British Journal of Anaesthesia, 2000, Vol. 84, No. 5 686-687
© 2000 The Board of Management and Trustees of the British Journal of Anaesthesia
Abstract |
Are we any better at predicting pulmonary artery occlusion pressure?
1 Department of Anaesthetics, Intensive Care Unit, Royal Infirmary of Edinburgh, Edinburgh EH3 9YW, UK
Abstract
The usefulness of pulmonary artery catheterization in the critically ill is questionable,1 but the measurement of haemodynamic variables permits a rational approach to management. The placement of pulmonary artery flotation catheters (PAFCs) is influenced by the perceived inability of the clinician to determine haemodynamic status from clinical evaluation alone.2 After 20 yr of experience in the use of the pulmonary artery catheter are we any better at predicting what we are measuring?
Four intensive care units in the South East of Scotland participated over a 6-month period. Pulmonary artery occlusion pressure (PAOP) and cardiac index were estimated immediately before PAFC placement. This abstract is confined to the results of PAOP estimations and actual values. In advance of analysis a difference of ±3 mmHg was considered clinically insignificant. Actual values were divided into three bands: <8 mmHg, 8–16 mmHg and >16 mmHg.
Data on 92 patients were suitable for analysis. Four patients were excluded due to incomplete information. The mean age was 61 yr. The PAFC was placed within 48 h of admission to the Intensive Care Unit in 87 patients. In 70 patients the placement of the PAFC was consultant directed. Sixty-one patients (66.3%) had a clinically insignificant difference between predicted and actual PAOP. Of the remaining 31 patients (33.6%), 11 patients were just out with the set criteria (4 mmHg difference) and this did not affect their management. Thirteen patients had estimations and actual measurements which were in different bands. Actual PAOP measurement revealed nine of these patients to be underfilled (six septic, three others). Two septic patients, estimated to be very underfilled, had elevated actual PAOP measurements (predicted PAOP 5 and 12 mmHg, actual PAOP 25 and 20 mmHg respectively). Prediction of pulmonary artery occlusion pressure is generally accurate, however clinical judgement may fail to diagnose significant underfilling in patients with sepsis. It is debatable whether PAOP accurately reflects volume status, but five of these patients had a measured PAOP of 4 mmHg or less, and only one of these patients was judged to be underfilled. Two patients who were shocked (meningococcal septicaemia and bowel obstruction) were judged to be underfilled, but both had a PAOP 
20 mmHg. It is hard to believe that this did not represent undiagnosed myocardial dysfunction.
PAFC placement remains a useful diagnostic and management tool in a small number of shocked septic patients.