Centre of Evidence Based Dermatology
  • Print

Pyoderma Gangrenosum Study (STOP GAP)


Pyoderma gangrenosum is a rare skin disease. It causes painful, rapidly spreading ulcers on the skin, and can take many months to heal. It is often associated with other conditions such as rheumatoid arthritis or inflammatory bowel disease.

Better evidence to help patients and clinicians to make decisions about their treatment is urgently needed. We conducted two studies as part of this project. The first was a randomised controlled trial to compare how effective the two most commonly used tablet treatments for pyoderma gangrenosum are. The second study measured how well people respond to treatments applied to the skin.

The STOP GAP study was led by Professor Anthony Ormerod (University of Aberdeen) and Professor Hywel Williams (University of Nottingham), and was run in collaboration with the Nottingham Clinical Trials Unit, the UK Dermatology Clinical Trials Network, the Robertson Centre for Biostatistics and the Centre for Healthcare Randomised Trials (CHaRT).

Sian Bennett, a STOP GAP trial participant, talks about her experiences of living with pyoderma gangrenosum.

A short video about the results of the STOP GAP trial. 


Key Facts

1. What treatments did the STOP GAP studies test?

The STOP GAP study consisted of two separate clinical studies – a randomised controlled trial of tablet treatments for pyoderma gangrenosum, and a prospective observational study looking at treatment response in patients treated with creams and ointments.

In the STOP GAP randomised controlled trial, the two most commonly used tablet treatments for pyoderma gangrenosum were tested:

  • Prednisolone tablets (0.75mg per kg per day, to a maximum dose of 75mg/day)
  • Ciclosporin tablets (4mg per kg per day, to a maximum dose of 400 mg/day)

In the study of creams and ointments, patients used many different treatments, but the most commonly used ones were potent topical corticosteroids (usually clobetosol propionate 0.05%) or calcineurin inhibitors (tacrolimus 0.03% or 0.1%).


2. How many patients took part?

Fifty three UK hospitals took part in the study. In total, 121 patients were recruited into the randomised controlled trial, and 66 were recruited into the observational study; making these the largest clinical trials of pyoderma gangrenosum ever done.


3. What did the studies involve?

Participants in the randomised controlled trial were randomised to receive either prednisolone tablets or ciclosporin tablets, and were followed up for a period of six months. Participants were seen by a dermatologist at the start of the trial, after 2 weeks, 6 weeks and 6 months (or sooner if the ulcer had healed).

Both participants and their doctors knew which treatment they had received and so improvement in the pyoderma gangrenosum was assessed using digital photographs of the ulcer. This meant that the treatment response could be assessed in an unbiased way by investigators who did not know which treatment the participants had received. The two most important outcomes were velocity of healing at 6 weeks and time taken for the ulcer to heal.

Participants in the observational study were not allocated to specific treatments. If they were not suitable (or unwilling to take) tablet treatments, then their doctor prescribed whatever cream or ointment that was felt to be most suitable for that patient. Participants were then followed up in the same way with hospital clinic visits at 2 weeks, 6 weeks and 6 months.


4. What were the key results?

We found no difference between ciclosporin and prednisolone in the speed of healing over 6 weeks, with very narrow confidence intervals: adjusted mean difference 0.00 cm 2 /day (95% CI: -0.20, 0.21; p = 0.975). Similarly, there was no difference in the median time to healing: 134 days for ciclosporin compared to 112 days for prednisolone (p=0.84). In both groups, fewer than 50% of ulcers had healed by 6 months, and almost 30% of participants had a recurrence of their pyoderma gangrenosum after initial healing.

Approximately two-thirds of patients experienced adverse reactions, with more serious infections occurring in the prednisolone group and more hypertension and renal dysfunction occurring with ciclosporin. Forty (67.8%) of participants in the ciclosporin group and 35 (66%) in the prednisolone group experienced at least one adverse reaction.

Overall, there were no between group differences for any of the collected outcomes measures.

The primary outcome of speed of healing was assessed using digital images of the ulcers and measuring the size of the ulcers at the start of the trial compared to the size after 6 weeks of treatment. This was felt to be the best measure of treatment success for the STOP GAP trial  as it could be assessed by assessors who did not know what treatment the patient had received (which was important for a trial like this in which the patients and doctors knew what treatments the patients received); because its collection at 6 weeks meant that most people could be included in the analysis; and because previous work in venous leg ulcer patients suggests that speed of healing is a good indicator of subsequent healing. Nevertheless, time to healing of the ulcer was also very important and was included as a secondary outcome. 


5. What do the STOP GAP study findings mean?

The STOP GAP trial found no difference between ciclosporin and prednisolone in the speed of healing over 6 weeks and in the median time to healing. In both groups, fewer than 50% of ulcers had healed by 6 months, and almost 30% of participants had a recurrence of their pyoderma gangrenosum after initial healing.

These results are important as they suggest that clinicians can choose to prescribe either ciclosporin or prednisolone for their patients, with the knowledge that they are likely to be of comparable efficacy.

This means that shared treatment decisions can be made on an individual basis—informed by patient preference and the side-effect profiles of the two drugs.

It is reassuring to know that a bigger trial is unlikely to have shown different results. The confidence intervals around the primary outcome were narrow, suggesting that a clinically meaningful difference between the two interventions is unlikely to have been missed in error.

In a blog in the BMJ following the publication of the STOP GAP trial results, Richard Lehman commented:

"This is the kind of trial we yearn to see. No pharma funding, using cheap old drugs, answering an important clinical question, designed with patients, pragmatic in that it allowed dose adjustment but valid in that it used blinded assessment. More of these, oh many more please."




4 December 2015: Putting UK DCTN studies into practice: Blister and STOP GAP Trials. A free meeting giving patients and clinicians the chance to hear about the latest developments in the management of bullous pemphigoid and pyoderma gangrenosum, and discuss the implications of the STOP GAP trial.

28 June 2015: STOP GAP trial results and their implications are featured in an NIHR Signal.

12 June 2015: The results of the STOP GAP trial have now been published in full as an open access paper in the BMJ. There is also an editorial about the trial and a blog by Richard Lehman. See press release.





Trial documents



Telephone: 0115 8844921



We would like to say a heartfelt thank you to everyone who contributed their time and expertise to the STOP GAP Trial (View contributors PDF format).


This website presents independent research funded by the National Institute for Health Research (NIHR) under its Programme Grants for Applied Research Programme (RP-PG-0407-10177). The views expressed in this publication are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.


Centre of Evidence Based Dermatology

The University of Nottingham
King's Meadow Campus
Lenton Lane
Nottingham, NG7 2NR

telephone: +44 (0) 115 823 1048