Hypoglycemia Raises In-Hospital Mortality Risk

Miriam E. Tucker

December 26, 2012

Hypoglycemia, both insulin-induced and spontaneous, increased the risk for mortality among hospitalized patients, a retrospective cohort study has found.

The findings were published online December 17 in Diabetes Care by Rajesh Garg, MD, from the Division of Endocrinology, Diabetes, and Hypertension and the Center for Clinical Investigation, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, and colleagues.

"[H]ypoglycemia in hospitalized patients, whether insulin induced or spontaneous, is associated with increased mortality as compared with absence of hypoglycemia," the authors write.

Study participants included 1096 hospitalized patients who had at least a single blood glucose value of 50 mg/dL or less during their hospital stay. Of those patients, 961 were insulin-treated and 135 were not. None were receiving oral diabetes medications while in the hospital.

Controls were 1794 patients hospitalized during the same period and matched for sex, race, and age decade who had not experienced any blood glucose value less than 70 mg/dL during their hospital stay. Of those patients, 736 were insulin-treated and 1058 were not.

In-hospital mortality was 22.1% among the total group that experienced hypoglycemia compared with just 2.5% among those who did not, a highly significant 9-fold difference (P < .0001).

Among all the patients experiencing hypoglycemia, mortality was significantly lower among those treated with insulin, at 20.3%, compared with 34.8% of the patients with spontaneous hypoglycemia (P < .0001).

However, among all the patients treated with insulin, mortality was significantly greater among those who experienced hypoglycemia than among those who did not (20.3% vs 4.5%; P < .0001).

The non–insulin treated patients who experienced spontaneous hypoglycemia were the sickest overall, with an average Charlson Comorbidity Index (CCI) of 2.4, followed by the insulin-treated group with hypoglycemia (CCI, 1.8) and the insulin-treated patients without hypoglycemia (CCI, 1.5). The non–insulin treated patients without hypoglycemia were the least sick, with a CCI of 1.1.

Admission to the intensive care unit (ICU) was greatest for the insulin-treated group with hypoglycemia (62.0%) followed by the non–insulin treated group with hypoglycemia (51.1%; P = .01 for the difference between the insulin-treated and spontaneous hypoglycemia groups). The patients without hypoglycemia were less likely to be admitted to the ICU: 29.1% of those treated with insulin and 10.1% of those not treated with insulin (P < .0001 for the difference between the 2 types of non-hypoglycemia treated groups).

On univariate regression analysis, both the CCI and ICU admission were associated with higher mortality, whereas female sex and insulin treatment reduced the risk, the authors note.

In a multivariate analysis adjusting for CCI, age, sex, and admission to the ICU, the hazard ratio for death after hypoglycemia for those treated with insulin compared with those not treated with insulin was 0.34 (95% confidence interval, 0.25 - 0.47; P < .0001).

The authors acknowledge several study limitations including the use of only a single study center, a small sample size in the spontaneous hypoglycemic group, the possibility of unmeasured cofounders such as use of other medications such as steroids, and a lack of data on nutritional status, duration of diabetes, insulin regimen, glucose control, or temporal relationship of hypoglycemic event to use of insulin.

"Hypoglycemic patients often have more severe illnesses. Most studies, in the inpatient setting as well as in the outpatient setting, have demonstrated that spontaneous hypoglycemia is associated with increased mortality as compared with insulin-induced hypoglycemia, and our study supports these data," Dr. Garg and colleagues write. "However, our study also shows that mortality in the [insulin-treated hypoglycemic] group was still very high, suggesting only a partial, if any, protective effect of insulin. We think insulin administration may unmask hypoglycemia more easily in sicker patients. Whether hypoglycemia causes increased mortality remains unclear."

This study was supported by the Brigham and Women's Hospital and Harvard Catalyst, the Harvard Clinical and Translational Science Center (National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health ,Award and financial contributions from Harvard University and its affiliated academic healthcare centers). The authors have disclosed no relevant financial relationships.

Diabetes Care. Published online December 17, 2012. Abstract