Assessment of clinical features and determinants of mortality among cancer patients with septic shock of pulmonary origin: a prospective analysis. Issue 1135 (4th November 2019)

• Record Type: Journal Article
• Authors: Rasheed, Maria; Khan, Yusra Habib; Mujtaba, Ghulam; Mallhi, Tauqeer Hussain; Saadullah, Malik; Saifullah, Amna
• Is Part Of: Postgraduate medical journal. Volume 96:Issue 1135(2020)
• Page Start: 277
• Page End: 285
• Publication Date: 2019-11-04
• Subjects: cancer -- pneumonia -- septic shock -- pneumonia associated septic shock -- SOFA score -- mortality; Medicine -- Periodicals; 610
• DOI: 10.1136/postgradmedj-2019-136987
• Languages: English
• ISSNs: 0032-5473

Abstract

Abstract: Background: Pneumonia-associated septic shock (PASS) in patients with cancer inflicts healthcare burden attributed to high morbidity and mortality. Current study was aimed to evaluate the clinical outcomes, microbiological characteristics, risk factors and impact of life-support interventions on 28-day mortality among cancer patients with PASS. Methods: A prospective observational study was conducted among cancer patients with PASS admitted to intensive care unit (ICU) of 'Shaukat Khanum Memorial Cancer Hospital'. Data were analysed using appropriate statistical methods. Results: Out of 100 patients who sought medical care during the study period, 59 (59%) were male and majority had solid tumour than haematological malignancies (68% vs 32%). Nosocomial pneumonia was most frequent (90%) followed by healthcare-associated pneumonia (HCAP) (9%) and community-acquired pneumonia (CAP) (1%). The most common causative pathogen was Pseudomonas aeruginosa, 21 (32%). Overall mortality rate was 76% including 15% hospital and 61% ICU mortality. Sequential Organ Failure Assessment (SOFA) score at first day (HR 3.8; 95% CI 1.7 to 8.9; p=0.002), SOFA score at seventh day (HR 8.9; 95% CI 3.6 to 22.7; p=<0.001), invasive mechanical ventilation (HR 8.0; 95% CI 3.2 to 20; p<0.001) and performance status (HR 5.4; 95% CI 2.5 to 11.3; p<0.001) were found to be independently associated with 28-day mortality. Receiver operating characteristic curve analysis accentuates the excellent predictive accuracy of Cox regression model for mortality indicated by area under the curve of 0.892 (95% CI 0.801 to 0.983, p<0.001). Conclusion: Our analysis demonstrates substantial mortality associated with PASS among patients with cancer. Timely recognition of patients with high predilection of increased mortality could be of value in improving the disease burden.