Thiamine Supplementation for the Treatment of Heart Failure: A Review of the Literature


Address for correspondence: James J. DiNicolantonio, PharmD, Wegmans Pharmacy, 500 South Meadow Street, Ithaca, NY 14850



A systematic review of the literature was performed by searching Pubmed and EMBASE databases using the terms “thiamine,” “vitamin B1,” “heart failure,” “systolic dysfunction,” “ventricular dysfunction,” “cardiomyopathy,” “ventricular failure,” and “systolic failure.” Relevant trials and articles were evaluated pertaining to thiamine deficiency in patients with heart failure (HF) and references were searched for further inclusion of articles. A total of 20 articles were reviewed and summarized in detail. While more research is needed to fully elucidate the clinical thiamine deficiency in HF patients, recent evidence has indicated that supplementing with thiamine in HF patients has the potential to improve left ventricular ejection fraction. Thiamine deficiency appears to be not uncommon in patients with HF, and supplementation with thiamine has been shown to improve cardiac function, urine output, weight loss, and signs and symptoms of HF. Therefore, this simple therapy should be tested in large-scale randomized clinical trial to further determine the effects of thiamine in HF patients.

Heart failure (HF) is a serious public health concern, with an estimated 5 million people in the United States having the condition.[1, 2] Additionally, more than 550,000 people are diagnosed with this condition each year.[1, 2] The availability of effective and proven therapies for HF has been steadily increasing the life expectancy of these patients, and thus the prevalence of chronic HF has risen. Although angiotensin-converting enzyme inhibitors, β-blockers, aldosterone antagonists, implantable cardiodefibrillator/cardiac resynchronization pacemaker devices, and omega-3 fatty acids have improved morbidity and mortality in this population, mortality rates remain high.[3] Thiamine supplementation may be an adjunct therapy in order to improve the prognosis and quality of life of patients with HF. For several reasons, as discussed in this review, patients with HF may have micronutrient deficiencies, one of which is thiamine deficiency. Thus, a systematic review of articles and trials pertaining to thiamine deficiency in HF patients was performed. The literature search yielded 154 titles, of which 22 were reviewed in full text on the basis of the inclusion criteria (Figure1). Of these, 20 articles were deemed eligible for inclusion and appropriately summarized. The articles included 8 reviews,[4-11] an editorial,[12] a hypothesis article,[13] two case reports,[14, 15] and eight research articles.[16-23]

Figure 1.

Process for selecting included trials.


Thiamine was discovered in 1926 by two Dutch scientists, Drs Jansen and Donath, and was isolated in a pure form and synthesized in a laboratory by Dr Williams.[4] Throughout history, thiamine has been given multiple names: vitamine, aneurin, and antineuritic vitamin. The public health prominence of thiamine, however, pre-dates its discovery. The disease caused by thiamine deficiency, beriberi, was widespread in the Asian countries as early as the 19th century. It was postulated even before the discovery of thiamine that beriberi was a nutritional deficiency that was caused by the ingestion of polished rice. The origin of the term beriberi is unclear. Two possible sources of the term may be from the Sinhalese meaning “I cannot” or from the Arabic meaning “sailor's asthma.”[11] It was only in the 1950s that interventions to encourage the use of whole grains were developed.

Thiamine deficiency was, at one time, widespread in the developing world as a result of the exclusive use of polished rice as a staple diet in many Asian countries. With the realization that polished rice could lead to thiamine deficiency, it was somewhat controlled. Until recently, thiamine deficiency was considered to be a disease of historical importance only in the developed world. However, it is now realized that a large number of certain populations may be at high risk for developing this deficiency, including HF; therefore, the interest in thiamine and thiamine deficiency has recently been reemerging.


There are two major sources of thiamine: dietary intake and bacterial production.[11] Bacterial production of thiamine is extremely small and is not as clinically relevant as dietary intake. Thiamine is present in small quantities in several food groups: wheat, rice, yeast, beef, pork, poultry, fish, milk, green leafy vegetables, nuts, and seeds.[4, 5] Unfortunately, however, thiamine can be stripped away from the whole grains during processing. Thiamine is a water-soluble vitamin and therefore the quantity of thiamine reserves in the lipid structures of body cells is quite low, with the maximum storage capacity being 30 mg.[4, 12] Thiamine stored in the body is depleted within 2 weeks of a thiamine-deficient diet[4] and clinical signs and symptoms appear in almost 3 months of a thiamine-deficient diet.[5] However, the features of Wernicke-Korsakoff syndrome can appear as early as 3 to 4 weeks after the onset of a thiamine-deficient diet.[8] Excess thiamine is excreted in the urine, whereas in thiamine deficiency, it is generally absent from the urine. However, a patient can be clinically (or subclinically) thiamine-deficient despite a “normal” serum and urinary thiamine excretion level. Therefore, a steady supply of thiamine intake is required regardless of whether urinary thiamine excretion is high. The sources of thiamine are listed in Table 1.

