observational studies⁹⁴ and supplementation was associated with improved disease measures,¹³⁰ but no patient-oriented outcomes have been demonstrated.¹³¹

The takeaway: The administration of vitamin K 0.5 to 1 mg intramuscularly (IM) to newborns is standard of care for the prevention of VKDB. Vitamin K may lead to a reduction in fracture risk, but the FDA considers the evidence insufficient. Vitamin K’s potential link to a lowered risk of cardiovascular disease has not been demonstrated with patient-­oriented outcomes. Vitamin K has low potential for toxicity, although its interaction with vitamin K antagonists (ie, warfarin) is clinically relevant.

Continue to: MULTIVITAMINS

Multivitamins are often defined as a supplement containing 3 or more vitamins and minerals but without herbs, hormones, or drugs.¹³² Many multivitamins do contain additional substances, and some include levels of vitamins that exceed the RDA or even the established tolerable upper intake level.¹³³

Safe medication storage should be practiced, as multivitamins with iron are a leading cause of poisoning in children.

A 2013 systematic review found limited evidence to support any benefit from multivitamin supplementation.⁴¹ Two included RCTs demonstrated a narrowly significant decrease in cancer rates among men, but saw no effect in women or the combined population.^134,135 This benefit appears to disappear at 5 years of follow-up.¹³⁶ RCT data have shown no benefit of multivitamin use on cognitive function,⁹⁵ and high-quality data suggest there is no effect on all-cause mortality.¹³⁷ Given this lack of supporting evidence, the USPSTF has concluded that there is insufficient evidence (grade I) to recommend vitamin supplementation in general to prevent cardiovascular disease or cancer.⁴¹

The use of prenatal multivitamins is generally recommended in the pregnancy and preconception period and has been associated with reduced risk of autism spectrum disorders, pediatric cancer rates, small-for-gestational-age infants, and multiple birth defects in offspring; however, studies have not examined if this benefit exceeds that of folate supplementation alone.^138-140 AAP does not recommend multivitamins for children with a well-balanced diet.¹⁴¹ Of concern, children taking multivitamins were often found to have excess levels of potentially harmful nutrients such as retinol, zinc, and folic acid.¹⁴²

The takeaway: There is limited evidence to support any benefit from multivitamin supplementation. Prenatal multivitamins are generally recommended in the pregnancy and preconception period. Overall, the risks of multivitamins are minimal, although that risk is dependent on the multivitamin’s constituent components.¹⁴³ Components such as vitamin K may interact with a patient’s medications, and multivitamins have been shown to reduce the circulating levels of antiretrovirals.¹⁴⁴ Specifically, multivitamins with iron should be avoided in men and postmenopausal women, and safe medication storage should be practiced as multivitamins with iron are a leading cause of poisoning in children.²

SUMMARY

Vitamin supplementation in the developed world remains common despite a paucity of RCT data supporting it. Supplementation of folate in women planning to conceive, vitamin D in breastfeeding infants, and vitamin K in newborns are well supported by clinical evidence. Otherwise, there is limited evidence supporting clinically significant benefit from supplementation in healthy patients with well-balanced diets—and in the case of vitamins A and E, there may be outright harms.

CORRESPONDENCE
Joel Herness, MD, 4700 North Las Vegas Boulevard, Nellis AFB, NV 89191; [email protected]