Haemoglobin A1c (HbA1C) in non-diabetic and diabetic vascular patients. Is HbA1C an independent risk factor and predictor of adverse outcome?

Background. Plasma Haemoglobin A1c (HbA1c) reflects ambient mean glycaemia over a 2-3 months period. Reports indicate that patients, with and without diabetes, with an elevated HbA1c have an increased risk of adverse outcome following surgical intervention. Our aim was to determine whether elevated plasma HbA1c level was associated with increased postoperative morbidity and mortality in patients undergoing vascular surgical procedures. Methods. Plasma HbA1c was measured prospectively in 165 consecutive patients undergoing emergency and elective vascular surgical procedures over a 6-month period. Patients were categorized into four groups depending on whether their plasma HbA1c was <= 6%, 6.1-7%, 7.1-8% or >8% and clinical data was entered into a prospectively maintained database. Patients were also classified by diabetic status with suboptimal HbA1c in a patient without diabetes being >6 to <= 7% and suboptimal HbA1c in a patient with diabetes being >7%. Patients with plasma HbA1c >7% were reclassified as having undiagnosed diabetes mellitus. Composite primary endpoints were all cause 30-day morbidity and mortality and all cause 6-month mortality. Composite secondary endpoints were procedure specific complications, adverse cardiac events, stroke, infection and mean length of hospital stay. Results. Of the 165 patients studied, 43 (26.1%) had diabetes and the remaining 122 (73.9%) did not. The mean age was 72 years and 59% were male. Suboptimal HbA1c levels were found in 58% patients without diabetes and in 51% patients with diabetes. In patients without diabetes those with suboptimal HbA1c levels (6-7%) had a significantly higher incidence of overall 30-day morbidity compared to patients with HbA1c levels <= 6% (56.5 vs 15.7%, p<0.001). Similarly, for patients with diabetes those with suboptimal HbA1c levels (HbA1c >7%) had a significantly higher incidence of 30-day morbidity compared to those with HbA1c levels <= 7% (59.1% vs 19%, p=0.018). Multivariate analysis revealed that a plasma HbA1c level of >6 to <= 7% was a significant independent predictor of overall 30-day morbidity in patients without diabetes undergoing vascular surgical procedures. No difference in mortality, composite secondary endpoints, procedure specific complications, stroke or mean length of hospital stay was observed between any of the groups in the study. Conclusion. Suboptimal HbA1c levels may hold prognostic significance in patients without diabetes undergoing vascular surgery.