Diagnostic challenges of venous thromboembolism in primary care

Abstract

Diagnosing venous thromboembolism (VTE) is often difficult for general practitioners (GPs) because symptoms can mimic other less life-threatening diseases. GPs can use clinical decision rules to determine whom has a high or low risk of VTE. Nevertheless, there is room for improvement in the diagnostic management of VTE.

In the thesis we described challenges in diagnosing VTE. With a systematic review and meta-analysis we showed that there is on average 6.3 days of delay between start of symptoms and the diagnosis of pulmonary embolism (PE). In another study we showed that when decision rules for VTE are correctly used in daily primary care, few (1.5%) diagnoses are missed (safety) and that the majority (58%) of patients suspected of VTE do not need to be referred (efficiency). However, in nearly a quarter of patients, decision rules are incorrectly applied, leading to a higher failure rate and lower efficiency.

Interestingly, we could show that the diagnostic accuracy of physicians' intuitive assessment of the presence of PE was good. Based on an analysis of multiple international studies including in total 20,770 patients suspected of PE, we showed that the accuracy of this intuitive estimation remained stable in different patient types and in different healthcare settings.

Finally, the PECAN study evaluated in primary care the YEARS algorithm, a clinical decision rule for PE developed in the hospital setting. The study showed that the YEARS algorithm was as safe as the existing rule, but more efficient with as a result less unnecessary referrals.