[The lung in peritonitis. Contribution of data processing to the evaluation of pulmonary risk: practical conclusions].

A retrospective study of 99 case-reports of patients with generalized peritonitis, including 38 postoperative cases involving peritoneum and lungs, was conducted with the aid of two files based on documented data containing 168 and 204 analytical criteria respectively. Data processing and statistical analysis alone were capable of establishing possible relationships between isolated or associated peritoneal factors and respiratory manifestations. Out of the vast quantity of data obtained by using this new approach, only factors significantly affecting pulmonary complications were retained: some relations could not be established because of the small number of patients studied. Respiratory complications were frequent (60%) and severe (58% mortality). They occurred in 96% of postoperative peritonitis cases, with 77% of deaths, and in 44% of primary peritonitis cases, with 36% of deaths. In contrast, when absent the mortality is reduced to 4%. Frequency distribution of clinical signs, pulmonary radiological images and the effects of gasometry are analyzed. Obstruction is a frequent and early symptom. The presence of bilateral diffuse lung shadows and their shunt effect are narrowly correlated with the presence of postoperative peritonitis. Respiratory complications were correlated with peritonitis criteria, particularly with age, the sub-mesocolon origin of the peritonitis, the rapidity of operative intervention, and surgical excision rather than shunt formation or suturing, and to the development of criteria of repeat abdominal or extra-abdominal surgery. Initial antibiotherapy and digestive suction have, on the contrary, no effect. Many other criteria were studied. A "profit" of patients liable to develop these complication is described, the mechanisms outlined, and the means of prevention and of rapid detection discussed, particularly the important role played by early gasometry.