Efficacy and safety of Propionibacterium extract gel versus glyceryl trinitrate ointment in the treatment of chronic anal fissure: a randomized controlled trial

Chronic anal fissure (CAF) is an extremely frequent finding in clinical practice. Several topical agents have been proposed for its treatment with the common goal of increasing anodermal blood flow to promote healing. The aim of this study was to compare the efficacy and safety of a Propionibacterium extract gel (PeG) and 0.4% glyceryl trinitrate ointment (GTN) in patients with CAF.


INTRODUC TI ON
The concept of anal fissure was first introduced by Lockhart-Mummery in 1934 as a linear or oval-shaped tear in the anoderm that can potentially extend from the anal verge to the dentate line [1].
Anal fissures can be categorized as acute or chronic based on their morphology and time of onset, with a 6-week cut-off often used to label anal fissures as chronic. While acute anal fissures typically present as linear lesions with clear margins, chronic anal fissures (CAFs) are typically relatively wider and deeper with granulation tissue at the base and potential exposure of the internal sphincter.
The pathophysiology of CAFs is not yet well understood, with the primum movens still being a topic of debate. The ischaemic theory proposed by Schouten et al. in 1996 [2], highlighting the role of a high resting sphincter tone and decreased anodermal blood flow, especially at the posterior midline, is the most widely accepted.
Although several therapeutic strategies have been proposed to correct the pathophysiological alterations underlying CAFs [i.e. fibre or sitz baths [3], topical nifedipine/lidocaine [4] and glyceryl trinitrate (GTN) ointments [5], and botulinum toxin [6]], a partial lateral internal sphincterotomy is still considered the gold standard despite possible detrimental sequelae, such as soiling and some degree of anal incontinence [7,8]. Conservative medical approaches continue to play an important role in this scenario because of their low cost and safety, with surgery being exclusively recommended after the failure of multiple lines of medical therapy [9]. 0.4% GTN ointment is a well-known nitric oxide donor that promotes CAF healing by decreasing resting anal pressure and increasing anodermal blood flow via the stimulation of intracellular cyclic guanosine monophosphate, resulting in a consequent reduction in cytosolic calcium [10]. The success rate is variable, with 28% of patients experiencing transient headaches; this often leads to drug discontinuation and poor compliance with treatment [7].
Propionibacterium extract gel (PeG) is a topical product that protects skin and mucous membranes from external agents [11,12]. Its film-forming property on the epidermis helps reduce inflammation, itching and pain, while the adjunct of antioxidant ingredients helps promote the healing process [13,14]. This product is currently available only in Italy for this indication.
This study aimed to compare the efficacy and safety of PeG (Emorsan Rag®) and GTN (Rectogesic®) in patients with CAF.

Study design
This was an open-label, randomized parallel-group controlled trial conducted between October 2021 and March 2022 across five high-volume tertiary referral centres for proctological disorders. A power analysis was performed to determine the number of patients enrolled in each arm. The protocol was approved by all local ethics committees in accordance with the Declaration of Helsinki (1996) and the International Conference on Harmonization Good Clinical Practice guidelines. All the patients received full information and provided informed consent for inclusion in the study.

Study population
Consecutive patients diagnosed with CAF in our outpatient clinics and aged 18-75 years were included in this study. Patients with faecal incontinence, other proctological diseases, inflammatory bowel disease, a history of anal surgery or previous or concomitant treatment for anal fissures, a sexually transmitted disease or cancer, who were undergoing immunosuppressive treatment, were pregnant or breastfeeding or had a known allergy to one of the agents contained in the evaluating drugs were excluded. Patients who were unable to return for postoperative follow-up visits or showed an unwillingness to sign the informed consent form were also excluded.
Digital rectal examination (DRE), anoscopy and anorectal manometry were performed at each follow-up visit unless they could not be tolerated by the patient. In cases where it was impossible to perform both a DRE and anoscopy during enrolment (baseline), the CAF site was determined by inspecting the anal region by asking the patient to bear down during defaecation while spreading the glutei.
In cases with a suspicious and unusual location, a colonoscopy was performed to rule out other neoplastic or inflammatory disorders.
After enrolment, patients were followed up for 10 (visit 1), 20 (visit 2) and 40 days (visit 3). The duration of the therapy and last follow-up visit were defined based on the results of the study, where no difference was found between 40 and 80 days in terms of healing rate and pain [15].
In addition to an onset more than 6 weeks previously, the presence of at least two of the following features proposed by Scholefield et al. [16]. were considered when diagnosing CAF: a sentinel skin tag, hypertrophic anal papillae, an exposed internal anal sphincter, a fibrotic lateral fissure or a fibrotic anal sphincter.

Outcome measures
A recently developed and validated five-item score, REALISE, was used from baseline to visit 3 to assess pain (score range 0-10), quality of life, duration of pain, intake of analgesics and bleeding. The latter four items were rated on a scale of 1-5 [17].
The degree of epithelialization of the fissure was determined at each visit and stratified into three levels, corresponding to <50% (i.e. nearly no change from baseline), >50% healing and complete healing.
Itching and burning were assessed using two visual analogue scales (VASs) (minimum score = 0, maximum score = 10) at each time point.
The Bristol stool chart (BST), a seven-point scale, was used to evaluate stool shape and consistency [18] from baseline to visit 3.
Quality of life was evaluated at baseline and visit 3 using Short-Form 12 (SF-12), a 12-item subset of SF-36 that includes both physical (PCS) and mental component scores (MCS) [19,20].

