Lung graphic About Pulmonary ReviewsFeatured IssuesEditorial BoardPublishing StaffAdvertising InformationSubscription InformationOnline CME from Jobson Medical Group Classifieds

Search:
Sort by:


Pulmonary Reviews.Com

Home  |  Contact Us  |  Archives


Vol. 11, No. 12
December 2006


NEW SERUM MARKER PREDICTS DEATH FROM PNEUMOCOCCAL BACTEREMIA

Key Point
Among individuals younger than 75, soluble CD163 is a better predictor of mortality in pneumococcal bacteremia than are other macrophage serum markers, including C-reactive protein.

AARHUS, DENMARK—The soluble CD163 (sCD163) level predicts mortality in pneumococcal bacteremia better than any of the established macrophage serum markers, including the best one, C-reactive protein, Holger J. Møller, MD, and colleagues have found.1 Studies of these markers should now focus on following sCD163 levels during the course of pneumococcal bacteremia, they suggested. Lead investigator Dr. Møller is in the Department of Clinical Biochemistry at Aarhus University Hospital in Denmark.

sCD163 is a new serum marker released by monocytes and macrophages, related the investigators. They pointed out that sCD163 is a scavenger receptor for haptoglobin-hemoglobin complexes, is expressed mainly on macrophages with an anti-inflammatory phenotype, and might possess its own anti-inflammatory properties.

"The serum level of sCD163 is highly increased in diseases with intense involvement of macrophages such as hemophagocytic syndrome, Gaucher disease, and fulminant hepatitis," the investigators added. "Preliminary results have indicated that sCD163 may be a marker also for severe infection."

The serum sCD163, neopterin, ferritin, transcobalamin, and soluble urokinase plasminogen activator receptor levels of 133 patients with Streptococcus pneumoniae bacteremia were compared to those of an equal number of age- and gender-matched controls. In the bacteremia group, serum biochemical analysis was performed at the time of first positive blood culture for pneumococcal bacteremia.

The bacteremia group ranged in age from 23 to 99 years, and 95 of these patients were younger than 75. Twenty-three died in the hospital, for a case fatality rate of 17%.

"An underlying chronic illness was present in 65% of the patients," the investigators noted. "All patients received adequate antibiotic treatment (eg, penicillin or cephalosporin) within a few hours of admission and before the result of the positive blood culture was available."

Male and female bacteremia patients showed no significant difference in the median sCD163 level (5.3 and 4.3 mg/L, respectively). However, that level was negatively correlated with age, with much lower values in the 75-and-older age-group.

Overall, the bacteremia group had a significantly higher median sCD163 level than did the controls (4.6 vs 2.7 mg/L). This difference was related mainly to the very high median level of 5.4 mg/L in the under-75 age-group, compared to only a slightly elevated level of 3.5 mg/L in older patients. Extremely high concentrations, surpassing 20 mg/L, were seen in seven patients in the former group; five of those patients died.

Remarkably high sCD163 levels occurred in a number of other subgroups as well. These included a median concentration of 13 mg/L among hemodialysis recipients, 9.3 mg/L for those requiring pharmacologic treatment of hypotension, 7.5 mg/L in mechanically ventilated patients, and 5.3 mg/L in those with underlying chronic disease.

In those younger than 75, "all macrophage markers were increased in patients who died from their infection compared with survivors, whereas no significant difference was observed in any of the markers in very old age," the investigators said. "sCD163 showed the most pronounced increase, with three times higher levels in nonsurvivors."

Receiver operating characteristic analyses limited to the under-75 age-group showed that all of the macrophage serum markers studied had an area under the curve that was well above 0.5. Indeed, the area under the curve ranged from 0.69 for transcobalamin to 0.82 for sCD163.

"Using the best cutoff point for each marker, we found that increased sCD163 and [C-reactive protein] involved the highest relative risk (10.1 and 7.0, respectively)," said the investigators. "The risk of dying from infection was 47% when sCD163 was greater than 9.5, whereas the risk was only 4.6% when the sCD163 was less than 9.5."

BETTER THAN C-REACTIVE PROTEIN?

At the 9.5-mg/L cutoff, sCD163 was 82% sensitive and 79% specific for death in the bacteremia group while C-reactive protein was 82% sensitive and 69% specific at its optimal cutoff of 1,360 nmol/L. In a multivariate model excluding patients 75 and older, a high sCD163 level separately predicted death with a concordance of 85.6% versus 83.2% for C-reactive protein. None of the other macrophage serum markers provided significant information on survival odds when the sCD163 and C-reactive protein levels were known, the investigators emphasized.

They concluded that the macrophage marker response to pneumococcal bacteremia was compromised in patients 75 and older. However, their findings suggest that sCD163 could replace C-reactive protein as the most accurate mortality predictor in pneumococcal bacteremia patients who are younger than 75.

—Timothy Begany

Reference
1. Møller HJ, Moestrup SK, Weis N, et al. Macrophage serum markers in pneumococcal bacteremia: prediction of survival by soluble CD163. Crit Care Med. 2006;34:2561-2566.

Return to table of contents