A Randomized Controlled Trial of Long-Term (R)-α-Lipoic Acid Supplementation Promotes Weight Loss in Overweight or Obese Adults without Altering Baseline Elevated Plasma Triglyceride Concentrations (2024)

Study design, funding, and ethical approval

This double-blind, randomized, placebo-controlled, parallel study (NCT00765310) was conducted between 2009 and 2013 at the Oregon Clinical and Translational Research Institute's Clinical and Translational Research Center at Oregon Health & Science University (OHSU) in Portland, Oregon. The study protocol complied with the Declaration of Helsinki and was approved by the Institutional Review Boards of both Oregon State University and OHSU.

Participants and recruiting

Participants were recruited from the greater Portland metropolitan area via websites, flyers, ads, and physician referrals to OHSU starting in October 2008. Recruiting terminated in February 2013 when funding was no longer available to recruit more participants. After an initial phone screen by a nurse, eligible participants were assessed for eligibility during a screening visit (Figure 1). Initial inclusion criteria were as follows: 18–50 y of age; weight-stable BMI ≥ 27(in kg/m²) (weight stable within ±1.0kg for the last 3 mo and at maximum lifetime body weight); fasting plasma triglycerides ≥ 150mg/dL; exercise limited to 30 min for ≤3 times/wk; and adherence to a prudent diet—as assessed using the Dietary Habit Survey (15)—at study start. Owing to low enrollment numbers, inclusion criteria were modified in January 2012 to increase the maximum age of enrollment (60 y), decrease the lower BMI limit (>25), and lower the threshold for triglycerides (>100mg/dL).

Exclusion criteria were formulated to exclude subjects with risk factors that could confound an analysis of R-LA's effect on otherwise relatively healthy obese individuals: C-reactive protein (CRP) concentration at baseline >10mg/L; smoking within the last 3 mo; consuming >2 alcoholic drinks per day; currently taking lipid-lowering drugs, antihypertensive drugs, insulin or oral hypoglycemic agents, anti-inflammatory drugs, weight loss medications, or hormone replacement therapy; diagnosed as having hyperglycemia (fasting glucose≥125mg/dL at baseline), cardiovascular disease, congestive heart failure, angina, thyroid disorders, cancer, inflammatory disorders, or renal, hepatic, or hematological abnormalities; having had acute medical conditions, such as hospitalizations or surgeries, ≥3 mo before entry into the study; and being pregnant, breastfeeding, or planning to become pregnant before the end of the study. Furthermore, volunteers were excluded if they were taking vitamin or antioxidant supplements, including R-LA, except standard multivitamin/mineral supplements containing not more than the Daily Value of each vitamin and mineral.

In a parallel-group design, participants were assigned to consume either a daily supplement containing 600mg (R)-α-lipoic acid (R-LA group) as a sodium salt or a matching inert tablet (Placebo group) for 24 wk (see details below). Participants were assigned to treatment groups by a covariate adaptive randomization method. The covariate blocks were set based on sex, baseline BMI (25.0–29.9; 30.0–34.9; 35.0–39.9; ≥40.0), and baseline plasma triglyceride concentrations (100–149, 150–199, 200–249, ≥250mg/dL) to ensure random allocation. CP at Oregon State University performed the block assignments and treatment allocations. Treatment group was communicated to the study site pharmacist to ensure proper blinding. All other investigators, staff, and participants remained blinded to the study supplement assignments until study completion.Table 1 shows the baseline demographic characteristics.

Supplementation, adverse effect reporting, and compliance

Participants were advised to take either 2 capsules containing 300mg LA each or matched inert tablets, both provided by GeroNova Research, on an empty stomach 30min before breakfast with plenty of fluids. The 600-mg dose of LA was chosen for this study in order to reduce minimal adverse events (gastrointestinal upset) compared with higher dosages, and is similar to that approved in Europe for treatment of diabetes-induced polyneuropathies (16–18). This dosage is also consistent with current doses in over-the-counter supplements in the United States.

