HbA1c Minimally Improves CVD Risk Prediction in Patients With, Without Diabetes

Inclusion of hemoglobin A1c as a factor in cardiovascular disease risk prediction scores only had modest benefit.

Hemoglobin A1c (HbA1c) minimally improves cardiovascular disease (CVD) risk prediction in patients with prediabetes and in those without diabetes, according to study results published in Diabetes Care.

The European Systematic COronary Risk Evaluation (SCORE) CVD risk score, QRISK3 risk score, and American College of Cardiology/American Heart Association (ACC/AHA) CVD risk score do not include any specific measure of glucose control, but only include diabetes as a categorical entity. The goal of this study was to explore the added value of inclusion of HbA1c in risk prediction.

The researchers used data from the UK Biobank and focused on 357,833 participants without baseline CVD or known diabetes, including 343,595 patients (96%) with normal HbA1c (<6.0%), 11,665 patients (3.3%) with prediabetes (HbA1c 6.0%-6.4%), and 2573 participants (0.7%) with undiagnosed diabetes (HbA1c ≥6.5%). Furthermore, data on 16,596 patients with known baseline diabetes were collected.

Among patients without diabetes at baseline, during a median follow-up time of 8.9 years, the QRISK3-based fatal/nonfatal CVD outcome occurred in 12,877 patients (3.6%), ACC/AHA-based CVD outcomes occurred in 6608 (1.9%), and SCORE-based fatal CVD occurred in 1803 (0.5%). HbA1c levels were higher in patients who subsequently had an incident CVD event (5.4% among patients without a CVD event compared with 5.5% among those with CVD events).

For patients with diabetes, during a median follow-up of 8.7 years, the QRISK-based CVD outcome occurred in 1472 subjects (8.9%) and fatal CVD occurred in 306 participants (1.8%).

Compared with patients with normal HbA1c, risk for QRISK3-based fatal/nonfatal CVD was higher among patients with prediabetes (hazard ratio [HR], 1.83; 95% CI, 1.69-1.97) and among those with undiagnosed diabetes (HR, 2.26; 95% CI, 1.96-2.60). After adjustment for classical risk factors, these HRs fell to 1.11 (95% CI, 1.03-1.20) and 1.20 (95% CI, 1.04-1.38), respectively.

Related Articles

The addition of log HbA1c to a model of CVD prediction based on QRISK3 CVD outcome among participants without known diabetes only slightly improved the predictive value (C-index increase of 0.0004; 95% CI, 0.0001-0.0007). Patterns were similar for the other outcomes of interest, including the SCORE and the ACC/AHA CVD risk scores.

The researches acknowledged several limitations to the study, including those secondary to the UK Biobank cohort, which includes that the population is somewhat healthier than the average population in the United Kingdom, and a lack of other measures to assess glucose homeostasis.

“Our results add to existing literature by validating prior results published by the Emerging Risk Factors Collaboration (ERFC) that demonstrated that while HbA1c levels better predicted incident CVD events than fasting and postprandial glucose levels in those without prior diabetes or CVD, the added predictive gain from inclusion of HbA1c in risk prediction was modest,” wrote the researchers.

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.

Follow @EndoAdvisor


Welsh C, Welsh P, Celis-Morales CA, et al. Glycated hemoglobin, prediabetes, and the links to cardiovascular disease: data from UK Biobank [published online December 18, 2019]. Diabetes Care. doi:10.2337/dc19-1683