Literature search

Keywords like “diabetic-neuropathy” AND “alpha lipoic acid” OR “ALA” OR “thioctic acid” AND “randomized controlled trials” were used to search the literature for research published up until December 2022 in a number of online databases, including PubMed, Embase, and Google Scholar. The emphasis was only on published RCTs related to the topic. The publications that satisfied the inclusion criteria were taken into consideration for an SR/MA after all of the articles were reviewed at the abstract and full-length article levels. SD and GLV have performed a literature search, and co-authors were consulted prior to the final selection of articles for the SR and MA.

Inclusion and exclusion criteria

The population, intervention, comparison, outcome, and study design (PICOS) framework was used to define the inclusion and exclusion criteria, summarized in Table 1.

Parameters

Primary outcomes included glycated hemoglobin (HbA1C), NIS, NIS-low limb (NIS-LL), neuropathy symptoms and change (NSC), TSS, motor nerve conduction velocity (MNCV), neuropathy symptom score (NSS), NDS, vibration perception threshold (VPT), and global satisfaction. The secondary outcomes, which are adverse drug reactions, are summarized to conclude the safety of ALA.

Risk of bias assessment

The methodological quality of the included studies was assessed using the Cochrane risk of bias tool (Revman version 5.4.1), which identifies six potential bias issues. Biases were classified as high, low, or unknown for each item, with studies considered to have poor quality if they presented three or more significant risks of bias. Three reviewers (SD, GLV, and AK) independently assessed each study’s profile, and a fourth reviewer was consulted to resolve any discrepancies in the bias rankings.

Quality of evidence assessment (GRADE analysis)

Evidence quality for each parameter was assessed using the GRADE Profiler version 3.6 and the consensus of SD and GLV [14]. Initially rated as high, RCT quality could be downgraded based on factors like inconsistency, publication bias, imprecision, indirectness, and risk of bias. The system classifies evidence into four levels: high, moderate, low, and very low [14–16].

Data collection and analysis

Numerical data was extracted directly from tables, while data from graphs was obtained using the built-in measurement tool in Adobe Acrobat XI. Additionally, a default standard deviation of 10% of the mean was applied for both the ALA and control when the standard deviation was unavailable.

Statistical analysis

MA was conducted using RevMan version 5.4.1, with a fixed/random effects model and mean difference (MD)/standardized MD (SMD) for continuous variables, estimating inverse variance (IV). The pooled prevalence was calculated at a 95% CI, and I² was used to assess heterogeneity. A sensitivity analysis was performed, incorporating only high-quality studies to enhance the robustness of the conclusions.

Selection of articles

After removing duplicates, 146 articles were retained from a total of 354, sourced from databases such as PubMed (121), Embase (166), and Google Scholar (67). Of these, 137 articles were excluded for various reasons, including parenteral routes (10), combination routes (1), studies other than RCTs (45), other than English language (7), ALA given with other drugs (16), irrelevant topics (48), and no placebo group (5). Additionally, only abstracts were available (5). Ultimately, 9 RCTs involving 1,345 participants were included for this MA. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) is depicted in Figure 1.

Display full size

Figure 1. PRISMA flow chart. From 354 initially identified articles (PubMed: 121, Embase: 166, Google Scholar: 67), 208 duplicates were removed. After screening, 137 articles were excluded, resulting in 9 randomized controlled trials. Source: https://www.prisma-statement.org/. This work is licensed under CC BY. PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses; ALA: alpha lipoic acid.

Study characteristics

The SR and MA comprised nine RCTs comparing intervention and placebo groups; the sample sizes ranged from 20 to 460. Participants aged 18 to 74 were included, with studies published between 1999 and 2021 from the USA, India, Egypt, Saudi Arabia, Israel, and Germany. The review focused on individuals with DN, excluding other conditions listed in Table 1. The attributes of the included studies are shown in Table 2.

Two studies, Ametov et al. (2003) [10] and Ziegler et al. (1999) [17], were excluded from the analysis because Ametov et al. (2003) [10] was an intravenous route study, while Ziegler et al. (1999) [17] administered ALA intravenously for an initial 3 weeks followed by oral route exposure up to 6 months (combination of oral and intravenous routes).

Among the included trials, Reljanovic et al. (1999) [12] involved an initial 5-day intravenous course followed by 24 months of oral treatment. As the oral phase represented the primary intervention period, the study was included to capture one of the longest-duration RCTs of oral ALA.