Table 1. Sources of Thiamine[4, 5]
Whole grains and unprocessed wheat
Green leafy vegetables
Orange juice
Tomato juice
Intestinal flora (minimal)
Supplemental drugs


Thiamine exists in 4 forms in the human body: unphosphorylated thiamine, thiamine monophosphate, thiamine diphosphate, and thiamine triphosphate.[14] Thiamine acts as a coenzyme for oxidation-reduction reactions in the body, especially glucose metabolism, the pentose shunt, and the citric acid cycle.[4, 5, 14] Thiamine diphosphate, also called thiamine pyrophosphate, is involved in oxidative decarboxylation in the mitochondrion[6] and is involved in more than 24 enzymatic reactions in the body.[8] Most importantly, it acts as coenzyme in the reactions catalyzed by the enzymes pyruvate dehydrogenase, alpha-ketoglutarate dehydrogenase, and transketolase (TK).[13] Thiamine, in the form of free thiamine and thiamine monophosphate, is actively transported into the central nervous system and the nerves, where it is involved in maintaining sodium and potassium gradients required for conducting nerve impulses.[4, 6]

Absorption and Excretion

Thiamine is absorbed in the jejunum and ileum by active as well as passive uptake.[4, 11] At low levels of thiamine concentration that are normally found in the diet, most of the uptake is by active transport. Decline in thiamine absorption occurs at intakes above 5 mg/d, and the recommended dietary intake is 1 to 1.5 mg/d.[4] After absorption, thiamine travels to the liver from where it enters the red blood cells (RBCs) by facilitated transport. Since thiamine is not protein-bound, it is easily filtered at the glomerulus. In conditions of excess of thiamine, a major determinant of thiamine excretion is urine flow; the higher the flow, the higher amount of thiamine will be excreted.[12] Excretion of thiamine takes place in the distal nephron, and any factors that increase the urine flow rate would increase thiamine excretion and may predispose individuals to thiamine deficiency.

Thiamine Deficiency

Thiamine deficiency can lead to two very different disorders: dry beriberi and wet beriberi. Dry beriberi involves neurological complications, whereas wet beriberi involves cardiovascular (CV) complications. Wernike-Korsakoff syndrome is closely related to dry beriberi. Within 1 week of the depletion of thiamine stores of the body, the blood-brain barrier is interrupted and cerebral hypoperfusion occurs, thus giving rise to the signs and symptoms of Wernike-Korsakoff syndrome.[8]

The possible pathogenesis behind the development of wet beriberi from thiamine deficiency is based on the depletion of adenosine triphosphate (ATP) from the cardiac myocytes and the enhanced production of adenosine in cardiac myocytes leading to its release in the plasma.[13] The ATP depletion causes weakening of the cardiac muscle function, eventually leading to HF. Failure of the myocytes to produce ATP causes accumulation of adenosine monophosphate, which is converted to adenosine. This enhanced production of intracellular adenosine causes its buildup in the cells, eventually causing its release into the plasma by a nucleoside transporter in the cell wall. Adenosine in the plasma manifests as systemic vasodilatation, flushing, and headache. The impairment of reactions (from thiamine depletion) causes a blockade in the citric acid cycle, thus preventing the conversion of pyruvate to acetyl-CoA and ultimately ATP formation, causing cellular acidosis and increasing intracellular free fatty acid levels.[6, 13] The lack of ATP forces the body to upregulate glycolysis and use body fat resources, such as subcutaneous fat, to fulfill its energy requirements. Even though the thiamine-based reactions are still blocked, the use of body fat resources proceeds to not only provide energy to the body but also produce ketones, which can be used for production of acetyl-CoA for extra-hepatic tissues. However, these body fat resources eventually run out in chronic thiamine deficiency if other sources of acetyl-CoA such as ethanol are not being used. Therefore, stopping alcohol ingestion in patients with thiamine deficiency may actually result in rapid death of the patient. The pyruvate accumulation, resulting from increased production (caused by enhanced glycolysis) and decreased utilization (caused by impaired conversion to acetyl-CoA), results in its conversion to lactate. The increased production of lactate, in turn, can lead to lactic acidosis. This entire process severely disturbs the mechanics of the normal CV system. The ventricular filling pressures are increased along with increased oxygen consumption.[5] The resistance vessels are damaged, which causes decreased peripheral vascular resistance, leading to arteriovenous shunting of blood, increased cardiac output, and venous congestion.


