To PERC, or not to PERC?

1) What clinical prediction rule can be used to identify a group of patient at very low risk for PE (pulmonary embolism) for whom no additional diagnostic workup is required? 

2) Evidence behind the PERC rule? 

3) Downfalls to this rule?

• PEs are scary and dangerous

• They are common but also uncommon

• Presentation of a patient with a PE can be ambiguous

• Working up a patient who you think might have a PE is not innocent ie. cost, radiation, time.

Risk factors put you at higher risk, and like a teeter-totter, patients present on a spectrum. Taking a thorough history and combining that with your clinical gestalt can parse out how you work up patients with a PE. In adult patients with suspected acute pulmonary embolism (PE), can a clinical prediction rule be used to identify a group of patients at very low risk (<15% risk of PE) for the diagnosis of PE for whom no additional diagnostic workup is required? ACEP’s policy is clear and follows with Level B Recommendations: For patients who are at low risk for acute PE, use the Pulmonary Embolism Rule-out Criteria (PERC) to exclude the diagnosis 

Remember, using the PERC rule is only appropriate in patients who is at very low risk (<15% risk of PE) for the diagnosis of PE. This can be done through clinical gestalt, or through using the two or three-tiered Wells score. There have been studies which have compared clinical gestalt to these decision rules, and clinical gestalt assessment seems to perform better than clinical decision rules (Wells, rGeneva) at estimating the pretest probably of PE and classifying patients with low and high clinical probability [Penaloza A, Verschuren F, Meyer G, et al. Comparison of the unstructured clinician gestalt, the Wells score, and the revised Geneva score to estimate pretest probability for suspected pulmonary embolism. Ann Emerg Med. 2013;62(2):117-124.e2].

The PERC study has been strongly validated; the PERC rule was validated in 8138 patients at 13 emergency departments and, combined with a gestalt clinical suspicion for PE <15%, was found to exclude PE in 20% of cases with a false negative rate of 1% [Kline JA, Courtney DM, Kabrhel C, et al. Prospective multicenter evaluation of the pulmonary embolism rule-out criteria. J Thromb Haemost. 2008;6(5):772-780]. 

PERC studies excluded these patients. 

• Known thrombophilia

• Strong family history of thrombosis

• Concurrent beta-blocker use (could blunt reflex tachycardia)

• Transient tachycardia

• Patients with amputations

• Massively obese patients in whom unilateral leg swelling could not be assessed

• Patients with baseline SaO2 of < 95%

First, the PERC rule should not be used in isolation to rule out PE in pregnant or postpartum patients. These are a special population of patients who are already in a hypercoaguable state, research is still in the works to develop clinical decision rules for this population however there is no definite way to approach these patients yet. And heading more into a statistical realm, PE prevalence must be 7% or less before the PERC rule can be applied. Per the authors of EMDocs, the meta-analysis pooled negative LR is 0.17, which gives you a maximum pretest probability of about 15% to apply the PERC rule to risk stratify your patient down to the standard risk of 2%.  So, caveat is depending on prevalence, you could get burned more then you suspect if prevalence is high enough within the population you serve. The threshold provided by the authors was 7%. How that applies to the population within your community and the application of the PERC rule makes things even more complicated, garnering the PERC rule a tool which is even more complicated than we initially imagined.

Before its application you have to appreciate the importance of clinical judgement and history taking prior to its use. Risk stratify, and apply the rule appropriately based on stratification. The PERC rule requires a clinical suspicion of <15% before it can be applied; it should not be applied to all patients in whom you are considering PE. Similarly, the Wells score is not meant to be used on all patients with chest pain or dyspnea; you must first have a genuine clinical suspicion for PE. And finally, these tools do not force you to order any diagnostic testing. A positive PERC is not an indication for ordering a d-dimer, and a high-risk Wells score does not necessarily mean you must order a CTPE. Think through each individual patient, be mindful of how you are working up these patients. These tools can be helpful in finding the appropriate balance between being too conservative and too cavalier about PE.

• How do you yourself as a clinician assess risk for PE/DVT?

• What are some experiences using PERC that come to your mind when you think about it? (successes and downfalls)

• How has your suspicion for PE changed with the COVID-19 pandemic?