Risk of bias assessment

The risk of bias assessment indicated that only 1 out of the 9 studies included had low-quality evidence [17]. The risk of bias has been illustrated in Figure 2.

Display full size

Figure 2. Risk of bias graph. The risk of bias assessment indicated that only 1 out of the 9 studies included had low-quality evidence.

Quality of evidence for outcomes (GRADE analysis)

The GRADE analysis revealed that the included studies ranged from moderate to high quality, and the analysis is presented in Table 3.

Outcome measures

HbA1C

This test measures glucose control over the past three months, aiding in diabetes diagnosis and monitoring glycemic control [18, 19]. Maintaining controlled HbA1C levels reduces the likelihood of DN. In this context, ALA was linked to a significant reduction in HbA1C levels [IV: –0.66, (–0.81 → –0.51) at 95% CI, p < 0.00001, I² = 92%] compared to placebo (Figure 3).

Display full size

Figure 3. Effect of ALA on glycated hemoglobin (HbA1C) levels in diabetic patients. ALA supplementation significantly reduced HbA1C compared to placebo (IV: –0.66; 95% CI: –0.81 to –0.51; p < 0.00001; I² = 92%), indicating improved glycemic control. ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

NIS

NIS evaluates neuropathy progression, with scores ranging from 0 to 244, where higher scores reflect more impairment [20]. The current MA found that ALA significantly reduced the NIS score [IV: –1.19, (–2.28 → –0.09) at 95% CI, p = 0.03, I² = 58%] compared to placebo (Figure 4).

Display full size

Figure 4. Effect of ALA on neuropathy impairment score (NIS) in diabetic patients. ALA treatment significantly reduced NIS compared to placebo (IV: –1.19; 95% CI: –2.28 to –0.09; p = 0.03; I² = 58%), indicating improvement in neuropathic symptoms. ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

NIS-LL

NIS-LL assesses lower limb neuropathy with 14 evaluations, scoring from 0 to 88, where higher scores denote more impairment [20]. This MA indicated that ALA administration showed a trend toward reducing NIS-LL scores, but the effect was not statistically significant [IV: –0.58, (–1.35 → 0.19) at 95% CI, p = 0.14, I² = 0%] compared to placebo (Figure 5).

Display full size

Figure 5. Effect of ALA on neuropathy impairment score-lower limb (NIS-LL) scores assessing lower limb neuropathy. ALA treatment showed a non-significant trend toward score reduction compared to placebo (IV: –0.58; 95% CI: –1.35 to 0.19; p = 0.14; I² = 0%), suggesting limited impact on lower limb neuropathy. ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

MNCV

This test evaluates nerve conduction capacity by passing the electric impulses through the nerves. In general, the slower and weaker signal indicates that there is damage in the nerve. The available evidence suggests that ALA is not beneficial in improving the MNCV [IV: 0.26, (–0.44 → 0.96) at 95% CI, p = 0.47, I² = 49%] compared to placebo [21] (Figure 6).

Display full size

Figure 6. Effect of ALA on motor nerve conduction velocity (MNCV). ALA showed no significant improvement in MNCV compared to placebo (IV: 0.26; 95% CI: –0.44 to 0.96; p = 0.47; I² = 49%), indicating limited impact on nerve conduction function. ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

NSC number

This score is derived based on the responses given by the patients to a questionnaire containing 38 items. A higher score indicates more severity [22]. This MA found that ALA significantly reduced the NSC number [IV: –0.18, (–0.35 → –0.01) at 95% CI, p = 0.04, I² = 0%] compared to placebo (Figure 7).

Display full size

Figure 7. Effect of ALA on neuropathy symptoms and change (NSC) number. ALA significantly reduced NSC scores compared to placebo (IV: –0.18; 95% CI: –0.35 to –0.01; p = 0.04; I² = 0%), suggesting a modest improvement in patient-reported neuropathic symptoms. ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

NSC severity

It is another parameter related to NSC, where severity is assessed. In this study, ALA significantly improved the NSC severity score [IV: –0.65, (–0.83 → –0.48) at 95% CI, p < 0.00001, I² = 89%] compared to placebo. The outcomes are shown in Figure 8.

Display full size

Figure 8. Effect of ALA on neuropathy symptoms and change (NSC) severity. ALA significantly reduced NSC severity scores compared to placebo (IV: –0.65; 95% CI: –0.83 to –0.48; p < 0.00001; I² = 89%), indicating an improvement in symptom severity. ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

TSS

The TSS is a questionnaire used to assess the frequency and severity of sensory neuropathy symptoms [23]. Notably, ALA significantly improved the TSS [IV: –0.43, (–0.59 → –0.27) at 95% CI, p < 0.00001, I² = 98%] compared to placebo (Figure 9).

