Identification of Low-Risk Patients with Acute Symptomatic Pulmonary Embolism for Outpatient Therapy.

Abstract

Patients with acute symptomatic pulmonary embolism (PE) deemed to be at low risk for early complications might be candidates for partial or complete outpatient treatment. To develop and validate a clinical prediction rule that accurately identifies patients with PE and low risk of short-term complications and to compare its prognostic ability with two previously validated models (i.e., the Pulmonary Embolism Severity Index [PESI] and the Simplified PESI [sPESI]) Multivariable logistic regression of a large international cohort of patients with PE prospectively enrolled in the RIETE (Registro Informatizado de la Enfermedad TromboEmbólica) registry. All-cause mortality, recurrent PE, and major bleeding up to 10 days after PE diagnosis were determined. Of 18,707 eligible patients with acute symptomatic PE, 46 (0.25%) developed recurrent PE, 203 (1.09%) bled, and 471 (2.51%) died. Predictors included in the final model were chronic heart failure, recent immobilization, recent major bleeding, cancer, hypotension, tachycardia, hypoxemia, renal insufficiency, and abnormal platelet count. The area under receiver-operating characteristic curve was 0.77 (95% confidence interval [CI], 0.75-0.78) for the RIETE score, 0.72 (95% CI, 0.70-0.73) for PESI (P < 0.05), and 0.71 (95% CI, 0.69-0.73) for sPESI (P < 0.05). Our RIETE score outperformed the prognostic value of PESI in terms of net reclassification improvement (P < 0.001), integrated discrimination improvement (P < 0.001), and sPESI (net reclassification improvement, P < 0.001; integrated discrimination improvement, P < 0.001). We built a new score, based on widely available variables, that can be used to identify patients with PE at low risk of short-term complications, assisting in triage and potentially shortening duration of hospital stay.