Different Effects of Lifestyle Intervention in High- and Low-Risk Prediabetes: Results of the Randomized Controlled Prediabetes Lifestyle Intervention Study (PLIS)
Published: Diabetes, December 2021
Where: University Tübingen, Tübingen, Germany
If some exercise is good for metabolic health, then more exercise should help people more at risk of metabolic dysfunction, right? Maybe.
This hypothesis is seldom directly tested, nor do we have data specifically for prediabetes (although ongoing studies exist). Furthermore, it’s clear that not all people with prediabetes are equally…prediabetic. Some have several risk factors, while others may only have glucose metabolism issues. It seems plausible that different degrees of disease may need different amounts of exercise.
Previous studies looking at preventing prediabetes from progressing to Type 2 diabetes have used an amount of exercise based on the Diabetes Prevention Program and showed a return to normal glucose metabolism in 40% of the subjects with prediabetes three years after the start of the study.
This study was interested in the 60% that remain prediabetic. Researchers asked them to do twice as much exercise and found that this led to a significantly higher probability of returning to normal glucose function.
Researchers took a group of people with prediabetes and separated them into high-risk and low-risk groups based on several related physiological measures. The high-risk group then received a conventional lifestyle intervention (diet + 3 hours of exercise/week) or an intensive lifestyle intervention (diet + 6 hours of exercise/week).
This more intense intervention was significantly more effective at improving both glucose levels and the other measures that made them a high risk in the first place. Twice as many subjects in the intensive group returned to normal glycemia than in the conventional lifestyle intervention as well.
In summary, identifying “high-risk” people with prediabetes and providing a higher dose of exercise is an effective strategy for reverting to normal glucose and reducing some of the other associated pathologies.
What It Looked At:
The authors of this study recently performed a randomized clinical study looking at lifestyle interventions preventing prediabetes from progressing to diabetes. After the study was complete, they performed a retrospective analysis and found that specific metabolic characteristics made a person less likely to benefit from lifestyle interventions. This finding formed the basis for the current study to test lifestyle interventions in patients prospectively identified as low-risk or high-risk. These so-called high-risk individuals had a higher degree of pancreatic beta-cell dysfunction or excessive liver adipose tissue as part of non-alcoholic fatty liver disease.
This study was multi-center (took place at multiple institutions) in which patients were screened for prediabetes as determined by their 2-hour glucose levels during an oral glucose tolerance test (OGTT). If they had prediabetes, they were stratified into high-risk or low-risk groups. The high-risk group met criteria such as increased liver fat (MRS scan), reduced insulin secretion (during the OGTT) from beta-cells, or increased insulin resistance (using an index from fasting values of insulin and glucose).
The low-risk group was randomly split into the following groups.
- Low-Risk Control (LR-CONT): This group received one 30-minute consultation about nutrition and exercise after their prediabetes diagnosis. This is, sadly, the current “standard of care” for a majority of those diagnosed with prediabetes in the United States.
- Low-Risk Conventional Lifestyle (LR-CONV): This group had a diet and exercise goal aimed at 5% bodyweight reduction for those overweight subjects. They also got exercise and nutritional coaching eight times over a year.
- The diet goal was to reduce fat intake (<30% of daily calories), saturated fat (<10% of daily calories), and increase fiber intake (>15g/1000 calories).
- The exercise goal was to perform three hours of exercise weekly of a similar type as the performed in the Diabetes Prevention Program and other diabetes prevention interventions.
The high-risk group was randomly split into the following groups.
- High-Risk Conventional Lifestyle (HR-CONV): Identical intervention as the LR-CONV group.
- High-Risk Intensive Lifestyle (HR-INT): Same goal of 5% bodyweight reduction for those overweight subjects as in the LR-CONV group, with the same nutritional guidance. However, these subjects received 16 coaching sessions during the year, and their exercise goal was six hours of exercise weekly (of the same type as the 3-hour group).
Researchers tested subjects at baseline, 6 months of treatment, 1 year of treatment, and 2 years after treatment ended to measure long-term changes.
The study’s primary outcome was a 2-hour glucose level during an OGTT after one year of treatment. Secondary outcomes were liver fat, insulin sensitivity, insulin secretion, and cardiovascular risk. Tertiary outcomes were adherence to the five lifestyle goals (three diet-based, one exercise, one weight loss goal).
What It Found:
The stratifying of patients resulted in a high-risk group who were slightly older, had higher BMIs, worse fasting glucose, lipid, insulin sensitivity, insulin secretion levels, and just overall worse health.
