An interview with Dr Ally Speight, consultant gastroenterologist at Newcastle upon Tyne NHS Trust, UK, on managing fistulating perianal Crohn's disease as part of our monthly topic, IBD.
Could you tell us what type of patients you might see in your specialist practice and the clinical challenge that these patients pose?
Thank you for talking to me today about an area of Crohn's disease management that is currently under-researched and is one of the hardest clinical manifestations of the disease to manage. The James Lind Alliance recently identified this as one of the top 10 priorities for research in IBD, and so this area is clearly important to both clinicians and patient stakeholders.
Perianal fistulising Crohn's disease (pCD) is relatively common developing in about 20% of Crohn's patients, with recurrent disease subsequently in about a third of these. Patients with colonic and especially active proctitis are those that are more likely to develop pCD. It’s also important to recognise that the development of pCD is a negative prognostic indicator and predicts poor overall long-term outcomes. These are often young patients with significant morbidity including perianal sepsis, pelvic or perianal pain, active pus drainage, faecal incontinence, sexual dysfunction and fertility issues. In severe, refractory cases patients may end up requiring a defunctioning stoma and/or proctectomy, with the inherent risks and psychosocial issues that this kind of surgery may include.
How might a patient present and what initial investigations would you suggest?
Patients may present either with a known diagnosis of Crohn's disease (CD) or as their index presentation. When patients present de-novo with a perianal abscess, the diagnosis of CD should always be considered. Chronicity, family history, the complexity of the perianal disease and active proctitis are obvious clues as to the aetiology. Contrary to this, I have been referred patients with known Crohn's disease, who have struggled for years before being diagnosed. Patients may not disclose their perianal symptoms and so, as ever, a careful history asking specifically about “tail-end problems” (e.g. proctalgia, discharging pus, ‘openings’, incontinence etc.) is important.
Pelvic MRI is the investigation of choice, allowing a highly accurate and non-invasive diagnosis. Clearly, if a patient is septic and/or a perianal abscess is suspected, then examination under anaesthesia (EUA) should not be delayed pending the MRI. EUA, performed by an experienced, if possible, specialist IBD/colorectal surgeon, will provide an accurate diagnosis with the added benefit of allowing therapeutic intervention for abscess drainage and the insertion of a loose-fitting, non-cutting seton. In practice, an MRI to map and classify disease followed by EUA is often the usual approach. Video-assisted fistulography to help map complex fistula tracts and side branches is practiced in some large centres, but is not currently widely available.
Flexible sigmoidoscopy, with biopsies, at the time of EUA is recommended to assess for luminal disease, the presence of internal openings, rectal strictures or even cancer. The formation of fistulas, like all Crohn’s pathophysiology, is poorly understood. However, penetrating inflammation at areas of high intraluminal pressure (e.g. proximal to a stricture) would seem to be important.
How is pCD classified and how does this differ from cryptoglandular (non-Crohn’s) related perianal fistula?
The obvious and most familiar classification is anatomical, using either that described by Parks in 1976 or the American Gastroenterological Association’s ‘simple’ and ‘complex’ descriptors. The Parks’ classification describes the course of the fistula tract as either superficial, intersphincteric, trans-sphincteric, suprasphincteric or extrasphincteric. It can be further modified with the description of secondary tracts (infralevator, supralevator, horizontal or horsehoe) and the location of abscesses (perianal, intersphincteric, ischiorectal, suprasphinteric). [For a good diagram, and indeed overall review of the subject, see Gecse et al, Gut 2014]
Missing from the anatomical classification, of course, is a description of disease activity. This can be done clinically (quality of life score, pain and induration, discharging fistula) and/or radiologically, with MRI based scoring systems described in the literature but not widely adopted in clinical practice.
The anatomical classification is not different for Crohn's disease versus cryptoglandular disease, although obviously, complex fistulas are more common in Crohn's. The pathophysiology is of course different in Crohn's and the detail is all in the clinical presentation.
On presentation, what should the initial management be? How soon should I involve gastroenterologists?
In terms of management, the first point I would like to make is that there is a large degree of heterogeneity in management, certainly in the UK, and I would suspect across Europe. This is really an area where research, education and improvement in practice could improve outcomes.
On initial presentation, antibiotics such as metronidazole or ciprofloxacin should be considered, especially if there is the clinical suggestion of pelvic sepsis. There is evidence to suggest that both of these antibiotics will improve symptoms, but may also contribute to healing. EUA, with incision and drainage of any abscesses shouldn’t be delayed for an MRI, but an MRI prior is useful if this is readily available.
