BackgroundMortality, morbidity, and organ failure are important and common serious harms after surgery. However, there are many candidate measures to describe these outcome domains. Definitions of these measures are highly variable, and validity is often unclear. As part of the International Standardised Endpoints in Perioperative Medicine (StEP) initiative, this study aimed to derive a set of standardised and valid measures of mortality, morbidity, and organ failure for use in perioperative clinical trials. MethodsThree domains of endpoints (mortality, morbidity, and organ failure) were explored through systematic literature review and a three-stage Delphi consensus process using methods consistently applied across the StEP initiative. Reliability, feasibility, and patient-centredness were assessed in round 3 of the consensus process. ResultsA high level of consensus was achieved for two mortality time points, 30-day and 1-yr mortality, and these two measures are recommended. No organ failure endpoints achieved threshold criteria for consensus recommendation. The Clavien–Dindo classification of complications achieved threshold criteria for consensus in round 2 of the Delphi process but did not achieve the threshold criteria in round 3 where it scored equivalently to the Post Operative Morbidity Survey. Clavien–Dindo therefore received conditional endorsement as the most widely used measure. No composite measures of organ failure achieved an acceptable level of consensus. ConclusionsBoth 30-day and 1-yr mortality measures are recommended. No measure is recommended for organ failure. One measure (Clavien–Dindo) is conditionally endorsed for postoperative morbidity, but our findings suggest that no single endpoint offers a reliable and valid measure to describe perioperative morbidity that is not dependent on the quality of deli-vered care. Further refinement of current measures, or development of novel measures, of postoperative morbidity might improve consensus in this area.
Read full abstract