[The prognostic relevance of preoperative pulmonary function tests]

At Leipzig University, preoperative pulmonary function testing has been performed for about 3 years in order to detect and classify patients at high pulmonary risk. During the postoperative period, the risk of developing pulmonary complications is particularly high due to factors influencing respiratory mechanics such as the supine position, pain, residual effects of narcotic drugs, etc. It has often been emphasised that an underlying ventilatory disturbance such as obstructive lung disease or smoking may enhance the postoperative pulmonary risk, although the extent of the influence of preoperative pulmonary diseases on the postoperative complication rate is still controversial. The prediction of postoperative lung function from preoperative spirometric values is complicated by factors such as patient cooperation, pulmonary complications secondary to aspiration, infection, peritonitis, etc., and by differing and therefore non-comparable postoperative care. For this reason, the criteria for assessing pulmonary risk vary widely. METHODS. We examined 339 patients (mean age 59.3 years) preoperatively by quiet and forced spirometry; in most cases we also measured airway resistance and functional residual capacity. We estimated the postoperative lung function using the quadrant scheme of Miller and compared this risk class with our spirometric diagnosis and the postoperative clinical course. RESULTS. According to our results, Miller's classification seems inadequately differentiated for patients with mild to moderate ventilatory disturbances. A relatively high percentage of these patients were considered to have normal postoperative lung function. Some patients with severely diminished pulmonary function were classified as having sufficient postoperative lung function. The number and severity of pulmonary complications also corresponded better with the spirometric diagnosis, which was made using all spirometric parameters and not only vital capacity (VC) and 1-s forced expiratory volume (FEV1). We found that the percentage of primary respiratory complications increased with deterioration of the preoperative spirometric values. To provide a prognostic model combining both the advantages of using only a few parameters (FEV1, VC) and appropriate risk assessment, we propose a modification of the Miller scheme consisting of five risk classes. The analysis of the respiratory therapy regimen was unsatisfactory because of discrepancies between the predicted pulmonary risk, the use of respiratory therapy, and the occurrence of pulmonary complications. CONCLUSIONS. For minimising perioperative pulmonary complications, respiratory care (prophylaxis and therapy) adequate for the functional risk of the patient is necessary. We assume that intensive pre- and postoperative respiratory care and therapy in patients with underlying reductions in ventilatory function can help to avoid or reduce respiratory complications. The modification of Miller's scheme proposed after evaluating the postoperative course of our patients provides a differentiated prognostic model that allows the establishment of an appropriate and economical therapeutic regimen of perioperative pulmonary care.