Stopping anticoagulation for isolated or incidental subsegmental pulmonary embolism: the challenges and lessons from the STOPAPE RCT.

Background: The increasing use of computed tomography pulmonary angiography to investigate patients with suspected pulmonary embolism has led to an increase in diagnosis of small subsegmental pulmonary embolism, which is rarely detectable with nuclear medicine-based imaging, the standard imaging modality prior to the development of computed tomography pulmonary angiography. The case fatality of pulmonary embolism has fallen in line with the increase in subsegmental pulmonary embolism diagnoses from computed tomography pulmonary angiography suggesting that we may be over-diagnosing pulmonary embolism (i.e. we may be diagnosing mild forms of pulmonary embolism which may not need any treatment). Given that full anticoagulation has significant side effects of bleeding and subsegmental pulmonary embolism was not commonly diagnosed previously with nuclear medicine imaging (and therefore left predominantly untreated prior to computed tomography pulmonary angiography scanning), there is growing equipoise about the value of full anticoagulation for patients with subsegmental pulmonary embolism.

Methods: We tried to undertake an open randomised trial with blinded end-point adjudication that recruited patients diagnosed with subsegmental pulmonary embolism without evidence of thrombus in the leg veins, termed 'isolated subsegmental pulmonary embolism'. We allocated patients with isolated subsegmental pulmonary embolism to either continuing with at least 3 months of full-dose anticoagulation (standard care) or stopping anticoagulation completely, unless they had a temporary hospital admission where prophylactic (i.e. preventative doses) of anticoagulation is standard practice. In addition, we interviewed patients and clinicians about their views on stopping anticoagulation for isolated subsegmental pulmonary embolism which would be a substantial change from current practice. We planned to assess the accuracy of isolated subsegmental pulmonary embolism diagnoses from computed tomography pulmonary angiographies.

Results: The trial was stopped prematurely due to low recruitment. This was due to a combination of insufficient trial sites, problems with identifying patients who were suitable to be recruited at the time of acute assessment in hospital, the impact of COVID-19 on research infrastructure and a lower prevalence than had been predicted based on published studies. Our interview study showed that the intervention (i.e. changing practice to stopping treatment) is feasible, although there were concerns raised about safety, which a trial would be needed to address. We did not have sufficient trial participants to determine accuracy of initial isolated subsegmental pulmonary embolism diagnoses.

Conclusion: Although we were not able to answer the question of whether it is clinically effective and cost-effective to stop anticoagulating patients with isolated subsegmental pulmonary embolism, we developed a protocol which can be used by future trialists who can successfully attract funding to address this research question, which remains important and an ongoing uncertainty for clinicians and patients.

Future work: Trialists attempting to answer this research question should plan for longer recruitment times and ensure there is sufficient resource for a large number of recruiting centres.

Limitations: There were insufficient recruits to progress from the pilot phase to the full STOPAPE trial.

Funding: This synopsis presents independent research funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme as award number NIHR128073.