Hypoglycemia and Cardiovascular Events: Direct Relationship or Explained by Co-Morbid Illness?
Editor's Note: Commentary based on Goto A, Arah OA, Goto M, Terauchi Y, Noda M. Severe hypoglycaemia and cardiovascular disease: systematic review and meta-analysis with bias analysis. BMJ 2013;347:f4533.
Hypoglycemia in persons with type 2 diabetes has been implicated as a risk factor for the occurrence of cardiovascular events but this association is controversial. In particular, it is thought that the higher prevalence co-morbid conditions in persons with diabetes may explain the observed association.
Goto et al. undertook a systematic review and meta-analysis to summarize the literature on severe hypoglycemia and cardiovascular risk.
The authors identified and synthesized results from six prospective studies with relevant data on severe hypoglycemia and cardiovascular risk. They report a pooled relative risk of 2.05 (95%CI 1.74 to 2.42) for the association of hypoglycemia with the occurrence of cardiovascular disease and a population attributable fraction (PAF) of 1.56% (95%CI 1.32% to 1.81%). A bias analysis evaluating whether residual confounding due to comorbid illness was also reported; the authors found that severe comorbid illness was unlikely to explain the observed association.
This study reports that persons experiencing severe hypoglycemia are at significantly higher risk of cardiovascular events and that this association is not likely to be explained by confounding by co-morbid illness.
Meta-analysis is a powerful and informative approach to more precisely summarizing an association between and exposure and outcome in the existing literature. Nonetheless, a meta-analysis is only as good as its constituent studies; the limitations of the meta-analysis will reflect the limitations of the literature. A concern here is the potential for residual confounding of the association of hypoglycemia and cardiovascular events by the presence of co-morbid conditions. The authors address this issue using a "bias analysis."
Goto et al. find that the pooled relative risk of 2.05 is unlikely to be explained by residual confounding because the prevalence of such comorbid conditions would need to be high and the associations with cardiovascular disease strong to substantially attenuate the observed association. While the abstracted RRs from the included studies were adjusted for various risk factors, the number of covariates, their ascertainment and measurement, and methods of adjustment (e.g. time-varying vs not) varied widely across studies. This is important since conditions that can contribute to the risk of hypoglycemia such as low kidney function, cognitive impairment, frailty, co-existing cardiovascular disease and other comorbidities, longer duration or difficult to control diabetes, and polypharmacy are very common in elderly persons with diabetes. Indeed, the prevalence of kidney disease among persons 65 or older with diabetes alone is over 50%1 and rates of multiple medication use are extremely high.2
The study by Goto et al. also provides an estimate of the population attributable fraction (PAF) of 1.6%, interpreted as the risk of cardiovascular events in the total population of persons with type 2 diabetes that is directly attributable to hypoglycemia. Attributable risk calculations are tricky because they involve a number of assumptions. First, the PAF assumes a direct causal relationship between hypoglycemia and cardiovascular events. Second, it assumes that the relative risk has been correctly specified. The PAF is also dependent on the prevalence of exposure in the population. Here, the authors assumed a "prevalence" of hypoglycemia of 3.1%, which is undoubtedly an overestimate since the condition is transient and cannot be interpreted without a clearly specified time interval. It is unknown if low blood glucose was present immediately prior to any of the cardiovascular events identified in these studies.
Ultimately, this study highlights the paucity of data on hypoglycemia and cardiovascular risk and our lack of understanding of the potential mediators of this association. The aging of the population and the epidemic levels of diabetes in this country highlight the critical need for further studies to inform how best to prevent cardiovascular outcomes in persons with diabetes, particularly older adults. The authors' conclusions regarding individualization of therapy among elderly persons with diabetes are well taken. Elderly persons with diabetes are a heterogeneous population3 and much of the existing evidence underlying diabetes management guidelines are based on evidence from middle-aged adults. There is growing recognition that less stringent glycemic targets may be appropriate among elderly persons and/or those with substantial co-morbid conditions due to the risk of hypoglycemia.4 Additional work is needed to help identify the best approaches to diabetes management in older persons, particular those with co-morbid conditions and at high risk for hypoglycemia.
- de Boer IH, Rue TC, Hall YN, Heagerty PJ, Weiss NS, Himmelfarb J. Temporal trends in the prevalence of diabetic kidney disease in the United States. JAMA 2011;305:2532-9.
- Stuart B, Simoni-Wastila L, Yin X, Davidoff A, Zuckerman IH, Doshi J. Medication use and adherence among elderly Medicare beneficiaries with diabetes enrolled in Part D and retiree health plans. Med Care 2011;49:511-5.
- Selvin E, Coresh J, Brancati FL. The burden and treatment of diabetes in elderly individuals in the U.S. Diabetes Care 2006;29:2415-9.
- Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al. Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 2012;35:1364-79.
< Back to Listings