Display full size

Figure 9. Effect of ALA on total symptom score (TSS) for sensory neuropathy. ALA significantly reduced TSS compared to placebo (IV: –0.43; 95% CI: –0.59 to –0.27; p < 0.00001; I² = 98%), indicating a notable improvement in symptom frequency and severity. ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

NDS

The NDS is a common tool for screening DN, evaluating vibration, pain, temperature perception, and ankle reflex, with scores from 0 to 10. Scores of 3–5 indicate mild symptoms, 6–8 moderate, and 9–10 severe [24]. This MA found that ALA significantly improved NDS compared to placebo [IV: –0.72, (–1.03 → –0.40) at 95% CI, p < 0.00001, I² = 98%] (Figure 10).

Display full size

Figure 10. Effect of ALA on neuropathy disability score (NDS). ALA significantly improved NDS compared to placebo (IV: –0.72; 95% CI: –1.03 to –0.40; p < 0.00001; I² = 98%), indicating reduced neuropathic disability in diabetic patients. ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

VPT

It is a common method for evaluating large fiber function in diabetic patients is the VPT. Serious DN problems are indicated by a high VPT [25]. Notably, ALA treatment significantly improved VPT compared to placebo [IV: –0.35, (–0.50 → –0.19) at 95% CI, p < 0.0001, I² = 96%] (Figure 11).

Display full size

Figure 11. Effect of ALA on visual perception threshold (VPT). ALA treatment significantly improved VPT compared to placebo (IV: –0.35; 95% CI: –0.50 to –0.19; p < 0.0001; I² = 96%), indicating enhanced large fiber nerve function in diabetic patients. ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

Global satisfaction

Global satisfaction was considered an important parameter for assessing ALA’s therapeutic benefits. Two studies [26, 27] collected global satisfaction ratings from participants using a 3-point scale: insufficient, satisfactory, and good/very good. Our analysis focused on the percentage of ratings categorized as “good/very good” as reported in the sources. These percentages were used to determine the number of occurrences of “good/very good” ratings relative to the total number of subjects. Interestingly, the ALA treatment has shown significant improvement in global satisfaction rate when compared to placebo [Mantel-Haenszel (M-H): 3.51, (1.06 → 11.61) at 95% CI, p = 0.04, I² = 72%] (Figure 12).

Display full size

Figure 12. Effect of ALA on Global satisfaction ratings. ALA treatment significantly increased the proportion of participants rating their experience as “good/very good” compared to placebo (M-H: 3.51; 95% CI: 1.06 to 11.61; p = 0.04; I² = 72%), reflecting enhanced perceived therapeutic benefit. ALA: alpha lipoic acid; M-H: Mantel-Haenszel; CI: confidence interval.

Sensitivity analysis

To enhance the robustness of our analysis, the MA was re-performed, excluding studies identified as having a high risk of bias. Specifically, the studies by Reljanovic et al. (1999) [12], Vijayakumar et al. (2014) [18], and Siddique et al. (2021) [28] were deemed to have a high risk of bias. Sensitivity analyses were conducted for the parameters MNCV and TSS to evaluate the influence of these high-risk studies on the study’s overall outcomes. However, sensitivity analysis cannot be performed for HbA1C because only four studies are available.

The sensitivity analysis revealed minor changes in effect magnitudes when comparing the MA results excluding vs. including high-risk studies for MNCV [IV: 0.32, (–0.42 → 1.07) at 95% CI, p = 0.39, I² = 82% vs. IV: 0.26, (–0.44 → 0.96) at 95% CI, p = 0.47, I² = 49%], and TSS [IV: –0.20, (–0.37 → –0.04) at 95% CI, p = 0.02, I² = 78% vs. IV: –0.43, (–0.59 → –0.27) at 95% CI, p < 0.00001, I² = 98%]. Therefore, the overall conclusions for the parameters HbA1C, MNCV, and TSS remain unchanged. Particularly for TSS, the sensitivity analysis revealed a reduced effect size for TSS when high-risk studies were excluded (IV: –0.20 vs. –0.43). While the p-value changed from p < 0.00001 to p =  0.02, the result remains statistically significant. Additionally, the heterogeneity was notably reduced from 98% to 78%, indicating improved consistency among the included studies. Sensitivity analyses were not conducted for the remaining parameters (NIS, NIS-LL, NSC number, NSC severity, NDS, VPT) because each of these parameters had only two high-quality studies (Figure 13).

