- Yesterday, JAMA finally published new long-term mortality data from the DCCT/EDIC cohort showing a modest but statistically significant reduction in mortality with early intensive glycemic control after 27 years of follow-up (HR = 0.67; 95% CI = 0.46-0.99; p=0.045).
- These results offer compelling evidence for an extended “legacy effect” of early intensive control on long-term outcomes. We continue to await the publication of the full cardiovascular outcomes analysis; data we saw at an NIH workshop last fall showed continued (though modest) benefits in the intensive group.
Yesterday, JAMA published a portion of the long-awaited extended follow-up results from DCCT/EDIC showing a modest but statistically significant reduction in mortality with early intensive glycemic control compared to conventional treatment. This is the long-awaited mortality data from this trial; before 2013, there had not been enough events for sufficient statistical power at the time of previous analyses (the data is over a year old and was not immediately published for reasons that have eluded us to date). Over an average of 27 years of follow-up (and ~20 years after the end of the randomized treatment period), the risk of all-cause mortality was significantly lower in the intensive group than in the conventional group (hazard ratio = 0.67; 95% CI = 0.46-0.99; p=0.045), though the absolute risk difference was fairly small (~1/1,000 patient-years). Absolute mortality rates were relatively low in both groups (6.0% in the intensive group and 8.8% in the conventional group) and did not begin to diverge until 15 years after the beginning of the trial – that isn’t so surprising given that the 1,441 participants in DCCT weren’t particularly old to begin with in this trial. The most common causes of death were cardiovascular events (22.4%), cancer (19.6%), acute diabetes complications (17.8%), and accidents or suicide (16.8%; none attributed to hypoglycemia). Mortality risk was higher in patients with albuminuria or end-stage renal disease, higher mean A1c levels, and a history of hypoglycemia.
These results offer compelling continued evidence of a “legacy effect” of early intensive glycemic control on long-term outcomes in type 1 diabetes – it is quite impressive to see the benefits persist more than 20 years after the end of randomized treatment – and we hope that they will serve to emphasize the importance of aggressive, optimized treatment early in the progression of the disease. These results should also help alleviate any lingering concerns about intensive control possibly increasing the risk of mortality. As Drs. Michelle Katz (Joslin Diabetes Center, Boston, MA) and Lori Laffel (Massachusetts General Hospital, Boston, MA) note in an editorial published in the same issue of JAMA, the results illustrate the need for continued focus on improving glycemic control and reducing the rates of complications in this patient population. We also continue to eagerly await the publication of the ~30-year cardiovascular outcomes analysis from this trial; the topline CV data we saw at an NIH meeting last fall demonstrated a continued (though perhaps slightly diminished) benefit of intensive control on cardiovascular risk. Demonstrating a link between glycemic control and long-term CV benefit in type 1 diabetes has been challenging given the relatively small size and young average age of the patient population, and these results should go a long way toward cementing that relationship in patients’ and providers’ minds. We continue to applaud the DCCT/EDIC investigators and participants for their incredible dedication; the study boasts a remarkable 95% retention rate after 30 years (hats off, of course, to the patients in the trial).
- Results demonstrated significantly lower mortality in the intensive group compared to the conventional group over an average of 27 years since the beginning of the trial. As of the end of 2012, 43 deaths had occurred in the intensive group and 64 in the conventional group (mortality rates of 6.0% and 8.8%, respectively), leading to a hazard ratio of 0.67 in favor of the intensively managed group (95% CI = 0.46-0.99; p=0.045) and an absolute risk difference of 109/100,000 patient-years. The overall mortality rate was relatively low (0.29% per year) and was greater in men than women (HR = 1.61; 95% CI = 1.09-2.39; p=0.02) – it will be key to see how this changes over time as the cohort ages. Patients who died were older, had a later onset of diabetes, were more likely to be smokers, and had higher baseline systolic blood pressure, cholesterol, and A1c levels. There was no significant difference in mortality between the original primary prevention (those without microvascular complications at the beginning of the trial) and secondary intervention (those with microvascular complications at the beginning of the trial) cohorts.
- These results offer compelling evidence of a “legacy effect” of early intensive glycemic control on long-term outcomes in type 1 diabetes. Though the absolute risk reduction was modest relatively speaking, the fact that an average of 6.5 years of intensive control led to a net mortality benefit almost 20 years after the end of randomized treatment is very encouraging. The positive mortality results should certainly bolster the link between good glycemic control and long-term benefits in providers’ minds and serve to emphasize the importance of aggressive treatment early in the progression of the disease. As the article notes, the results should also alleviate any concerns about intensive therapy potentially increasing the risk of mortality – higher rates of hypoglycemia in the intensive group, as well as results from the ACCORD trial showing higher mortality rates with intensive therapy in type 2 diabetes, have prompted such concerns in the past.
- The results also illustrate the need for a continued focus on improving glycemic control and reducing complication rates in patients with type 1 diabetes. As Drs. Michelle Katz (Joslin Diabetes Center, Boston, MA) and Lori Laffel (Massachusetts General Hospital, Boston, MA) note in an editorial published in the same issue of JAMA, glycemic control remains suboptimal for the majority of patients with type 1 diabetes, and high rates of complications contribute to reduced life expectancy compared to the general population. The DCCT/EDIC results provide plenty of evidence to support that claim: two of the leading causes of death were cardiovascular events (22.4%) and acute diabetes complications (17.8%), and mortality risk was higher among patients with albuminuria (HR = 2.20; 95% CI = 1.46-3.31; p<0.01) and higher mean A1c levels (HR = 1.56 per 10% relative increase in A1c; 95% CI = 1.35-1.81; p<0.001).
- There is currently some disagreement in the field over how elevated the risk of mortality is in patients with type 1 diabetes compared to the general population. For example, the JAMA editorial cites two registry-based studies (by Livingstone et al and Lind et al) showing that mortality from cardiovascular disease is significantly higher in patients with type 1 diabetes than in those without, even in the absence of renal disease. However, it also cites two studies (by Orchard et al and Groop et al) that contradict that conclusion, finding that in the absence of renal disease, mortality rates are similar in people with and without type 1 diabetes.
- We continue to await the publication of the latest cardiovascular outcomes results from this trial; topline data presented at an NIH workshop last fall showed a continued (though slightly diminished) benefit in the intensive group. As a reminder, a 2004 analysis of EDIC data concluded that intensive treatment and lower A1c were associated with a significant 47% reduction in risk of a CVD composite (MACE + congestive heart failure, confirmed angina, silent MI, revascularization), but there was not enough statistical power at that time to draw full conclusions about the risk profile. The latest analysis, which included data collected through the end of 2013, found a significant reduction (with a numerically more modest point estimate) in the same CVD composite. The benefit with regard to MACE was no longer statistically significant, although to us the trend still appeared compelling. Notably, a multivariate model of 63 potential risk factors found that A1c was one of the top two or three predictors of CV and MACE risk, ahead of many traditional risk factors like microalbuminuria, blood pressure, and LDL cholesterol. Full CV results are expected in the near future and should continue to bolster the case for the benefits of intensive glycemic control in type 1 diabetes.
-- by Emily Regier, Manu Venkat, and Kelly Close