Cardiovascular Disease (CVD) Limbo: How Soon and Low Should We Go to Prevent CVD in Diabetes?

P eople with both type 1 and type 2 diabetes (T1D and T2D, respectively) continue to have increased rates of morbidity and mortality from cardiovascular disease (CVD) compared with people without diabetes, ^1,2 despite numerous advances in the past decades to better understand the pathophysiology, prevention, and treatment of CVD in diabetes. In adults with diabetes, an “ABC” approach has been advocated by the American Diabetes Association and American Heart Association to emphasize the importance of controlling blood glucose (glycated hemoglobin [A1c]), blood pressure (BP), and cholesterol, supported by numerous clinical trials. ³ Examples include the Diabetes Control and Complications Trial (DCCT)-Epidemiology of Diabetes Interventions and Complications (EDIC) data that intensive insulin therapy (and lower A1c) compared with conventional treatment over 6.5 years reduced CVD by 57% over a 17-year period. ⁴ Similarly, the United Kingdom Prospective Diabetes Study (UKPDS) reported continued reduction in CVD among patients with T2D who received intensive glucose therapy after 10 years of follow-up. ⁵ These data and others have led to widespread awareness of the importance of treating CVD risk factors in adults with diabetes, although more work needs to be done as macrovascular outcomes in diabetes have not improved as much as for microvascular disease. ⁶ It is possible there will be a time lag before treatment improvements become evident in CVD outcomes, but earlier onset of both T1D and T2D and increasing obesity (for example, in youth with T1D as well as T2D ⁷ ) may attenuate these advances.

Questions have been raised as to how low to aim for A1c, BP, and lipid control in adults with T2D. ⁸ In addition to the ABCs, other modifiable CVD risk factors exist in people with both T1D and T2D, including obesity, insulin resistance, kidney disease, inflammation, diet, exercise, and smoking, as well as non-modifiable risk factors, such as genetics and diabetes duration. Of the non-modifiable CVD risk factors, research continues to try to identify genetic markers to better stratify CVD risk. On the other hand, diabetes duration is consistently and strongly related to CVD in diabetes, and given the life-time increased risk for CVD, it raises the question of how soon and how low should we go in treatment of CVD risk factors, especially in the case of adolescents with diabetes. ⁹ In addition to these risk factors, glycemic variability ^10,11 and more recently hypoglycemia ^12,13 have been implicated as contributing to CVD. In this issue of Diabetes Technology & Therapeutics, Peña et al. ¹⁴ report that hypoglycemia but not glycemic variability (during 48 h of continuous glucose monitoring) was associated with impaired endothelial function in children with T1D. Future studies are needed to test the hypotheses that glycemic variability and hypoglycemia affect CVD in larger, longitudinal samples with longer duration of continuous glucose monitoring measurement. In addition, Theilade et al. ¹⁵ from the Steno Diabetes Clinic report on use of a more user-friendly device for 24-h ambulatory BP monitoring.

Although some uncertainties exist for CVD care in adults with diabetes, the current situation in adolescents and younger adults is much more speculative. The past decade has seen the first set of guidelines for treatment of CVD in pediatric diabetes from the American Diabetes Association, ¹⁶ the American Heart Association, ¹⁷ the American Academy of Pediatrics, ¹⁸ and the International Society of Pediatric and Adolescent Diabetes. ¹⁹ As a result, awareness of CVD as a health problem in adolescents with diabetes has increased. Data to support these guidelines were largely extrapolated from adults or from treatment of youth with other conditions (i.e., familial hypercholesterolemia for dyslipidemia medications or end-stage renal disease for BP) or epidemiologic studies such as the Pathobiologic Determinants of Atherosclerosis in Youth ²⁰ or the Bogalusa Heart Study. ²¹ What is also clear from the literature in the decade after these pediatric guidelines is that few youth with diabetes are receiving BP or dyslipidemia medications as recommended. For example, the SEARCH for Diabetes in Youth study (n=2,165 with T1D and n=283 with T2D) reported 15% and 24% of T1D and T2D youth, respectively, with low-density lipoprotein of >130 mg/dL, but only 1% received lipid-lowering medications ²² ; the German DPV study (n=27,358) reported 28.6% of youth up to 26 years of age had dyslipidemia, but only 0.4% were on lipid-lowering medications. ²³ Similar data from these studies for hypertension and BP medications exist with a hypertension prevalence of 5.9% in T1D and 23.7% in T2D youth but only 1.5% received BP-lowering medications in the SEARCH investigation, ²⁴ similar to the values of 8.1% with hypertension and 2.1% receiving medication in the DPV report. ²³ In defense of pediatric providers, these CVD guidelines are relatively new, few pediatric subspecialists entered the field with an interest or training in “adult” conditions such as CVD, and the guidelines themselves—despite being written with the best available evidence by experts in the field—lack the level of data that would be required for similar adult guidelines.

