Unless you’re good at guessing, it’s not much use being a detective – Agatha Christie
Introduction
- This study sought to compare the accuracy of pre-operative subjective assessment (METS) with cardiopulmonary exercise testing (CPET, the Duke Activity Index Score (DASI) questionnaire and pre-operative NT pro-BNP levels.
- The prognostic accuracy of these values for predicting post-operative morbidity and mortality was also assessed.
Methods
- An International, multi-centre, prospective cohort study conducted in 25 hospitals (Canada, UK, Australia and New Zealand).
- Primary outcome = Death or myocardial infarction within 30 days after surgery
- Secondary outcome = Death within one year after surgery
- Other outcomes:
- Death or myocardial injury within 30 days
- Moderate or severe post-operative complications defined by the Clavien-Dindo classification
- Inclusion criteria were patients >40 years of age undergoing elective non-cardiac surgery (with an anticipated overnight stay or longer) with at least one risk factor for cardiac complications.
- Exclusion criteria included; planned endovascular approach, pregnancy, cardiac conditions or absolute contraindications to CPET.
- All patients recruited underwent pre-operative subjective functional assessment (by an anaesthetist), DASI questionnaire, CPET and NT Pro-BNP measured. Anaesthetist s blinded to these results (except whereby adverse events during CPET warranted unmasking eg. arrhythmia).
- Patients were assessed with daily ECG and Troponin until day 3 post-operatively. Further follow up was conducted at 30 days and 1 year.
Results
- 6548 met inclusion criteria with 1741 consenting to participate.
- 1401 underwent both surgery and CPET and were included in analysis.
- Subjective assessment of functional capacity had a sensitivity of 19.2% for identifying peak oxygen consumption of less than 14 mL/kg per min and a specificity was 94·7%.
- Peak oxygen consumption was positively correlated with DASI scores (p<00001), and negatively correlated with NT proBNP concentrations (p<0·0001).
- Subjectively assessed preoperative functional capacity had no significant adjusted association with the four main study outcomes.
- DASI scores were associated with the primary outcome (OR 0.91 CI 0.83-0.99) and 30-day death or myocardial injury (OR 0.96 CI 0.92-0.99)
- NT Pro-BNP was associated with 30-day death or myocardial injury (OR 1.78 CI 1.21-2.62) and 1-year death (OR 2.91 CI 1.54-5.49).
- Peak oxygen consumption was associated with in-hospital moderate or severe complications (OR 0.86 CI 0.78-0.97).
Discussion
- Only 27% of patients approached consented to the study. Those that did consent may be a skewed cohort of patients ie. those willing to undergo pre-operative CPET for the purpose of the study.
- Study did not adequately explain what happened to patients who were deemed unfit for surgery based on subjective assessment.
- The sample size was based on a 5% primary outcome event rate. The actual event rate only 2% meaning the study was underpowered for this.
- No adjustments were made for multiple testing.
- Subjective assessment may influence complication rates in terms of minimising post-operative risks through augmentation of anaesthetic conduct. An effect that cannot be measured.
- The author’s state that functional assessment should not be used for peri-operative risk evaluation. However, CPET failed also failed to predict MI, myocardial injury or death.
- CPET did predict moderate to severe post-operative complications based on the Clavien-Dindo classification.
- DASI scores proved a useful predictor of cardiac complications and are easily implemented.
Change to clinical practice?
- Greater use of the DASI for pre-operative risk prediction should be considered.
- Additionally pre-operative NT Pro-BNP should be considered for prediction of post-operative cardiac complications.
Summary by Dr Sam Curtis
