Consultant surgeon survey

Methods

Survey design and development

A survey was developed to identify the most frequently used interventions for specific clinical scenarios in current PSD practice. As this was a novel survey, it was designed by study collaborators and followed the CHERRIES statement checklist of recommendations.22 The survey included questions on the following: the mean number of primary elective procedures performed annually, factors affecting choice of procedure, treatment choice for recurrent disease presentation and the factors affecting treatment choice for recurrent disease treatment. The survey was piloted to determine the clinical sensibility.

Data collection

To maximise completion rates, the survey could be completed online or on paper. The online survey was hosted on the Research Electronic Data Capture^TM (REDCap) system managed centrally by the University of Sheffield Clinical Trials Research Unit (CTRU). REDCap (Research Electronic Data Capture) is a secure, web-based software platform designed to support data capture for research studies, providing 1) an intuitive interface for validated data capture; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for data integration and interoperability with external sources.23,24 Paper surveys were returned by post or via e-mail. The questionnaires were anonymised at the respondent level.

Sampling

The survey was disseminated via the UK surgical trainee research collaboratives, led jointly by the South Yorkshire Surgical Research Group and the North-West Research Collaborative. Collaborators were asked to deliver the questionnaire to consultant colorectal surgeons in their units. The first point of contact was made through the National Research Collaborative e-mail lists, and electronic contact to local collaborative leads was cascaded locally. The collaborators were asked to circulate the survey locally to three consultants and thereafter return the completed questionnaires to the REDCap system.

Data analysis

All aspects of data management were provided by the CTRU in accordance with their standard operating procedures. The data emanating from this survey were captured and stored in the REDCap software.

Results

The link was followed by 200 surgeons and completed by 113 participants. Of these, 109 routinely cared for patients with PSD. These 109 were entered into the final analysis, giving a final response rate of 54.5%.

Respondent practice overview

Respondents reported a median caseload of 15 patients per year [interquartile range (IQR) 10–20 patients] and indicated that recurrent disease accounted for 20% of overall workload (IQR 10–30%). Of those estimating their recurrence rates (n = 97), 19 (19.5%) were unaware of their recurrence rate, 14 (14.4%) estimated their rate to be <5%, 36 (37.1%) to be in the 6–15% range, and 28 (28.8%) in the 16–30% range.

With regards to hair management, depilation was recommended by 54 (49.5%), laser hair removal by 32 (29.4%), salt baths by 14 (12.8%), shaving by 52 (47.7%) and waxing by 32 (29.4%).

Operative strategies employed

A wide range of treatment strategies were employed by responding surgeons, summarised in Table 1. Excision of disease with wound left open was the most frequently used strategy (71 responses; 65.1%), followed by Karydakis flap (62 responses; 58.1%). Curettage with phenol injection (1 response; 0.9%) and endoscopic pilonidal sinus treatment (EPSiT) (2 responses; 1.8%) were the least frequently performed interventions.

Participants were asked to provide a first-, second, and third-choice preference for their interventions. Karydakis was the first-preference treatment for 24/96 respondents (25.0%), followed by Bascom’s II (n = 18; 18.7%), and curettage and glue (n = 15; 15.5%). For second-preference treatments, local excision with wound left open was the most popular with 21/85 participants (24.7%), followed by local excision with midline closure for 15 (17.6%) and Karydakis procedure for 14 (16.4%). The most popular third-preference treatment was local excision with wound left open (27/32; 84.4%), followed by local excision with midline closure for 12 (37.5%), and Bascom’s II for 7 (21.9%) respondents (see Appendix 3, Figure 12).

Case vignettes

Case vignettes demonstrated heterogeneity across respondents. Case one (recurrent disease) showed a preference for rhomboid flap or ‘other’ procedures (22.6% and 25.5%, respectively). For case two (female with primary disease and cosmesis concerns), preferences turned to favour conservative management (21.6%), followed by excision and primary closure (16.0%) and cleaning/curettage of tracts (14.1%). Case three assessed recurrent disease and requirement for minimal time off work. For this scenario, most respondents opted for conservative management with hair removal (25.4%), followed by curettage of tracts (16.0%). Of note, 15.1% would offer a Karydakis procedure in this setting. Responses are summarised in Table 2.

