Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Protocol - Intervention

Interventions for maintenance of mucosal healing in ulcerative colitis

Authors' declarations of interest

Version published: 20 February 2014 Version history

https://doi.org/10.1002/14651858.CD010998

This is not the most recent version

view the current version 22 November 2019
Collapse all Expand all

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

The primary objective of this review is to evaluate the efficacy and safety of medical therapies used to maintain mucosal healing in UC.

Background

Description of the condition

Ulcerative colitis (UC) is an idiopathic inflammatory bowel disease that results in diffuse mucosal inflammation in the colon. The pathogenesis of UC remains unclear, but appears to involve alterations in cellular and humoral immunity in genetically predisposed individuals in concert with a complex interplay of environmental factors and the colonic microbiome.

The clinical course of UC is generally characterized by periods of asymptomatic remission interspersed by periods of abdominal cramping, urgency, and bloody diarrhea caused by colonic inflammation. The inflammation arises just above the anorectal junction and, while it can be limited to the the rectum (i.e. proctitis), it often extends proximally to the splenic flexure (i.e. left sided colitis) or can involve the more proximal colon (i.e. pancolitis). Mucosal abnormalities that can be seen endoscopically begin as mild changes such as erythema and a diminished vascular pattern and, as inflammation continues, can develop into marked erythema, an absent vascular pattern, friability, and mucosal erosions. Severe UC is characterized by mucosal friability, spontaneous bleeding and significant ulceration. While UC can often be managed with medical therapies, severe, life‐threatening attacks may require a colectomy (Caprilli 2007).  

Ulcerative colitis activity is typically assessed using clinical, biochemical, and endoscopic measures. The traditional approach to the medical management of UC is to use medications both to induce remission and to then maintain disease quiescence. Induction therapies typically include corticosteroids (e.g prednisone, prednisolone, methylprednisolone, and budesonide) and aminosalicylates (e.g. mesalazine or sulfasalazine), which can be used both topically and systemically. Severe attacks of UC can also be treated with biologic agents (e.g. infliximab). Immunosuppressants (e.g. azathioprine, 6‐mercaptopurine) can be used as maintenance therapy to augment aminosalicylate therapy. Unlike Crohn’s disease, UC is limited to the colon so a colectomy may be performed in cases that fail to respond to standard medical therapy.

Historically the goal of treating UC was to induce clinical remission. However, the idea of treating to resolve the underlying inflammation is appealing because of mounting evidence that mucosal healing (MH) may alter future disease course (Froslie 2007; Armuzzi 2012; Ananthakrishnan 2012; Colombel 2012). In the ACT‐1 and ACT‐2 trials endoscopic improvement at week 8 was associated with sustained symptomatic and corticosteroid‐free remission, sustained mucosal healing, and reduced need for colectomy through 54 weeks (Colombel 2012). In one Norwegian cohort study MH was also shown to decrease the risk of future colectomy (Froslie 2007). Further evidence for treating underlying inflammation comes from studies that demonstrate the degree of colonic inflammation is an important risk factor for colorectal neoplasia (Rutter 2004a; Rutter 2004b; Gupta 2007).The current review will assess the efficacy and safety of medical therapies for maintenance of MH in UC.

Description of the intervention

A variety of medical agents have been studied as potential therapies for maintenance of remission in UC. Traditional, maintenance therapies include oral or topical 5‐aminosalicylic acid medications or immunosuppressant medications such as methotrexate (oral, intramuscular, or subcutaneous) or oral thiopurines including azathioprine and 6‐mercaptopurine. In recent years biologic medications including infliximab, adalimumab, and vedolizumab have been developed. These medications are administered either intravenously or subcutaneously on a regular dosing schedule.

