nonpharmacological diabetes remission decreases CKD and CVD

 A post hoc analysis of the Look AHEAD study, which involved extensive nonpharmacologic therapies for those with diabetes, found that individuals achieving diabetes remission had significantly decreased incidence of chronic kidney disease (CKD) and cardiovascular disease (CVD): see dm remission dec CKD CVD Diab2024 in dropbox, or doi.org/10.1007/s00125-023-06048-6

Background information on the Look AHEAD study:

-- the Look AHEAD study is a multicenter RCT assessing the effect of a 12 year intensive lifestyle intervention in individuals with diabetes, finding that diabetes remission occurred in 12% of all intervention participants and also that 21% of those with fewer than two years of diabetes duration achieved remission the first year and 10% achieved two years of remission (remission was assessed annually)

-- the Look AHEAD study, a multicenter RCT assessing the effect of a 12 year intensive lifestyle intervention in individuals with diabetes, compared intensive lifestyle intervention (ILI) versus diabetes support and education (DSE) on CVD and other long-term health conditions 

-- the study included 5145 adults with diabetes and BMI >25 in individuals not on insulin , or BMI>27 if on insulin, aged 45-76.

-- Exclusion criteria included >1g/d protein excretion, serum creatinine >1.4 in women/>1.5 in men; A1c levels >11%

-- ILI included weekly group and individual sessions in the first six months, followed by two group sessions and one individual session per month for the second six months, two contacts per month (at least one in person) for years 2-4, and participants were encouraged to attend monthly support groups from years 4-12. The goal was to reduce total caloric intake to 1200-1800 kilocalories per day based on their initial weight, to reduce total fat and saturated fat intake to less than 30% and 10% respectively, increase physical activity to 135 minutes per week using brisk walking and other moderate intensity activities. Behavioral support included self-monitoring and problem-solving to assist in behavioral, weight, dietary, and physical activity goals 

-- DSE participants were offered three group sessions each year focusing on diet, physical activity, and social support, but no behavioral support 

-- Participants had annual follow-up for four years followed by visits every two years thereafter for 12 years. At these visits medication assessment was done along with health status by questionnaire, body weight, height, and labs for A1c, creatinine, urinary albumin/creatinine ratio 

-- diabetes remission was assessed at each visit, though specific dates of medication cessation was not elicited; therefore in their analysis of patients with one year of diabetes remission for example, the actual extent could have been significantly less than one full year to potentially up to almost 3 years 

    -- diabetes remission was defined as taking no diabetes medications and having a hemoglobin A1c <6.5% 

 

Details:  

-- this current post hoc analysis assessed the risk of CKD and CVD by this lifestyle intervention, where CKD high-risk was defined according to the KDIGO criteria (eGFR <45 based on creatinine, eGFR <60 with a urinary albumin-to-creatinine ratio of at least 30 mg/g, or just a urinary albumin-creatinine ratio >300 mg/g); CVD was defined as a composite of CVD death, nonfatal acute MI, nonfatal stroke, or admission for asthma 

-- for more details about the Look AHEAD study, see the commentary below

-- 58% female, mean age 59, duration of diabetes 6 years, BMI 36, Education level <13 years in 21%/13-16 years in 38%/ more in 40%, current smoking 4% 

-- the following were statistically significant differences at baseline in the group without remissions vs those with remissions: weight loss (-2.2kg vs -7kg), history of CVD (15% versus 12%), insulin use (95% versus 70%), diabetes duration (6 yrs versus 3 yrs), A1c (7.4% versus 6.8%) 

 

Results: 

-- Overall, 12.7% of the participants met their definition for remission in at least one follow-up visit 

-- at year one:  

    -- ILI group: the prevalence of remission was 11.2%, declining about 0.7 percentage points every year 

    -- DSE group: remained at about 2%  

        -- by year 4: the prevalence of any remission was 3.5 times as high in the ILI group (7.2%) versus the DSE group (2.1%) 

        -- by year 12: the prevalence of any remission was twice as high in the ILI group (3.7%) versus the DSE group (1.95%) 

-- Those patients who achieved remission were less likely to have prior CVD, and had lower baseline levels of hemoglobin A1c, fasting glucose, systolic blood pressure, and shorter duration of diabetes 

-- There were also major differences in weight loss between the groups, as noted above 

 

 

-- patients with any evidence of remission during follow-up, adjusting for hemoglobin A1c, blood pressure, lipids, CVD history, diabetes duration, and intervention arm, compared with participants without remission: 

    -- 33% lower rate of CKD, HR 0.67 (0.52-0.87) 

    -- 40% lower rate of the CVD composite, HR 0.60 (0.47-0.79) 

 

-- there was a dose response relationship in the rates of CKD and CVD, with each being lowest among participants who had remission for at least four visits: 

