Pneumonia in the surgical intensive care unit. Immunologic keys to the silent epidemic.

OBJECTIVE

The authors undertook a prospective study of trauma victims in the intensive care unit (ICU) to investigate the clinical course of pneumonia and the local and systemic immune responses to the pneumonia.

SUMMARY BACKGROUND DATA

The silent epidemic of pneumonia has been an "unappreciated killer" in terms of being overlooked in surgical ICUs for the past 5 years, and specifically, the most common major infection after severe trauma. Little is known about the immune response to an acute pulmonary infection.

METHODS

The authors studied 50 consecutive, critically ill trauma patients, with a mean injury severity score of 28 +/- 2, who developed pneumonia while ventilated mechanically. Patients were observed clinically, and specific immunologic parameters, including major histocompatibility antigen HLA-DR, complement receptor (CR3), and Fc receptor (FcRIII), were measured in circulating and local alveolar leukocytes for up to 30 days. Eleven patients provided unique clinical data via bronchoscopy for unilateral pneumonia, with collection of bronchoalveolar lavage (BAL) fluid from both the infected and uninfected sides.

RESULTS

Patients developed clinical pneumonia 5.3 +/- 0.4 days after admission to the ICU. At diagnosis, mean temperature was 101.4 F, white blood cell count was 16,000/mm3, arterial oxygen tension was 104 +/- 14, fraction of inspired oxygen was 0.47, and positive end-expiratory pressure was 5. Thirty patients (Group A) recovered relatively promptly; 20 patients had prolonged illnesses (Group B), 15 of whom ultimately survived, and five of whom died. Patients with poor outcomes had greater leukocytosis (p < 0.05) and temperature elevation (p < 0.05) after 5 days of pneumonia. Immunologically, peripheral leukocyte expression of HLA-DR, FcRIII, and CR3 was equivalent in both groups. However, the expression of all three antigens on local alveolar leukocytes was decreased to a greater extent in the poor outcome group compared to the good outcome group, evident before any clinical differentiation between the two outcome groups.

CONCLUSIONS

Pneumonia prolonged duration of mechanical ventilation, ICU and hospital stay, and overall infectious morbidity. Although immune suppression has been recognized as a result of initial injury, the development of pneumonia coincided with the nadir of immune function. Poor outcome patients were clinically identifiable 5 days after pneumonia and immunologically identifiable within 2 days. Moreover, there was localized suppression of pulmonary leukocytes at the site of the infiltrate compared to the uninfected lobes. This same alteration was noted in experimental Klebsiella pneumoniae pneumonia. This evidence suggests that there is active immune participation within the respiratory system. It also suggests that there are predispositions to pulmonary infections, and it may allow immune modulation targeted to pulmonary leukocytes to hasten clinical recovery and minimize pulmonary dysfunction.