Placebo and R-LA capsules were identical in size, appearance, and taste. Compliance was measured by counting the returned capsules at the 3-, 12-, and 24-wk study visits. Subjects were advised to continue eating a prudent diet while taking the supplements, and continue their habitual exercise regimen during the study. At the start of the study, participants were given information on signs and symptoms of adverse events associated with LA supplementation (i.e., modest indigestion) and instructions on whom to contact if they suspected they were experiencing an adverse event. Participants were asked about adverse events every 3 wk via phone call as well as at their 3-, 12-, and 24-wk visits by a physician or nurse. A generalized severity scale and a 5-point attribution scale were used for adverse events; adverse events were reviewed by the OHSU Institutional Review Board to ensure participant safety. Supplemental Table 1 shows the results for compliance and adverse events.

Activity monitoring and dietary assessment

Physical activity was monitored for 2–4 wk around baseline, 12, and 24 wk using the Actical^® accelerometer (Philips Respironics). The device was worn at the hip and logged the estimated total energy expenditure (kcal/d) as well as the amount of time (min/d) spent doing no, mild, moderate, and vigorous activity. For statistical analysis, we calculated the median of 7 consecutive days (minimum: 4 d) for these measures. To avoid days when the device was not worn in most of the wakeful hours, we excluded from the calculation those days with <5% of time spent being physically active, as well as the first and the last day the device was worn.

Food intake was assessed using three 24-h food recalls around baseline as well as 24 wk. The food recalls were done via phone by the Arizona Diet, Behavior, and Quality of Life Assessment Center (University of Arizona Cancer Center, Tucson, AZ). Nutrient intakes from each 24-h recall were calculated using the Minnesota Nutrient Data System for Research (University of Minnesota Nutrition Coordinating Center). For statistical analysis, we calculated the 3-d mean for these measures.

Sample collection and anthropometric analysis

Anthropometric measures and vital signs (heart rate, blood pressure) were determined after overnight fasting at baseline, 12, and 24 wk by a trained nurse. Height, circumference of the waist (horizontal plane around the abdomen at the level of the iliac crest), and circumference of the hip (horizontal plane around the hips at the level of maximum prominence of the buttocks) were measured in centimeters. Hip and waist measures were repeated until duplicates agreed to within 0.5cm. Total body weight and body fat mass were measured in duplicate within 10g by air-displacement plethysmography using the BOD POD^® 2000A from Life Measurement, Inc. Body fat percentage was calculated as body fat mass divided by body weight multiplied by 100.

Overnight fasting blood samples were collected at baseline, 12, 23 (for triglyceride measurements only), and 24 wk and stored at −80° C until analysis. Plasma concentrations of triglycerides, cholesterol (total, HDL, VLDL, and LDL), and glucose and hematological measures were analyzed using standard clinical assays in the Clinical Core Laboratory at OHSU. Morning urine samples were collected at baseline, 12, and 24 wk and stored at −80° C until analysis. Table1 shows anthropometric measures, vital signs, and metabolic profiles of the study participants at baseline.

Plasma biomarkers

Plasma concentrations of acute-phase proteins, cytokines, and soluble adhesion molecules were measured using commercially available ELISA kits (R&D Systems). The interassay and intra-assay CVs were as follows: CRP, 7.0% and 3.8%; vascular cell adhesion molecule 1 (VCAM1) , 7.7% and 3.5%; intercellular adhesion molecule 1 (ICAM1), 4.4% and 5.0%; soluble E-selectin (SELE), 8.7% and 5.7%; IL-6, 7.2% and 7.8%; TNFA, 7.2% and 4.3%; and CCL2, 5.8% and 4.7%, respectively. Plasma total antioxidant capacity was measured using the ferric ion reducing antioxidant potential (FRAP) assay. The interassay and intra-assay variations between the samples were <3.0% and <1.0%, respectively.

