Skip to main content
  • Poster presentation
  • Published:

Predicting failure to survive cardiopulmonary resuscitation in intensive care: evaluation of two morbidity scores


Major advances have been made in providing prompt cardiopulmonary resuscitation (CPR) to cardiac arrest patients within the hospital. As patients undergoing cardiac surgery become older and sicker with concomitant comorbidities, the quality of immediate and long-term postoperative care as well as resuscitation are likely to become increasingly important. Inappropriate and unsuccessful resuscitation of these patients is associated with a large expenditure of health care resources. CPR is a time-consuming and resource-consuming procedure that was never originally recommended for all patients. Gillon [1] described CPR as 'violent, painful and undignified'. Therefore, the goal of CPR should be to 'reverse premature death not prolong inevitable death' [2].


The aim of this study is to evaluate the usefulness of two morbidity scores – prognosis after resuscitation score (PAR) and modified PAM index (MPI) – in predicting failure to survive following cardiopulmonary resuscitation (CPR) in our ICU.


We reviewed the records of postoperative cardiac adult patients who underwent CPR in our ICU from April 1999 to March 2003 at the Georg-August University Hospital, Göttingen, Germany.


During the study period, 169 patients had cardiac arrest for which CPR was instituted. Eighty (53.7%) survived to discharge from the ICU and 71 (47.7%) survived to be discharged from the hospital. The mean time interval until ROSC was reestablished and the duration of CPR differed significantly in survivors and nonsurvivors (P = 0.008 and P = 0.0002, respectively). The PAR score identified 77 patients with a score >7 and the MPI score identified 74 patients with a score >7 none of whom survived to be discharged from the ICU. The sensitivities of the PAR score and MPI for predicting failure to survive following CPR were 86.5% and 83.1%, respectively. The PAR and MPI scores did not incorrectly identify a patient as a nonsurvivor who subsequently survived.


The PAR and MPI scores are useful in identifying patients in whom CPR may be unsuccessful and could additionally provide useful information to physicians and patients involved with decisions about do-not-attempt-resuscitation orders.


  1. Gillon R: Resuscitation policies – action required. J Med Ethics 1992, 18: 115-116. 10.1136/jme.18.3.115

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  2. Cummins RO: Ethical Aspect of Emergency Cardiac Care. Textbook of Advanced Cardiac Life Support. Dallas, TX: AHA; 1997:1-71.

    Google Scholar 

Download references

Author information

Authors and Affiliations


Rights and permissions

Reprints and permissions

About this article

Cite this article

Enohumah, K., Mörer, O., Hinz, J. et al. Predicting failure to survive cardiopulmonary resuscitation in intensive care: evaluation of two morbidity scores. Crit Care 10 (Suppl 1), P384 (2006).

Download citation

  • Published:

  • DOI: