Send to

Choose Destination
Ann Thorac Surg. 1998 Jul;66(1):220-4.

Predictive respiratory complication quotient predicts pulmonary complications in thoracic surgical patients.

Author information

Department of Anesthesiology and Critical Care Medicine, Memorial Sloan-Kettering Cancer Center, New York, New York 10021, USA.

Erratum in

  • Ann Thorac Surg 1998 Dec;66(6):2164.



This study was designed to develop an accurate preoperative index of prediction of outcome and hospital charges after lung resection with standard available pulmonary tests in a tertiary cancer center.


Sixty-one consecutive patients undergoing pulmonary resections were evaluated. All patients underwent spirometry, carbon monoxide diffusion capacity, split lung function testing, and room air blood gas analysis at rest and after a 2-minute step climb. The thoracic prospective data base and patient charts were reviewed for length of hospitalization, postoperative length of stay, and complications requiring therapy. Logistic regression analysis of the preoperative data, operation and postoperative outcome was used to develop a new postoperative predictive index: the predictive respiratory complication quotient (PRQ). We describe the design of the equation for the probability of serious pulmonary complications, hospital stay, and hospital charges based on PRQ.


Ten of 12 patients with a PRQ less than 2,200 suffered serious pulmonary complications of pneumonia, respiratory insufficiency, hypoxemia, and death. Forty-nine patients with a PRQ more than 2,200 did not experience any pulmonary complications. Postoperative length of stay and hospital charges correlated with the PRQ.


A construct such as the PRQ may provide a better prediction of outcome than its individual parts. We identified an important underlying relationship between intensive care unit stay, hospital stay and charges, and our index. A PRQ of less than 2,200 was associated with an increased risk of pulmonary complications and mortality.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center