    -- CKD: 55% decreased risk, HR 0.45 (0.25-0.82) 

    -- CVD: 49% decreased risk, HR 0.51 (0.30-0.89) 

 

-- Comparing the 2 groups of patients, independent of their remission status:

    -- ILI had 40% lower CKD rate, HR 0.60 (0.44-0.82)

    -- ILI had 23% lower CVD rate, HR 0.77 (0.58-1.03), not quite statistically significant

 

-- Risk factor changes from 0 to 1 year in the study, comparing those with no remission to those with any number of years with remissions: 

    -- hemoglobin A1c -0.4% versus -0.7%, weight -4.0 kg versus -10 kg, fitness (METs) 11 versus 25 

-- risk factor changes from 0-4 years: 

    -- hemoglobin A1c -0.4 percent versus -0.7%, weight -2.2 kg versus -7 kg, fitness (METs) 0.6 versus 9 

-- there were small increases in LDL and decreases in HDL found in those having remissions

  

Commentary:  

-- as a background, there was an observational study (ie not assessing intentional weight loss) from the UK National Diabetes Audit of  2.3 million people with diabetes finding that diabetes remission was 1% in patients with diabetes and 3% in those with recent diabetes diagnosis (though in 8% in those with lots of weight loss, with a gradient of diabetes remission with increased weight loss): https://diabetesjournals.org/care/article/45/5/1151/144853/Incidence-and-Characteristics-of-Remission-of-Type

-- other studies (DiRECT and DIADEM-I) have found even more impressive diabetes remission rates than the Look AHEAD post hoc study; these studies were more based on diet than exercise and with a goal of more aggressive dietary restriction (eg 800-850 kcal/d). However, these studies were shorter-term and had a more active protocol directed to medication removal (Look AHEAD had a 1200-1800 kcal goal and left medication changes up to individual practitioner decision)

 

--this was an impressive study reaffirming the value of aggressive nonpharmacologic therapy in creating remission of diabetes and dramatically decreasing the evidence of clinical CKD and CVD. 

-- It does raise the issue of diabetes remission associated with medications. In particular, the GLP-1 receptor agonists as well as tirzepatide can lead to dramatic decreases of the hemoglobin A1c, down to levels in the low 5% range (ie diabetes remission but with meds)

    -- though, unfortunately, it was very hard for many people to maintain their diabetes remission over time nonpharmacologically: about a 4% decrease in the percent of those in remission by the 8^th year of the study (though there continued to be a decrease in CKD/CVD with any interval of remission). But, for those unable to achieve a sustainable remission, there may be an important role for metformin or GLP-1 agonists/tirzepatide

-- and, as per the last blog, a significant number patients can maintain their levels of weight reduction (and likely diabetes remission) even when that medication is stopped, especially with continued nonpharmacologic therapies (diet and exercise): https://gmodestmedblogs.blogspot.com/2024/02/obesity-weight-changes-when-stop.html  

-- it is also notable that independent of the diabetes remission rate, there were major decreases in CVD and CKD in those with ILI, suggesting that the intensive diet/exercise program with intensive reinforcement with this protocol consisting of much instruction and phone call/in-person reinforcement is likely in itself important, though the numbers of people with diabetes remission is much less in those with DSE (which is no doubt part of the issue)

    -- but it is clear from a multitude of studies that healthy diet and exercise is really important for health overall (ie, not just targeting diabetes). For example, the Mediterranean diet decreases not just CVD but also CKD: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9824533/ . so how much of the benefit is specifically from achieving diabetes remission vs just losing weight vs the broad benefits of a healthier diet and more exercise???

-- Those individuals  with the highest remission rate tended to be those with a short duration of diabetes, lower baseline A1c levels, and those achieving a more impressive weight loss

    -- that being said, I have (anecdotally) had several patients over the years who have quite remarkable diabetes (A1c >14%), who have had complete long-standing remissions with just diet and exercise. A few comments:

        -- they achieved these quite high A1c levels slowly, so did not notice significant symptoms until their A1c levels were very high (at which point they started to feel fatigued)

        -- when they improved their glucose control by meds (metformin and insulin, the latter being used to get a more rapid blood sugar reduction), they could be titrated off the insulin and then the metformin

            -- this was, I suspect, related to glucotoxicity of the high blood sugars: even non-diabetics getting high blood sugars through high glucose loads in an experimental setting have decreased insulin effect, as measured by impaired beta-cell function, as well as decreased insulin-mediated glucose uptake (ie insulin resistance peripherally)

        -- these patients were able to maintain A1c levels in the 5-5.5% just with aggressive diet/exercise/weight loss for years (though one stopped the diet/exercise, re-developed diabetes, and is now back on diet/exercise without a problem for a few years now)