The features of thiamine deficiency (Table 2) in any patient may take one of two forms depending on whether the patient has dry or wet beriberi.[4] Patients with dry beriberi have neurologic symptoms in the central and peripheral nervous systems, manifesting as Wernike-Korsakoff syndrome or a peripheral neuropathy, respectively. Patients with Wernike-Korsakoff syndrome typically present with occulomotor abnormalities, ataxia, delirium, and global confusion.[4, 8] On the other hand, patients with wet beriberi have CV symptoms, which manifest as high-output or low-output cardiac failure (generally the former), systemic vasodilatation, lactic acidosis, edema, and fluid retention.[4, 12, 13] Additionally, these patients can have signs and symptoms of increased catecholamine levels,[13] low diastolic pressure, and a widened pulse pressure resulting from the severe CV disturbances.[5] These typical features of wet beriberi may be modified in the presence of other cardiac pathologies or in severe cases where hypotension ensues.[16]

Table 2. Features of Thiamine Deficiency[4, 5, 8, 12, 13]
General FeaturesDry Beriberi/Wernike Korsakoff SyndromeWet Beriberi
Psychological symptoms: depression, emotional instability, mood lability, uncooperative behavior, fearfulness and agitationOcculomotor abnormalities: ophthalmoplegia and optic neuropathyHigh-output or low-output cardiac failure
Neurological symptoms: weakness, dizziness, insomnia, memory loss, peripheral neuropathy, pain sensitivity and sonophobiaAtaxia and gait abnormalitiesSystemic vasodilatation, pulmonary and peripheral edema, and fluid retention
Musculoskeletal symptoms: backache, myalgia, and muscular atrophyDelirium, global confusion, psychosis, and comaMetabolic acidosis, lactic acidosis
Gastrointestinal symptoms: anorexia, nausea, vomiting and constipationPeripheral neuropathyPalpitations, widened pulse pressure, hypotension, bradycardia at rest, and sinus arrhythmia
 Movement disorders: myoclonus and chorea 

Wet beriberi can also present as an acute form, called shoshin beriberi. Patients with shoshin beriberi present with acute CV collapse and metabolic acidosis and may lead to death of the patient if thiamine is not urgently injected. It is not uncommon for patients with thiamine deficiency to present with worsening of HF symptoms. In a case reported in the literature,[14] a thiamine-deficient patient with congestive HF (CHF) presented with dyspnea, orthopnea, edema, paroxysmal nocturnal dyspnea, bilateral pulmonary rales, and a decreased left ventricular ejection fraction (LVEF; 31%). Serum thiamine level was low while urinary thiamine excretion was high, despite the patient consuming more than 30 servings of thiamine-rich foods each week, which is considered to be adequate. This may be attributed to the patient taking furosemide for 2 years. The patient showed a dramatic improvement in LVEF (49%) after a week of intravenous (IV) thiamine of 200 mg/d and re-adjustment of his CV medications.

A similar case reported in the literature presented with HF secondary to idiopathic-dilated cardiomyopathy, which, despite the optimal medical management, was progressively worsening.[15] The patient had orthopnea, hypotension, raised jugular venous pressure, a loud third heart sound, a holosystolic murmur, laterally displaced apical beat, warm extremities, and peripheral edema. The patient's LVEF had decreased to 15% from a 5-month previous measurement of 25%, and his right ventricular systolic pressure was 55 mm Hg. The patient was screened for thiamine deficiency using thiamine pyrophosphate effect (TPPE), which showed severe thiamine deficiency. The patient was treated with IV thiamine 100 mg/d, after which there was a substantial improvement in the patient's condition. Three days after the IV thiamine was started, the LVEF increased to 25% with disappearance of the loud third heart sound. This case points out the importance of considering thiamine deficiency in patients presenting with worsening CHF symptoms and the dramatic improvement that can be achieved by thiamine supplementation in patients with unknown thiamine deficiency. It may be suggested that patients presenting with HF that is refractory to treatment should empirically be given thiamine supplementation. Other nonspecific features of thiamine deficiency may include depression, weakness, dizziness, insomnia, back pain, myalgia, muscular atrophy, anorexia, nausea, vomiting, weight loss, constipation, memory loss, peripheral neuropathy, pain sensitivity, sonophobia, emotional instability, mood lability, uncooperative behavior, and fearfulness with agitation.