Improved care in adults with T1D since publication of the DCCT results in 1993 are suggested by data from the Epidemiology of Diabetes Complications and the Coronary Artery Calcification in Type 1 Diabetes (CACTI) cohorts with a reduction in mean A1c and increases in use of BP- and lipid-lowering medications. ^25,26 As already mentioned, in adolescents with diabetes use of such medications is rare, although glycemic control usually exceeds the target of 7.5% and on average approaches 9%—similar to the mean A1c in the conventionally treated arm of the DCCT. It is encouraging that recent data indicate some improvement in A1c in youth with T1D in the past decades from Australia, ^27,28 Norway, ²⁹ Denmark, ³⁰ and the DPV study. ³¹ Unfortunately, mean A1c values remain above 8%, and more improvements are needed. In light of the DCCT/EDIC data on “metabolic memory” viewed either as the persistently protective effect of intensively managed diabetes (with mean A1c=7.3% for an average of 6.5 years) or alternately as the persistently deleterious effect of conventionally managed diabetes (with mean A1c=approximately 9.1%), ³² these data on youth with T1D heighten concern for future vascular disease. Continued benefit of intensive therapy in T2D has also been reported by the UKPDS. ⁵ In youth with T2D data are more limited, but the SEARCH study reported a mean A1c of 7.99% with 27%>9.5%, ³³ and therefore reports of vascular complications in diabetes may shift thinking on treatment of CVD risk factors in youth with T2D. ^{34 –36}

Unfortunately, both T1D and T2D are increasing worldwide, as is obesity—a risk factor for hypertension, dyslipidemia, insulin resistance, inflammation, and CVD ³⁷ —which can be a perceived barrier to, or consequence of, intensive glycemic control. The various organizational guidelines for CVD risk factors in youth with diabetes are relatively consistent, as are the data for prevalence of dyslipidemia and hypertension and the infrequency of pharmacologic treatment. What then is needed to advance the field of primary prevention of CVD in youth with diabetes? Certainly, awareness of these guidelines among patients and their families should be improved. Adolescents and their parents are accustomed to the message of improved glycemic control, but efforts to lower BP and lipids are often new concepts and historically not emphasized until adulthood. More data are needed on appropriate targets for BP and lipids and how age, gender, and pubertal status influence these. Few providers, after appropriate evaluation and efforts at therapeutic life-style change, would hesitate to recommend an adolescent start a statin for a low-density lipoprotein level of >160 mg/dL or an angiotensin converting enzyme inhibitor for a BP of >99^th percentile, but what about other values above target but less extreme? How should family history, diabetes duration, other CVD risk factors, and family/adolescent preference affect this? How well do diet and exercise lower these risk factors (and how frequently does this succeed)? If the A1c is above target, is this the sole therapeutic focus, and if so after how many visits with an elevated A1c does one decide a once-daily pill for hypertension or dyslipidemia might lend itself to better compliance than (and in addition to) multiple daily doses of insulin? While commonplace in adult diabetes (for both provider and patient), lowering CVD risk is often a topic ignored or deferred given the other challenges of adolescence.

Fortunately, data on treatment of CVD risk factors from randomized clinical trials in adolescents are expected in the next few years in both T1D ³⁸ and T2D. ³⁹ Although these studies will lack data on long-term health outcomes, they will provide data on safety and surrogate markers of CVD risk for treatment with BP- and lipid-lowering medications (in addition to glycemic control). These data may provide initial results to improve compliance with CVD risk reduction guidelines.

Although clinical trial data with outcomes are needed (but perhaps may never be available) on how soon and how low to go to prevent CVD in youth with diabetes, other forms of data are accumulating since these first guidelines on CVD health in youth with diabetes. ⁴⁰ If primary prevention of atherosclerotic lesions—in contrast to clinical CVD—is the goal, then more aggressive treatment of glycemia, BP, and dyslipidemia are needed. Prior to such data, a balance is required between overly aggressive therapy and potential harm (such as angiotensin converting enzyme inhibitors and congenital malformations ⁴¹ ) and potential missed opportunity to improve the future vascular health of adolescents with diabetes. At present, for adolescents with diabetes, data suggest CVD risk factor prevention is in limbo between the publication and adoption of CVD risk factor reduction guidelines. For current practice, the likely answer for CVD prevention in adolescents with diabetes is as low and as early as safety allows, and in the absence of more compelling data this will likely vary for each patient and provider. In contrast, in adults CVD risk factor treatment goals will likely be refined and targeted based on recent data. Clearly data from DCCT-EDIC, UKPDS, and other studies indicate intensive glucose control decreases CVD events. However, there are no data to show how low we should aim. In fact, ACCORD indicated increased mortality in the intensive (<6%) versus the standard (7.0–7.9%) glycemic control, ¹³ even though the exact cause of the increased mortality is unknown. To decrease CVD safely we need to balance potential adverse and beneficial effects, and future studies are required to refine and individualize CVD recommendations.