[Hospitalization for renal colic in the emergency room of a large central hospital in Israel: epidemiological features and clinical manifestations].

The medical records of all patients referred to the emergency department (ED) of Sheba Medical Center for renal colic during 1996 were analyzed. Patients discharged from the ED and those hospitalized were compared. There was no significant difference between the 2 groups with regard to average age or sex distribution. Statistically significant differences were found with regard to frequency of chills and fever, history of renal colic, referral for renal colic during that year or hospitalization for renal colic or nephrolithiasis, previous positive imaging, stone removal by surgery or extracorporeal shock wave lithotripsy, fever exceeding 37.5 degrees and administration of fluids, pethidine or pramin in the ED, prolonged stay in the ED and previous appendectomy. A conditional regression model tested the predictive value of each of those factors. Inclusion of independent variables into the model led to an overall correct classification rate of 84.43%, with 44.83% sensitivity and 93.16% specificity. There were correlations between referrals for renal colic, overall renal colic rate and average monthly temperature, so there was no pure correlation between average monthly temperature and referrals to the ED for renal colic. The major indications for hospitalization were actually the clinical ones, indicating either an active metabolic disease or suspected obstruction of the urinary tract. Treatment in the ED and duration of the visit indicated disease severity.