Prior studies have detailed single-center experiences of PERT implementation with heterogenous results, often suggesting that hospital adoption of PERT could improve patient outcomes by rapidly identifying intermediate and high-risk cases and increasing institutional knowledge of PE evaluation and management, regardless of PERT consultation [3, 13, 14]. Though few have assessed the direct impact of PERT consultation on clinical outcomes and hospital utilization metrics.

Consistent with prior studies, we found no significant difference in 30-day mortality between the pre- and post-PERT cohorts [15,16,17,18,19,20]. However, when assessing the impact of PERT consultation, there was a markedly reduced 30-day mortality for the PC cohort, despite PCs having similar PESI scores. Furthermore, PCs were more likely to have intermediate-high and high-risk PEs when compared to NPCs (p< 0.004). PC patients had a slightly lower 30-day mortality than the 6.5% reported by The PERT Consortium™ for PC patients with intermediaterisk PE [21].Of the two prior studies comparing PC to NPC cohorts, neither demonstrated a significant difference in mortality [18, 20]. Notably, these studies included patients who were admitted with low-risk PE, which may have obscured any potential benefit for higher risk populations.

We propose several factors that may explain the difference in mortality seen in our study. PC patients were much more likely to receive a complete guideline-directed PE risk stratification evaluation. This, coupled with reduced TAC, reduced in-hospital bleeding complications, and fewer procedures may suggest a more effective evaluation of patient-specific factors and therapy-selection following the multidisciplinary PERT discussion.

Reduced TAC, as seen in the PC cohort may play an important role. Professional guidelines recommend prompt initiation of anticoagulation in acute PE, even prior to diagnostic confirmation when there is high clinical suspicion [4]. Prior research has shown an increase in mortality for every hour delay in diagnosis, and that patients who achieve therapeutic anticoagulation within 24 h have a reduced in-hospital and 30-day mortality [22, 23]. This suggests a time-sensitive benefit to reperfusion. The noted reduction in TAC may be partially driven by the PC cohort being more likely to receive enoxaparin as the initial anticoagulant,

which frequently achieves therapeutic levels more rapidly than heparin infusions. Notably, patients in the PC cohort were less likely to have in-hospital bleeding complications despite being more likely to be initiated on enoxaparin. A potential confounder is differences in baseline comorbidities between PCs and NPCs, with PCs being more likely to have COPD, congestive heart failure, and thrombophilia, but less likely to have active cancer and cirrhosis. Despite higher rates of active cancer, NPC patients rarely had documented goals of care that would preclude PERT consultation or consideration of procedures, which may suggest perceived futility of consultation in this population.

When comparing the pre- vs post-PERT eras, our study demonstrated no significant difference in the outcomes of interest noted above; however, there were significantly fewer procedures in the post-PERT era. This was driven largely by a statistically significant reduction in catheter-based procedures as well as decreased IVC filter placement. There was also a non-significant trend towards a concomitant increase in thromboaspiration procedures. There were no institutional changes in the types of available therapies during our study period. Notably, there was no difference in rates of systemic lysis in the pre- vs post-PERT eras. In contrast, most studies including a recent meta-analysis demonstrate an overall increase in advanced therapies following PERT implementation, which is typically defined as both procedural interventions and.

systemic lysis.10,17,19–23.

The reduction in procedures observed following initiation of PERT at our center stands in.

contrast to prior studies that reported an increase in procedures following PERT initiation [6, 20, 23, 24]. Importantly, this reduction in procedures was not associated with a significant difference in mortality, HLOS, or TAC. There are several factors likely contributing to the reduction in procedures noted in our study. For one, this difference may be due our institution’s “gatekeeper” PERT model: PERT consults are initially evaluated by a pulmonologist on the Pulmonary Hypertension Service prior to involvement of interventional specialties. In this “gatekeeper” model, the pulmonologist determines whether involvement of other services is warranted, and the decision regarding advanced therapies is then made with the full multidisciplinary PERT. As a result, interventional specialties are involved in a minority of cases. Not all PERTs are structured in this way, with some centers activating all PERT services simultaneously for a coordinated multidisciplinary evaluation or have non-pulmonary services in the gatekeeper role. This heterogeneity of PERTs at different institutions may limit the generalizability of these results to other centers. Secondly, thromboaspiration has been an increasingly appealing option compared to CDT due to decreased need for ICU monitoring and concerns for non-trivial bleeding risk associated with CDT. PC patients were more likely to undergo thromboaspiration or receive systemic thrombolysis than NPC patients, while having reduced rates of bleeding during hospitalization. This may suggest more effective therapy selection following the multidisciplinary discussion with PERT involvement. The significant reduction in IVC filter placement likely reflects a change in global practice pattern after large randomized-controlled trials, published after the first PERT was established, showing no significant benefit in routine IVC filter placement in select populations [24, 25]. This is supported by the nearly ubiquitous demonstration of reduced IVC filter use in studies comparing pre- vs post-PERT eras, including a recent meta-analysis by Sosa et al. [9, 10, 15] As further evidence, the rates of IVC filter placement did not vary amongst PC and NPC cohorts in the PERT era.

Importantly, this two-tiered analysis highlights possible confounding factors present in prior prepost PERT studies. The fact that PERT consultation, rather than PERT presence, was associated with significant benefit to patients, suggests that PERT consultation may have previously unappreciated benefits. Studies of the effect of PERTs with a pre-post design may be confounded by global practice changes, non-consultation, or lack of involvement by the PERT.

There are several limitations to our study, including the retrospective nature and single-center design Despite the authors attempts to address potential confounding with the use of propensity score matching, PC patients and NPC patients may have intangible differences that we were not able to capture through review of the medical records. Additionally, our extensive institutional experience with interventions such as thromboaspiration and catheter-based therapies, and comparatively limited experience ECMO, may limit the generalizability of our conclusions to centers with similar experiences and resources. Furthermore, the significant heterogeneity of PERTs in terms of PE severity, frequency of activation, interventions available, and team composition may explain the subsequent heterogeneity in reported outcomes [10]. Thus, our results may only be generalizable to institutions with similar PERT composition and resources.