Study objective: To determine how risk factors could be combined to be
st predict the development of a postoperative pulmonary complication (
PPC) following abdominal surgery. Design: Prospective model-building s
tudy. Logistic regression models were developed using significant risk
factors identified in the univariate analysis. Setting: Four midweste
rn hospitals. Patients: Convenience sample of 400 patients who underwe
nt abdominal surgical procedures between January 1993 and August 1995.
Measurements and results: Multicriteria outcome for postoperative pul
monary complication used to collectively assess atelectasis and pneumo
nia. Twenty-three risk factors were assessed. Six risk factors were id
entified as independent by logistic regression: age greater than or eq
ual to 60 years (adjusted odds ratio [Adj OR], 1.89); impaired preoper
ative cognitive function (Adj OR, 5.93); smoking history within the pa
st 8 weeks (Adj OR, 2.27); body mass index: greater than or equal to 2
7 (Adj OR, 2.82); history of cancer (Adj OR, 2.23); and incision site-
upper abdominal or both upper/lower abdominal incision (Adj OR 2.30).
Conclusions: These results provide a framework for identifying patient
s at risk of developing a PPC following abdominal surgery. A reliable
and valid risk index could be used clinically to guide preoperative an
d postoperative pulmonary care and target limited resources for patien
ts at risk.