Hyperglycemia as a predictor for mortality in veterans with pneumonia

The elderly may have multiple problems including diabetes mellitus and pneumonia. This study was undertaken to evaluate the effect of hyperglycemia on the length of stay (LOS) and mortality of elderly veterans admitted for pneumonia. Forty-seven and forty-eight patients over the age of 60 admitted for the diagnosis of pneumonia at the Lexington, Kentucky and Iron Mountain, Michigan VAMC's respectively were evaluated. The mean age at Lexington was 72 and at Iron Mountain was 73.9. At Lexington the mean LOS was 39.5 days with a standard deviation of 9.2 for those with hyperglycemia (n = 18) and 16.5 days with a standard deviation of 2.5 for those without hyperglycemia (n = 29). At Iron Mountain the mean LOS was 21.5 days with a standard deviation of 3.3 for those with hyperglycemia (n = 10) and 13.6 days with a standard deviation of 1.2 for those without (n = 38). These differences were significant at p less than .01 using the 2-sample independent t-test. At Iron Mountain 5 of 38 (13%) of the patients with normoglycemia died during the hospital stay compared to 3 of 10 (30%) those with hyperglycemia. For Lexington the figures were 4 of 29 (13%) and 6 of 18 (33%) respectively. These differences were significant at p less than .05. Age did not have any significant effect on either mortality or length of stay. Hyperglycemia appears to be a significant predictor of a longer hospital course and increased mortality in elderly veterans hospitalized for pneumonia.