Colonoscopy to evaluate abdominal pain in clinical practice: How often is significant neoplasia found?

D. A. Lieberman, D. E. Fleischer, G. M. Eisen, P. DeGarmo, M. Helfand

Research output: Contribution to journalArticlepeer-review

Abstract

Colonoscopy is often performed to evaluate patients with abdominal pain, bloating, constipation or diarrhea to rule out significant neoplasia. The purpose of this study was to utilize a national physician network to determine 1) how often colonoscopy is performed to evaluate vague abdominal symptoms (abd-sx); 2) the rate of significant colon neoplasia in such patients and 3) to compare the rates of significant neoplasia to rates in patients undergoing colonoscopy because of a positive fecal occult blood test (FOBT). Methods: A computerized endoscopic report generator was used at 18 sites by 99 physicians from April, 1997 until November 1997. Endoscopy data was sent electronically to a central databank for analysis. A cohort with only vague abd-sx and no other reason for colonoscopy was compared to a cohort undergoing colonoscopy to evaluate a positive FOBT. Significant neoplasia was defined as colonic polyps or masses ≥ 1cm. Results: 5073 colonoscopy reports were received during the study period. 388 patients (43% male) had colonoscopy to evaluate vague abd-sx (8%). 659 patients (52% male) had colonoscopy to evaluate (1) FOBT, accounting for 13% of procedures. Rates of significant neoplasia are shown in the table: Indication for Colonoscopy n Mean age (±SD) Polyp > 9mm Suspected Tumor TOTAL Vague Abd-sx 388 59 (±17) 37 12 49 (12.6%) (+) FOBT 659 65 (±13)*107 36 143 (22%)*p < 0.001 Patients with vague abd-sx had a significantly lower rate of serious neoplasia compared to patients with a (+) FOBT in clinical practice settings. Patients with vague abd-sx were younger, which may contribute to the lower rate of neoplasia. Conclusion: 8% of colonoscopy procedures were performed to evaluate vague abd-sx. The yield of serious neoplasia is similar to results in asymptomatic subjects. Further study is needed to determine which patients with vague abd-sx are most likely to benefit from a complete colon evaluation. This study highlights how a national endoscopic database can be used to evaluate endoscopic outcomes.

Original languageEnglish (US)
Pages (from-to)AB99
JournalGastrointestinal endoscopy
Volume47
Issue number4
StatePublished - Dec 1 1998

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Gastroenterology

Fingerprint Dive into the research topics of 'Colonoscopy to evaluate abdominal pain in clinical practice: How often is significant neoplasia found?'. Together they form a unique fingerprint.

Cite this