Systemic steroids are not routinely recommended, as they may worsen pCD and precipitate abscess formation. Steroids may become a necessary evil for induction of active luminal CD, but should be avoided if possible.
I would like to be informed as near to diagnosis as possible. A close-knit, collaborative approach is vital. At this stage, consideration should already be given to a pre-immunosuppression screen (e.g. latent TB, prior viral exposure, thiopurine metabolism). Multimodal therapy, that is combined medical and surgical treatment, is the best way to improve outcomes.
Tell us more about the evidence for multimodal therapy? When should the seton come out? What predicts treatment failure?
Multimodal treatment, with drainage of sepsis and then early antiTNF, is associated with improved outcomes. A recent retrospective study demonstrated an association between fistula healing and multimodal therapy, over surgical or medical management alone. With good combined management, healing rates of about 50% are achieved. Clearly there is still work to be done to improve this further.
In terms of medical therapy, the best evidence is for infliximab. Our practice is to optimise dosing to the upper limit of the therapeutic range, aiming for a trough level of 6 to 7 microg/ml (higher serum trough concentrations are associated with fistula healing). The co-prescription of immunomodulator (such as azathioprine or methotrexate) helps the efficacy of antiTNF therapy, reducing the rates of immunogenicity and loss of response to antiTNF.
Factors that are associated with treatment failure are complex fistula, active proctitis and anal stenosis.
Increasingly, the Newcastle IBD service has moved towards the early removal of seton(s). Whilst leaving a seton in-situ gives confidence that sepsis can drain, it also impairs healing of the fistula. In our unit, we aim to down stage and then remove any seton(s) during induction with antiTNF.
What are the common mistakes you encounter in the management of these patients?
There are three main areas;
- The inadequate drainage of sepsis. Several EUAs may be required to gain adequate drainage and, in patients who are failing to respond, repeating the EUA should always be considered. Timely MRI is important in assessing for undrained sepsis. Also, surgical experience is vital at EUA when exploring fistulous tracts and trying to avoid the formation of false tracts.
- A delay in starting antiTNF. The whole clinical pathway from presentation, EUA and then medical management can sometimes take months. A simple service improvement would be to take a look at the length of your own clinical pathway. Consider pre-booking theatre/MRI/infusion unit time in order to shorten the duration from sepsis drainage to first dose of antiTNF. We know that in luminal CD, early treatment reduces disease complications and I suspect that similar is true for pCD.
- Inadequate control of luminal disease. Active proctitis predicts poor outcomes for perianal disease. Optimising medical therapy (combined immunosuppression, therapeutic drug monitoring), smoking cessation and even temporary defunctioning should be considered.
What new treatments are on the horizon?
The injection of mesenchymal stem cell (MSC) therapy is the potential new kid on the block. Yet to be licensed in Europe, initial RCT trial results are relatively promising; 50% of fistulas healed up with this surgical approach, with the potential for higher rates of healing, perhaps, if combined with antiTNF treatment. There was an impressive ‘placebo’ response, with a 34% healing rate in the sham treatment group, presumably due to the beneficial effects of closing the internal orifice and curetting the tract, procedures performed in both arms of the study. Caution is clearly required when interpreting the results of this published data as it excluded patients with active proctitis, strictures or rectovaginal fistulas and so isn’t quite applicable to most of the patients I see regularly in the clinic, and there have been many false starts with overly optimistic rates achieved with some techniques in specialist centres, which have failed to reach repetition in the general setting.
It is likely also that the stem cells may well prove to be prohibitively expensive, although the surgical technique is relatively straight forward.
Different modes of mesenchymal stem cell delivery are being studied, with promising preliminary results using stem cell coated, fibrin plugs inserted into fistulas as a scaffold for the cells. However, results of early phase 1 studies are not due to be reported until 2019.
Dr R A Speight
MA (Cantab) MBBS (hons) MD MRCP (Gastro)
Ally Speight is a consultant gastroenterologist at the Newcastle upon Tyne NHS Trust. He is a graduate of Girton College, Cambridge (Natural Sciences, Genetics) and completed his MBBS, with honours, at Newcastle University in 2005. He completed specialty training in gastroenterology in North East England.
He subspecialises in IBD and was awarded an MD by Newcastle University for his research in metabolic signalling pathways in Crohn's disease in 2015. He has a clinical interest in IBD, therapeutic drug monitoring, coeliac disease and GI complications from oncology treatments. His research interests include therapeutics in IBD and immune oncology associated colitis.
He is a PI on several commercial trials and investigator lead, portfolio studies in IBD. He was awarded an NIHR Greenshoots award in 2017.
He is a member of the BSG IBD research group and is currently contributing to the BSG position statement on immune inhibitor checkpoint colitis.