For plasma ascorbic acid and uric acid analyses, samples were stabilized by the addition (1:1, vol:vol) of 15% perchloric acid (EMD Millipore) containing 1mM diethylenetriaminepentaacetic acid (DTPA) (Sigma-Aldrich), a metal chelator, and analyzed using HPLC. Briefly, extracts were diluted in a sodium acetate:methanol:water mobile phase (0.3%:7.5%:92% wt:vol) containing Q12 ion-pairing reagent (Regis Technologies) and pH-adjusted with KH₂PO₄ (pH 9.8). Samples were separated on a Waters 2695 instrument using an LC-8 Supelco column (Sigma-Aldrich) under an applied potential of 600mV using an electrochemical detector (BAS). Samples were run in duplicate, and uric acid concentrations were used to normalize ascorbic acid values in order to minimize any variations detected. Under these conditions, the intra-assay CV was <5%, whereas the interassay CV was <15%. Supplemental Table 1 shows the baseline concentrations of plasma biomarkers.

Cellular biomarkers

Peripheral blood mononuclear cells (PBMCs) were isolated from heparinized venous blood of the subjects. Briefly, 20mL peripheral venous heparinized blood was collected and PBMCs were isolated by gradient centrifugation at room temperature (25° C) on Lymphoprep (Sigma-Aldrich) at 1200× g for 20min and then washed and resuspended to 1×10⁷ cells/mL in HBSS. Monocytes were identified by May-Grunwald-Giemsa staining.

For gene expression analysis, total RNA was isolated using TRIzol Reagent (Invitrogen). cDNA synthesis was performed using the high-capacity cDNA archive kit (Applied Biosystems). All primers and probes for the human genes IL1B, IL6, CCL2, TNFA, HMOX1, GCLC, NQO1, and GAPDH were purchased as Assays on Demand kits from Applied Biosystems. TaqMan qPCR was performed as described previously using TaqMan Universal PCR Master Mix (Applied Biosystems) in 96-well plates with the ABI Prism 7500 Sequence Detection System (Applied Biosystems) (19, 20). To obtain relative quantification, 2 standard curves were constructed in each plate with 1 target gene and GAPDH as an endogenous control. Standard curves were generated by plotting the threshold cycle number values against the log of the amount of input cDNA and used to quantify the expression of the various target genes and GAPDH in the same sample. After normalization to GAPDH, the results were expressed as fold of baseline within the same group.

To determine intracellular glutathione and glutathione disulfide concentrations, we adapted the method of Fariss and Reed (21). Briefly, PBMCs were isolated from freshly drawn human blood, immediately acidified with an equal volume of 15% perchloric acid (wt:vol) containing 10mM DTPA, and stored at −80° C until ready for use. After deproteinization, iodoacetic acid was added to acetylate acid–soluble thiols, and pH-adjusted PBMC samples were derivatized with 1-fluoro-2,4-dinitrobenzene (0.1%, vol:vol). Reduced and oxidized glutathione were subsequently measured by a Shimadzu HPLC (Shimadzu Scientific Instruments, Inc.) using UV detection at 365nm with a Hypersil 3-aminopropyl column (Thermo Scientific). For intracellular ascorbate and urate analysis, acid extracts were analyzed using HPLC as described already for the plasma ascorbic acid and uric acid analyses. Supplemental Table 1 shows the baseline concentrations of cellular biomarkers.

Urinary biomarkers

Urine F2-isoprostanes, including 8-iso-PGF2α and its metabolites, and PGE₂ concentrations were analyzed by a solid phase extraction process coupled to LC-tandem MS, and standardized to urinary creatinine concentrations. The assay for urinary 8-iso-PGF2α has a within-day precision of ±7% and an accuracy of 100% (22, 23). PGE₂ and F2-isoprostanes are expressed as ng/mg creatinine. Supplemental Table 1 shows the baseline concentrations of urinary biomarkers.

Statistical analysis

The primary objective of this clinical trial was to evaluate changes in plasma triglycerides. The study was designed to show a statistically significant (P≤0.025) 14.6% decrease in triglycerides with 80% power as the main effect. The power calculation was based on an SD of 46.6mg/dL for triglycerides and 50 subjects/treatment group. Secondary aims were to determine if R-LA affects body weight and body fat mass, the cellular and plasma antioxidant pool and capacity, gene expression of antioxidant enzymes, lipid peroxidation, cellular and plasma markers of inflammation, and blood markers of immune surveillance, which were all defined at the study design stage.