        -- this issue of glucotoxicity is also present in some patients who develop an insulin-resistant state, such as a urinary tract infection, get high A1c levels that remain high after the UTI is treated, get meds for awhile to decrease their blood sugars, then no longer need meds…

        -- the point here is that promoting aggressive diet and exercise does work in some individuals who do not fit the above phenotype, but seem to have very bad diabetes but are still responsive to aggressive diet/exercise

-- so, the goal of trying to achieve diabetes remission is not simply to improve the diabetes, which does help with several pathologic hyperglycemia-related issues. Many of the adverse effects are related to insulin resistance, which is associated with  high insulin blood levels, high thrombotic risk (insulin is the main mediator of plasminogen activator inhibitor, PAI-1, levels), hypertension, hyperuricemia, hyperlipidemia and endothelial dysfunction. And many of these risk hyperinsulinemia effects are associated with CVD and CKD even in the absence of diabetes. This suggests that the role of diet and exercise in decreasing CVD and CKD may well be at least in part independent of their role for diabetes remission

-- and there may be an important psychological benefit to the aggressive diet/exercise in terms of self-esteem and motivation, as well as stress-reduction, which themselves may decrease bad outcomes (eg stress may well be a more important cause of cardiovascular events in those with CVD than exercise: https://gmodestmedblogs.blogspot.com/2022/01/stress-induced-cardiovascular-disease.html ). This Look AHEAD study did not measure these factors of stress, quality-of-life, depression, etc, which may well be as important as the CVD/CKD outcomes they did measure….

 

Limitations:

-- as noted above, the definitions of the length of time in diabetes remission are quite broad, since they were based on a single A1c value on a yearly exam, so the 1-year remission category may reflect almost 2 years of remission, but also only a few months

-- we do not have granular data on the actual diets and exercise done in those randomized to the 2 groups. It is certainly possible that some in the DSE group initiated much more aggressive lifestyle changes than some in the ILI group, obscuring the above results (ie, we do not have analysis of the actually achieved diet and exercise and the diabetes remission rates)

-- not enough granular data to see if there were important subgroups with different results, such as by sex, or comorbid conditions (only current smoking and history of CVD were included in their analysis: not a very extensive list and in both cases were pretty low numbers suggesting this was a quite healthy population, also limiting generalizability of their results)

-- there was no information in the article about medications taken. For example, the unexpected increase (though small) of LDL and decrease of HDL in the ILI group may have been because more of the DSE group had more adverse lipids and were put on statins?

-- there was not even granular information on what diabetes meds people were taking or what criteria the medical providers used to titrate down or stop the medications

-- there is not enough information to determine whether it was improvement in the diabetes control vs controlling the CVD or CKD risk factors themselves that led to the decrease in CVD/CKD events

-- there was a limited approach to healthy “lifestyle changes” as only diet/exercise in leading to decreased CVD/CKD. It is very clear in the medical literature that other factors are really important, including stress levels, depression, air pollution, smoking, chronic inflammatory conditions, access to good food/exercise venues, psychological support systems, etc (see https://gmodestmedblogs.blogspot.com/2023/10/update-ascvd-risk-factor-critique.html ). And these were not evaluated or addressed in this study

-- there were exclusion criteria that would exclude many patients, particularly those with relatively small increases in their eGFR (at least a very large % of my patients with diabetes could not have been in this study).

    -- and, as noted in many blogs, eGFR has a poor correlation with measured GFR (https://gmodestmedblogs.blogspot.com/2022/07/egfr-not-such-great-estimate-of-renal.html ), and cystatin-based eGFR seems to be a better predictor of bad clinical outcomes than creatinine-based eGFR (https://gmodestmedblogs.blogspot.com/2023/12/cystatin-c-better-predictor-of-bad.html )

 

So,

-- a very useful study promoting the very real importance of lifestyle changes (specifically diet and exercise in this study) as a means to achieve diabetes remission (with the apparent major driver being weight loss)

-- though it is clear here that these lifestyle changes are transient for many people, there seems to be benefit even with short intervals of diabetes remission

-- it is also clear in the medical literature that these lifestyle changes are important for a large variety of clinical outcomes, not just diabetes remission, but also many of our chronic medical conditions (hypertension, hyperlipidemia, dementia, heart disease, CKD,  etc etc)

-- many people with short-term benefit from lifestyle changes may well also need medications to control their diabetes (and other medical conditions that might get better with weight loss, or healthier diet/exercise), which reinforces these still important lifestyle changes even if adding back diabetes meds (esp metformin, GLP-1 agonists, tirzepatide, SGLT-2’s)

    -- and, our approach to many of the chronic diseases in our society really should be an increasingly holistic one, not just meds but also the social/economic/political/lifestyle issues creating most of these chronic diseases….

 

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

  

to get access to all of the blogs:

 

 go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them

 

if you would like to see the articles in this blog, please email me. 

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list