Cause-Specific Mortality Due to Hypoglycemia

A retrospective observational study on hypoglycemia and mortality seeks to determine absolute risk of death due to CVD, cancer, and other causes in patients who experienced a severe hypoglycemic event. 

While it is recognized that severe hypoglycemia can lead to fatal complications, such as arrhythmias, seizures, and coma, the long-term profile of severe hypoglycemia is unknown. The 2008 Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial showed that there was lower risk of death due to severe hypoglycemia in patients receiving intensive control than in conventional treatment patients. However, the Action in Diabetes and Vascular Disease: Preterax and Diamicron Controlled Evaluation (ADVANCE) trial showcased that patients with hypoglycemic events requiring medical assistance also had increased risk of developing cancer and skin, respiratory, and digestive complications. Thus, this study seeks to assess the long-term risk of CVD and non-CVD deaths associated with hypoglycemia.  

This five-year retrospective observational study aims to determine absolute risk of death due to CVD, cancer, and other causes in patients who experienced a severe hypoglycemic event. Patient data from the Clinical Practice Research Datalink (CPRD) database containing a record from 11 million patients from 674 practices in the UK was utilized for this study. Patients that were 18 years or older with a type 2 diabetes mellitus (T2DM) diagnosis between January 1, 1998, and January 1, 2011, were included in the cohort. Patients without data linking to the Hospital Episode Statistics database or the Office for National Statistics were excluded from this study. The primary outcome was the incidence of severe hypoglycemia resulting in hospital admission. The secondary outcome was the index date of myocardial infarction, stroke, peripheral artery disease, heart failure, atrial fibrillation, or cancer.  

A total of 74,610 subjects were used for the final cohort, of which 388 (0.5%) subjects had at least one hospital admission for hypoglycemia. The average age of patients admitted for severe hypoglycemia was 73.9 (60.3-81.8), while the average age of patients who were not admitted was 67.7 (58.4-76.5). The HbA1c for those admitted was 7.5% (6.4-8.8) versus 6.9% (6.3-7.7) in those not admitted for severe hypoglycemia. In hospitalized patients, comorbidities included myocardial infarction (14.7%), stroke (25%), peripheral arterial disease (12.9%), heart failure (20.6%), atrial fibrillation (23.3%), and cancer (16.2%). On the other hand, in patients that were not hospitalized, the comorbidities differed: myocardial infarction (9%), stroke (9.9%), peripheral arterial disease (3.8%), heart failure (6.1%), arterial fibrillation (10.3%), and cancer (13.5%). With the 74,610 subjects’ data, the Royston-Parmar-Lambert parametric survival model was used to determine the five years of cause-specific death.   

Through this cohort, it was observed that subjects who experienced a severe hypoglycemic event were more likely to be older with white ethnicity (93.3% in hospitalized vs. 89.7% in non-hospitalized), had a lower BMI (28.3 vs. 30.1), eGFR (63 vs. 77 mL/min/1.73m2), and poor glycemic control (7.5% vs. 6.9%). The risk of the severe hypoglycemic event was observed with the use of insulin (absolute difference 28.7%) as well as a medical history of cancer (absolute difference 2.7%), myocardial infarction (5.7%), peripheral artery disease (9.1%), atrial fibrillation (12.9%), or stroke (15.1%). The mortality increased over five years for all ages, with the main difference resulting in the cause of death. The CVD mortality risk for 50-year old individuals with hypoglycemia was calculated to be 5.6%, which increased to 6.6% at age 60, 7.6% at age 70, and 12.6% at 80. However, in patients without hypoglycemia, it was calculated that the CVD mortality risk was 1.0% at age 60, 1.9% at age 70, and 8.8% at age 80.  

In patients with and without hypoglycemia, death’s most common cause was not a CVD event regardless of hospitalization. In contrast, the most common cause of death was cancer, along with other reasons, which accounted for 71% of the deaths associated with this study. This study was one of the few that quantified the absolute risk of hypoglycemia outcomes due to T2DM. However, this study defined hypoglycemia as a hospitalization event and utilized a database from the UK. Similar studies using the database from other countries should be conducted to determine generalizability.   

Practice Pearls: 

• Patients that are older, of white ethnicity, with a low BMI and low eGFR, with poor glycemic control are at high risk of severe hypoglycemia. 
• The risk of hypoglycemia increases with the use of insulin and medical history of cancer. 
• The CVD mortality in patients with hypoglycemia is 5.6%. 6.6%, 7.6%, and 12.6% at age 50, 60, 70, and 80, respectively. 

Bonds, D. E, et al.”  “The Association between Symptomatic, Severe Hypoglycaemia and Mortality in Type 2 Diabetes: Retrospective Epidemiological Analysis of the ACCORD Study” BMJ 

Zaccardi, Francesco, et al.”  “Severe Hypoglycaemia and Absolute Risk of Cause-Specific Mortality in Individuals with Type 2 Diabetes: a UK Primary Care Observational Study” Diabetologia,  

Shana Indawala, PharmD Candidate, University of South Florida Health, Taneja College of Pharmacy