Training

Surgical training programmes were the key training setting for commonly offered procedures. These included training in wide local excision with wound left open or closed for 59/71 (83.1%) and 36/48 (75.0%) of those offering the procedures, respectively. Similar numbers were seen for Bascom’s I (21/27; 77.7%) and Karydakis procedure (49/62; 79.0%). For some procedures, no formal training was reported by 5–10% of respondents. Courses, observation of colleagues and reference material such as text or video was also variably used. A summary of training experiences is presented in Table 3.

Discussion

Overview

The key finding of this survey is the heterogeneity and variation in practice of consultant colorectal surgeons who treat PSD. It demonstrates a relatively low annual volume of operative procedures (around 15) when compared to other conditions such as colorectal cancer surgery. This number is slightly higher than the median of four cases per year identified through HES.

One in four surgeons perceived they had treatment failure rates approaching 30% in this study. This is somewhat at odds with the published literature, which often claims cure in > 90% of cases.13 This demonstrates dissonance between published reports and real-world experience of clinicians. Conversely, 1 in 10 respondents reported recurrence rates of < 5%, which is concordant with the literature. This gap in outcomes may arise from issues with the quality of the literature, where the often surprisingly high quality of outcomes has been challenged.14 Alternatively, it may reflect a small group of clinicians with a high volume of practice and associated good outcomes.25 This poses three key questions. First, should complex pilonidal disease, or even all pilonidal disease, be managed by a group of high-volume surgeons? This is an approach that is advocated in other aspects of surgery such as rectal cancer and inflammatory bowel disease.26,27 The second question to ask is whether we should improve training opportunities (highlighted here as limited) for colorectal surgeons to improve their skill set. Finally, we should ask whether there is a need for better monitoring of outcomes in PSD, as with registries established for other conditions. The findings of this study largely match those from a previous survey conducted in 2011,28 suggesting little has changed in a decade, making these questions more important to improve care.

Surgeons expressed preferences for some treatments such as excise and leave open. The literature suggests these should be considered outdated as they are associated with significant wound morbidity.15 Sixty-five per cent of surgeons used the leave open technique with healing occurring by secondary intention, and 44% used a midline closure technique. Surgeons expressed a stronger preference for asymmetric closure than when previously assessed,28 in keeping with global trends.29 In contrast, minimally invasive techniques such as EPSiT and pit picking appear to be relatively unpopular treatments. This suggests that surgeons are focused on cure rather than symptomatic relief. This may contrast with stated patient preferences where they are willing to trade a less major procedure in exchange for a higher risk of recurrence.30

The survey does have limitations. Surveys of this nature can present artificial choices and do not permit qualification of answers. The use of vignettes does, however, allow some direct comparison of preferences. The survey may have drawn in experts and enthusiastic practitioners in pilonidal disease. However, the heterogeneity of responses does not reflect unity of thought, and responses are in keeping with published surveys. The response rate of 54.5% should reassure us as to representativeness of the survey.

Implications for policy-makers

This study presents three key actions for policy-makers. First, there is a need to agree a general framework for interventions to standardise pathways of care. This should be supported by best available evidence. Where such evidence is not available, funding should be secured to inform such guidance. Secondly, this is an area with no clear registry or oversight. Policy-makers should consider whether collection of granular data on practice and outcomes might aid initiatives to improve care, or even justify specialisation or centralisation of practice. Finally, surgeons’ practice is driven by their postgraduate training and persists into their independence. Therefore, it is important to offer opportunities for further training in new techniques. This would allow implementation of new techniques and may support the decommissioning of outdated procedures.

Implications for researchers

The level of heterogeneity likely speaks in part to uncertainty. Researchers should consider whether the findings presented here would support the delivery of specific procedures in a head-to-head RCT, or whether a ‘bucket’ approach would be more pragmatic.

Conclusion

This survey demonstrates significant heterogeneity in surgeon practice preference. It suggests that limited access to training opportunities may impede efforts to improve practice in the area.