Treatment plan
Patients were instructed to squeeze out approximately 1.5 mg from an aluminium tube containing GTN and apply it to the distal anal canal and perianal area with a gloved finger every 12 h for 40 days, as described elsewhere [22].
In the PeG group, 3 g of gel (3 cm) was applied twice daily for 40 days to the distal anal canal and perianal area. Both groups were instructed not to use any other topical preparations until study completion.
Patients in both groups were encouraged to prevent passing hard stools and constipation by using laxatives (macrogol twice or three times a day) and a recommended oral dose of ketorolac tromethamine (10 mg every 6 h) on an as-needed basis, not exceeding 40 mg per day.
At each follow-up visit, patients were asked about their willingness to continue treatment and the number of tubes/boxes consumed.

Safety
Safety was evaluated by reporting adverse drug effects (ADEs), adverse events (AEs), serious AEs and toxicity after each topical drug application. Toxicity was defined using the World Health Organization toxicity scale [23]. The AEs were stratified as none, remote, possible, probable or not assessable based on their relationship with the drug.

Sample size and randomization
Assuming a compound symmetry covariance structure, a withinpatient autocorrelation of 0.50, a common 20% value for the standard deviation and a 10% noninformative dropout rate, a minimum of 59 patients per arm were required to test an average REALISE score improvement of at least 10% over time under the alternative hypothesis using a repeated measurement design (1β = 0.80, α = 0.05). The expected value under the alternative hypothesis was derived from the literature data using a linear interpolation function. The patients were allocated to either group using a blocked randomization scheme with a fixed block size of six (Table S1).

Statistical analyses
The analyses were conducted using SAS 9.4 according to intention to treat principles. Data are presented as per cent or mean and median, along with standard deviation (SD) and interquartile range (IQR). Differences between treatment arms of categorical variables were tested using Fisher's exact test. Within treatment arms, score changes from baseline were estimated at all visits using the least squares means method and tested for multiple comparison.
Multivariate analysis of variance was performed to test any possible association between some evaluated outcomes and factors such as age, score at baseline, fissure localization, bowel habit and visit. All tests were two-tailed and considered significant at the 5% level.

RE SULTS
The CONSORT diagrams [24] are shown in Figure 1, and the patient characteristics and procedures at the time of enrolment are listed in Table 1. All follow-up visits were completed by 53 and 43 patients in the PeG and GTN groups, respectively. The number of patients included in the analysis of the clinical outcomes at each visit is detailed in Table 2.

Primary outcome
A steady average decrease in the REALISE score over time was observed in both groups (Table 3).
Significant mean decreases of 6.2 and 12.8 points for PeG and 6.2 and 13.7 points for GTN were observed from baseline at visits 1 and 3, respectively. Factors included in the multivariable analysis were not significantly associated with the observed score changes except for the REALISE score at baseline and visit (Table S2).

VAS for burning
A steady average decrease in the VAS burning score over time was observed in both groups (Table 3). Significant mean decreases of 2.5 and 5.2 points for PeG and 2.3 and 5.6 points for the GTN group were obtained from baseline at visits 1 and 3, respectively (p < 0.001). Factors included in the multivariate analysis were not significantly associated with the observed score changes except for the VAS score at baseline and visit (Table S3).

VAS for itching
A steady average decrease in the VAS itching score over time was observed in both groups (Table 3) (Table S4).

SF-12: PCS-12 and MCS-12
There was an increase in the PCS-12 score at visit 3 for both groups.

Satisfaction
There were no statistically significant differences between the two arms in terms of patient satisfaction (p = 0.25) ( Table 5). After pooling TA B L E 3 Within-treatment differences from baseline (Δ) a in the REALISE and VAS scores of both treatment arms.

Adverse events
The frequency distribution for AEs according to treatment arm and visit is shown in

Treatment boxes and cost analysis
The cumulative average number of treatment boxes and costs per patient are shown in Table 7. Although these increased in both groups, a significant difference was observed between the treat- Assuming that prices were fixed at €32.60 and €73.09 per box for PeG and GTN, respectively, the between-treatment cumulative average costs per patient were significantly higher for GTN than those for PeG at each follow-up visit (p < 0.001

Limitations
This study has some limitations that need to be considered. Given Finally, the cost analysis can be considered partial because it only included the cost per patient. The most appropriate methodology for cost assessment would have been a health technology assessment study, but the latter was beyond the scope of the present study.
The greatest strength of this randomized controlled trial was the use of a validated scoring system for anal fissures, which provided a comprehensive evaluation of pain, quality of life, analgesic intake and bleeding.

CON CLUS IONS
Although there was no difference in healing rates between the two treatments, PeG was more cost-effective, within the limits of the economic analysis, and associated with fewer adverse events.
Future prospective, relatively larger, trials with longer follow-up periods are needed.

ACK N O WLE D G E M ENTS
The authors would like to thank Paola Gallon for her support with the statistical analysis. Open Access Funding provided by Universita degli Studi di Roma La Sapienza within the CRUI-CARE Agreement.

FU N D I N G I N FO R M ATI O N
None.

CO N FLI C T O F I NTER E S T S TATEM ENT
All authors declare no personal conflict of interest.

E TH I C S S TATEM ENT
This study was approved by all ethics committees at all study centres and written informed consent was obtained from all patients. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.

I N FO R M ED CO N S ENT
Informed consent was obtained from all individual participants included in the study.

TR I A L R EG I S TR ATI O N
NCT05616455.

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.