(Before going into the results, it’s important to note that 1,105 subjects were randomized to a group, and only 908 [201 low-risk, 707 high-risk] of them finished the 1-year study. This dropoff is in line with other studies but still a possible source of survivorship bias: those that don’t complete the intervention and don’t respond positively are excluded from the final result.)
- Absolute change in 2-hour glucose following an OGTT: Researchers saw little to no difference within the high-risk and low-risk cohorts. Both saw similar results whether they did the conventional or intensive intervention. (Though it’s worth noting the low-risk group was a lot smaller than the high-risk group).
- Conversion from prediabetic to normal glucose tolerance (so an all-or-nothing measurement rather than an absolute change): The high-risk cohort with the more intensive program was 1.57 times more likely to move to a normal glucose tolerance than the conventional cohort. Within the low-risk group, those who did the intervention were 2.02 more likely to move into the normal glucose tolerance than the cohort with no intervention. (We can’t really compare results from the high-risk group to results from the low-risk group since the “controls” for each group were different.)
- Other physiological markers—liver fat, insulin secretion, insulin sensitivity, BMI, and cardiovascular risk factors: The high-risk intensive cohort had significantly more improvement than the conventional cohort in all measures except insulin secretion. For the low-risk group, only BMI and fasting glucose improved in those with the intervention.
- Adherence to lifestyle goals (important to see whether subjects could actually stick with more exercise and whether more exercise aided in weight loss): Interestingly, subjects who met their weight reduction and exercise goals were more likely to improve their 2-hour OGTT glucose. None of the dietary goals (less fat, less saturated fat, more fiber) were associated with 2-hour glucose improvements. Thus it appears exercise and weight loss mattered more than these specific diet changes.
Why It Matters:
Chronic diseases cost the medical system billions of dollars a year, and figuring out ways to prevent people on the verge of developing a chronic disease like Type 2 diabetes is extremely important to society’s health and economic health, not to mention that individual’s health. Given that medical systems are always resource-constrained, a screening approach that determines where and how to allocate resources best to prevent high-risk individuals from developing Type 2 diabetes is valuable.
Furthermore, even individuals diagnosed with prediabetes look different, with some being more or less likely to progress to Type 2 diabetes. Individualizing treatments, whether lifestyle or pharmacological, will hopefully improve the cost-effectiveness of therapies.
This study shows that high-risk individuals benefit from a more intensive lifestyle approach (6 versus 3 hours of exercise and 16 versus 8 visits with counselors) compared to the conventional approach and may need additional lifestyle interventions.
While this is a high-quality study with sufficiently large sample size, limitations and additional questions remain.
For instance, it would be helpful to have more detail on the exercise. Were outcomes associated with how much time people exercised (say, 6 hours vs. 7 or 8 or 10)? What modes of exercise: aerobic or resistance? At what intensity?
The dietary changes in this study focused on reducing fat and saturated fat and increasing fiber. It is possible that a lower carbohydrate diet may have further improved the lifestyle group and led to more weight loss. A design that was diet macronutrient agnostic and more focused on whole foods may have led to more pronounced effects in the lifestyle group as long as weight loss was achieved (through dietary adherence).
One interesting finding was that the more counseling sessions a subject in the high-intensity group attended, the better the outcomes. Is that because counseling is essential or because only the most motivated subjects go to counseling?
Differentiating between clinical significance and statistical significance is always a challenge, and additional measures may help clarify that. For instance, the number needed to treat is a valuable metric defined as the number of subjects required to treat for one additional positive event to happen (or adverse event avoided). In this study, “how many additional high-risk diabetics need to be treated for 1 year with an intensive lifestyle intervention versus a conventional lifestyle intervention for 1 additional high-risk subject to revert to normal glycemia?”. Running the calculations on my own for every 10 high-risk patients treated with the intensive lifestyle, one more will revert to normal glycemia. Whether this is efficacious enough given the cost and time of the treatment is not clear.
Related to this number needed to treat future work looking at the economic cost-benefit of treating 10 more high-risk prediabetics with 3 hours of additional exercise and 8 more counseling sessions to convert one more subject to normal glycemia are needed.
People with prediabetes who have impaired insulin sensitivity, insulin secretion, or increased liver fat benefit from a doubling of weekly exercise from 3 hours to 6 hours per week with additional counseling sessions. These are promising results even with less-than-perfect dietary guidelines and an exercise program lacking resistance training, which suggests that lifestyle interventions have the potential to be even more effective.