Use and definitions of perioperative mortality rates in low-income and middle-income countries: a systematic review.

Background: Aggregate and risk-stratified perioperative mortality rates (POMR) are well-documented in high-income countries where surgical databases are common. In many low-income and middle-income country (LMIC) settings, such data are unavailable, compromising efforts to understand and improve surgical outcomes. We undertook a systematic review to determine how POMR is used and defined in LMICs and to inform baseline rates.

Methods: We searched PubMed for all articles published between Jan 1, 2009, and Sept 1, 2014, reporting surgical mortality in LMICs. Search criteria, inclusion and exclusion criteria, and study assessment methodology are reported in the appendix. Titles and abstracts were screened independently by two reviewers. Full-text review and data extraction were completed by four trained clinician coders with regular validation for consistency. We extracted the definition of POMR used, clinical risk scores reported, and strategies for risk adjustment in addition to reported mortality rates.

Findings: We screened 2657 abstracts and included 373 full-text articles. 493 409 patients in 68 countries and 12 surgical specialties were represented. The most common definition for the numerator of POMR was in-hospital deaths following surgery (55·3%) and for the denominator it was the number of operative patients (96·2%). Few studies reported preoperative comorbidities (41·8%), ASA status (11·3%), and HIV status (7·8%), with a smaller proportion stratifying on or adjusting mortality for these factors. Studies reporting on planned procedures recorded a median mortality of 1·2% (n=121 [IQR 0·0-4·7]). Median mortality was 10·1% (n=182 [IQR 2·5-16·2) for emergent procedures.

Interpretation: POMR is frequently reported in LMICs, but a standardised approach for reporting and risk stratification is absent from the literature. There was wide variation in POMR across procedures and specialties. A quality assessment checklist for surgical mortality studies could improve mortality reporting and facilitate benchmarking across sites and countries.

Funding: None.