Display full size

Figure 13. Sensitivity analysis excluding high-risk bias studies for MNCV and TSS. (A): MNCV; (B): TSS. Results showed minor changes in effect sizes between analyses, excluding vs. including high-risk studies, MNCV (IV: 0.32 vs. 0.26), and TSS (IV: –0.20 vs. –0.43), remaining consistent in direction and significance, indicating robustness of the overall findings. Sensitivity analyses were not performed for other parameters due to either the limited number of available studies or the high quality of the included studies. MNCV: motor nerve conduction velocity; TSS: total severity score; ALA: alpha lipoic acid; IV: inverse variance; CI: confidence interval.

Context and rationale

This article includes most studies focusing on diabetic polyneuropathy, a common complication in diabetics that can lead to foot ulcers and amputations [10]. While no definitive treatment exists to halt neuropathy progression, management options are available [5]. Literature shows significant improvements in various parameters with ALA administered at 1,200 mg or 600 mg daily over 6 months to 4 years. ALA is noted for its antioxidant properties and its ability to boost levels of reduced glutathione, thereby potentially delaying or reversing peripheral DN symptoms. Previous clinical trials have demonstrated that ALA improves neuropathic deficit symptoms [32, 33].

Comparison with previous reviews

Further, Hsieh et al. (2023) [9] endeavored to conduct a comprehensive analysis and draw conclusions regarding the favourable function of ALA in managing DN. Though the published MA by Hsieh et al. (2023) [9] is noted to have several methodological flaws, as highlighted in the introduction section. This provides an opportunity for a robust and comprehensive SR and MA.

Aim of this review

This SR/MA aims to assess the effectiveness of oral ALA in treating DN through available RCTs [5], while also evaluating its efficacy and safety profile.

This study used parameters like HbA1C, NIS, NIS-LL, MNCV, NSC number and severity, TSS, NDS, VPT, and global satisfaction to highlight the benefits of ALA in managing DN.

Outcome-specific findings

The HbA1C that assesses glucose control over the past 90 days and serves as a diagnostic tool for diabetes [19] and neuropathy severity. HbA1C > 6.5% markedly heightens the risk and offers a quantitative measure of neuropathy severity in diabetic individuals [34]. Furthermore, a 1% increase in HbA1C is correlated with a roughly 10–15% elevation in the occurrence of DN [1]. In a previous multivariate analysis study, a significant association was discovered between elevated 3-year mean HbA1C levels and the presence of DPN in patients [35]. Studies have shown that ALA aids in reducing HbA1C levels by increasing the GLUT-4 transportation to muscle and fat, suppressing gluconeogenesis, and enhancing specific proteins within the insulin signaling pathway. Thus, ALA was classified as an insulin-mimetic agent. Hence, oral ALA supplementation has improved overall DN [36]. In this MA, ALA has significantly alleviated the elevated HbA1C levels compared to placebo/control.

Further, NIS is used to evaluate the neuropathy progression through 37 bilateral assessments covering cranial nerves, muscle weakness, reflexes, and sensation loss. Ratings range from 0 to 4 for muscle weakness and cranial nerves (0 being normal and 4 indicating paralysis) and 0 to 2 for reflexes or sensation (0 for normal and 2 for absence). The total score ranging from 0 to 244 for both sides of the body indicates the degree of neuropathic impairment; the higher scores imply higher significant impairment [10, 20]. This SR and MA revealed that ALA is favorable in ameliorating the NIS compared to placebo.

The NIS-LL is specifically designed to evaluate lower limb neuropathy with parameters such as weakness of muscle, reflexes, and sensation. A high NIS-LL score signifies greater impairment [20]. This MA found that ALA ameliorated the NIS-LL compared to placebo/control. This MA observed a numerical improvement in NIS-LL scores among ALA-treated patients compared to placebo; however, the difference did not reach statistical significance, indicating that ALA may have a limited impact on lower limb neuropathic impairment.

The MNCV is a test that assesses the flow of electrical signals via the nerves. The treated group didn’t show any significant changes in this parameter, which indicates nerve damage. A slower and weaker signal indicates nerve damage. The ALA-treated group showed no substantial changes in this parameter [20, 37]. The MA concluded that ALA is not helpful in improving the MNCV compared to placebo/control.

The NSC number and severity are used to rate muscle weakness, nerve sensitivity, and autonomic symptoms. These are subject-based assessments carried out using a standard questionnaire. A higher score indicates more neuronal injury and pain [22]. This MA revealed that ALA has the capability to correct the NSC number and severity compared to placebo/control.

The TSS questionnaire asks patients to rate the intensity (absent, mild, moderate, severe) and frequency (occasional, often, continuous) of four symptoms: pain, burning, paresthesia, and numbness. The score ranges from 0, indicating no symptoms, to 14.64, signifying severe and nearly constant presence of all four symptoms [38]. This MA exposed that ALA has a beneficial role in improving the TSS when compared to placebo/control.

The NDS encompasses assessments of vibration sensation, temperature sensation, pinprick, and ankle reflex on both sides of the great toes, with a highest score of 10 [39]. Neuropathy severity is classified into three categories: mild (scores 3–5), moderate (scores 6–8), and severe (scores 9–10) [24]. This MA revealed that ALA could significantly improve the NDS when compared to placebo/control.

The VPT is a testing tool used to screen for DN, despite being somewhat painful. It assesses the function of large myelinated nerve fibers, which often exhibit poor vibration sensation in DN patients. Higher VPT values indicate DN symptoms [25]. Testing is usually performed on the tips of the great toes using a neurothesiometer or a 128-Hz tuning fork [11]. Responses are categorized as abnormal (no vibration perception), present (vibration perceived within 10 seconds after the patient reports loss of perception), or reduced (vibration perceived after more than 10 seconds) [9]. VPT is crucial for early DN detection, helping to mitigate related complications. VPT values are classified as severe (> 25 V), moderate (20–25 V), and mild (15–20 V) [40]. This study demonstrated that ALA significantly improved NSC and VPT compared to placebo.

Global satisfaction scores in DN trials often involve asking participants to rate their satisfaction with treatment using a scale that ranges from “insufficient” or “poor” to “satisfactory” and “good” or “very good”. This assessment provides insights into how well the treatment alleviates symptoms, improves the quality of life, and meets patients’ expectations [5, 26, 27, 29]. Researchers analyze these scores to understand the broader impact of the treatment beyond specific clinical measures. They help gauge patient-reported outcomes and satisfaction levels, crucial for evaluating the effectiveness and acceptance of therapeutic interventions in managing DN [27, 29]. This MA shows that ALA has significantly improved global patient satisfaction compared to placebo/control.

Dose-dependent patterns

Our analysis indicates a dose-dependent trend in both efficacy and tolerability outcomes. Trials utilizing higher ALA doses (1,200–1,800 mg/day) demonstrated more pronounced improvements in neuropathic symptom scores such as TSS, NIS, and NSC compared to lower doses (600 mg/day). However, higher doses were also associated with increased gastrointestinal side effects, particularly nausea, as reported in Ziegler et al. (2006) [26]. These findings suggest that while the therapeutic benefits of ALA increase with dose, 600–1,200 mg/day represents an optimal balance between efficacy and tolerability for most patients. Future studies should further clarify the dose-response relationship and long-term safety implications of high-dose ALA therapy.

Summary of findings

In conclusion, based on the available moderate-high-quality evidence, the outcomes of this MA state that ALA is beneficial in improving HbA1C, NIS, NSC, TSS, NDS, VPT, and global satisfaction. However, ALA has not shown a significant effect on NIS-LL and MNCV when compared to the placebo/control. Hence, ALA can be a promising adjuvant in managing DN. Improvements in HbA1C, TSS, and NDS were supported by high-quality GRADE evidence, whereas outcomes such as global satisfaction and NSC severity were rated as moderate in quality. Thus, while ALA demonstrates consistent benefits in glycemic control and symptom reduction, conclusions regarding patient-reported and exploratory measures should be interpreted with greater caution until confirmed by further trials.

Combination therapy insights

ALA’s possible synergistic effects when combined with other DN treatments have attracted a lot of attention. According to the literature currently in publication, ALA may improve treatment effectiveness when combined with other therapeutic drugs by taking advantage of their complementary modes of action. Vitamin B12 and ALA together reduce oxidative stress and improve nerve myelination, making them a strong therapeutic option for DN, according to an SR and MA on their combined use in diabetes mellitus (DM) and DN. Similar to this, the MAINTAIN study showed that patients with type 2 diabetes-associated peripheral neuropathy benefit greatly from the addition of methylcobalamin (a type of vitamin B12) and ALA to pregabalin in terms of pain management and nerve function, as indicated by increased nerve conduction velocity [41, 42].

