What Is Alpha-Lipoic Acid? Neuropathy Evidence and Safety

At a glance

A Cochrane review pooling placebo-controlled studies lasting at least 6 months (3 RCTs, 816 people) concluded that ALA probably has little or no effect on diabetic neuropathy symptoms. ^[1]
In the 5-week SYDNEY 2 trial, the total symptom score fell more than with placebo. ^[3]
In type 2 diabetes, the mean HbA1c reduction was found to be 0.32 points; the authors found the effects below clinically important thresholds. ^[5]
ALA-associated insulin autoimmune syndrome is a rare cause of hypoglycemia; the SmPC approved by TİTCK lists this syndrome in its warnings section. ^[10]
In Türkiye, ALA is regulated both as a licensed drug and as a food supplement; the supplement upper limit is 600 mg/day for ages 11 and older. ^[11]
This guide does not recommend individual doses.

What is alpha-lipoic acid and what does it do in the body?

According to the SmPC (Turkish KÜB) approved by TİTCK (Turkish Medicines and Medical Devices Agency), alpha-lipoic acid (thioctic acid) is a vitamin-like endogenous (produced in the body) substance that has a coenzyme function in the oxidative decarboxylation of alpha-keto acids. ^[10] The pharmacodynamics section of the SmPC reports that in diabetes, advanced glycation end products and free oxygen radicals resulting from hyperglycemia have been linked to nerve damage, and that in rat experiments ALA interacted with these processes and showed an antioxidant effect in diabetic nerve; these experimental results are said to support the theory that human nerve function may improve. ^[10] Hsieh and colleagues also state, as an introductory sentence, that ALA has been found to improve symptoms by reducing oxidative stress and improving microcirculation. ^[2] This is a mechanistic hypothesis; the sections below address the results of human studies.

According to the SmPC, the absolute bioavailability of oral ALA is about 20% relative to IV administration; the plasma half-life is about 25 minutes, and peak plasma values of about 4 µg/mL are measured 0.5 hours after a 600 mg oral dose. The SmPC also notes that the relative bioavailability of solid pharmaceutical forms is higher than 60% compared with oral solutions, and that intake with food can reduce absorption. ^[10] ALA is not considered an essential nutrient like a vitamin or mineral; there is no defined daily requirement or deficiency.

Does alpha-lipoic acid help diabetic neuropathy?

Short-term studies. The SYDNEY 2 trial by Ziegler and colleagues is a multicenter, double-blind, placebo-controlled RCT: 181 people with diabetes and distal symmetric polyneuropathy in Russia and Israel received ALA 600 mg (45 people), 1,200 mg (47), or 1,800 mg (46) once daily for 5 weeks, or placebo (43). The total symptom score (TSS) did not differ between groups at baseline; it fell by a mean of 4.9 points (51%) in the ALA600 group, 4.5 points (48%) in ALA1200, 4.7 points (52%) in ALA1800, and 2.9 points (32%) in the placebo group (all p < 0.05 versus placebo). The proportion of patients achieving at least a 50% reduction in TSS was 26% in the placebo group, 62% in ALA600, 50% in ALA1200, and 56% in ALA1800. The neuropathy impairment score (NIS) fell numerically; in the safety analysis, dose-related increases in nausea, vomiting, and dizziness were seen. ^[3]

The 2023 meta-analysis by Hsieh and colleagues pooled 10 RCTs (1,242 patients) comparing oral ALA with placebo. ALA gave positive results for the total symptom score (with a dose-dependent trend), the neurological impairment score (NDS), and the overall satisfaction score; it gave no positive results for VAS, vibration perception threshold, NIS-LL, and nerve conduction studies. ^[2] According to the authors’ declaration, these results come from a review with no conflicts of interest.

Cochrane review (at least 6 months). The 2024 Cochrane review by Baicus and colleagues included RCTs comparing ALA with placebo for at least six months; three studies (816 participants) entered the analysis, and all studies had a high risk of attrition bias. Compared with placebo, ALA probably had little or no effect on neuropathy symptoms measured by TSS at six months (mean difference −0.16 points; 95% CI −0.83 to 0.51; 1 study, 330 participants; moderate-certainty evidence); the CI did not include 0.97 points, the minimal clinically important difference. ALA may have little or no effect on the lower-limb neuropathy impairment score (NIS-LL) at six months (mean difference −1.02; 95% CI −2.93 to 0.89; 1 study, 245 participants; low certainty); however, the authors noted that a meaningful benefit cannot be excluded because the lower limit of the CI exceeds the minimal important difference of 2 points. There was probably no difference in adverse events leading to discontinuation of treatment within six months (RR 1.48; 95% CI 0.50–4.35; 3 studies, 1,090 participants; moderate certainty). No study reported quality of life or neuropathy complications. ^[1]