The level of thiamine in the body can be measured by either directly detecting the presence of thiamine or by measuring the activity of enzymes dependent on thiamine. Several methods are available for this purpose: serum thiamine level, urinary thiamine level, and erythrocyte transketolase and thiamine pyrophosphate analysis.[4] However, a reliable test for measuring the thiamine stores of the body is still not available. This, coupled with the low risk of adverse events from thiamine supplementation, suggests that empirical thiamine supplementation may be a more practical approach than measuring thiamine and treating only those with documented low values. Shortly after the absorption of thiamine from the jejunum, thiamine travels to the liver from where it enters the erythrocytes. The concentration of thiamine in blood is 60 to 120 μg/L, with almost 80% of this present in erythrocytes.[6] However, because of its short duration of stay and the presence of only 0.8% of the body thiamine stores in the blood, serum thiamine is not a reliable indicator of the body thiamine stores and depends only on the recent thiamine intake. The results of a serum thiamine test also normalize rapidly, as soon as thiamine supplementation has initiated, therefore the level needs to be obtained before the start of treatment and the flaws of this test should be taken into consideration.[11] Since urinary thiamine excretion is directly related to the concentration of thiamine in the blood, it is also dependent only on the recent thiamine intake and does not indicate the body thiamine stores. Measurement of urinary thiamine levels requires a 24-hour urinary sample, which is inconvenient for the patient.

The erythrocyte transketolase activity (ETKA) assay measures the activity of the enzyme transketolase. This test involves the addition of thiamine to the erythrocytes in vitro and then measuring the transketolase activity. Since transketolase is dependent on thiamine diphosphate as a coenzyme for its activity, the action of transketolase reflects the presence of thiamine diphosphate.[5] The results of this assay are expressed as a percentage: 0% to 15% is considered the normal range, 15% to 24% reflects marginal deficiency, and >25% reflects severe deficiency. This is currently the most reliable measurement tool available to diagnose thiamine deficiency. It is a very sensitive test because erythrocytes are among the first cells to be affected in states of thiamine deficiency.[6]

The erythrocyte thiamine pyrophosphate (TPP) analysis directly measures the concentration of thiamine pyrophosphate in an erythrocyte using high-performance liquid chromatography, as opposed to providing indirect evidence as in the ETKA assay, which is a functional test of transketolase enzyme. This test appears to be more specific than the ETKA assay. High-performance liquid chromatography has also been used for measuring thiamine, thiamine diphosphate, and its esters in the erythrocytes,[5, 6] and this test is more sensitive than any of the preceding tests because a drop in the quantity of thiamine esters occurs much earlier than that in unphosphorylated or phosphorylated thiamine.[6] The high cost and prolonged time until the receipt of results of these tests dictate that thiamine supplementation should perhaps be started immediately if thiamine deficiency is suspected.[11] The various methods to measure the thiamine status of the body, along with their advantages and disadvantages, are listed in Table 3.

Table 3. Assays to Measure Thiamine Stores [4-6, 11]
Serum thiamine level

Readily available


Represents short-term thiamine intake

Does not represent long-term thiamine intake

Does not indicate body thiamine stores

A normal level does not rule out thiamine deficiency

Urinary thiamine level

Readily available


Depends on the serum thiamine levels

Does not represent body thiamine stores

Inconvenient because the patient has to collect a 24-hour urine sample

Test results depend on the renal function of the patient

Normal level does not rule out thiamine deficiency

Erythrocyte transketolase activity

Thiamine pyrophosphate effect measures the erythrocyte transketolase activity

Very sensitive testIndirect evidence of the function of transketolase vs direct evidence
Erythrocyte thiamine pyrophosphate analysis

Directly measures thiamine pyrophosphate

Good indicator of body stores as depletion of thiamine occurs at a similar rate in red blood cells as other major organs

Very specific test

Not readily available


High-performance liquid chromatography

Most sensitive test

Measures thiamine status in red blood cells or whole blood

Reproducible results

Not readily available


Risk Factors

Thiamine deficiency in HF patients is multifactorial. It may be present in these patients as a result of reasons unrelated to HF, such as inadequate dietary intake, advancing age, trauma, surgery, fever, institutionalization, and alcohol excess, as well as in patients with malabsorption syndromes, severe infections, eating disorders, cancer, acquired immunodeficiency syndrome, inborn errors of metabolism, gastrointestinal surgery, persistent diarrhea or vomiting, or in those taking drugs such as diuretics.[4, 8, 12] The use of phenytoin, penicillins, cephalosporins, aminoglycosides, tetracycline derivatives, fluoroquinolones, sulfonamide derivatives, and trimethoprim is also associated with thiamine deficiency.[8] The factors associated with thiamine deficiency in patients with HF are the use of diuretics, malnutrition, non-use of thiamine supplements, preserved renal function, severe HF, advanced age, and frequent hospitalizations.[7] The risk factors for thiamine deficiency are listed in Table 4.