A number of endoscopic activity indices have been developed to assess disease activity. These include the Baron score (Baron 1964), the modified Baron score (Feagan 2005), the St Mark's index (Powell‐Tuck 1982), the Rachmilewitz endoscopic index (Rachmilewitz 1989), the Mayo score (Schroeder 1987), the Lémann endoscopic index (Lemann 1995), the Sutherland index (Sutherland 1987), the ulcerative colitis endoscopic index of severity (UCEIS) (Travis 2012), and Truelove and Witt’s sigmoidoscopic assessment (Truelove 1955). Since each tool uses a slightly different definition for scoring mucosal abnormalities, there is no standard definition of MH. Generally, it is accepted that MH refers to the complete endoscopic healing of mucosal abnormalities that were observed at baseline assessment.

How the intervention might work

Anti‐inflammatory and immunosuppressive medications are generally used to maintain remission in UC with the goal of suppressing gastrointestinal tract inflammation sufficiently to allow the mucosa to heal. The specific mechanism of action of 5‐aminosalicylic acid medications is unknown. Current evidence points to an anti‐inflammatory effect through alteration of local inflammatory mediators and inhibition of T‐cell proliferation (MacDermott 2000). Corticosteroids decrease inflammation through a number of mechanisms, including inhibition of pro‐inflammatory cytokines and arachidonic acid metabolites, and by effects on vascular permeability an the migration of circulating leukocytes (Rubin 1999). Systemic corticosteroids are avoided as maintenance therapy because of a significant adverse effect profile.

The thiopurine medications, azathioprine and 6‐mercaptopurine, are purine analogues that exert an immunosuppressive effect by disrupting DNA synthesis and inducing apoptosis (Sahasranaman 2008). Methotrexate is a folate antagonist that also exerts an immunosuppressive effect. Although the exact mechanism of action is unknown, it likely involves decreased production of leukotrienes and the induction of adenosine (Cronstein 1993).

Infliximab and adalimumab are tumor necrosis factor‐alpha (TNF‐α) antagonists that bind to soluble and cell‐bound TNF and induce apoptosis of activated lymphocytes (Van den Brande 2003). Vedolizumab is a monoclonal antibody against the alpha‐4‐beta‐7 integrin that is relatively specific for leukocytes in the gastrointestinal tract and is thought to prevent the migration of leukocytes into inflamed tissue (Gerner 2013).

Why it is important to do this review

Ulcerative colitis is a chronic disease characterized by flares of increased disease activity interposed between periods of disease quiescence. Over the past century the use of effective medical therapies has dramatically reduced the mortality associated with UC (Caprilli 2007). While most patients present with mild disease, approximately 15% of patients with UC will develop an attack severe colitis and 30% of these patients will fail medical therapy and require a colectomy (Caprilli 2007). At 10 years post‐diagnosis the overall colectomy rate for patients with UC is 24% (Langholz 1994). While a colectomy may be considered to be curative, it does not necessarily restore pre‐morbid quality of life. 

There is mounting evidence that MH correlates with improved long‐term outcomes in UC (Froslie 2007; Baert 2010; Armuzzi 2012; Colombel 2012). This review will assess all randomized controlled trial data for interventions that have evaluated MH in UC. Evaluating the ability of UC therapies to maintain MH may be clinically relevant for the management of UC as the ability to alter the natural history of the disease, achieve sustained remission, and improve patient outcomes are the ultimate goals of therapy.

Objectives

The primary objective of this review is to evaluate the efficacy and safety of medical therapies used to maintain mucosal healing in UC.

Methods

Criteria for considering studies for this review

Types of studies

Randomized controlled trials (RCTs) assessing long‐term maintenance of endoscopic remission will be considered for inclusion. 

Types of participants

Participants of all ages previously diagnosed with UC using standard clinical, endoscopic, radiologic, and histologic criteria who are determined to be in endoscopic remission will be eligible for inclusion in this review. Due to the expected heterogeneity in diagnostic criteria and assessment of disease activity, the definitions used by the authors of each trial will be accepted for this review. Eligible trials will include a baseline endoscopic assessment to confirm endoscopically quiescent UC.

