Strength training beats cardio for blood sugar in lean diabetics

Strength training lowered long-term blood sugar more than cardio in people who have type 2 diabetes without being overweight, a randomised trial found. In a nine-month trial of 186 adults with normal-weight type 2 diabetes, published in Diabetologia in 2023 by a Stanford University team led by Yukari Kobayashi and Latha Palaniappan, the strength group’s HbA1c fell by 0.44 percentage points, compared with 0.24 points in the aerobic (cardio) group. HbA1c is a blood test that reflects average blood sugar over the past few months.

Strength training was also the only program that added muscle relative to fat, and that shift predicted how much blood sugar fell, the paper reports. Both the strength group and the aerobic group lost some body fat, and the paper does not report a test showing that one group lost more fat than the other.

Why normal-weight type 2 diabetes is a special case

Most people with type 2 diabetes are overweight, but about 20% have a body mass index (BMI, weight relative to height) under 25, the healthy-weight range, according to the paper’s introduction. The same paper notes that this group is more common among Asian people and older adults, and that people diagnosed at a normal weight have a higher risk of death than those diagnosed while overweight or obese.

People with normal-weight type 2 diabetes often have sarcopenia (low muscle mass), the authors explain. That matters for blood sugar: skeletal muscle takes up about 80% of the sugar that insulin clears from the blood, the authors write, so having less muscle leaves less room to store it.

Two earlier large trials, DARE and HART-D, tested the same three programs in people with obesity (average BMI 33 to 35) and found combined training worked best, the paper explains. The Stanford team expected people of normal weight, who have less muscle, to respond better to strength training.

How the strength training vs cardio trial worked

The STRONG-D trial (Strength Training Regimen for Normal Weight Diabetics) recruited adults from the San Francisco Bay Area between November 2016 and December 2019, the paper reports. To join, people had to be 18 to 80 years old, with an HbA1c of 6.5% to 13.0% and a BMI of 18.5 to 25, according to the same paper. The study was registered on ClinicalTrials.gov under the number NCT02448498, with Stanford University as sponsor, and the paper names the National Institutes of Health (NIH) as funder.

Participants were randomly assigned to strength training, aerobic training, or both, three days a week for nine months, at community fitness centres, the paper reports. Each strength session included two sets of four upper-body exercises (bench press, seated row, shoulder press, and pull-down), three sets of three leg exercises (leg press, leg extension, and leg curl), and back and abdominal work, according to the same paper. Loads started light and were raised over time, with a target of eight to 12 repetitions per exercise, the paper explains.

The aerobic group used a treadmill, an elliptical, or a stationary bike at 50% to 80% of their peak capacity, measured on a treadmill test before the trial, the same paper reports. Their weekly dose was set at 12 kilocalories (50.2 kilojoules) per kilogram of body weight, and the authors designed both programs to take about the same time. The combined group did two strength sessions a week with one set of each exercise, plus aerobic work of 10 kilocalories per kilogram per week, according to the paper. The staff who measured the results did not know each person’s group, but the participants did, and there was no group without exercise, the authors explain.

Strength training lowered blood sugar the most

The final analysis included 186 people: 63 in the strength group, 58 in the aerobic group, and 65 in the combined group, the paper reports. Their median age was 59, 60% were men, 83% were Asian, and 84% took glucose-lowering drugs, according to the same paper. At the start, average HbA1c was 7.6% and people had had diabetes for 9.8 years on average, the paper found.

After nine months, HbA1c fell by 0.44 percentage points in the strength group (95% confidence interval -0.78 to -0.12), the only drop large enough to rule out chance, the paper reports. The aerobic group fell by 0.24 points and the combined group by 0.35 points, and neither change was statistically significant, according to the same paper. The strength group improved more than the aerobic group (p = 0.01), while strength and combined training did not differ significantly, the authors report.

Among people who attended at least half of their sessions, HbA1c fell by 0.51 points with strength training, 0.20 with aerobic training, and 0.45 with combined training, the same paper found. Participants attended a median of 2.5 to 2.7 sessions a week, with no significant difference between groups, the paper reports. Changes in diabetes medication and in diet did not differ between the groups either, according to the authors.

Strength training shifted the balance between muscle and fat

Body composition came from whole-body scans on a Hologic Horizon A scanner, a low-dose X-ray that separates fat from lean tissue, and 84 people had a scan at nine months, the paper reports. The strength group lost 0.99 kg of fat on average and gained 0.30 kg of lean mass, a gain too small to rule out chance, according to the same paper. The aerobic group lost 0.70 kg of fat and 0.37 kg of lean mass, the paper’s table shows, while the combined group’s changes were not significant.

The measure the authors focus on is lean mass in the arms and legs relative to fat mass, adjusted for height, age, sex, and ethnicity. It rose only in the strength group, by 0.27 standard units, and it was the only body measure that independently predicted the drop in HbA1c, the paper reports. The aerobic group was the only one to lose a significant amount of body weight, but weight change was not linked to the change in HbA1c, according to the same paper.

Among people who attended at least half of their sessions, only the strength group gained significant leg strength, an extra 15.8 newton-metres of knee torque (turning force) on a test machine, the paper found. The authors write that weight or fat loss through aerobic exercise usually comes with some loss of muscle, which they also saw in this trial.

Limits of the trial

The trial planned 282 participants to have enough statistical power, the paper explains, but the COVID-19 shelter-in-place orders of March 2020 closed the gyms and the study site. Of 289 people randomly assigned, 103 were withdrawn, 84 of them because of these restrictions, according to the same paper. The authors write that the follow-up rate was about 45% and that the study was underpowered for conclusive findings.

The strength program grew harder over the nine months, while the aerobic work stayed at about 70% of peak capacity, which the authors say may have influenced the result. Most participants were Asian (83%), and 79% had at least a bachelor’s degree, the paper reports. Body scans were missing for more than half of the people in the main analysis, the same table shows.

What this means for lean people with type 2 diabetes

For adults with type 2 diabetes and a BMI under 25, the authors conclude that strength training alone lowered HbA1c more than aerobic training alone, and write that the results could be applied right away to exercise advice for this group. The combined program landed between the two, and the trial found no significant difference between it and strength training alone. Only one serious adverse event, a shoulder repair in the combined group, was judged possibly linked to the exercise, the paper reports. For long-term health beyond blood sugar, an earlier Gromeus article explains how cardio fitness and muscle strength both predict a longer life.

Update (2024): in a large study of adults in England and Scotland, any strength training is tied to a 23% lower risk of early death, as a later Gromeus article reports.

Update (2025): a later mouse study pointed the same way for blood sugar, as Gromeus reported in strength training beats running for diabetes benefits in mice.

Sources and related information

Diabetologia – Strength training is more effective than aerobic exercise for improving glycaemic control and body composition in people with normal-weight type 2 diabetes: a randomised controlled trial – 2023

Kobayashi, Long, Dan, Johannsen, Talamoa, Raghuram, Chung, Kent, Basina, Lamendola, Haddad, Leonard, Church, and Palaniappan, Diabetologia 2023;66(10):1897-1907, published online on 26 July 2023, DOI 10.1007/s00125-023-05958-9. Source for all trial figures in this article (design, 186 participants, HbA1c changes, body composition, limits), with the author manuscript free on PubMed Central.

ClinicalTrials.gov – Strength Training Regimen for Normal Weight Diabetics (STRONG-D) – 2015

The trial registration NCT02448498 names Stanford University as sponsor and Latha Palaniappan as principal investigator.

Leave a Reply

Your email address will not be published. Required fields are marked *