Table 4. Risk Factors for Thiamine Deficiency[4, 8, 12]
Diet-RelatedComorbid ConditionsOthers
Inadequate dietary intake, excess alcohol ingestion, malabsorption syndromes, eating disorders, and drugs such as diuretics, phenytoin, penicillins, cephalosporins, aminoglycosides, tetracycline derivatives, fluoroquinolones, sulfonamide derivatives, and trimethoprimHeart failure, severe infections, trauma, surgery, cancer, acquired immunodeficiency syndrome, inborn errors of metabolism, gastrointestinal surgery, fever, and persistent diarrhea or vomitingAdvancing age, institutionalization, and frequent hospitalizations

Thiamine and HF

In addition to the use of diuretics, the presence of advanced age, certain dietary factors, and the presence of some comorbid conditions also predispose patients with HF to thiamine deficiency. Patients with HF experienc early satiety and cachexia, both of which may be responsible for low dietary thiamine intake.[6] Many foods containing sodium are also potential sources of thiamine and end up being avoided by the HF patients when they are advised to follow a low-sodium diet.[6] Patients with HF also have a higher than normal basal metabolic rate, which may, at least theoretically, contribute to low thiamine levels.[16] The increase in venous pressure that is characteristic of CHF increases lymphatic production causing lymphatic obstruction,[17] which impairs absorption from the intestines and thus accelerates thiamine deficiency.

Thiamine has multiple effects on the CV system. It has important hemodynamic effects on the circulatory system as well as direct positive pharmacologic effects on the heart.[16] Thiamine deficiency has been shown to cause cardiac hypertrophy, depressed cardiac contractility, and dysrhythmias.[19] Several studies have examined the role of thiamine supplementation in patients with HF.[5] Clinical trials in patients with CHF have shown that thiamine supplementation increases the systolic, diastolic, and central venous pressures, with a decline in heart rate and increase in LVEF. Thiamine acts as a vasodilator and reduces the afterload on the heart, thus improving cardiac function.[5, 6] Thiamine has also been reported to increase diuresis and natriuresis in patients with HF receiving diuretics—an effect that is of considerable benefit in this population.[9]


Patients with dry and wet beriberi require thiamine supplementation.[4] The recommended dosages are detailed in Table 5. In dry beriberi, the daily dose of thiamine depends on the severity of neuropathy, ranging from 10 to 20 mg/d for 2 weeks for mild neuropathy and up to 20 to 30 mg/d for several weeks for more severe neuropathy. A patient may also need to be placed on IV thiamine supplementation in cases of wet beriberi, typically 100 mg/d IV, for several days. A rapid improvement in the condition of a patient is generally noted after thiamine supplementation. However, a patient may develop temporary hypertension during treatment with thiamine as a result of the sudden closure of arteriovenous shunts causing volume overload.[5] Thiamine supplementation should also be considered prophylactically even before the occurrence of symptoms if additional risk factors for thiamine deficiency are present in an HF patient.[6] It has been recommended that patients at risk for developing thiamine deficiency should be given 100 mg of thiamine 3 times a day, which should then be increased to 200 mg of thiamine 3 times a day in cases of proven deficiency.[11]

Table 5. Dosage for Thiamine Supplementation[4, 6, 11]
Recommended Daily Allowance1.1–1.2 mg/d Orally
Risk of thiamine deficiency100 mg 3 times daily until thiamine levels normalize
Proven thiamine deficiency200 mg 3 times daily until thiamine levels normalize
Alcoholics without encephalopathy50 mg/d orally
Patients on a refined grain diet5–15 mg/d orally
Mild neuropathy10–20 mg/d orally for 2 weeks
Severe neuropathy20–30 mg/d orally for several weeks
Wet beriberi100 mg/d intravenous for several days
Prophylactic dose in HF10–20 mg/d orally