Types of interventions

Eligible trials will include an intervention that consist of a medical agent compared to a control therapy, which, can be either a placebo or an alternate medical therapy used for the treatment of UC. We will consider all studies with a minimum treatment duration of 6 months and endoscopic evaluation as a primary or secondary outcome.

Types of outcome measures

Primary outcomes

The primary outcome will be the proportion of participants successfully maintaining MH after at least six‐months follow‐up. This will be expressed as a percentage of the total number of patients randomized to each treatment arm (i.e intention‐to‐treat analysis). The ideal definition of MH would be the complete absence of mucosal abnormalities that were observed at baseline assessment. A number of indices have been developed to assess endoscopic disease activity during colonoscopy or flexible sigmoidoscopy. These include the Baron score (Baron 1964), the modified Baron score (Feagan 2005), the St Mark's index (Powell‐Tuck 1982), the Rachmilewitz endoscopic index (Rachmilewitz 1989), the Mayo score (Schroeder 1987), the Lémann endoscopic index (Lemann 1995), the Sutherland index (Sutherland 1987), the ulcerative colitis endoscopic index of severity (UCEIS) (Travis 2012), and Truelove and Witt’s sigmoidoscopic assessment (Truelove 1955). Generally, a score of “0” using these indices corresponds to normal endoscopic appearance. However, there is inconsistency in the literature with scores of “0” and “0 or 1” often being used to define endoscopic remission. Due to a lack of uniformity in the definition of MH we will accept the author definition of MH as determined by a recognized or validated endoscopic disease activity tool.

Secondary outcomes

Secondary outcomes will include the following:

  1. Mean time to endoscopic relapse;

  2. Adverse events;

  3. Serious adverse events; and

  4. Adverse events requiring withdrawal from the study.

Search methods for identification of studies

Electronic searches

A comprehensive literature search will be performed to identify all relevant RCTs. There will be no language restrictions and translation of articles will be performed as required. Trials published as abstracts only will be considered for inclusion and any additional information required to assess study quality will be sought from the abstract authors.

We will search the following databases from inception to date:

  1. MEDLINE;

  2. EMBASE;

  3. Cochrane Central Register of Controlled Trials (CENTRAL);

  4. Cochrane Inflammatory Bowel Disease and Function Bowel Disorders (IBD/FBD) Group Specialized Register.

  5. Ongoing trials will be identified from the registry http://ClinicalTrials.gov.

The search strategies for each database are reported in Appendix 1, Appendix 2, Appendix 3, and Appendix 4.

Searching other resources

Other search resources will include the following.

  1. Manual search of reference lists of trials and review articles identified by a computer‐assisted search.

  2. Proceedings from major meetings in gastroenterology will be manually searched. This will include the annual meeting of the American Gastroenterology Association, the British Society of Gastroenterology, and the United European Gastroenterology Week, the European Crohn’s and Colitis Organisation meeting, and the annual meeting of the American College of Gastroenterology, and the Canadian Digestive Disease Week, from 2002 to present.

  3. Personal contact with experts in the field as well as representatives of pharmaceutical companies involved in the development of medical therapies for UC.

  4. In addition to the main search strategy we will perform a “grey literature” keyword search using Google Scholar. 

Data collection and analysis

Selection of studies

All studies identified by the electronic searches will be independently reviewed by two authors (MJS and MK). If appraisal of the study title or abstract clearly identifies that the study does not meet the pre‐defined selection criteria it will be excluded from further analysis. Otherwise, the full text article will be retrieved and assessed for inclusion. In the case of disagreement a third author (CHS) will be consulted. If the full text article does not contain sufficient information to assess for inclusion, we will contact the authors of the trial for additional information. The study‐selection process will be outlined using a flow‐chart as suggested by the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) statement (Moher 2009).

