Helical CT pneumocolon to assess colonic tumors: radiologic-pathologic correlation.

OBJECTIVE

The objective of this study was to determine the accuracy of helical CT pneumocolon in the staging of colonic carcinomas.

SUBJECTS AND METHODS

Fifty-two patients (27 men, 25 women; age range, 40-88 years old; mean age, 67 years old) with known or strongly suspected colonic disorders underwent CT pneumocolon. After bowel cleansing, administration of smooth muscle relaxant, and rectal air insufflation, CT pneumocolon was performed with 5-mm collimation, 2.5-mm reconstruction interval, and a pitch of 1.5. Two contrast administration protocols were used. Twelve patients received 100 ml of i.v. contrast material that was injected at 3 ml/sec; scan delay was 45 sec. The other 40 patients received 150 ml of contrast material at 5 ml/sec; scan delay was 25 sec. Images were prospectively evaluated. All patients had pathologic confirmation: 46 patients had resections and six patients had endoscopic biopsies.

RESULTS

Diagnostic images were obtained in 47 of 52 patients. In the 47 patients, there were 38 colonic carcinomas (one synchronous), nine diverticular strictures, eight polyps, one ischemic stricture, and one normal study. Thirty of 38 carcinomas were correctly staged by CT. Sensitivity and specificity for serosal infiltration were 100% (35/35 carcinomas) and 33% (one of three carcinomas), respectively; sensitivity and specificity for lymph node involvement were 56% (nine of 16 carcinomas) and 95% (21/22 carcinomas), respectively. Four polyps that were smaller than 5 mm in diameter were not revealed by CT. Twelve of 14 benign lesions were correctly diagnosed and two were believed to be malignant. No malignant lesion was misdiagnosed.

CONCLUSION

CT pneumocolon gave an overall staging accuracy of 79% in 38 carcinomas.