High Glucose Levels Predict Mortality in Pneumonia

Jenny Powers

September 12, 2012

September 12, 2012 (Vienna, Austria) — Nondiabetic patients who have elevated serum glucose levels when they are admitted to the hospital for community-acquired pneumonia (CAP) have an increased risk of dying within 90 days, compared with similar normoglycemic patients.

Acute hyperglycemia associated with CAP can identify at-risk patients who need additional care to reduce their risk for mortality, according to findings presented here at the European Respiratory Society 2012 Annual Congress.

Studies have suggested that high serum glucose levels predispose people to CAP by increasing the risk for aspiration, decreasing immunity, and causing impaired lung function.

Philipp M. Lepper, MD, from the University Hospital of Saarland in Homburg, Germany, headed a team of investigators who evaluated whether acute dysglycemia could predict a poor outcome in patients with CAP who had not been diagnosed with diabetes.

Their study used data from 6891 adults with CAP enrolled in the prospective CAP Competence Network (CAPNETZ) study from 2003 to 2009. Uni- and multivariable hazard ratios (HR) were adjusted for sex, age, body mass index, current smoking status, and CRB-65 (confusion, recent; respiratory rate, 30 breaths/min or greater; systolic blood pressure of 90 mm Hg or less or diastolic blood pressure of 60 mm Hg or less; and age, 65 years or older). CRB-65 is a clinical prediction rule that grades the severity of CAP in terms of 30-day mortality. The investigators also calculated various comorbidities for 28-, 90-, and 180-day mortality on the basis of serum glucose levels at hospital admission.

"CAPNETZ is the largest trial to look at hyperglycemia as an independent risk factor for increased risk of death from pneumonia," said Dr. Lepper.

At the time of hospital admission, patients who had glucose levels from 6 to 11 mmol/L were considered to have mild acute hyperglycemia and patients who had glucose levels of 14 mmol/L or more were considered to have acute hyperglycemia. In all, 40% of CAP patients presented with hyperglycemia.

The majority (62%) of the patients were male, and average age was 60 years.

On multivariate analysis, it was determined that an elevated glucose level at hospital admission was an independent predictor of 28-, 90-, and 180-day mortality in CAP patients. In fact, increasing glucose levels corresponded to increasing risk for death from CAP.

Miguel Ángel Martinez-Garcia, MD, from the pneumology unit at the General Hospital of Requena in Valencia, Spain, who was not involved in the study, told to Medscape Medical News that this "study concludes that patients with elevated serum glucose on admission to hospital for CAP have an increased risk of CAP mortality at short- and long- term timepoints."

"The study has several strengths, the most important being the large number of patients in the study, but it also has some limitations." For example, it is not clear whether the investigators measured basal glucose, which is important because glucose levels can change for many reasons other than infection, he said.

"Second, we have some prognostic scores for pneumonia that are widely used, such as Fine's score or CURB-65," he noted. Fine's score (or the Pneumonia Severity Index score) considers high glucose levels to be associated with an increase in 30-day mortality. Dr. Lepper and colleagues "should discuss how their findings add to the current literature on this topic," Dr. Martinez-Garcia added.

Patients with mild to moderate hyperglycemia had a significantly higher HR for mortality at 90 days (1.55; 95% confidence interval [CI], 1.18 to 2.04; P < .001) than patients with normal glucose levels at hospital admission.

In patients presenting with acute hyperglycemia, the HR increased to 6.04 (95% CI, 4.18 to 8.74; P < .001).

Sensitivity analyses confirmed the predictive value of admission glucose levels for 28- and 180-day mortality. However, the study did not establish a causal relation between glucose levels and increased mortality risk in patients with CAP.

Patients with increased serum glucose were treated at the discretion of each physician, so changes in glucose levels during the hospital stay could not be evaluated in a standardized fashion.

In patients previously diagnosed with diabetes, there was no association between glucose level at admission and mortality (P = .18 for interaction). However, there was a nearly 2-fold increased risk of dying in the 16% of patients with preexisting diabetes (adjusted HR, 1.88; P < .001).

"Increased serum glucose levels at admission is a risk factor for death among patients with community-acquired pneumonia. The risk for mortality starts to increase when serum glucose levels are slightly increased but remain below the defined threshold for overt diabetes," Dr. Lepper explained.

"Oral glucose tolerance testing and close glucose monitoring after discharge are necessary to diagnose overt diabetes and to prevent subacute or long-term complications.... Large-scale trials are needed to define optimal serum glucose levels and to determine if a drug intervention is suitable," he noted.

"With this study, we cannot conclude a causality relationship, only an association between high levels of serum glucose and increased mortality, because the main objective of the CAPNETZ study did not study this relationship," said Dr. Martinez-Garcia. "Further studies are necessary to corroborate these results and to establish a causality relationship."

The CAPNETZ study was funded by the German Ministry of Education and Research, BundesministeriumfürBildung und Forschung. Dr. Lepper and Dr. Martinez-Garcia have disclosed no relevant financial relationships.

European Respiratory Society (ERS) 2012 Annual Congress: Abstract OC2814. Presented September 2, 2012.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.