As previously discussed, thiamine deficiency can lead to de novo HF and adversely affect pre-existing HF. Therefore, the prevalence of thiamine deficiency represents a dangerous clinical status. It is known that a substantial portion of the general population experiences thiamine deficiency, which is even more prevalent in patients with HF. In fact, the prevalence of thiamine deficiency in HF patients varies with the individual conditions of each population, their nutritional status, drug use, and the presence of comorbid conditions. Thiamine deficiency is estimated to range from 21% to 98% in patients with HF,[4] with this wide range of prevalence of thiamine deficiency in patients with HF reflecting the differences in the population studied and the thiamine assays used in various surveys. The prevalence is higher in patients with advanced age, taking diuretics, more severe HF, and multiple diseases. The prevalence of thiamine deficiency in patients with HF, as seen in various studies, is given in Table 6.

Table 6. Prevalence of Thiamine Deficiency in Heart Failure
Wooley et al[4]21%–98% in patients with heart failure
Sica[6]3%–96% in patients with heart failure
Allard et al[7]13%–93% in patients with heart failure
Lee et al[10]13%–33% in patients with heart failure
Brady et al[17]21% in patients with heart failure
Levy et al[22]0% in younger patients with stable heart failure
Seligmann et al[23]91% in patients with heart failure and taking diuretics

Loop Diuretics

Loop diuretics have been shown to deplete the body of water-soluble vitamins such as thiamine (also known as thiamin or vitamin B1), which can result in refractory edema and worsening of HF (ie, CV beriberi).[9] The prevalence of thiamine deficiency in HF has been estimated to be anywhere from 21% to 98%.[23-26] There are a plethora of reasons why clinicians may place an individual on a loop diuretic, eg, to eliminate peripheral edema, achieve “dry weight” (ie, prevent or treat weight gain from fluid accumulation), prevent or treat pulmonary congestion, decrease preload, reduce peripheral resistance or afterload, increase cardiac output, relieve dyspnea (on exertion or orthopnea), reduce hospitalizations, or improve survival. While the latter has never been proven in a randomized controlled trial, it is apparent that loop diuretics are mainly used for the symptomatic treatment of patients with CHF. Despite the varying reasons behind the use of loop diuretics, it is important that the appropriateness of their use be substantiated, as these medications are not without side effects. If a loop diuretic is prescribed because it is “common clinical practice,” then continuing treatment should perhaps be questioned. However, if the prescription of a loop diuretic is deemed appropriate and necessary, then an astute clinician must be aware of the side effects of these medications and be able to manage and/or prevent them (one example being thiamine deficiency).

The use of diuretics also leads to thiamine deficiency, because the excretion of thiamine in urine is directly proportional to urine flow.[12] Therefore, any increase in the urine flow, as occurs with the use of diuretics, may lead to an enhanced excretion of thiamine, potentially causing thiamine deficiency. Furosemide has also been shown to decrease the uptake of thiamine by the cardiac myocytes.[6] Other effects of furosemide contributing to the development of thiamine deficiency include furosemide-induced anorexia and decreased intestinal thiamine absorption and cellular uptake caused by furosemide-induced hyponatremia or hypomagnesemia.[23] There has been an association between the use of loop diuretics and thiamine deficiency. Patients who are taking furosemide >80 mg/d or bumetanide >2 mg/d for prolonged durations (>6 months), as are many patients with CHF, are at an especially increased risk for developing thiamine deficiency.[14] Thiamine deficiency produced by loop diuretics is dose-dependent; therefore, patients receiving higher doses of the diuretic are at higher risk for developing the deficiency.[6] In addition, patients without HF taking diuretics and those with HF and polyuria (but not taking diuretics) are also at increased risk for developing thiamine deficiency.

While diuresing a CHF patient with a loop diuretic may be considered clinically appropriate, increased diuresis may predispose the patient to a greater risk for thiamine deficiency, as increased diuresis has been associated with increased renal losses of thiamine.[26-30] Often when prescribing loop diuretics, a side effect commonly recognized by physicians is hypokalemia (or low potassium levels). However, magnesium is required to drive potassium into the cell, which is also depleted by loop diuretics (more so with furosemide vs torsemide).[9] Thus, achieving a normal potassium level with the use of potassium (K) salts or a K-sparring diuretic (ie, eplerenone or spironolactone) seems to be a mainstay in these patients. Despite hypokalemia being a well-recognized side effect of loop diuretics, all of these agents (torsemide, furosemide, and bumetanide) can deplete the body of thiamine, with seemingly greater depletions upon greater increases in diuresis.[26, 28-30] Yet, thiamine depletion caused by loop diuretics seemingly goes unnoticed, as clinicians do not generally measure evidence of thiamine deficiency.

