Efficacy of Permacol injection for perianal fistulas in a tertiary referral population: poor outcome in patients with complex fistulas

Abstract Aim Injection of Permacol collagen paste can be used as a sphincter‐sparing treatment for perianal fistulas. In a tertiary referral population we aimed to evaluate the efficacy of Permacol injection and the clinical and fistula‐related factors associated with recurrence. Method This was a retrospective analysis of consecutive patients with perianal fistulas treated with Permacol injection at a specialist centre between June 2015 and April 2019. Endoanal ultrasonography was systematically reanalysed, blinded to treatment outcome. Rectovaginal, anovaginal and Crohn's disease fistulas were excluded. Healed fistulas were defined as absent anal symptoms and a closed external opening on physical examination at a minimum follow‐up of 6 months. Regression analyses were performed to identify factors associated with unhealed fistulas. Results A total of 90 patients (51 men; median age 45 years) were analysed. Seventy‐two (80.0%) patients had complex perianal fistulas (greater than one‐third sphincter involvement or multiple tracts). After a single Permacol injection, fistulas were healed in 20 (22.2%) patients at 3 months follow‐up and in 18 (20.0%) patients at a median follow‐up of 30 months (interquartile range 17–37). Eight (11.1%) patients with unhealed fistulas had significant improvement in their symptoms. Complex fistulas were significantly associated with unhealed status [OR 3.53 (95% CI 1.12–11.09); p = 0.031]. Twenty patients underwent subsequent Permacol injections, which were successful in six (30.0%) patients after one (n = 3) or two (n = 3) additional injections. Conclusion This largest study to date in patients with mainly complex perianal fistulas, demonstrated that the efficacy of a single Permacol injection was only 20%. Complex fistulas were associated with a poor outcome.


INTRODUC TI ON
Perianal fistulas are defined as abnormal tracts between the anorectum and the perineum. A variety of concepts describing their underlying pathophysiology have been described [1][2][3][4]. The incidence of fistula-in-ano in European countries is estimated to be between 1.2 and 2.8 per 10 000 per annum [5].
Symptoms from perianal fistulas may have a profound impact on a patient's quality of life [6], and the majority of patients require surgery to eradicate the primary tract. In low fistulas, fistulotomy is the most commonly performed procedure and is considered the gold standard by most surgeons [7], resulting in low recurrence rates [8,9]. Nevertheless, such interventions may jeopardize anal sphincter function in patients with high or complex fistulas, as a larger part of sphincter muscle is involved [10,11]. Indeed, several sphinctersparing techniques have been developed as alternative treatment options [12][13][14][15][16]. Biomaterials have also been shown to have a limited impact on anal function, although recurrence rates remain significant and do not improve upon traditional surgical techniques [17,18].
Permacol is a recently introduced biomaterial in the treatment algorithm for perianal fistulas, consisting of an acellular crosslinked porcine dermal collagen matrix suspension. It is injected into the fistula tract, enabling the fistula to heal without damaging the anal sphincter complex. The MASERATI100 study showed a healing rate of 53.5% in a large population of patients with primary or recurrent fistulas at 12 months follow-up [19]. To date, only a small number of other studies have been published on the efficacy of Permacol, all suggesting similar healing rates [20][21][22][23][24].
Nonetheless, outcome data in complex anal fistulas are scarce, and the available evidence is based on studies which are limited in sample size (i.e. fewer than 50 patients) [21,22,24]. Hence, we analysed a series of consecutive patients referred to a specialist centre for treatment of mainly complex perianal fistulas with the aim of evaluating the efficacy of Permacol injection and the clinical and fistula-related factors associated with unhealed fistula status after Permacol injection.

Study sample
A retrospective review of medical records was performed for consecutive patients (aged ≥18 years) who underwent perianal fistula treatment with Permacol injection at a tertiary referral centre Follow-up of at least 6 months after Permacol injection was required for study inclusion.

Three-dimensional endoanal ultrasonography and fistula characterization
Three-dimensional images of the anal canal were acquired by endoanal ultrasonography with a 10-16 MHz transducer (Hawktype 2050, B-K Medical, Naerum, Denmark). Three per cent hydrogen peroxide was introduced into the external opening of the fistula using an intravenous cannula in order to visualize the fistula tract. To reduce reporting bias, all endoanal ultrasonography recordings were systematically reanalysed by a single senior investigator with over 30 years' experience (RFB) who was blinded to preoperative assessment and treatment outcome. Fistulas were classified according to the Parks classification [2]. Transsphincteric fistulas were subclassified as being low (lower third of the sphincter complex), mid (middle third of the sphincter complex) or high (upper third of the sphincter complex). Other systematically collected fistula characteristics were the number of tracts, height (in mm) of the internal opening measured from the anal verge and the distance (in mm) of the furthest external opening measured from the anodermal junction. All fistulas were further classified as being simple (low, and single tract) or complex (mid or high, or multiple tracts). Finally, patients were classified as presenting with a primary or recurrent fistula, with the latter characterized as those with a previously failed surgical intervention.

