Early Diagnosis of Postoperative Infection Using Serum Total Cholesterol Level -Mutivariate Analysis Study-

Yukihiro Akiya, Masahiko Onda, Kiyonori Furukawa, Hideyuki Suzuki

First Department of Surgery, Nippon Medical School

The purpose of the current study is designed to determine whether it is possible to predict the incidence of postoperative infection. One hundred and eleven cases of gastric and colorectal operations, which were carried out between July 1995 and June 1997, were divided into two groups as follows. Thirteen cases suffered from postoperative infection (infection group), ninety-eight were infection free until the 10th postoperative day (POD10) (no infection group). We analyzed examinations of both groups on POD3 with univariate analysis. Lymphocyte counts (Ly), total cholesterol (TC), and total protein (TP) of the infection group were obviously lower than those of the no infection group. Body temperature, pulse and CRP were higher in the infection group than in no the infection group. Then, we utilized logistic regression analysis in multivariate analysis to evaluate which parameters (e.g. WBC, Ly, TC, triglyceride, TP, albumin, CRP) should be employed. Consequently, in gastric operations, 8 parameters were analyzed including TC, and in colon operations, 3 parameters were analyzed including TC, which produced the best overall percent correct and sensitivity between prediction and observation. The overall percent correct for analysis was 95.1% for gastric operations, and 95.4% for colon operations, and the sensitivity of analysis was 80.0% for both gastric and colon operations. However, in rectal operations, the overall percent correct was satisfactory, but the sensitivity was lower than that of gastric and colon operations. Taken together, we conclude that early diagnosis of postoperative infection is possible by utilizing logistic regression analysis of clinical data on POD3 including the serum total cholesterol level.

Key words
postoperative infection, early diagnosis, serum total cholesterol, multivariate analysis, digestive tract

Reprint requests
Yukihiro Akiya First Department of Surgery, Nippon Medical School
1-1-5 Sendagi, Bunkyo-ku, Tokyo, 113-8603 JAPAN

Accepted
March 11, 1998

To read the PDF file you will need Abobe Reader installed on your computer.