Purpose: To analyze the cause of GI symptoms after pelvic radiotherapy (RT) in a consecutive series of patients with symptoms beginning after RT. A striking lack of evidence is available concerning the optimal treatment for the 50% of patients who develop permanent changes in bowel habits affecting their quality of life after pelvic RT. As a result, in the UK, most such patients are never referred to a gastroenterologist.
Methods and materials: All diagnoses were prospectively recorded from a consecutive series of patients with symptoms that started after RT and who were referred during a 32-month period to a gastroenterology clinic. Patients either underwent direct access flexible sigmoidoscopy or were investigated in a standard manner by one gastroenterologist after first being seen in the clinic.
Results: A total of 265 patients referred from 15 institutions were investigated. They included 90 women (median age, 61.5 years; range, 22-84 years) and 175 men (median age, 70 years; range, 31-85 years). RT had been completed a median of 3 years (range, 0.1-34 years) before the study in the women and 2 years (range, 0-21 years) before in the men. Of the 265 patients, 171 had primary urologic, 78 gynecologic, and 16 GI tumors. The GI symptoms included rectal bleeding in 171, urgency in 82, frequency in 80, tenesmus, discomfort, or pain in 79, fecal incontinence in 79, change in bowel habit in 42, weight loss in 19, vomiting without other obstructive symptoms in 18, steatorrhea in 7, nocturnal defecation in 8, obstructive symptoms in 4, and other in 24. After investigation, significant neoplasia was found in 12%. One-third of all diagnoses were unrelated to the previous RT. More than one-half of the patients had at least two diagnoses. Many of the abnormalities diagnosed were readily treatable. The symptoms were generally unhelpful in predicting the diagnosis, with the exception of pain, which was associated with neoplasia (p < 0.001).
Conclusion: The results of our study have shown that radiation enteritis is not a single disease entity. More than one-half of the patients had more than one GI diagnosis contributing to their symptoms. After pelvic RT, specific GI symptoms were not a reliable measure of the underlying diagnoses, and the evaluation of new GI symptoms is worthwhile. An algorithm for this purpose is proposed.