APACHE scoring and prediction of survival in intensive careBMJ 1995; 310 doi: https://doi.org/10.1136/bmj.310.6988.1197b (Published 06 May 1995) Cite this as: BMJ 1995;310:1197
- S N Pilkington
- Senior registrar Department of Intensive Care Medicine, Queen Alexandra Hospital, Cosham, Portsmouth PO6 3LY
EDITOR,—We are disappointed that Minerva has fallen into the same trap as others before her in misconstruing the role of the acute physiology and chronic health evaluation (APACHE) score in critically ill patients.1 Although the APACHE systems can be used to provide information on the risks of death for a group of patients suffering from a specific disease category that may require admission to an intensive care unit, they cannot be used as predictors of the risk of death in individual patients and consequently should not be used to make decisions concerning withdrawal of treatment in seriously ill patients.
For example, the APACHE score on admission for a patient with diabetic ketoacidosis may be high because of considerable physiological disturbance, but the mortality from this condition is low. On the other hand, a low score on admission may be obtained after resuscitation of a patient with a catastrophic intracranial haemorrhage, while the prognosis may remain extremely poor. Additionally, APACHE scores on admission may be altered considerably by the interventions performed before admission to the intensive care unit—the so called “lead time bias.”2
Misconceptions surrounding the use of the APACHE scoring system are common. Despite the contribution that APACHE scoring has made in clarifying severity of illness and to the concept of standardised mortality ratios it was not intended to be a method of predicting outcome in individual patients. Thus curtailing costs by limiting admission or continuing treatment according to scoring is invalid.3 In addition, APACHE scores cannot be used to decide who is to be admitted to an intensive care unit and should not be used to decide on a patient's discharge from an intensive care unit to either a ward or a high dependency area, as frequently patients' scores are similar in each of these areas.4 Finally, recent evidence suggests that the ability of the APACHE system to predict outcome may be no better than the clinical judgment of physicians and nurses.5