A 12-month study on the combination of superoxide dismutase, ALA, vitamin B12, and carnitine reported significant improvements in various peripheral neuropathy indices, including pain, quality of life, sensory nerve action potential (SNAP), and sensory nerve conduction velocity (SNCV). However, some measures, like CARTs and MNSIE, did not show improvement [43].

The combination of ALA with epalrestat, an aldose reductase inhibitor, has also been explored. This combination was found to increase motor and SNCVs and improve neuropathic symptom scores more effectively than epalrestat monotherapy. The MA of 20 RCTs confirmed the superiority of ALA plus epalrestat combination therapy over monotherapies in clinical efficacy and nerve conduction velocities [44, 45].

Methylcobalamin plus ALA for 2–4 weeks produced superior results in NCV and neuropathic symptoms than methylcobalamin alone, according to another MA. A promising treatment approach was also suggested by a study on the effects of baicalin capsules combined with ALA, which found that patients with diabetic peripheral neuropathy experienced significant improvements in nerve conduction velocity, oxidative stress, and inflammatory injury without an increase in adverse reactions [46, 47].

However, not all combination therapies show added benefits. The PAIN-CARE trial indicated that while pregabalin alone was superior to ALA in treating neuropathic pain, there was no added benefit of combining ALA with pregabalin for this purpose [48].

In summary, while there is compelling evidence supporting the synergistic effects of ALA with other treatments in improving clinical outcomes for DN, further research is needed to optimize these combinations and understand their mechanisms fully. Future studies should focus on long-term effects, potential adverse reactions, and the comparative efficacy of different combination therapies to establish comprehensive treatment protocols.

Limitations of the study

The study’s focus on English-language publications may have excluded important data from non-English sources. Moreover, the inclusion of only RCTs led to some studies being of lower quality due to a lack of blinding and insufficient outcome data. The limited number of studies and varying parameters caused inconsistent assessments. Factors such as HbA1C, NSC severity, TSS, NDS, and VPT exhibited significant heterogeneity, indicating varied results. While study designs were consistent, there were differences in sample sizes and age groups. The primary source of heterogeneity appeared to be the variability in ALA dosage (ranging from 600 to 1,800 mg/day) and treatment duration (from 3 weeks to 4 years). Moreover, the analysis was restricted by excluding studies that combined ALA with other treatments. Thus, further well-designed, randomized, placebo-controlled trials are needed to assess the long-term effects of oral ALA on DN and improve data reliability.

Authors prospective and Future direction

The potential of ALA as a cost-effective treatment for nerve issues in diabetics has generated interest and hope. Our study aimed to thoroughly evaluate the safety and effectiveness of oral ALA for DN. The results supported ALA’s use in this context; however, the lack of documented side effects in the selected studies limited our ability to conduct a definitive quantitative safety analysis, despite a systematic evaluation of safety criteria. Nine RCTs were included based on specific inclusion and exclusion criteria. This review demonstrated ALA’s effectiveness in improving neuropathic impairment, although improvements in MNCV and NIS-LL were not conclusive. Notably, we found a relationship between ALA’s effects, duration, and dosage, with significant benefits seen at higher short-term doses or shorter extended regimens. Our findings enhance understanding of ALA’s potential in treating DN and underscore the importance of balanced administration. To validate these results further, larger, long-term clinical trials are needed to explore the synergistic effects of ALA with other treatments, determine optimal dosages and durations, and investigate its mechanisms of action for more targeted therapies.

Conclusion

The findings of this MA suggest that oral ALA can significantly improve parameters such as HbA1C, NIS, NSC, TSS, NDS, global satisfaction, and VPT. No significant effects were noted on MNCV or NIS-LL, reflecting ALA’s limitations in reversing structural nerve damage. Improvements in individuals with DN occur with higher doses, such as 1,800 mg/day for at least three weeks, or lower doses (600–1,200 mg/day) over extended periods ranging from three weeks to four years. Our SR indicates that a daily dose of 600 mg for up to 24 months is considered safe, as no adverse effects were reported at this level. However, specific side effects were noted when ALA was administered at doses between 1,200 and 1,800 mg/day for five weeks or longer. Overall, ALA appears to be a promising adjuvant therapy for DN, but larger and more standardized trials are needed to confirm its long-term efficacy, optimal dosing, and safety profile [A preprint of this manuscript has previously been published in Research Square for peer review and feedback from the research community (https://doi.org/10.21203/rs.3.rs-3897905/v1)].