The original statistical analysis plan specified 3-visit repeated-measures-in-time analysis. During the primary study analysis, it became apparent that several Placebo participants showed pronounced changes to body weight in the first 12 wk that were not sustained and often reversed after 24 wk. By contrast, gradual changes were observed in R-LA participants. As such, we analyzed changes per time point and modified the statistical analysis plan accordingly. In addition, several participants withdrew consent from the study and were effectively lost to follow-up for any endpoints. Because no data for these participants existed, we analyzed all participants with follow-up data instead of utilizing a method to impute the missing data.

All statistical analyses were performed as intention-to-treat analysis using SAS (SAS/Stat version 9.2, SAS Institute). Fisher's exact test was used to compare the numbers of participants experiencing adverse events. We calculated absolute and percentage changes from baseline at 12 wk and at 24 wk of treatment. Generalized linear models (PROC GLM) were used to evaluate treatment effects on the percentage changes from baseline as well as for absolute changes from baseline for primary and secondary endpoints. Fixed effects in the model were supplementation (Placebo, R-LA), sex (male, female), baseline BMI (25.0–29.9; 30.0–34.9; 35.0–39.9; ≥40), age at baseline (21–30, 31–40, 41–50, 51–60 y), and plasma triglyceride concentrations at baseline (<150; 150–199; 200–249; ≥250mg/dL). Changes in circulating metabolites, vital signs, and body weight and composition have been reported as absolute change and proportional change with similar P values. We chose proportional change for the other endpoints to clarify the magnitude of the changes and make the extent of the changes more comparable.

We tested for effect modification by sex, baseline BMI, age, and plasma triglycerides. Significant interactions were observed only for sex (male compared with female) and for baseline BMI (<35.0 compared with ≥35.0). If significant interactions were observed, the data were analyzed stratified by sex or baseline BMI. A stratified analysis but no power analysis of a stratified analysis were part of the statistical plan at the study design stage.

The results are reported as least-squares means±SEMs. Reported P values refer to differences between the Placebo and R-LA groups. Statistical significance was set at P≤0.05. No adjustments for multiplicity were performed because this was not part of the statistical plan at the study design stage.

Baseline characteristics, compliance, and adverse events

Demographics and initial clinical measures were all collected from study participants at baseline (Table1). Blood samples from these participants were also evaluated for biomarkers of inflammation and redox status (Supplemental Table 1). Overall, the population groups appeared evenly matched for the comparison of treatments.

During the study, the median compliance based on pill recount for the R-LA group was 97% (range: 79%–100%) and for the placebo group 94% (68%–100%). Most subjects tolerated R-LA well. The major adverse event in subjects supplemented with R-LA was heartburn, which persisted throughout the study (R-LA compared with placebo: 33%compared with 5%; P=0.003) (Supplemental Table 2). Some subjects receiving R-LA reported a strong odor of their urine (18%compared with 0%; P=0.01). A total of 13 subjects withdrew (Figure1), 6 of them (4 subjects in the R-LA group, 2 in placebo) because of adverse events (placebo: heartburn, headache; R-LA: heartburn, hospitalization, lower back pain). No deaths occurred during the study. All subjects that withdrew did so before week 12 and their samples were not analyzed. One participant was excluded from the final analysis when it was determined retrospectively that they met 1 of the exclusion criteria for the protocol.

LA supplementation did not change fasting triglyceride, glucose, or cholesterol status

The primary endpoint, changes in fasting plasma triglyceride concentrations, did not differ between the R-LA and placebo groups (Table 2). Similarly, fasting blood concentrations of glucose, total, HDL, and LDL cholesterol did not differ between the R-LA and placebo groups.

LA supplementation reduces BMI and body fat in women and severely obese participants

At week 24, the decrease in BMI was greater in the R-LA group than in placebo (P=0.04) (Table 3). Body-weight loss occurred in a sex-specific manner (see below). This change in body weight was not explainable by changes in energy intake (caloric intake) or physical activity (Supplemental Table 3). This suggests that any observed changes in body weight or fat mass (see below) were independent of these factors. In study participants receiving R-LA, weight loss was linearly correlated with lower plasma triglyceride concentrations (r = +0.50, P=0.004) and lower VLDL-cholesterol concentrations (r = +0.44, P=0.01; not shown). This correlation was not observed in participants receiving placebo (Figure 2).