4-year study. In the NATHAN 1 trial by Ziegler and colleagues, 460 patients with mild-to-moderate diabetic sensorimotor polyneuropathy received ALA 600 mg/day (233) or placebo (227). In the primary composite outcome (NIS-LL and seven neurophysiological tests), there was no significant difference between groups at 4 years (p = 0.105); among secondary outcomes, NIS (p = 0.028), NIS-LL (p = 0.05), and the NIS-LL muscle weakness subscore (p = 0.045) favored ALA. The rate of serious adverse events was 38.1% in the ALA group and 28.0% in the placebo group; the tolerability assessment and tolerability-related discontinuations did not differ between groups. According to the full-text disclosure, three of the authors are employees of the company that manufactures the drug, and most of the other authors received consulting fees or grants from the same company. ^[4]

Overall picture: short-term studies showed a fall in symptom score; the Cochrane review pooling studies lasting at least 6 months found probably little or no effect, and the primary outcome of the 4-year study was not significant. For this reason, the positive results are regarded as having limited certainty (grade C), and the long-term findings form a counterweight (grade E).

Does alpha-lipoic acid affect blood sugar and weight?

The dose-response meta-analysis by Jibril and colleagues covered 16 studies (1,035 patients) of oral ALA in adults with type 2 diabetes; each 500 mg/day increase significantly reduced HbA1c, body weight, CRP, fasting glucose, and triglycerides. A J-shaped effect was seen for HbA1c (mean difference −0.32 points; 95% CI −0.45 to −0.18); above 600 mg/day there was a linear decrease in body weight (−0.30 kg; 95% CI −0.57 to −0.04). The authors noted that for all outcomes the point estimates were below the minimal clinical importance thresholds, and that despite significant improvements, the effects were not clinically important. ^[5] The NCCIH summary, in turn, reports that a 2019 review of 10 studies (553 participants) showed ALA was no better than placebo at lowering blood sugar, cholesterol, and triglyceride levels. ^[7] The results of the two reviews do not point in the same direction; there is no common framework for combining them.

The meta-analysis by Kucukgoncu and colleagues pooled 10 double-blind RCTs comparing ALA with placebo; ALA produced a mean 1.27 kg (95% CI 0.25–2.29) greater weight loss and a −0.43 kg/m² (95% CI −0.82 to −0.03) greater BMI reduction than placebo. In meta-regression, dose did not significantly affect weight and BMI change. The authors interpreted this as a small but significant short-term weight loss and noted that more research is needed on doses and long-term outcomes. ^[6]

Does alpha-lipoic acid affect other diabetes complications?

The NCCIH summary reports that in a study in which 467 people with type 2 diabetes received 600 mg/day ALA or placebo for 2 years, there was no statistical difference in the development of clinically significant diabetic macular edema. NCCIH also reports that a 2022 review evaluating 12 studies (653 participants) found that ALA supplementation did not improve kidney function in diabetic nephropathy, with the authors noting that the evidence is limited by the small number of studies and participants. For diabetic neuropathy, NCCIH also reports that in a 2022 review of eight studies (1,500 participants), three studies found improvement in symptoms and five found none; and that another 2022 review of 9 studies (2,062 participants) found that ALA may help reduce pain in diabetic neuropathy. ^[7] NCCIH writes in general that there is weak evidence of possible benefit for a few supplements, that there are some studies suggesting ALA may be useful in reducing pain due to diabetic neuropathy, and that most studies do not lead to a clear conclusion because of size and methodological limitations. ^[7]

Is alpha-lipoic acid safe? Side effects, hypoglycemia, and insulin autoimmune syndrome

According to the SmPC, common side effects are dizziness and nausea; very rarely, taste changes, headache, excessive sweating, visual disturbances, vomiting, stomach and bowel pain, diarrhea, and allergic reactions such as urticaria and itching are reported. Because of increased glucose utilization, a drop in blood glucose (hypoglycemia) can occur very rarely; the frequency of insulin autoimmune syndrome is listed as “not known.” ^[10] The NCCIH summary lists the most common side effects as headache, heartburn, nausea, and vomiting. ^[7]