Clinical Trials

A double-blind, randomized, controlled trial was conducted to assess the effect of thiamine supplementation on thiamine status and cardiac muscle function in patients with CHF.[16] The study enrolled 30 patients with CHF and New York Heart Association (NYHA) class II to IV who had been prescribed 80 mg/d furosemide or more for at least 3 months. These patients were admitted to the hospital and randomized to receive 1 week of either IV thiamine 200 mg/d or placebo. After discharge, both groups of patients received oral thiamine 200 mg/d for 6 weeks. Thiamine status of the patients was measured by TPPE. Baseline thiamine levels were not significantly different in both groups (14.0%±10.0% [range, 2.7%–34.4%] in the placebo group; 11.7%±6.5% [range, 1.6%–27.8%] in the IV thiamine group; P=not significant). There was no change in the TPPE or plasma thiamine levels in the placebo group. In the IV thiamine group, the TPPE decreased to 5.4%±3.2% (P~.01) and the plasma thiamine levels increased to 90.4±63.2 μg/L (P<.001) after the intervention. The blood pressure, heart rate, body weight, norepinephrine and atrial natriuretic peptide levels, and end-diastolic volume did not show any change at the end of the intervention. There was a nonsignificant decrease in the end-systolic volume (155±63 mL to 139±67 mL; P<.1) and a significant increase in the mean LVEF (28±11% to 32±9%; P<.05) in the group receiving IV thiamine but not in the placebo group. However, this increase in the LVEF did not correlate with the TPPE levels. At the end of the 7 weeks of the study, the LVEF increased by 22% (33%±11%; P<.01 compared with baseline) and the mean NYHA class decreased from 2.6±0.6 at baseline to 2.2±0.7 (P<.01). TPPE decreased to 4.2%±4.4% (P<.01 compared with baseline) and plasma thiamine levels increased to 50.2±32.4 μg/L (P<.001 compared with baseline). This study demonstrates that thiamine supplementation significantly increases LVEF in patients with HF. Another important finding in this study is that thiamine supplementation in patients with clinically normal thiamine status may be able to reverse underlying subclinical thiamine deficiency, which may improve cardiac function.

A cross-sectional study was conducted to assess the prevalence and predictors of thiamine deficiency in patients with CHF who take loop diuretics.[17] TPPE was used to assess the thiamine stores in these patients and a semiquantitative food frequency questionnaire was used to assess the amount of dietary thiamine intake. The study enrolled 38 patients with CHF with an average LVEF of 22%±8% and an average furosemide dose of 242±216 mg/d. The mean TPPE was 14%±33% and thiamine deficiency was seen in 21% of the patients. The mean thiamine intake was estimated at 0.996 mg/d. Patients with biochemical thiamine deficiency had a lower LVEF (P=.07) than those with a normal thiamine status, and patients with more severe HF were more likely to have thiamine deficiency.

A randomized, double-blind, crossover study was performed to assess the effects of thiamine supplementation in patients with HF.[18] The study enrolled 9 patients who were receiving diuretics as a part of their management for CHF. These patients were randomly assigned to receive either high-dose thiamine supplementation (300 mg/d) or placebo for 28 days. After this time period, there was a wash-out period of 6 weeks, following which the groups were crossed over. Plasma-free thiamine increased from 73±10 ng/mL (range, 60–91 ng/mL) on day 1 of thiamine supplementation to 132±29 ng/mL (range, 102–199 ng/mL) on day 8 (P<.001) of thiamine supplementation. In the group that received thiamine as the first treatment, plasma-free thiamine decreased to 75±4 ng/mL (range, 70–81 ng/mL) in the placebo period. The decrease in the plasma-free thiamine levels in these patients was associated with a decrease in LVEF from 33.8%, during the thiamine supplementation period, to 30.6% in the placebo period. During each period, the LVEF was significantly higher in the thiamine group as compared with the placebo group (32.8% with thiamine and 28.8% with placebo, P=.024). The right ventricular area was also significantly decreased in the group receiving thiamine from its pre-intervention area (16.7–14.8 cm2, P=.026). This study is significant because it indicates that the beneficial effects of thiamine supplementation in patients with severe CHF may also hold true in cases of mild to moderate HF.

