The records of 22 consecutive patients with a non-traumatic ill-defined acute abdomen having diagnostic closed peritoneal lavage, were reviewed retrospectively. All patients had one or more risk factors. Peritoneal lavage was performed under visual control and its diagnostic value was based on the comparison with the definitive diagnosis, based on laparatomy and complementary tests. There were 12 positive results with 11 true positives (7 cases of peritonitis, and 4 of visceral necrosis) and 1 false positive. There were 10 negative results with 9 true negatives and 1 false negative. Sensitivity of peritoneal lavage was 0.91%, specificity 0.90%, positive predictivity 0.91% and negativity 0.90%. Overall mortality was 45%. The limits of the procedure appear to be the infections or localized abscesses. Peritoneal lavage as a simple and easy method, is suitable for high-risk patients when the choice is between immediate operation with possible side-effects or delay with possible adverse course of the abdominal lesion, and permits immediate indication for surgery in case of positivity.