As an exploratory analysis of the data, we tested for effect modification of change in BMI by sex, age, baseline BMI, and baseline plasma triglycerides. Of these, only female sex (Supplemental Table 4) and baseline BMI (Supplemental Table 5) independently modified the effect of R-LA supplementation on anthropometric measures. Women, but not men, lost more body weight in the R-LA group than in placebo (Figure 3). Women supplemented with R-LA lost body weight (−1.4%±0.5% and −3.2%±0.8% at 12 and 24 wk, respectively) (Figure3A), which was primarily body fat (−2.3%±1.2% and −6.5%±1.9% at 12 and 24 wk, respectively) (Figure3C, Supplemental Table 4). By contrast, body fat or body weight did not decrease in men in the R-LA group (Figures3B, ​,D);D); body weight, but not body fat, decreased in men in the placebo group by 24 wk, specifically in those with <30% body fat (n=5).

Clinically meaningful weight loss, defined as ≥5% decline in body mass, was achieved by 42% (8 of 19) of women after 24 wk of R-LA supplementation. This was significantly greater than for women in the placebo group (interaction P=0.004). The weight loss associated with R-LA supplementation was more pronounced after 24 than after 12 wk of the study. Women in the R-LA group decreased waist and hip circumference after 24 wk (Supplemental Table 4), but no group differences in waist-to-hip ratio were observed (results not shown).

To define the effect of initial BMI status in more detail, we separated participants into 2 groups: those with severe obesity (BMI≥35) and those with a lower BMI (BMI<35) (Figure 4). Severely obese participants lost more body weight after 24 wk of supplementation on R-LA than on placebo (−2.4%±1.0%) (Figure4B), which was primarily body fat (−4.3%±2.4%) (Figure4D, Supplemental Table 5). In these participants, 38% (5 of 13) of those taking R-LA showed clinically significant weight loss, which was greater than placebo participants who showed none of these changes over time. On the other hand, participants who were less heavy (BMI<35) in the R-LA group did not lose more body weight or fat mass than placebo-supplemented participants (Figure4A, ​,C).C). Conversely, body fat mass decreased in the placebo-treated group at 12 wk (Figure4C), which may have been driven by the male participants who also decreased body weight (Figure3B) in that treatment arm.

LA supplementation has mixed effects on vital signs and blood hematology

Whereas vital signs—blood pressure and heart rate—did not differ between treatment groups, participants on R-LA had greater leukocyte and platelet count decreases than those on placebo (Supplemental Table 6). However, leukocyte numbers remained within normal reference ranges.

LA supplementation has mixed effects on plasma, urinary, and cellular biomarkers of inflammation or redox status

The participants uniformly had low concentrations of inflammatory biomarkers, which was anticipated from the subject inclusion criteria. R-LA supplementation did not affect the concentrations of many inflammatory biomarkers (Table 4, Supplemental Table 7). However, the R-LA group had greater increases in plasma concentrations of TNFA than placebo had (P=0.003) (Table4). It is worth noting that in all participants TNFA stayed within normal reference ranges (24).

In contrast, participants taking R-LA compared with placebo controls had greater decreases in circulating concentrations of soluble ICAM1 (P=0.04), a biomarker of immune cell recruitment, endothelial cell adhesion, and tissue infiltration (Table4). Although SELE did not change at 12 or 24 wk in the R-LA group when compared with placebo (Table4), subgroup analysis revealed that women had a greater decline in SELE at 24 wk in the R-LA group than in placebo (P=0.01).