Another study investigated the potential use of thiamine in cases of acute decompensated HF.[19] The study enrolled 49 patients who were randomized to receive either 100 mg IV thiamine within 30 minutes of arrival to the hospital or saline, which was used as placebo. There was no difference in the change in dyspnea in the first 4 hours (P=.052) or the duration of hospitalization (P=.721) between the two groups. This indicates that thiamine may not have the same beneficial effect in the management of acute HF as it does in chronic HF.

A study conducted in 23 patients with CHF taking furosemide and 16 controls without CHF and not taking furosemide compared the prevalence of thiamine deficiency in the two groups.[23] TPPE and urinary thiamine excretion were used to measure body thiamine stores. The mean TPPE in patients with HF taking furosemide was significantly higher (27.7%±2.5%) than that in the controls (7.1%±1.6%, P<.001). A similar statistically significant difference could not be seen in the urinary thiamine excretion. In 6 patients with HF who were taking furosemide, IV thiamine was given 100 mg twice a day for 7 days, which was later continued as 200 mg/d oral thiamine supplements. In all patients receiving the thiamine supplements, the elevated TPPE was normalized (27.0%±3.8% to 4.5%±1.3%, P<.001), functional capacity increased by one NYHA class, systolic and diastolic blood pressure increased by 10 mm Hg each, urine output increased by a mean of 500 mL/24 hours, and LVEF increased by 24.0%±4.3% to 37.0%±2.4% in 4 of 5 patients.

Another study sought to investigate the prevalence of thiamine deficiency in younger patients with HF.[22] Thirty-eight patients, with a mean age of 47±10 years and a mean NYHA class of 2.5±0.6, were included in this study. Erythrocyte transketolase activity was used to assess the presence of thiamine deficiency in the patients. However, only 1 patient of the 38 patients enrolled in this study had thiamine deficiency. This could possibly be a simple result of a different age group or it could also be explained on the basis of a different mean duration of use of diuretics in both age groups.

Six patients admitted in the intensive care unit who were being mechanically ventilated because of pulmonary, cardiac, or renal impairment were given thiamine over a period of 2 minutes in a study.[21] A total dose of 50 mg/kg was given incrementally to the patients. A mean rise in the systolic blood pressure of 20 mm Hg and a mean rise in the central venous pressure of 3 mm Hg could be seen after thiamine IV infusion.

The importance of micronutrient deficiencies in patients with HF is demonstrated by a double-blind study where 30 HF patients were randomized to receive either a capsule of high-dose micronutrients (calcium, magnesium, zinc, copper, selenium, vitamin A, thiamine, riboflavin, vitamin B6, folate, vitamin B12, vitamin C, vitamin E, vitamin D, and coenzyme Q10) or placebo for 9 months.[20] Quality of life of patients in both groups was measured using a questionnaire. In comparison with the placebo group, the intervention group showed decreased left ventricular volumes (−13.1% [17.1%] vs ±3.8% [10.0%]; P<.05], increased LVEF [5.3%±1.4%, P<.05), and higher quality-of-life score (+9.5% [1.6%]; P<.05]. The patients taking placebo, however, had a slight deterioration (−1.1% [0.8%]; P=.12] in their quality-of-life score. The results of this study emphasize the importance of maintaining normal micronutrient levels in patients with HF.

A recent meta-analysis of randomized, double-blind, placebo-controlled trials has indicated that thiamine supplementation results in a significant improvement in net change in LVEF (3.28%; 95% confidence interval, 0.64%–5.93%) in patients with systolic HF.[31] Despite the fact that further trials should be performed to confirm the effect of thiamine on morbidity and mortality, thiamine seems to improve LVEF in systolic HF patients and may also provide symptomatic improvement in these individuals.[23, 31]


For many reasons, patients with HF may be at risk for developing certain micronutrient deficiencies, including thiamine deficiency.[32] The focus of management of HF may need to be broadened to include the normalization of these nutritional abnormalities. Patients with HF, especially those in advanced stages, may benefit from regular thiamine supplementation. Because of the low risk of toxicity of thiamine supplementation and its widespread deficiency, this seems to be a reasonable approach. Patients with refractory HF or those with additional risk factors should be screened for thiamine deficiency. Even though thiamine deficiency is known to have negative effects on cardiac function, it is not known whether supplementation in these patients is produced by the elimination of thiamine deficiency or by a positive effect of thiamine on the CV system. Experimental data are not sufficiently strong to prove any possible positive effect of thiamine supplementation in patients with normal thiamine levels. Finally, large-scale randomized controlled clinical trials are needed to further determine the optimal use of thiamine supplementation in various groups of patients with HF.


The authors have no conflicts of interest to declare.