Large-Bore Mechanical Thrombectomy Versus Catheter-Directed Thrombolysis in the Management of Intermediate-Risk Pulmonary Embolism: Primary Results of the PEERLESS Randomized Controlled Trial

Abstract

Background: There are a lack of randomized controlled trial data comparing outcomes of different catheter-based interventions for intermediate-risk pulmonary embolism.

Methods: PEERLESS is a prospective, multicenter, randomized controlled trial that enrolled 550 patients with intermediate-risk pulmonary embolism with right ventricular dilatation and additional clinical risk factors randomized 1:1 to treatment with large-bore mechanical thrombectomy (LBMT) or catheter-directed thrombolysis (CDT). The primary end point was a hierarchal win ratio composite of the following (assessed at the sooner of hospital discharge or 7 days after the procedure): (1) all-cause mortality, (2) intracranial hemorrhage, (3) major bleeding, (4) clinical deterioration and/or escalation to bailout, and (5) postprocedural intensive care unit admission and length of stay. Assessments at the 24-hour visit included respiratory rate, modified Medical Research Council dyspnea score, New York Heart Association classification, right ventricle/left ventricle ratio reduction, and right ventricular function. End points through 30 days included total hospital stay, all-cause readmission, and all-cause mortality.

Results: The primary end point occurred significantly less frequently with LBMT compared with CDT (win ratio, 5.01 [95% CI, 3.68-6.97]; P<0.001). There were significantly fewer episodes of clinical deterioration and/or bailout (1.8% versus 5.4%; P=0.04) with LBMT compared with CDT and less postprocedural intensive care unit use (P<0.001), including admissions (41.6% versus 98.6%) and stays >24 hours (19.3% versus 64.5%). There were no significant differences in mortality, intracranial hemorrhage, or major bleeding between strategies or in a secondary win ratio end point including the first 4 components (win ratio, 1.34 [95% CI, 0.78-2.35]; P=0.30). At the 24-hour visit, respiratory rate was lower for patients treated with LBMT (18.3±3.3 versus 20.1±5.1; P<0.001), and fewer had moderate to severe modified Medical Research Council dyspnea scores (13.5% versus 26.4%; P<0.001), New York Heart Association classifications (16.3% versus 27.4%; P=0.002), and right ventricular dysfunction (42.1% versus 57.9%; P=0.004). Right ventricle/left ventricle ratio reduction was similar (0.32±0.24 versus 0.30±0.26; P=0.55). Patients treated with LBMT had shorter total hospital stays (4.5±2.8 overnights versus 5.3±3.9 overnights; P=0.002) and fewer all-cause readmissions (3.2% versus 7.9%; P=0.03), whereas 30-day mortality was similar (0.4% versus 0.8%; P=0.62).

Conclusions: PEERLESS met its primary end point in favor of LBMT compared with CDT in treatment of intermediate-risk pulmonary embolism. LBMT had lower rates of clinical deterioration and/or bailout and postprocedural intensive care unit use compared with CDT, with no difference in mortality or bleeding.

Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT05111613.

Keywords: percutaneous aspiration; pulmonary embolism; randomized controlled trial; thrombectomy; thrombolytic therapy.

Figure 1.

Win ratio end points and odds ratios for the win ratio components. Shown are the primary and secondary win ratio end points (A) and the odds ratios for the clinical outcome components of the win ratios (B), assessed at the sooner of discharge or 7 days after the procedure. P values for the win ratio end points are derived with a modified generalized Wilcoxon test (F-S test), and P values for clinical outcomes of the win ratio are derived from the 2-sided Fisher exact test. Odds ratios and 95% Wald CIs are calculated with large-bore mechanical thrombectomy (LBMT) as the reference. *N=275 because of 1 death. CDT indicates catheter-directed thrombolysis; and ICU, intensive care unit.

Figure 2.

Functional and imaging assessments at the 24-hour visit. Assessments at the 24-hour visit included the proportion of patients with a modified Medical Research Council (mMRC) dyspnea score <3 vs ≥3 (A), mean respiratory rate (B), proportion of patients New York Heart Association (NYHA) class I or II vs class III or IV (C), mean modified Borg dyspnea score at rest (D), proportion of patients with normal or mildly reduced vs moderately or severely reduced right ventricular (RV) function among those with echocardiograms at baseline and 24-hour visit (E), and mean reduction in RV/left ventricle (LV) ratio from baseline based on the same type of imaging (echocardiogram or computed tomography pulmonary angiogram) at baseline and 24-hour visit (F). P values are derived from the 2-sided Fisher exact test for categorical variables (mMRC score, NYHA classification, and RV function) and from Wilcoxon rank-sum tests with continuity correction for continuous variables (respiratory rate, modified Borg score, and reduction in RV/LV ratio). CDT indicates catheter-directed thrombolysis; and LBMT, large-bore mechanical thrombectomy.

Figure 3.

ICU use and length of stays. Shown is postprocedural intensive care unit (ICU) use (proportion of patients with no ICU admission, ICU admission lasting 0–24 hours, and ICU admission lasting >24 hours; A) and the mean postprocedural ICU length of stay in hours among all patients (B). ICU length of stay is measured from the end of the index procedure or the time of ICU admission, whichever is later, until the time of an order to discharge from the ICU or transfer to a standard or lower-acuity unit. Total hospital length of stay is reported by number of overnights (C); postprocedural hospital length of stay is reported in days (D). P values are derived from the 2-sided Fisher exact test for categorical variables (postprocedural ICU admission) and from Wilcoxon rank-sum tests with continuity correction for continuous variables (postprocedural ICU, postprocedural hospital, and total hospital lengths of stay). CDT indicates catheter-directed thrombolysis; and LBMT, large-bore mechanical thrombectomy.