PURPOSE: We sought to determine the independent predictors of hospital and within 120-day mortality among hospitalized patients with pulmonary embolism (PE).
METHODS: The study included all adult hospitalized patients with objective evidence for PE between 1999 and 2009. We excluded patients with PE if they were pregnant or treated as outpatient. We collected demographic data in addition to co-morbid conditions, physical examination, radiographic and laboratory data. Mortality was recorded at hospital discharge. We categorized troponin I into two groups using normal cutoff of 0.04. All variables that were statistically significant in uni-variate analysis were entered in the multivariate analysis by logistic regression. Variables that had a p value <0.05 were considered independent predictors of mortality.Analysis was performed for both hospital and within 120-day mortality.
RESULTS: 320 patients were analyzed. Mean age and Body mass index (BMI) were 56.76 and 31.75 respectively. Females comprised 52.5 % of study patients. Twenty patients (6.3%) died during hospital stay at a median of 4 days (2-8). Patient who died were more likely to be older than 70 (13/20, 65%), and had higher rates of renal failure, lower albumin, higher rates of abnormal troponin, lower systolic blood pressure and higher shock index. Presence or absence of diabetes, congestive heart failure, airway disease, obesity (defined as BMI ≥ 30), prior or current venous thromboembolism did not affect hospital mortality. Multivariate analysis revealed that troponin >0.04 was the only independent risk factors for hospital mortality [odds ratio 9.73 (1.14-83.1); p=0.038). Five more patients (total 7.8%) died after hospital discharge within 120 days. Multivariate analysis again revealed troponin the only independent predictor of mortality within 120 days (p=0.014).
CONCLUSION: Although baseline characteristics may affect mortality in uni-variate analysis, only troponin appears to be the independent risk factor that reliably predicts mortality during hospitalization and within 120 days.
CLINICAL IMPLICATIONS: Clinicians need to be aware of the value of troponin in predicting short and long term mortality. This my impact disposition and management of those with high troponin levels.
DISCLOSURE: Junaid Khan, Consultant fee, speaker bureau, advisory committee, etc. Ousama Dabbagh is on the speaker bureau : SANOFI -AVENTIS; No Product/Research Disclosure Information