EUSOS: Mortality after surgery in Europe

Pearse et al. Lancet 2012; 380:1059-65. doi:10.1016/S0140-6736(12)61148-9

Clinical Question

• What is the clinical outcome of patients after major non-cardiac surgery in Europe?

Design

• Prospective cohort design
• Participating hospitals were a voluntary convenience sample

Setting

• 498 hospitals across 28 European countries (56% university affiliated)
• 7 days in April 2011

Population

• Inclusion: all adults (over 16 years) undergoing elective or non-elective inpatient surgery
• Exclusion: day-case surgery; cardiac surgery; neurosurgery; radiological procedures; obstetric procedures
• 46,985 patients screened, 46,539 included in analysis (1% removed: duplicate, inconsistent or missing data)

Outcome

• Primary outcome:
  • Crude in-hospital mortality across Europe was 4.0%
  • Compared to UK as baseline, four countries had statistically significantly different and unexplained higher rates of mortality:
    • Poland — adjusted OR 6.92 (95% CI 2.37–20.27); p-value 0.0004
    • Romania — adjusted OR 3.19 (95% CI 1.61–6.29); p-value 0.001
    • Latvia — adjusted OR 4.98 (95% CI 1.22–20.29); p-value 0.025
    • Ireland — adjusted OR 2.61 (95% CI 1.30–5.27); p-value 0.007
  • Compared to UK, no country had statistically significantly better rates of mortality, although UK was 10th in rank order
• Secondary outcome:
  • Hospital stay 3.0 days [IQR 1.0 – 7.0]
  • Overall admission rate to ICU was 8%
    • Of which it was unplanned for 29%
      • Mortality in this unplanned Critical Care cohort was 23% compared to 10% in the planned Critical Care cohort
    • Median critical care stay was 1.2 days [IQR 0.9 – 3.6]
  • Of all the deaths, 73% of these patients never attended Critical Care
    • Of the deaths in the cohort that did attend Critical Care, 43% occurred after discharge from Critical Care
• Independent variables associated with increased mortality:
  • Country
  • Urgency
  • Grade of surgery
  • Surgical procedure category
  • American Society of Anesthesiologists physiology score (ASAps)
  • Metastatic disease
  • Cirrhosis

Summary of EuSOS Data

Choropleth map graphically illustrating primary and secondary outcomes by country

Authors’ Conclusions

• The findings demonstrate that 4% of patients undergoing inpatient surgery die before discharge from hospital, which was higher than expected and varies substantially across Europe.
• There is a systematic failure in the allocation of Critical Care resources

Strengths

• First, large, multi-centre, multi-national cohort survey
• Only small number of cases excluded due to missing data
• Outcomes are comparable to smaller national surveys

Weaknesses

• May not represent true outcomes across Europe, as only based on self-selecting hospitals
• Some countries reported only small numbers of patients, so the described practice and outcomes may not represent national practice
• True number of eligible patients not known, as some may not have been reported
• Risk-adjustments made for confounding variables were based upon subjective scores such as ASA and surgical urgency category
• As a cross-sectional survey, associations and causations are difficult / impossible to conclude with certainty