Medizinische Universität Graz Austria/Österreich - Forschungsportal - Medical University of Graz
Gewählte Publikation:
Kaiser, R.
Approaches to differentiate between colonization and infection of cultural confirmed Nocardia spp.
Humanmedizin; [ Diplomarbeit ] Graz Medical University; 2015. pp.
[OPEN ACCESS]
FullText
- Autor*innen der Med Uni Graz:
- Betreuer*innen:
- Flick Holger
- Krause Robert
- Altmetrics:
- Abstract:
- Background: Nocardiosis occurs mainly in immunocompromised patients, can affect virtually any organ and requires prolonged antibiotic treatment. However, cultural detection of Nocardia does not imply inevitably infection and therefore colonisation should also be considered. Methods: Clinical data from 62 patients with cultural Nocardia detection were reviewed and applied to a proposed classification for differentiation between colonization and infection. The classification includes clinical (symptomatic disease AND exclusion of other aetiologies), radiological (findings compatible with nocardiosis) and microbiological criteria (=2 positive cultures from sputum OR one positive culture from BAL, biopsy or sterile compartments OR one positive culture from sputum in case of immunosuppression). Nocardial infection was presumed when all three criteria were fulfilled. Results: 24 from 62 patients were excluded because of insufficient medical data. From the remaining 37 patients 13 (35%) were categorized as infected. Of these, 10 (77%) received a presumably effective antiinfective treatment and 3 (10%) of them died. From the three infected patients who did not receive appropriate antibiotics two (66%) died in the course of the disease. Overall 5 from 13 (38%) patients of the infection group died. 24 from 37 (65%) patients were classified as colonized. None of these died and only 4 (17%) received appropriate antibiotics. Therefore, the proportion of effective treatment and mortality were significantly less than in the infected group (p=0.002 and p=0.003, respectively) and the proposed classifications reached a negative predictive value of 84%. Conclusion: The evaluated algorithm detected all live-threatening infections and none of the untreated presumably colonized patients developed nocardiosis. Therefore, the proposed classification might prevent under- and overtreatment in case of Nocardia positive cultures.