The stakes of the question are considerable. Insulin resistance, in which the body’s tissues respond poorly to normal circulating insulin, affects an estimated 15.5 to 46.5 percent of adults worldwide, and recent NHANES data indicate that roughly 40 percent of American adults aged 18 to 44 show the condition even in the absence of overt diabetes. Because insulin resistance sits upstream of type 2 diabetes, cardiovascular disease, hypertension, polycystic ovarian syndrome, and nonalcoholic fatty liver disease, any cheap and safe dietary tool that could improve insulin sensitivity would carry enormous public health value. That promise is precisely what drew researchers to glucomannan, whose viscosity is reportedly five times greater than that of fibers like beta-glucan and guar gum.

The new analysis, registered prospectively in the PROSPERO database, followed the PRISMA 2020 framework and searched Scopus, Web of Science, and PubMed up to March 2025. From 289 initial records, the team winnowed the literature down to eleven eligible randomized controlled trials published between 1984 and 2023, conducted in the United States, Canada, China, Slovenia, Brazil, Thailand, and Taiwan. Participants fell into three broad groups: people with type 2 diabetes, people with overweight or obesity, and people with insulin resistance syndrome. Trials ranged from as few as 11 participants to as many as 80, with durations spanning 3 to 12 weeks and glucomannan doses typically between 3,000 and 10,000 milligrams per day, sometimes delivered as enriched biscuits or gummy candy rather than capsules.

Methodologically, the review was unusually thorough for a nutrition meta-analysis. The authors used the Cochrane ROB 2 tool to grade each trial’s risk of bias, applied the GRADE framework to rate the certainty of evidence for each outcome, and ran leave-one-out sensitivity analyses to test whether any single study was propping up the pooled results. They also performed meta-regression and fractional polynomial modeling to probe whether effects scaled with dose or duration, and used Egger’s regression test and funnel plots to detect publication bias. Only one of the eleven trials, a Chinese study of obese adults, was judged to have a low overall risk of bias; three were rated high risk and the remainder raised some concerns.

The headline finding concerned fasting blood glucose. Across seven pooled effect sizes, glucomannan supplementation lowered fasting glucose by a weighted mean of 7.02 milligrams per deciliter compared with controls, a result that just crossed the threshold of statistical significance. The proposed mechanism is rooted in the fiber’s extraordinary viscosity: by forming a gel in the stomach, glucomannan delays gastric emptying and slows the absorption of carbohydrates into the bloodstream, flattening the post-meal glucose spikes that chronically stress insulin-producing cells. Fermentation of the fiber by gut microbiota, which yields short-chain fatty acids, may also contribute to improved metabolic signaling.

Yet the glucose result comes with heavy caveats that the authors themselves emphasize. The pooled estimate carried substantial heterogeneity, and removing a single trial, a Taiwanese crossover study in older adults with type 2 diabetes, caused the effect to vanish entirely. More strikingly, when the researchers reran the analysis under an alternative statistical assumption about the correlation between baseline and final measurements in crossover trials, the fasting glucose benefit lost its significance altogether. The GRADE assessment downgraded the certainty of evidence for fasting glucose to low, alongside similarly low or very low ratings for most other outcomes.

The broader cardiometabolic picture was largely one of null results. Body weight showed no significant change across nine pooled effect sizes, and neither did body mass index, body fat percentage, fasting insulin, triglycerides, total cholesterol, LDL cholesterol, or HDL cholesterol. Blood pressure outcomes were similarly flat overall, though subgroup analyses hinted at context-dependent effects: significant systolic reductions appeared in Canadian trials, crossover designs, participants over 50, and doses above 3,000 milligrams per day, while some subgroups, including male-only trials and studies in the United States, paradoxically showed increases. The authors stress that these subgroup findings, drawn from small numbers of studies, should be treated as exploratory hypothesis-generating observations rather than clinical guidance.

The dose–response analysis added a further layer of nuance. Meta-regression found no significant linear relationship between glucomannan dosage or supplementation duration and changes in weight or any lipid parameter, and fractional polynomial modeling detected no meaningful non-linear dose effects. The single exception was total cholesterol, which showed a significant non-linear relationship with intervention duration. Subgroup patterns suggested that higher doses above 3,000 milligrams per day and shorter trials of four weeks or less were associated with cholesterol reductions, but the authors caution against reading these patterns as dosing recommendations given the fragility of the underlying data.

How do these results square with earlier reviews? A 2008 meta-analysis by Sood and colleagues reported significant improvements in weight, glucose, triglycerides, and cholesterol, while more recent analyses found lipid benefits in general adult populations. The discrepancies likely reflect differences in study populations and inclusion criteria. The new review is the first to focus exclusively on adults with insulin resistance–related conditions across the spectrum, rather than narrowing to type 2 diabetes or obesity alone, and its stricter risk-of-bias standards may have tempered earlier optimism. Notably, its null weight finding echoes a 2014 analysis in people with overweight and obesity that also found no significant effect of glucomannan on body weight.

The authors conclude that glucomannan may serve as a useful adjunct for glycemic control in insulin-resistant individuals, but that its broader cardiometabolic benefits appear inconsistent and context-dependent. With significant publication bias detected for systolic blood pressure and HDL cholesterol, low-to-very-low certainty across most outcomes, and sensitivity analyses revealing instability in several pooled estimates, the evidence base remains thin. The researchers call for large, well-designed randomized trials to clarify the optimal dose, duration, and patient populations most likely to benefit, and to illuminate the mechanisms behind the fiber’s modest glucose-lowering effect. Until then, the konjac root’s reputation as a metabolic multitool appears to rest, at best, on a single fragile number.

Subject of Research: Effects of glucomannan supplementation on cardiometabolic parameters in adults with insulin resistance-related conditions

Article Title: Effects of Glucomannan on Cardiometabolic Parameters in Adults With Insulin Resistance‐Related Conditions: A Systematic Review and Dose–Response Meta‐Analysis

Article References: Jazinaki, M. S., Rafiei, H., Bahari, H., & Pahlavani, N. (2026). Effects of Glucomannan on Cardiometabolic Parameters in Adults With Insulin Resistance‐Related Conditions: A Systematic Review and Dose–Response Meta‐Analysis. Food Science & Nutrition, 14(10), Article e72460. https://doi.org/10.1002/fsn3.72460

Image Credits: AI Generated

DOI: 10.1002/fsn3.72460

Keywords: glucomannan, konjac root, insulin resistance, fasting blood glucose, meta-analysis, dietary fiber, type 2 diabetes, obesity, lipid profile, blood pressure, cardiometabolic health, systematic review