Insulin autoimmune syndrome (IAS). The SmPC states that cases of IAS have been reported during thioctic acid treatment, and that patients carrying the HLA-DRB104:06 and HLA-DRB104:03 alleles are more predisposed to IAS; the predisposition odds ratio is 1.6 for HLA-DRB104:03 (particularly common in white people and in southern more than northern Europe) and 56.6 for HLA-DRB104:06 (particularly in Japanese and Korean patients). The SmPC states that IAS should be considered in the differential diagnosis of spontaneous hypoglycemia in patients using thioctic acid. ^[10] Gullo and colleagues reported IAS in six European white patients (median age 63; three men, three women) who were taking 600 mg/day ALA, not using insulin, and had no insulinoma: hypoglycemia symptoms appeared 30 to 120 days after starting ALA, hypoglycemic attacks decreased after ALA was stopped, and all patients were given oral or intravenous glucose and prednisone. HLA-DRB104:03 was detected in five patients and HLA-DRB104:06 in one. ^[8] The literature analysis by Li and colleagues covered 37 cases (median age 61; 28 women, 9 men); after ALA was stopped or after corticosteroid treatment, hypoglycemia disappeared within 5 days to 8 months, insulin autoantibodies turned negative at a median of 7 months, and no recurrent attacks were seen over a median follow-up of 7.25 months. The authors describe IAS as the most serious adverse reaction reported with ALA and as a rare autoimmune disease. ^[9]

Interactions and special groups (SmPC). According to the SmPC, the effect of cisplatin may be reduced with ALA; ALA is a metal chelator and should not be administered together with metal-containing products (iron and magnesium preparations, and dairy products because of their calcium content); if the daily dose is taken 30 minutes before breakfast, iron and/or magnesium preparations should be taken at lunch or in the evening. The blood-sugar-lowering effects of insulin and/or oral antidiabetic drugs may increase; blood glucose needs to be monitored regularly, especially at the start of treatment, and in some cases the doses of antidiabetic drugs may need to be reduced. It is contraindicated in children and adolescents because of insufficient clinical experience. There are no clinical data for pregnancy, animal studies have shown no direct or indirect harmful effects, and caution is needed when giving it to pregnant women; whether it is excreted in human milk is unknown. After accidental or suicidal intake of ALA doses of 10 to 40 g together with alcohol, severe intoxication (including seizures and lactic acidosis) that can end in death has been reported. ^[10]

What is the regulatory status in Türkiye?

In Türkiye, ALA falls under two separate frameworks; these two regulations are in separate documents and do not endorse each other.

Food supplement. In the September 21, 2026 version of the Ministry of Agriculture and Forestry’s Restricted Substances List for food supplements, “Alfa Lipoik asit” is listed with cattle and chemical synthesis as sources; in the 11 years and older column the upper limit is 600 mg/day, and the publication date is January 22, 2016. ^[11]

Drug. Alpha-lipoic acid is also available in Türkiye as a licensed drug. The 600 mg film-coated tablet SmPC published in the 2024 TİTCK archive states that it “is indicated in the treatment of symptoms of peripheral (sensorimotor) diabetic polyneuropathy,” defines the daily dose as a single dose equivalent to 600 mg alpha-lipoic acid, and states that the basis of diabetic polyneuropathy treatment is optimal control of diabetes; it also states that in cases with severe symptoms, treatment may be started with an infusion. ^[10] This paragraph relays the drug SmPC; drug use is a decision for a physician, and this guide does not recommend individual doses.

Reimbursement. The Turkish Pharmacists’ Association announcement dated September 28, 2026 reports that, through the Health Implementation Communiqué amendment by SGK (Social Security Institution) published in Official Gazette (Mükerrer) No. 32541 dated May 9, 2024, effective May 17, 2024, the wording “10.12.4. Alpha-lipoic acid (only in peripheral diabetic neuropathy)” in the List of Drugs Exempt from Patient Participation Share (EK-4/D) was changed to “10.12.4. Alpha-lipoic acid (Only G63.2*)”; the announcement states that SGK reported the G63.2* code covers the diagnoses in Article 4.2.35.A of the Communiqué. These sentences are the SGK text as relayed by the Association. ^[12]

This guide does not recommend products or places of supply.

Evidence table: claim, population, and limits

Claim	Group studied	Finding	Limit
Neuropathy symptoms (short-term)	181 people with diabetes (5 weeks); 10 RCTs, 1,242 patients	TSS fell; no positive result for VPT, NIS-LL, and nerve conduction	Short duration; symptom score is subjective
Neuropathy symptoms (≥6 months)	Cochrane, 3 RCTs, 816 people	Probably little or no effect; TSS difference −0.16 (no benefit shown)	All at high risk of bias (attrition)
4-year nerve damage	460 patients	Primary outcome not significant; favored NIS	Serious adverse events 38.1% vs 28.0%
HbA1c and lipids	16 RCTs, 1,035 people with type 2 diabetes	HbA1c −0.32 points	Below clinical importance threshold; another review found no difference
Weight	10 RCTs	1.27 kg greater loss	Small, short-term
Macular edema and nephropathy	467 people (2 years); 12 studies	No difference (no benefit shown)	NCCIH summary; small number of studies
Safety	SmPC, case series	Hypoglycemia; insulin autoimmune syndrome; contraindicated in children	IAS frequency not known

This guide does not recommend individual doses; the numbers above relay the content of studies and official documents.