R-LA supplementation did not change low molecular weight antioxidant status (ascorbate, glutathione, urate) of either plasma or PBMCs or the antioxidant potential (FRAP) of plasma (P>0.1 for all markers) (Table4, Supplemental Table 7). Antioxidant-responsive gene expression in PBMCs, as exemplified by HMOX1, GCLC, and NQO1 mRNA at 24 wk, tended to increase more in participants on R-LA than on placebo (P<0.1 for all), but only reached significance for HMOX1 (P=0.02) (Supplemental Table 7). No group differences were observed in genes involved in inflammatory responses at 24 wk. Lastly, participants on R-LA had greater decreases in urinary concentrations of F₂-isoprostanes, a well-established biomarker of lipid peroxidation (Table4).

nxaa203_Supplemental_File

2. Zehnpfennig B, Wiriyasermkul P, Carlson DA, Quick M. Interaction of α-lipoic acid with the human Na⁺/multivitamin transporter (hSMVT). J Biol Chem. 2015;290(26):16372–82. [PMC free article] [PubMed] [Google Scholar]

3. Butler JA, Hagen TM, Moreau R. Lipoic acid improves hypertriglyceridemia by stimulating triacylglycerol clearance and downregulating liver triacylglycerol secretion. Arch Biochem Biophys. 2009;485(1):63–71. [PMC free article] [PubMed] [Google Scholar]

4. Fernández-Galilea M, Pérez-Matute P, Prieto-Hontoria PL, Sáinz N, López-Yoldi M, Houssier M, Martínez JA, Langin D, Moreno-Aliaga MJ. α-Lipoic acid reduces fatty acid esterification and lipogenesis in adipocytes from overweight/obese subjects. Obesity. 2014;22(10):2210–15. [PubMed] [Google Scholar]

5. Shay KP, Moreau RF, Smith EJ, Smith AR, Hagen TM. Alpha-lipoic acid as a dietary supplement: molecular mechanisms and therapeutic potential. Biochim Biophys Acta. 2009;1790(10):1149–60. [PMC free article] [PubMed] [Google Scholar]

6. Kim MS, Park JY, Namkoong C, Jang PG, Ryu JW, Song HS, Yun JY, Namgoong IS, Ha J, Park IS et al.. Anti-obesity effects of α-lipoic acid mediated by suppression of hypothalamic AMP-activated protein kinase. Nat Med. 2004;10(7):727–33. [PubMed] [Google Scholar]

7. Namazi N, Larijani B, Azadbakht L. Alpha-lipoic acid supplement in obesity treatment: a systematic review and meta-analysis of clinical trials. Clin Nutr. 2018;37(2):419–28. [PubMed] [Google Scholar]

8. Kucukgoncu S, Zhou E, Lucas KB, Tek C. Alpha-lipoic acid (ALA) as a supplementation for weight loss: results from a meta-analysis of randomized controlled trials. Obes Rev. 2017;18(5):594–601. [PMC free article] [PubMed] [Google Scholar]

9. Gleiter CH, Schug BS, Hermann R, Elze M, Blume HH, Gundert-Remy U. Influence of food intake on the bioavailability of thioctic acid enantiomers. Eur J Clin Pharmacol. 1996;50(6):513–14. [PubMed] [Google Scholar]

10. Hermann R, Niebch G, Borbe HO, Fieger-Büschges H, Ruus P, Nowak H, Riethmüller-Winzen H, Peukert M, Blume H. Enantioselective pharmaco*kinetics and bioavailability of different racemic α-lipoic acid formulations in healthy volunteers. Eur J Pharm Sci. 1996;4(3):167–74. [Google Scholar]

11. Breithaupt-Grögler K, Niebch G, Schneider E, Erb K, Hermann R, Blume HH, Schug BS, Belz GG. Dose-proportionality of oral thioctic acid—coincidence of assessments via pooled plasma and individual data. Eur J Pharm Sci. 1999;8(1):57–65. [PubMed] [Google Scholar]

12. Zhang W-J, Frei B. α-Lipoic acid inhibits TNF-a-induced NF-κB activation and adhesion molecule expression in human aortic endothelial cells. FASEB J. 2001;15(13):2423–32. [PubMed] [Google Scholar]

13. Packer L, Witt EH, Tritschler HJ. Alpha-lipoic acid as a biological antioxidant. Free Radic Biol Med. 1995;19(2):227–50. [PubMed] [Google Scholar]