Data extraction and management

Data extraction will be completed using a tool specifically designed to capture relevant data relating to the primary and secondary outcomes of this review. Two authors (MJS and MK) will independently extract data and any discrepancies will be resolved through consensus with a third author. Extracted data will be entered into the Review Manager software (RevMan 2012).

We will extract the following data:

  1. General study information including the name of the first author, year of publication, and full title;

  2. Details of the participant population including the total number of patients screened, the number of patients randomized to each study group, the number of patients completing treatment in each study group, the number of dropouts or patients lost to follow‐up in each study group, countries in which the study was performed, years patients were entered into the study;

  3. Participant characteristics including age, sex, disease extent and disease severity; disease duration prior to enrolment;

  4. Methodological information including details of randomization, blinding, allocation concealment, inclusion and exclusion criteria;

  5. Details of the intervention including the medical agent, dose, route of administration, frequency of administration, and duration of treatment;

  6. Details of the control group(s) including the placebo or other medical agent, dose, route of administration, frequency of administration, and duration of treatment;

  7. Concomitant medication allowed and number of patients in each group using concomitant medication; and

  8. Details of study outcomes including the timing of study assessments, the definition of MH used in the trial and the proportion of participants achieving MH in each study arm.  If available, we will also extract information on the endoscopic disease activity scoring tool(s). Authors of studies in which endoscopic activity is assessed but not expressed in a way that allows for the proportion of participants maintaining MH to be calculated (e.g. combined clinical and endoscopic definition of relapse) will be contacted and further information sought.

Assessment of risk of bias in included studies

The quality of included trials will be independently assessed by two study authors (MJS and MK) using the Cochrane risk of bias tool (Higgins 2011a). Discrepancies will be resolved through consensus with a third author (CHS). Study authors will be contacted for further information when insufficient information is provided to determine the risk of bias.

We will assess the following factors:

  1. Sequence generation;

  2. Allocation concealment ;

  3. Blinding of participants, personnel and outcome assessors ;

  4. Incomplete outcome data ;

  5. Selective outcome reporting ; and

  6. Other potential sources of bias .

Each item will be rated as high, low or unclear risk of bias, and a justification from the study report will be supplied to support the judgement as appropriate. Studies will be considered to have a ‘low risk of bias’ if ‘low risk’ is reported for all six factors assessed.  If ‘unclear risk’ is reported for one or more domain than the study will be considered to have an ‘unclear risk of bias’. Studies will be considered to be at ‘high risk of bias’ if ‘high risk’ is reported for one or more of the six factors assessed.

We will use the GRADE approach to assess the overall quality of the body of evidence for the primary outcome and secondary outcomes of interest. Meta‐analyses of randomized trials start as high quality evidence, but may be downgraded due to: (1) risk of bias, (2) indirectness of evidence, (3) inconsistency (unexplained heterogeneity), (4) imprecision (sparse data), and (5) reporting bias (publication bias). The overall quality of evidence for each outcome will be determined after considering each of these elements, and categorized as high quality (i.e. further research is very unlikely to change our confidence in the estimate of effect); moderate quality (i.e. further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate); low quality (i.e. further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate); or very low quality (i.e. we are very uncertain about the estimate) (Guyatt 2008; Schünemann 2011).

Measures of treatment effect

All analyses will be performed using Review Manager 5.2 (RevMan 2012). For primary and secondary outcomes data will be arranged in two by two tables and synthesized into a summary test statistic. A random‐effects model will be used because it provides more conservative estimates than a fixed‐effects model. Studies will be weighted using the DerSimonian and Laird method.

Dichotomous data

We will calculate the risk ratio (RR) and corresponding 95% confidence interval (95% CI) for dichotomous outcomes.

Continuous data

We will calculate the mean difference (MD), or standardized mean difference (SMD), along with corresponding 95% CI for continuous outcomes.