What does the evidence say?

What it does and does not affect

Limited or insufficient evidence

CDiabetic neuropathy symptoms (short-term studies)In the 5-week SYDNEY 2 trial (181 people), the total symptom score fell more than with placebo; a meta-analysis of 10 RCTs of oral ALA reported positive results for symptom score and overall satisfaction, but not for nerve conduction and vibration threshold.
DHbA1c, glucose, and lipids in type 2 diabetes16 RCTs, 1,035 people: J-shaped mean HbA1c change of −0.32 points; the authors noted that all results stayed below clinical importance thresholds. A 2019 review cited by NCCIH found no difference from placebo.
DWeight lossIn a meta-analysis of 10 RCTs, mean weight loss was 1.27 kg (95% CI 0.25–2.29) greater than with placebo; the authors interpreted this as a small short-term effect.

No benefit shown

EDiabetic neuropathy symptoms and impairment (at least 6 months, Cochrane)no benefit shown3 RCTs, 816 people; compared with placebo, ALA probably has little or no effect on neuropathy symptoms at month 6 (TSS difference −0.16; 95% CI −0.83 to 0.51; moderate certainty) and may have little or no effect on the impairment score (low certainty).
EPreventing diabetic macular edemano benefit shownIn 467 people with type 2 diabetes over 2 years, there was no statistical difference between 600 mg/day ALA and placebo in the development of clinically significant macular edema (NCCIH summary).
EDiabetic nephropathy (kidney function)no benefit shownA 2022 review of 12 studies found no improvement in kidney function; the authors noted that the evidence is limited by the small number of studies and participants (NCCIH summary).

What do the A–E grades mean?

The letter describes the strength of evidence for a specific claim, not the size of the benefit or a usage recommendation. E does not mean ineffectiveness has been proven.

AStrong: Established physiology (e.g. correcting a deficiency) or strong, consistent, direct human studies
BModerate: Supportive human evidence; small effect or some limitations
CLimited: A single randomised trial, mixed results or low-certainty human findings
DInsufficient: Small, uncontrolled or indirect evidence; not enough to judge
ENo benefit shown / not studied in humans: No benefit shown for the claim, or only animal or cell data exist

These grades are Half-Life Journal’s editorial evidence summary; they are not a formal GRADE rating or independent medical approval.

The letter shows evidence strength; it is not a score for positive effect or a usage recommendation.

Frequently asked questions

What does alpha-lipoic acid do?

Alpha-lipoic acid (ALA, thioctic acid) is a vitamin-like substance that is also found in the body. In diabetic neuropathy, short-term studies reported lower symptom scores; the Cochrane review pooling placebo-controlled studies lasting at least 6 months concluded that ALA probably has little or no effect on symptoms.

Does ALA treat diabetic neuropathy?

In the 5-week SYDNEY 2 trial, the symptom score fell more than with placebo; in the 4-year NATHAN 1 trial, the primary composite outcome did not change significantly. In Türkiye, ALA 600 mg tablets are among the licensed drugs for treating symptoms of peripheral diabetic polyneuropathy; this is a separate regulation from supplement claims.

Does ALA lower blood sugar?

In type 2 diabetes, a dose-response meta-analysis of 16 studies found a mean HbA1c reduction of 0.32 points; the authors noted that the effect stayed below clinically important thresholds. The drug information states that ALA can increase the blood-sugar-lowering effect of insulin and oral antidiabetic drugs.

What are the side effects and risks of ALA?

Nausea and dizziness are common, and hypoglycemia is a very rarely reported effect. ALA-associated insulin autoimmune syndrome is a rare cause of hypoglycemia; case series have been reported. According to the drug information, it is contraindicated in children and adolescents.

Is ALA a drug or a supplement in Türkiye?

Both. The Ministry of Agriculture and Forestry's Restricted Substances List sets an upper limit of 600 mg/day for food supplements, while in the SmPCs (Turkish KÜB) approved by TİTCK (Turkish Medicines and Medical Devices Agency), ALA 600 mg tablets appear as a licensed drug.

Sources

Method: Institutional clinical guidance and relevant studies were compared. Each finding is presented with its population, outcome, sources and limitations. Independent medical review has not yet been completed.

Updated 18 October 2026

This content is for information only and is not medical advice. It does not recommend doses, products or vendors. Talk to your clinician before making any health decision.