Pre-arrest predictors of failure to survive after in-hospital cardiopulmonary resuscitation: a meta-analysis

Methods. We searched PubMed for studies published since 1985 and bibliographies of previous meta-analyses. We included studies with predominantly adult patients, limited to in-hospital arrest, using an explicit definition of cardiopulmonary arrest and CPR and reporting survival to discharge by at least one pre-arrest variable. A total of 35 studies were included in the final analysis. Inclusion criteria, design elements and results were abstracted in parallel by both investigators. Discrepancies were resolved by consensus. Results. The rate of survival to discharge was 17.5%; we found a trend towards increasing survival in more recent studies. Metastatic malignancy [odds ratio (OR) 3.9] or haematologic malignancy (OR 3.9), age over 70, 75 or 80 years (OR 1.5, 2.8 and 2.7, respectively), black race (OR 2.1), altered mental status (OR 2.2), dependency for activities of daily living (range OR 3.2-7.0 depending on specific activity), impaired renal function (OR 1.9), hypotension on admission (OR 1.8) and admission for pneumonia (OR 1.7), trauma (OR 1.7) or medical non-cardiac diagnosis (OR 2.2) were significantly associated with failure to survive to discharge; cardiovascular diagnoses and co-morbidities were associated with improved survival (range OR 0.23-0.53). Elevated CPR risk scores predicted failure to survive but have not been validated consistently in different populations. Conclusions. We identified several pre-arrest variables associated with failure to survive to discharge. This information should be shared with patients as part of a shared decision-making process regarding the use of do not resuscitate orders.