Perioperative glucose control: what is enough?

TYPE 2 diabetes mellitus, impaired fasting glucose/impaired glucose tolerance, and stress-induced hyperglycemia (SIH) are ubiquitous in the adult population and represent major public health concerns. Almost 10% of adult Americans have type 2 diabetes mellitus, an additional 20–25% have impaired glucose tolerance/impaired fasting glucose, and an unknown number develop SIH. Upwards of one third of affected patients are unaware of the presence of dysglycemia and its systemic effects. Projections predict a continued, dramatic increase in the incidence and prevalence of type 2 diabetes over the next several decades, with its deleterious impact on quality of life and life expectancy. In this issue of ANESTHESIOLOGY, Drs. Lipshutz and Gropper address the impact of dysglycemia on perioperative management. Patients with diabetes require acute and critical care, procedural interventions, and hospitalizations more commonly than those with normal glucose tolerance. When patients with diabetes require hospitalization or undergo certain procedures, they sustain greater morbidity and mortality. Studies from this decade have shown that a minimalist approach to glucose control in selected perioperative and critically ill patient populations is unwarranted, and improved glucose control leads to less morbidity and better outcomes, particularly in those with SIH. Key questions remain unanswered. How tight should glycemic control be? Are all hyperglycemic patients at equal risk for morbid and lethal events at a given degree of dysglycemia? What is the incidence and degree of morbidity when tight glycemic control (TGC) is universally applied? Identification of the dysglycemic patient and application of reliable glucose monitoring and glucose management techniques to a proper endpoint are crucial to achieving adequate perioperative glucose control. Identification of new-onset glucose intolerance in the perioperative patient should be followed by appropriate referral to the patient’s primary care provider for ambulatory unstressed diabetes testing. Drs. Lipshutz and Gropper emphasize that the current data reporting the benefits in reducing morbidity and mortality in intensive care unit patients using intensive insulin therapy to provide TGC be interpreted with care in light of risks reported when this approach is applied universally. They comment on the potential differences in glucose control and outcome related to type 1 versus type 2 diabetes or SIH, the effect of glucose variability during the course of intense monitoring and therapy, and the current risk-benefit data on TGC in various populations. They caution about extrapolating intensive care unit studies directly to the perioperative patient. We would go a step further and caution against a sudden call for intraoperative normalization of blood glucose (80–110 mg/dL; 4.4–6.1 mmol). Additional data should be obtained before implementing rigid perioperative standards of glucose management while tying reimbursement for care of the hyperglycemic perioperative patient to potentially unsubstantiated goals. This thorough review briefly comments on the importance of glucose monitoring, quality control of bedside glucose measurements versus laboratory techniques, and attempts at developing continuous and closed loop systems to control glucose. The reliability of glucose measurements is important to remember when controlling glucose levels during the dynamic perioperative period. Practical pitfalls in glucose monitoring secondary to sample site and source, technique of monitoring, impact of concurrent pathophysiologic states and interfering substances such as nonglucose sugars, and various medications are now recognized. The source of glucose monitoring, point-of-care device, blood gas analyzer, or central laboratory evaluation may explain some of the conflicting results reported when intensive insulin therapy and TGC protocols are instituted. Point-of-care glucose monitoring using finger-stick capillary blood, the most common approach to perioperative evaluation, is based on application of ambulatory technology using photoreflectometry or electrochemical reaction. The Food and Drug Administration mandates a 20% agreement between the pointof-care device and laboratory gold standard.§ Differences between laboratory and point-of-care–derived values are particularly important in intensive care unit patients who are anemic, hypothermic, or hypoperfused. Potentially critical disagreements between the central laboratory value and point-of-care measurement may lead to inappropriate insulin management. Certain operative patients, particularly those in shock or actively hemorrhaging, are likely to be affected. Multidisciplinary teams should develop glucose control protocols, set reasonable goals for control, monitor the effectiveness of controlling glucose, and recognize This Editorial View accompanies the following article: Lipshutz AKM, Gropper MA: Perioperative glycemic control: An evidence-based review. ANESTHESIOLOGY 2009; 110:408–21.