Unit of analysis issues

Cluster randomized trials

Trials that use cluster randomization (i.e. groups of individuals rather than individuals are randomized to different interventions) may be at risk of unit of analysis errors that can lead to inappropriate confidence intervals and weighting of studies. If appropriate, we will perform approximate analyses of cluster‐randomized trials by calculating effective sample sizes (Higgins 2011b).

Cross‐over studies

Trials in which participants are allocated to a sequence of interventions can also result in unit of analysis issues. Due to the nature of the outcomes we will be assessing, we will only include data from the first treatment period for cross‐over studies.

Within‐patient studies

For trials in which there are multiple intervention or control groups, or were repeated observations are made, we will combine study groups to create single pair‐wise comparisons (Higgins 2011b). 

Dealing with missing data

If a published article contains insufficient data for analysis then the original study authors will be contacted to provide further details.  If there is a discrepancy in the number of participants randomized and the number of participants analyzed in each treatment group, the percentage lost to follow‐up will be calculated for each group. For continuous data analyses will be based on those completing the trial. For dichotomous data, if the lost to follow‐up rate exceed 10% for any trial, we will assign the worst case outcome to those lost to follow‐up. The impact of this data imputation will be assessed in sensitivity analysis.

Assessment of heterogeneity

Clinical and methodological heterogeneity will be independently assessed by two authors (MJS and MK). Data will not be pooled for statistical analysis if significant clinical, statistical, or methodological heterogeneity is found. Statistical heterogeneity will be assessed by visual inspection of forest plots to determine if CIs over lap. Statistical heterogeneity will be formally assessed using the Chi2 test. A P value of 0.1 or less will be considered to be statistically significant and to be evidence of significant statistical heterogeneity. We will also evaluate the degree of between study variance using the I2 statistic (Higgins 2002). We will interpret the I2 value as follows:

  • 0% to 40%: might not be important;

  • 30% to 60%: may represent moderate heterogeneity;

  • 50% to 90%: may represent substantial heterogeneity;

  • 75% to 100%: considerable heterogeneity.

Assessment of reporting biases

If greater then 10 trials are available for pooled analysis we will use visual inspection of funnel plot symmetry to assess potential reporting bias.  Asymmetry in the plot may be attributed to publication bias, poor methodology quality, or due to true heterogeneity. We will use Egger’s methods to explore bias (Egger 1997).

Data synthesis

Data will be pooled for analysis if the interventions, outcomes and patient groups are sufficiently similar (to be determined by consensus). Data will not be pooled for meta‐analysis if significant clinical, methodological, or statistical heterogeneity is identified. We will perform a meta‐analysis on the results of included studies using a random‐effects model and the Mantel‐Haenszel method of weighting. We will calculate the pooled RR with corresponding 95% CI for dichotomous outcomes. For continuous outcomes when two or more studies present data derived from the same instrument of evaluation, and with the same units of measurement, the data will be pooled and results will be presented as MD with corresponding 95% CI. Conversely, when primary studies express the same variables through different instruments, and different units of measurement, we will present data as the SMD with corresponding 95% CI.

Subgroup analysis and investigation of heterogeneity

If a sufficient number of trials are included subgroup analysis will be performed based on the age of study subjects (pediatric versus adult); disease severity as determined by a recognized disease activity tool (e.g. mild, moderate, or severe); disease location (e.g. proctitis, left‐sided, or pancolitis); and duration of the intervention.

Sensitivity analysis

If an adequate number of studies are included in this review, we will perform sensitivity analyses to assess the consistency and robustness of the results of the meta‐analyses. Potential sensitivity analyses will include the following.

  1. Excluding studies judged to be at high risk of selection bias.

  2. Excluding studies judged to be at high risk of performance or detection bias.

  3. Excluding studies judged to be at high risk of attrition bias.

  4. Excluding studies published as abstracts.

  5. Excluding studies that utilize different definitions of MH (e.g. scores of "0" versus "0" and "1").