Community-acquired pneumonia occurring in immunocompromised older patients: incidence, causative organisms, and outcome

Abstract number: O30

Sousa D., Justo I., Manzur A., Izquierdo C., Ruiz L., Nebot M., Bayas J.M., Celorrio J.M., Varona W., Dominguez A., LLinares P., Carratala J.

Objectives: We sought to determine the incidence, causative organisms and outcome of community-acquired pneumonia (CAP) occurring in immunocompromised older patients.

Methods: Prospective observational multicentre study of a cohort of patients aged 65 years or older hospitalised with CAP from May 2005 to January 2007 in 5 teaching hospitals in Spain. A comparison between cases of CAP occurring in immunocompromised patients and the remaining cases was performed.

Results: Over the study period, we documented a total of 320 cases of CAP; 115 (36%) of which occurred in immunocompromised patients. Main underlying conditions among these patients included one or more of the following: solid cancer or haematological malignancy (97 patients), treatment with corticosteroids or other immunosuppressive drugs (44), solid organ or stem cell transplant (5), and other conditions (8). An aetiologic diagnosis of CAP was more frequently established in immunocompromised patients than in the remaining cases (44% vs 32%; p = 0.03). The most common causative organism was Streptococcus pneumoniae in both groups (29% vs 21%; p = 0.08). No significant differences were observed between groups regarding the incidence of Haemophilus influenzae (2% vs 1%), Legionella pneumophila (3% vs 6%) and atypical agents (1% vs 2%). Gram-negative bacilli were more frequently encountered among immunocompromised patients (5% vs 0.5%; p = 0.009), particularly Pseudomonas aeruginosa (3% vs 0%; p = 0.04). Nocardiosis was only observed among immunocompromised patients (2 cases). The occurrence of bacteraemia was similar in both groups of patients (12% vs 9%). Most patients were given initial empirical antibiotic monotherapy (61% vs 62%). No significant differences were found concerning the percentage of patients requiring ICU admission (8%, in both groups), and length of hospital stay (12.5 vs 10. 4 days). The early (<48 hours) (3.5 vs 0.5%; p = 0.05) and overall case-fatality rates (12% vs 3%; p = 0.003) were higher in immunocompromised patients.

Conclusions: A substantial number of older patients currently hospitalised for CAP are immunocompromised. Although relatively uncommon, CAP due to Gram-negative bacilli, including P. aeruginosa, is more frequent among these patients. CAP occurring in immunocomprised patients causes significant case-fatality rates.

Session Details

Date: 19/04/2008
Time: 00:00-00:00
Session name: 18th European Congress of Clinical Microbiology and Infectious Diseases
Location: Barcelona, Spain
Presentation type:
Back to top