Surgical procedure
Prior to the procedure a Microlax enema was used, and a combination of 1 g cefazoline and 500 mg metronidazole was administered intravenously. The patient was placed in the lithotomy position and the internal fistula opening was identified using a Czerny forceps.
If a seton was present it was removed. The fistula tract was debrided with a curette, followed by rinsing of the fistula tract with 0.9% NaCl. If required, the internal opening was debrided from hypertrophic tissue. The internal opening was sutured using Vicryl 2.0 full-thickness sutures without tying the sutures. A flexible cannula sheath was introduced in the external opening and the Permacol™ collagen paste (Medtronic) was installed in the fistula tract until the

What does this paper add to the literature?
In this largest study to date to report on Permacol injection for mainly complex perianal fistulas, we found a healing rate of 20.0% at a median follow-up of 30 months. In contrast to previous studies in less complex fistulas, these results suggest that Permacol injection is not effective in complex fistulas.
paste was visualized at the internal opening. The sutures were tied when the tract was completely filled, i.e. when the paste was visible at both the internal and external openings. Debridement of the external opening was performed and all hypertrophic tissue was excised. The external opening was approximated loosely with one suture, so that the paste stayed inside the fistula tract. A sterile dressing was applied.

Outcome measures
The efficacy of a single Permacol injection was evaluated in all patients at (a) the standard clinic appointment 3 months after surgery

Perioperative data
Prior to Permacol injection, seton drainage was performed in 56

DISCUSS ION
In 90 consecutive patients undergoing Permacol injection for perianal fistulas, we found a healing rate of 22.2% (n = 20) at 3 months follow-up and 20.0% (n = 18) at a median follow-up of 30 months.
Complex fistulas were associated with a success rate of only 15.3%, while fistula healing was achieved in 38.9% of patients with a simple fistula. This is the largest study to date reporting on Permacol injection for mainly complex fistulas, with the longest follow-up period.
The outcome of Permacol injection in our cohort was poor compared with other studies (summary of previous studies with a minimum of ≥10 patients; Table 3) [19][20][21][22][23][24]. The proportion of patients with healed fistulas was almost three-fold lower than in the MASERATI100 study at 12 months follow-up (56.7%) [19]. Their definition of fistula healing was identical to that in our study (absence of anal symptoms and a closed external opening confirmed on physical examination), making it possible to compare like with like.
However, the proportion of patients with a fistula with over onethird sphincter involvement in our study was almost two-fold higher than in the MASERATI100 study (66.3% vs. 35.6%). Indeed, a shorter fistula tract, which usually corresponds with less sphincter involvement, was associated with favourable outcome in the MASERATI100 study. Additionally, 18 (20.0%) patients in our study had multiple tracts, which might explain the poor outcome of our study as they are generally associated with recurrence [25], although such fistulas were excluded in the MASERATI100 study [19]. Nonetheless, our population reflects daily clinical practice in a tertiary care setting, which usually consists of a higher proportion of patients with complex fistulas.
Other studies reported efficacy rates much like the MASERATI100 study, with a range between 47.6% and 57.0% (Table 3) [21][22][23][24]. Only one study reported a much higher healing rate of 77.4% [20]. Nevertheless, patients in our cohort were more likely to have recurrent fistulas, a higher internal opening or multiple tracts and the follow-up period was longer than in the other studies, which might explain the difference in proportions of fistula healing. Recently, consensus groups have sought to develop core outcome sets for the evaluation of cryptoglandular [27] and Crohn's disease fistulas [28] in order to involve all stakeholders (researchers, patients and other healthcare professionals) in the measurement of treatment outcome. Indeed, once validated, these measurements will be likely to improve the evaluation of fistula surgery in future studies.
Many studies evaluating surgical techniques in perianal fistulas suggest high initial success rates, although this tends to decrease

CON CLUS ION
This study in patients with mainly complex perianal fistulas demonstrated that the efficacy of a single Permacol injection was only 20%.
Complex fistulas (more one third of sphincter complex involvement or multiple tracts) were associated with a poor outcome. Permacol injection should not be considered a treatment option in complex fistulas.

ACK N OWLED G EM ENTS
None.

CO N FLI C T O F I NTE R E S T
PFV, GJVM, CBHM and RJFF-B have no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
All authors conceived the study design. PFV collected the study data, analysed the data and wrote the manuscript. All authors edited the manuscript and approved the final version.

AUTH O R CO NTR I B UTI O N S
All authors conceived the study design. Paul F vollebregt collected the study data, analysed the data and wrote the manuscript. All authors edited the manuscript and approved the final version.

E TH I C A L A PPROVA L
The study was approved by the Medical Ethical Committee of the VU University Medical Centre (reference number 2013/21).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.