14. Smith AR, Shenvi SV, Widlansky M, Suh JH, Hagen TM. Lipoic acid as a potential therapy for chronic diseases associated with oxidative stress. Curr Med Chem. 2004;11(9):1135–46. [PubMed] [Google Scholar]

15. Connor SL, Gustafson JR, Sexton G, Becker N, Artaud-Wild S, Connor WE. The Diet Habit Survey: a new method of dietary assessment that relates to plasma cholesterol changes. J Am Diet Assoc. 1992;92(1):41–7. [PubMed] [Google Scholar]

16. Ziegler D, Ametov A, Barinov A, Dyck PJ, Gurieva I, Low PA, Munzel U, Yakhno N, Raz I, Novosadova M et al.. Oral treatment with α-lipoic acid improves symptomatic diabetic polyneuropathy. Diabetes Care. 2006;29(11):2365–70. [PubMed] [Google Scholar]

17. Koh EH, Lee WJ, Lee SA, Kim EH, Cho EH, Jeong E, Kim DW, Kim M-S, Park J-Y, Park K-G et al.. Effects of alpha-lipoic acid on body weight in obese subjects. Am J Med. 2011;124(1):85.e1–8. [PubMed] [Google Scholar]

18. Kim NW, Song Y-M, Kim E, Cho H-S, Cheon K-A, Kim SJ, Park JY. Adjunctive α-lipoic acid reduces weight gain compared with placebo at 12 weeks in schizophrenic patients treated with atypical antipsychotics: a double-blind randomized placebo-controlled study. Int Clin Psychopharmacol. 2016;31(5):265–74. [PubMed] [Google Scholar]

19. Lyons CL, Roche HM. Nutritional modulation of AMPK-impact upon metabolic-inflammation. Int J Mol Sci. 2018;19(10):3092. [PMC free article] [PubMed] [Google Scholar]

20. Carbonelli M, Renzo L, Bigioni M, Daniele N, DeLorenzo A, Fusco M. α-Lipoic acid supplementation: a tool for obesity therapy?. Curr Pharm Des. 2010;16(7):840–6. [PubMed] [Google Scholar]

21. Fariss MW, Reed DJ. High-performance liquid chromatography of thiols and disulfides: dinitrophenol derivatives. Methods Enzymol. 1987;143:101–9. [PubMed] [Google Scholar]

22. Taylor AW, Bruno RS, Traber MG. Women and smokers have elevated urinary F2-isoprostane metabolites: a novel extraction and LC-MS methodology. Lipids. 2008;43(10):925–36. [PMC free article] [PubMed] [Google Scholar]

23. Taylor AW, Traber MG. Quantitation of plasma total 15-series F2-isoprostanes by sequential solid phase and liquid-liquid extraction. Anal Biochem. 2010;396(2):319–21. [PubMed] [Google Scholar]

24. Todd J, Simpson P, Estis J, Torres V, Wub AH. Reference range and short- and long-term biological variation of interleukin (IL)-6, IL-17A and tissue necrosis factor-alpha using high sensitivity assays. Cytokine. 2013;64(3):660–5. [PubMed] [Google Scholar]

25. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA. 2014;311(8):806–14. [PMC free article] [PubMed] [Google Scholar]

26. Keith DJ, Butler JA, Bemer B, Dixon B, Johnson S, Garrard M, Sudakin DL, Christensen JM, Pereira C, Hagen TM. Age and gender dependent bioavailability of R- and R,S-α-lipoic acid: a pilot study. Pharmacol Res. 2012;66(3):199–206. [PMC free article] [PubMed] [Google Scholar]

27. Yanovski SZ, Yanovski JA. Long-term drug treatment for obesity: a systematic and clinical review. JAMA. 2014;311(1):74–86. [PMC free article] [PubMed] [Google Scholar]

28. Huerta AE, Prieto-Hontoria PL, Sáinz N, Martínez JA, Moreno-Aliaga MJ. Supplementation with α-lipoic acid alone or in combination with eicosapentaenoic acid modulates the inflammatory status of healthy overweight or obese women consuming an energy-restricted diet. J Nutr. 2016;146(4):889S–96S. [PubMed] [Google Scholar]