Algorithm-based management of patients with gastrointestinal symptoms in patients after pelvic radiation treatment (ORBIT): a randomised controlled trial

Summary

Background

Chronic gastrointestinal symptoms after pelvic radiotherapy are common, multifactorial in cause, and affect patients' quality of life. We assessed whether such patients could be helped if a practitioner followed an investigative and management algorithm, and whether outcomes differed by whether a nurse or a gastroenterologist led this algorithm-based care.

Methods

For this three-arm randomised controlled trial we recruited patients (aged ≥18 years) from clinics in London, UK, with new-onset gastrointestinal symptoms persisting 6 months after pelvic radiotherapy. Using a computer-generated randomisation sequence, we randomly allocated patients to one of three groups (1:1:1; stratified by tumour site [urological, gynaecological, or gastrointestinal], and degree of bowel dysfunction [IBDQ-B score <60 vs 60–70]): usual care (a detailed self-help booklet), gastroenterologist-led algorithm-based treatment, or nurse-led algorithm-based treatment. The primary endpoint was change in Inflammatory Bowel Disease Questionnaire–Bowel subset score (IBDQ-B) at 6 months, analysed by intention to treat. This trial is registered with ClinicalTrials.gov, number NCT00737230.

Findings

Between Nov 26, 2007, and Dec 12, 2011, we enrolled and randomly allocated 218 patients to treatment: 80 to the nurse group, 70 to the gastroenterologist group, and 68 to the booklet group (figure). Most had a baseline IBDQ-B score indicating moderate-to-severe symptoms. We recorded the following pair-wise mean difference in change in IBDQ-B score between groups: nurse versus booklet 4·12 (95% CI 0·04–8·19; p=0·04), gastroenterologist versus booklet 5·47 (1·14–9·81; p=0·01). Outcomes in the nurse group were not inferior to outcomes in the gastroenterologist group (mean difference 1·36, one sided 95% CI −1·48).

Interpretation

Patients given targeted intervention following a detailed clinical algorithm had better improvements in radiotherapy-induced gastrointestinal symptoms than did patients given usual care. Our findings suggest that, for most patients, this algorithm-based care can be given by a trained nurse.

Funding

The National Institute for Health Research.

Introduction

The number people who survive cancer has tripled in the past 30 years.¹ However, chronic physical consequences of treatment for cancer adversely affect on the quality of life of 20–25% of survivors.²

The largest group of patients reporting debilitating chronic side-effects are those treated with radiotherapy alone or in combination with other treatments for pelvic cancer. Gastrointestinal symptoms are the most common chronic physical side-effects and have the greatest effect on daily activity.3, 4 Overall, 50% of patients report that their gastrointestinal symptoms affect their quality of life and 20–40% say that this effect is moderate or severe.⁵ Such problems include chronic faecal incontinence (up to 60% of patients) after radiotherapy for prostate or rectal cancer,6, 7 and chronic loose stool (47%), defaecatory urgency (29%), or chronic abdominal pain (17%) after radiotherapy for gynaecological cancer.⁸

In 2010, the UK National Cancer Survivorship Initiative Vision challenged professionals to develop new models of care for these patients because “the needs of cancer survivors are not being met, that being ‘cured’ of cancer does not necessarily equate with being well and that chronic consequences of treatment can have a devastating impact on daily life”.⁹

To meet the challenge of the survivorship initiative and to develop a sustainable service to deal with the rapidly escalating demand for treatment, we explored the potential of nurse-delivered, algorithm-directed care for these patients.

Radiotherapy induces long-term changes in bowel function as a result of progressive endothelial dysfunction, which induces ischaemia and subsequent fibrosis. The same processes might cause dysfunction in other pelvic organs, a disorder defined as pelvic radiation disease.¹⁰ During therapeutic irradiation of a cancer in the pelvis, parts of distal small bowel, caecum, transverse and sigmoid colon, and rectum are often also irradiated. Additionally, the pancreas and proximal small bowel might also receive some irradiation if para-aortic nodes are treated. That even low dose radiation can cause substantial changes to gastrointestinal function is becoming increasingly recognised.¹¹

We know of no previous clinical trials investigating whether radiation-induced bowel injury is treatable, but a substantial amount of clinical research challenges the widely held view that nothing can be done to manage these patients' symptoms.12, 13 This research indicates that gastrointestinal symptoms arise because irradiation potentially induces a variety of abnormalities in physiological functioning in exposed areas of the gastrointestinal tract. Changes in different physiological functions can result in identical symptoms, so the presence of a specific symptom or cluster of symptoms does not predict the underlying cause.14, 15 Because more than one gastrointestinal physiological function is often affected, patients' symptoms could have more than one cause.

We postulated that when a patient develops new gastrointestinal symptoms after pelvic irradiation, systematic assessment should be able to identify which physiological abnormalities are contributing to these symptoms. If abnormalities are detected, then treatments prescribed to treat each abnormality systematically should be effective.

In a series of studies,16, 17, 18 we identified 23 symptoms that develop after radiotherapy. We defined the investigations needed to find out the cause or causes of each symptom and possible sequential treatments. We piloted and adjusted the resultant algorithm in our clinic.16, 17, 18 After adjustment, the algorithm provides a step-by-step approach along a care pathway from initial identification of symptoms to long-term management. We previously showed that our algorithm could be applied by a nurse and that it seemed to improve symptoms.¹⁹

We tested whether patients with new-onset gastrointestinal symptoms after previous pelvic radiotherapy could be helped if a practitioner followed our investigative and management algorithm, and whether a nurse could apply the algorithm in such a way that outcomes were not worse than when applied by a consultant gastroenterologist.