View Full Version : Tight Glucose Control Used Improperly in Intensive Care

Mon 27th April '15, 10:36pm
Jim Kling
November 06, 2014

AUSTIN — Physicians have been slow to stop using tight glycemic control in intensive care units (ICUs), despite a 5-year-old randomized controlled trial showing it is a risky practice.

The approach first gained momentum after the Leuven 1 study suggested an improvement in tightly controlled surgical patients (N Engl J Med. 2001;345:1359-1367 (http://www.nejm.org/doi/full/10.1056/NEJMoa011300)). But further investigation in the NICE-SUGAR study demonstrated increased mortality (N Engl J Med. 2009;360:1283-1297 (http://www.nejm.org/doi/full/10.1056/NEJMoa0810625)).

"Our study highlights the fact that you need to be careful which populations you're applying evidence to," said Daniel Niven, MS, an intensivist and health services researcher at the University of Calgary in Alberta, Canada.

"It's probably best to wait for a confirmatory study before the widespread adoption of a given therapy," he explained while presenting the results here at CHEST 2014.

Our research suggests that physicians in the ICU have a hard time giving up an intervention they believe is useful, he added.

"Leuven 1 was a good study, but it was performed in a narrow group of patients. Still, our profession jumped on it and extrapolated it to other groups, including medical patients," Niven told Medscape Medical News.

Leuven 1 suggested that tight glycemic control, defined as the use of intravenous insulin to maintain blood glucose levels at 80 to 110 mg/dL, improved outcomes in surgical patients. But the study was a single-center analysis of a specific group of patients, and subsequent evidence showed that the practice might in fact harm some patients.

In contrast, the multicenter prospective NICE-SUGAR trial (http://www.medscape.com/viewarticle/590137) demonstrated an association between tight glycemic control in general ICU patients and increased mortality.

In their study, Niven and colleagues used an interrupted time series analysis to evaluate 368,030 adults admitted to 102 ICUs from January 2001 to December 2012. They collected data from the APACHE III database (http://www.openclinical.org/aisp_apache.html).

The researchers analyzed the most physiologically extreme blood glucose level in the first 24 hours of ICU admission; a level of 80 to 110 mg/dL was considered tight glycemic control.

Before Leuven 1 was published, physicians employed tight glycemic control in 20.3% of patients admitted to the ICU (95% confidence interval [CI], 19.7 - 20.8). After the publication, tight glycemic control incidence increased by 0.50% per quarter (95% CI, 0.40 - 0.60; P < .0001). The increase was highest in surgical patients, who were the subject of the study (odds ratio per quarter, 1.04, 95% CI, 1.03 - 1.04; P < .0001).

After the publication of NICE-SUGAR, the rate of increase of tight glycemic control declined (P = .002 for change in quarterly rate); however, the overall incidence of tight glycemic control in the ICU did not decrease (P = .7 for quarterly trend). At the conclusion of the study, the rate of tight glycemic control was 36.0% (95% CI, 34.9 - 37.1).

These results suggest that the passive broadcast of recommendation changes might be insufficient to reverse a long-standing practice, and that more active means of communication might be necessary.

Inert Evidence Sharing

"The passive diffusion of evidence from the medical journals to practitioners isn't going to result in practitioners discontinuing a practice that they may have done for years," said Niven. "Rather, what's going to need to take place is a more active effort to distribute the evidence and implement change through changes in protocols. How that's going to take place isn't clear, but the active role, as opposed to a passive role, is going to be required."

The emotional attachment of physicians to procedures that might no longer be best practice can affect their choices and desire to take action.

"In critical care, I think this is a big one because our patients are often so sick and the chance of death is high. We tend to want to do things we think make sense, but what we're learning, more and more, is that doing less might be the way to go," said Niven.

It's easy to jump on the adoption bandwagon.

Deadoption is an ongoing problem, according to Harvey Reich, MD, director of critical care medicine at the Rutland Regional Medical Center and clinical associate professor of medicine at the University of Vermont College of Medicine, who attended the presentation.

"It's easy to jump on the adoption bandwagon — especially when we see something that appears to be efficacious and has a mortality difference — because everyone is trying to do the right thing for their patients. Often that is based on one study, and there are some limitations, but the data look good and it seems logical. But many times, the initial data look fantastic and as time goes on, things fade," he told Medscape Medical News.

Dr Reich said he also agrees that deadoption is too slow. "It tends to take about a decade, which is scary."

Mr Niven and Dr Reich have disclosed no relevant financial relationships.

CHEST 2014: the American College of Chest Physicians Meeting: Abstract 1991128. Presented October 29, 2014.

Forwarded from http://www.medscape.com/viewarticle/834515?nlid=69783_1842&src=wnl_edit_medp_wir&uac=180112PN&spon=17