CDR Blog

The original article by Kline et al. published in 2004 included the derivation and validation for the PERC rule.¹ In the derivation, eight variables were included in the block rule based on data collected on 3148 emergency department patients at 10 US hospitals. The rule was prospectively validated on both a low-risk group (n=1427) as well as a very low-risk group (convenience sample, n=382). Results from the validation in low‐risk and very low‐risk populations achieved sensitivities of 96% and 100% and specificities of 27% and 15%, respectively. One quarter of patients were PERC negative in the low risk population and had rates of PE/DVT below the predefined test threshold of 1.8%. Overall, this study is deemed to be taken from relatively diverse populations across the US, with respectable sample size. It demonstrates the potential of the PERC rule to complement clinical judgement in the assessment of low-risk patients for PE. The authors do acknowledge limitations with the derived rule in that it does not apply well in patients with comorbidities such as known thrombophilia or history of thrombosis. It would thus be the objective in subsequent studies to better define the “low risk” patient population eligible for the PERC rule to apply.

In our appraisal of the derivation methodology, we applied the criteria per McGinn et al.²

A. Were all important predictors included in the derivation process? The study included 21 variables which were thought to be all known significant predictors based off of previous studies on outpatients, aside from radiographic data.

B. Were all important predictors included in a significant proportion of the study population? The derivation database contained no missing data on any of 21 variables for 3148 patients.

C. Were all outcome events and predictors clearly defined? Certain predictors were removed from the final block rule as a result of interobserver variability that may stem from unclear definitions. These include “patient immobility” as well as “sudden onset of symptoms”. Quantifying the time frame for both may allow for more consistency amongst responses.

D. Were those assessing outcome events blinded to the presence of predictors & vice versa? Clinicians were required to prospectively complete a mandatory computerized data collection form on the presence of predictors prior to completion of diagnostic test results. However, it is unclear if the vice-versa was blinded as well.

E. Was sample size adequate to account for all potential outcomes? The sample size (n=3148) for the derivation process does appear adequate given the potential outcomes. However, it is of note that the convenience sample taken to assess for interobserver variability (n=129) of the 21 variables, was likely not adequately powered.

F. Does the rule make clinical sense? The rationale for the rule is to mitigate the detrimental increase in pulmonary vascular imaging as a result of high false-positive rate from d-dimer testing. These concerns support the clinical reasoning behind the development of a straight-shot decision rule to justify not initiating diagnostic work-up for PE. While some variables were excluded due to poor interobserver agreement and feasibility of negative predictors in a rapid clinical setting, the end product does make sense medically and practically.

The information provided by McGinn et al. was also used to determine the Clinical Decision Rule level of the PERC rule to be level 1. The original study by Kline et al. included both derivation and validation cohorts that identified the relevant factors to be used in the rule, as well as identifying the relevant sensitivities and specificities in the 2 different populations used for validation. Many future validation studies as well as a systematic review and meta-analysis of said studies re-assessed the usefulness of the PERC rule in low prevalence PE populations in different countries around the world and found results similar to the original validation study completed.^3-5 However, one moderately sized European validation study concluded that in populations with a relatively high prevalence of PE (21.3%), PERC rule could not safely rule out PE in very low risk patients.⁶ Interestingly, this study is included in a larger meta-analysis by Singh et al., and the meta-regression analysis of studies divided based on PE prevalence (< or >= 10%) found no significant difference between PE prevalence and PERC diagnostic performance.⁴ The PROPER RCT conducted by Freund et al. demonstrated the impact of the PERC rule, including that the usage of the PERC rule was associated with 10% fewer CT-PAs, decreased ED length of stay by 36 minutes, and decreased hospital admission rate by 3.3%. Based on the evidence in the literature, the PERC rule can be used in a wide variety of settings with confidence that they can change clinician behavior and improve patient outcomes.⁵

1. Kline JA, Mitchell AM, Kabrhel C, Richman PB, Courtney DM. Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. Journal of Thrombosis and Haemostasis. 2004 Aug;2(8):1247-55.

2. McGinn TG, Guyatt GH, Wyer PC, Naylor CD, Stiell IG, Richardson WS, Evidence-Based Medicine Working Group. Users' guides to the medical literature: XXII: how to use articles about clinical decision rules. Jama. 2000 Jul 5;284(1):79-84.

3. Kline JA, Courtney DM, Kabrhel C, Moore CL, Smithline HA, Plewa MC, Richman PB, O'NEIL BJ, Nordenholz K. Prospective multicenter evaluation of the pulmonary embolism rule‐out criteria. Journal of Thrombosis and Haemostasis. 2008 May;6(5):772-80.

4. Singh B, Mommer SK, Erwin PJ, Mascarenhas SS, Parsaik AK. Pulmonary embolism rule-out criteria (PERC) in pulmonary embolism—revisited: a systematic review and meta-analysis. Emergency Medicine Journal. 2013 Sep 1;30(9):701-6.

5. Freund, Y., Cachanado, M., Aubry, A., Orsini, C., Raynal, P., Féral-Pierssens, A., Charpentier, S., Dumas, F., Baarir, N., Truchot, J., Desmettre, T., Tazarourte, K., Beaune, S., Leleu, A., Khellaf, M., Wargon, M., Bloom, B., Rousseau, A., Simon, T. and Riou, B., 2018. Effect of the Pulmonary Embolism Rule-Out Criteria on Subsequent Thromboembolic Events Among Low-Risk Emergency Department Patients. JAMA. 2018 May; 319(6), p.559.

6. Hugli O, Righini M, Le Gal G, Roy PM, Sanchez O, Verschuren F, Meyer G, Bounameaux H, Aujesky D. The pulmonary embolism rule‐out criteria (PERC) rule does not safely exclude pulmonary embolism. Journal of thrombosis and haemostasis. 2011 Feb;9(2):300-4.