Amidst controversy about the pros and cons of tight glycemic control, an expert panel convened by The Endocrine Society is now recommending blood glucose testing for all patients on admission to a hospital (J Clin Endocrinol Metab 2012;97:16–38). The panel also proposed glycemic targets and focused on process and system improvements, with the goal of improving care for patients with hyperglycemia and diabetes. The guideline was intended to build upon a consensus statement on inpatient glycemic control issued jointly in 2009 by The Endocrine Society and American Diabetes Association, which focused more on critically ill patients (Diabetes Care 2009;32:1119–31).

The eight-member panel advised that all patients, whether diabetic or not, have laboratory-based blood glucose testing on admission. They also recommended bedside point-of-care testing (POCT) for at least 24–48 hours in patients without a history of diabetes but with blood glucose levels >140 mg/dL. In doing so, the panel cited observational studies that have found hyperglycemia in 32–38% of patients in community hospitals, but in whom a sizable minority had no history of diabetes.

"Hyperglycemia is very common in hospitalized patients, particularly among those with cardiovascular disease, and is a harbinger of poor prognosis. Initial assessment of glucose levels in all inpatients provides clinicians with prognostically useful information," said panel member Mikhail Kosiborod, MD, a cardiologist at Saint Luke's Mid America Heart and Vascular Institute and associate professor of medicine at the University of Missouri-Kansas City. "It can help identify patients who should be screened for diabetes, and direct decisions in regard to the intensity of subsequent glucose monitoring and, if necessary, glucose-lowering treatment."

Although the authors recommended bedside POCT for ongoing glycemic management using glucose meters that have demonstrated accuracy in acutely ill patients, they did not set specific analytical goals. The panel did warn that "the accuracy of most hand-held glucose meters is far from optimal." National Academy of Clinical Biochemistry (NACB) guidelines for laboratory analysis in the diagnosis and management of diabetes noted the lack of consensus on quality goals for glucose meters but suggested that meters should measure and report plasma glucose concentrations to facilitate comparison with assays performed in accredited laboratories (Clin Chem 2011;57:e1–47).

The chair of the NACB guidelines committee, David Sacks, MD, criticized the panel's process and recommendations. "I was very disappointed that no lab person actually read these guidelines before they were published. It seems we could have provided useful input," he observed. "For example, the panel said POCT has some advantages over lab venous glucose testing. I found that section very unpersuasive, and I don't think it has advantages, other than results being available immediately. However, if the results aren't accurate, then you can make treatment and management errors." Sacks is senior investigator and chief of clinical chemistry at the National Institutes of Health in Bethesda, Md.

NACB guidelines committee member, David Bruns, MD, shared Sacks' concerns. "These recommendations reinforce the need to study glucose measuring devices used in hospital locations outside the ICU. The guidelines mention accuracy issues, but accuracy requires quantification to be meaningful. The question is, how accurate must the devices be?" Bruns is professor of pathology, director of clinical chemistry, and associate director of the molecular diagnostics laboratory at the University of Virginia School of Medicine in Charlottesville.

The guideline calls for a pre-prandial glucose target <140 mg/dL and a random blood glucose <180 mg/dL for the majority of hospitalized patients who aren't critically ill. On the flip side, the panel cautioned about the need to avoid hypoglycemia, and recommended that hospitals implement a standardized, nurse-initiated protocol for immediate intervention when a patient's blood glucose level falls below 70 mg/dL. Sacks noted, however, that the guideline does not mention well-documented performance issues with glucose meters in the hypoglycemic range.

In an effort to help diabetics manage their disease better and tease-out undiagnosed diabetics from patients with stress-induced hyperglycemia, the panel recommended HbA1c testing in all hospitalized diabetics and others with glucose levels ≥140 mg/dL, if it has not been done within the preceding 2–3 months. The panel also suggested that hospital-wide protocols and systems are needed to effectively recognize and manage hyperglycemia in the hospital setting.