ABSTRACT

Hemoglobin A1c (A1C) or glycated hemoglobin reflects the levels of blood glucose during the previous 8–12 weeks duration. It also helps us to diagnose diabetes in some cases, during the preoperative screening, who were initially missed out. Although the number of patients with diabetes undergoing various surgeries has increased many times, the role of A1C as a predictor for the complications during the perioperative phase remains intriguing. This could be due to various factors such as lack of best shreds of evidence, various cut-off levels of target A1C, variations of the patient population, presence of other comorbid conditions, and so on. This narrative review article presents the role of A1C as a reflector of perioperative adverse events in various surgeries and discusses the controversies surrounding it. We searched “PubMed Central” database with search criteria of “hemoglobin A1c, glycated hemoglobin, and perioperative complications” with publication date from January 01, 2010, to January 31, 2020, and found a total of 214 articles. We included only the relevant articles to our topic and added a few more articles that we found as “secondary references” from those articles to suit the structured headings of our narrative review and made it a total of fifty. To our knowledge, the majority of the studies published on this topic are of the “Retrospective analysis” type of study, besides no narrative review article available to date in the literature. We suggest that assessment of A1C levels preoperatively can be used as a routine practice for major procedures in patients with diabetes and for patients who have persistent high glucose values during preoperative screening regardless of whether a diagnosis of diabetes is established or not. We found that a cut-off of 8% is acceptable for the majority of the surgical procedures. However, it is better to have a cut-off of 7% or lower for procedures such as spine and joint replacement surgeries, cardiac surgeries, and so on. Further prospective studies involving a large population preferably with a multicenter design would provide us more clarity on this topic.