Organizational setups and non-elective surgical delays. A single-center cohort study: institutional setups and
Maxime T Aparicio1, Jean Pasqueron2, Sylvain Diop3,4
1Department of Anesthesiology, Critical Care and Perioperative Medicine, DMU CARE, Assistance Publique-Hôpitaux de Paris (AP-HP), Hôpitaux Universitaires Henri Mondor, Créteil, France - maxime.theo.aparicio@gmail.com.
Background:
Extended delays in non-elective surgeries have been associated with suboptimal outcomes. The SARS-CoV-2 pandemic forced healthcare systems to adapt their setups for unscheduled procedures, leading, in our institution, to a reorganization from a setup with two dedicated operating rooms (ORs) at a central facility without dedicated teams to a temporary one with both dedicated teams and ORs during lockdown phase. This study evaluates the impact of this transitions on the time to surgery considering unscheduled procedures.
Methods:
We considered three periods: a historical cohort from the year preceding the first French lockdown, the lockdown period, and a post-lockdown era covering the four weeks immediately afterward. The ideal time to surgery (iTTS) was retrospectively determined using the non-elective surgery triage classification. The primary outcome focused on the proportion of patients operated after their iTTS.
Results:
Over study periods, 435 patients underwent non-scheduled surgery: 137,198 and 100 in the historical cohort, the lockdown, and post-lockdown period respectively. The proportion of out-timed patients was significantly lower during lockdown period than in the historical cohort (36.5% vs. 55.3%, P<0.001), driven by the less urgent patients (i.e., NEST 5-6 patients, 25.6% during lockdown vs. 58.2% in historical cohort). There was no significant difference between the lockdown era and the post-lockdown phase.
Conclusions:
Our findings suggest that dedicated teams and surgical suite may reduce time to surgery for non-scheduled procedures, particularly for less urgent cases. However, the retrospective monocentric design and our limited statistical power limit the extend of our conclusions.
