Study objective: Intra-abdominal operations are relatively high risk for pulmonary complications. Previous research has more intensely investigated cardiac operative risk, but recent work suggests that significant pulmonary complications may be more common than cardiac complications and associated with longer length of stay. This study identified risk indicators for pulmonary complications after elective abdominal operations.
Design: Nested case-control.
Setting: University affiliated Veterans Affairs hospital.
Patients: We used a computerized registry of all 2,291 patients undergoing elective abdominal operations from 1982 to 1991. Ascertainment and verification of pulmonary and cardiac complications were systematic and explicit. Charts of all 116 patients identified by the registry as having complications and 412 (19%) randomly selected from 2,175 remaining patients were reviewed to verify complications, using explicit criteria and independent abstraction of preoperative and postoperative components of charts. From 528 validated subjects (23% of the cohort), 82 cases and 82 control subjects were closely matched by operation type and age, ±10 years.
Measurements and results: The primary outcome measure was postoperative pulmonary complications. Among 82 cases with pulmonary complications, 27 (33%) also had cardiac complications. Preoperative variables independently associated with pulmonary complications by multivariable analysis (p≤0.05) included the following: Charlson comorbidity index (per point odds ratio [OR], 1.6; 95% confidence interval [CI], 1.004 to 2.6), Goldman cardiac risk index (per point OR, 2.04; 95% CI, 1.17 to 3.6), abnormal chest radiograph (OR, 3.2; 95% CI, 1.07 to 9.4), and abnormal findings on lung examination (OR, 5.8; 95% CI, 1.04 to 32). Equal proportions of cases and control subjects had preoperative diagnostic spirometry. No component of spirometry predicted complications, including severity of obstructive lung disease.
Conclusions: For pulmonary operative risk, abnormal results of lung examination and chest radiography plus cardiac and overall comorbidity were important. Spirometry was not helpful. Because 33% of cases had both cardiac and pulmonary complications, future studies should prospectively examine comparative incidence, outcomes, and predictors of both types of complications.