Guest Editorial – Pulmonary Embolism: the risk of what we don’t yet know!

Abstract

Pulmonary embolism (PE) is a common and potentially life-threatening condition encountered routinely in acute care.1, 2 The diagnosis and management of PE has been the topic of National Institute of Health Care Excellence3 and European Cardiology Society guidelines.4 The recommendations within these guidelines have allowed standardisation of care and have facilitated the delivery of protocolised care pathways. Whilst some elements of care are determined by consensus view, there have been large randomised controlled trials and well-designed observational studies which have helped us understand the role of risk factors for PE, short term risk-stratification after initial diagnosis and treatment options in hospital as well as in the months after discharge from Acute Medicine. Few other conditions in acute care are informed by the same levels of evidence, yet there are many unresolved questions. The Acute and General Medicine Getting it Right First Time (GIRFT) speciality report used PE as a target condition to quantify variation between NHS Trusts.5 The report described significant difference in the proportion of patients with PE discharged without the need for overnight stay between Trusts. The report concluded that variation in rates of zero-day length of stay could be partially addressed by more widespread adoption of the Pulmonary Embolism Severity Score (PESI). The PESI score uses 11 clinical variables to predict 30 day6, 7 and 1 year mortality.8 The GIRFT report described awareness of the score amongst clinicians, but heterogeneity in rates of adoption into formalised pathways. However, reducing variation may not be beneficial if all Trusts adopt an approach that does not optimise medical decision making – either about treatment or location of care. The PESI score may help standardise decision making in relation to the appropriateness of ambulatory outpatient pathways, but its construction limits application across the full risk spectrum of PE. The influence of cancer as a predictor of mortality in the derivation and validation studies does not distinguish between patients that die with, rather than of, PE. This will reduce the accuracy of the score in selecting patients who will benefit from admission to monitor for the development of haemodynamic instability. Furthermore, the prediction time horizon of 30 days of PE related harm is much longer than the average length of stay in patients admitted with a diagnosis of PE, so the monitoring period may not match the true risk period. Therefore the PESI score may predict mortality, but this information is not always directly relevant to decisions about the intensity and duration of monitoring in hospital, or the likelihood of requiring more aggressive forms of treatment to alter the trajectory of disease. Haemodynamic instability is a poor prognostic sign and is associated with a short-term mortality in excess of 15%.9 However, haemodynamic instability is uncommon and occurs in less than 5% of all cases. The use of thrombolytic therapy in patients with haemodynamic instability is recommended.3 There also exists a sub-group of patients which fall below the threshold for immediate systemic thrombolytic therapy but are at relatively high risk of deterioration. This intermediate risk sub-group is characterised by elevated troponins, B-type natriuretic peptide and evidence of right ventricular dysfunction on echocardiography.10 A systemic review and meta-analysis of patients with low-risk PE defined by PESI score found an elevated troponin was associated with a significant increase in 30 day all-cause mortality (Odds ratio 4.8).11 How can we identify and manage intermediate risk PE? This is not covered in current NICE guidelines[3] but the Pulmonary Embolism Thrombolysis (PEITHO) trial, randomized 1005 patients with right ventricular dysfunction on either CTPA or echocardiogram or an elevated troponin to receive thrombolysis in addition to unfractionated heparin or unfractionated heparin alone.12 The study found no overall difference in 7-day mortality between the two-groups. The intervention group demonstrated significant reductions in the composite endpoint of death and haemodynamic compromise at the expense of an increase in major bleeding. The use of rescue thrombolysis in the control group may explain the neutral primary outcome. These findings highlight the importance of close monitoring with the option to deliver systemic thrombolysis if haemodynamic instability develops. The current edition of Acute Medicine contains original research articles by Le Jeune et al clarifying the association between obesity and unprovoked PE and Overgaauw et al, describing process and clinical outcomes in patients with intermediate risk PE. Le Jeune et al report the findings from a single centre observational case-control study of 1137 patients diagnosed with PE between 2020 and 2021 with a sample of 540 randomly selected negative controls derived from 10073 CT pulmonary angiograms or ventilation and perfusion undertaken to investigate suspected PE.13 The authors estimated the risk of PE associated with obesity using a logistic regression model controlling for age, sex and index of multiple deprivation score. There was a significant association between PE and obesity. Overweight and obese patients had an increased likelihood of PE (Odds ratio 1.8 and 3.3 respectively). Stratified analysis based on the presence of classical risk factors for PE suggested the association between obesity and PE was particularly strong (Odds ratio 4.6) in those otherwise classified as unprovoked PE. Although the precision of these estimates may be affected by residual confounding and ascertainment bias the findings highlight the importance of considering obesity as a risk-factor for PE. Overgaauw et al present a single center prospective observational cohort study of 98 patients with intermediate-high risk PE from the Amsterdam University Medical Center. Systemic thrombolytic treatment was delivered in 3% (95%CI 0.6-8.5%) of patients.14 In all the reported cases thrombolytics were delivered within 24 hours of arrival. Interestingly the overall mortality was similar in groups with intermediate risk PE compared with the overall cohort. Most patients with intermediate risk PE (83%) were managed in high care units, care settings capable of providing continuous physiological monitoring. This invites comparisons with the current standard of care on UK Acute Medical Units. The authors argue current guidelines are not appropriately calibrated to risk and that more nuanced approaches to identifying patients likely to benefit from more intensive monitoring are required. Together these papers highlight areas of interest not well covered in current guidelines. It is imperative that the specialty of Acute Medicine drives the research agenda in PE management to ensure future large-scale studies focus on areas of clinical practice directly relevant to front door decision making. The drive to standardize care without appropriate evidence may not improve the clinical or cost effectiveness of care. We need to strengthen acute medicine research capability and undertake research that helps front door decision making, and then we will know how to get it right first time.