Background: The optimal risk stratification of unsuspected pulmonary embolism (UPE) in ambulatory cancer patients remains unclear. Existing prognostic models have been developed from retrospective datasets, mostly of suspected PE cases and have limitations. The UPE registry collected data prospectively from an ambulatory population with active cancer and a diagnosis of recent UPE.

Aims: To assess risk factors for recurrent venous thromboembolism (rVTE), major bleeding (MB), and death in cancer patients with UPE.

Methods: 695 patients with UPE were included in multivariate logistic regression and competing risk survival analyses to identify factors independently associated with proximate (30-day, 90-day and 180-day) mortality, overall survival, rVTE, and MB. The most consistent predictors found were subjected to an exploratory analysis using the Hull CPR, an existing 5-point prediction rule (1). Receiver operating characteristic analysis was used to assess discriminatory performance.

Results: During 12-month follow-up, 283 patients had died (cumulative incidence, 41%; 95% CI, 39 to 46) with 30-day, 90-day and 180-day mortality of 5.6% (n=39), 15.0% (n=104), and 25.2% (n=175), respectively. Most consistent predictors of mortality were patient-reported respiratory symptoms within 2 weeks before, and ECOG performance status score at the time of UPE diagnosis (Table). 
Exploratory application of these predictors to the Hull CPR produced a consistent correlation with 30-day, 90-day and 180-day mortality and overall survival (area under the curve [AUC] =0.70 [95% CI 0.63, 077], AUC =0.65 [95% CI 0.60, 070], AUC =0.64 [95% CI 0.59, 068], and AUC=0.61, 95% CI 0.57, 0.65, respectively)(Figure). No factors were associated with MB or rVTE.

The significant prognostic risk factors for mortality are summarised. Other clinical variables explored as predictors included age (with a cut-off of 70 years), gender, presence of abnormal vital signs (tachycardia, hypotension, atrial fibrillation) and the location and extent of imaged UPE. Only ECOG PS and symptoms within 14 days to UPE diagnosis performed consistently across all mortality risk categories. These risk factors were applied to the HULL CPR (1).Kaplan Meier survival curves (A) and events incidence (B) over 180 days generated from the exploratory application of the most consistent predictors to the Hull CPR [1]. Three risk clusters (Low Risk: 0, Intermediate Risk: 1-2, High Risk: 3-4).

Conclusions: In ambulatory cancer patients with UPE computation of ECOG performance status logged contemporaneously to the UPE diagnosis and self-reported respiratory symptoms prior to UPE diagnosis can stratify mortality risk. When applied to the HULL-CPR these risk predictors reproduced the risk stratification of the derivation cohort(1).

« Back to ISTH 2021 Congress

ISTH Congress Abstracts - https://abstracts.isth.org/abstract/the-prognostic-value-of-symptoms-and-performance-status-in-ambulatory-cancer-patients-and-unsuspected-pulmonary-embolism-analysis-of-an-international-prospective-observational-cohort-study/