Ligation of the intersphincteric fistula tract procedure and its modifications

Procedimento de ligadura de trato de fístula interesfincteriana e suas modificações

ABSTRACT

Purpose:

Treatment of anal fistulae is regarded as a challenge due to the diverse nature of this disease and its countless complications. Ligation of the intersphincteric fistula tract procedure and its modifications have been popularized among many surgeons worldwide due to their simplicity and promising outcomes. The main purpose of this article was to conduct a comprehensives review of the published literature on ligation of the intersphincteric fistula tract procedure and its modifications.

Method:

PubMed, the Cochrane database and Ovid were searched from January 2007 to June 2017. Fully published peer-reviewed studies which applied ligation of the intersphincteric fistula tract procedure and its modifications for the treatment of anal fistulae of cryptogenic origin with follow-up of median 12 months were eligible. Uncompleted studies, case reports, reviews, abstracts, letters, short communication, comments, and studies which did not fulfill inclusion criteria were excluded. The primary outcome was to measure primary healing, overall healing, failure, and recurrence of ligation of the intersphincteric fistula tract procedure and its modifications.

Results:

Twenty-two studies were identified with only ten studies meeting criteria of inclusion. Original ligation of the intersphincteric fistula tract was performed in five studies with a population of 199 patients while the remaining five studies showed four different modifications of the ligation of the intersphincteric fistula tract with a total number of 147 patients. Both original LIFT and its modifications have promising as well as potentially similar outcomes; primary healing in the original ligation of the intersphincteric fistula tract (73.95%) (95% CI 60.3-85.6) performed less than the modifications (82.3%) (95% CI 64.8-94.7). Overall healing in the original ligation of the intersphincteric fistula tract (78.9%) (95% CI 58.5-93.7) performed relatively less than in the modifications (93.6%) (95% CI 81.4-99.6). Failure in the original ligation of the intersphincteric fistula tract (17.9%) (95% CI 4.9-36.5) performed almost the same as the modifications (17.7%) (95% CI 5.3-35.2). Recurrence in the original ligation of the intersphincteric fistula tract was 9.7% (95% CI 1.7-23.2). However, there was no recurrence in the modifications.

Conclusion:

Ligation of the intersphincteric fistula tract and its modifications are effective and simple procedures in treating simple anal fistulae, especially high transsphincteric ones. However, more trials should be performed to evaluate its effectiveness regarding complex fistulae.

Keywords:
Ligation of intersphincteric fistula tract; Fistula; Fistula-in-ano; Anal fistula; Review

RESUMO

Objetivo:

O tratamento de fístulas anais é considerado um desafio devido à natureza diversa dessa doença e suas incontáveis complicações. O procedimento de ligadura do trato da fístula interesfincteriana e suas modificações foi popularizado entre cirurgiões em todo o mundo devido a sua simplicidade e desfechos promissores. O principal objetivo deste artigo foi conduzir uma revisão abrangente da literatura publicada sobre o procedimento de ligadura do trato da fístula interesfincteriana e suas modificações.

Método:

as bases de dados PubMed, Cochrane e Ovid foram pesquisadas de janeiro de 2007 a junho de 2017. Estudos publicados com revisão por pares que aplicaram o procedimento de ligadura do trato da fístula interesfincteriana e suas modificações para o tratamento de fístulas anais de origem criptogênica com acompanhamento de mediana de 12 meses foram elegíveis. Estudos incompletos, relatos de casos, revisões, resumos, cartas, comunicação breve, comentários e estudos que não preenchiam os critérios de inclusão foram excluídos. O desfecho primário foi medir a cicatrização primária, a cicatrização geral, falhas e recorrência do procedimento de ligadura do trato da fístula interesfincteriana e suas modificações.

Resultados:

Vinte e dois estudos foram identificados com apenas dez estudos atendendo aos critérios de inclusão. A ligadura original do trato da fístula interesfincteriana foi realizada em cinco estudos com uma população de 199 pacientes, enquanto os cinco estudos restantes apresentaram quatro modificações diferentes da ligadura do trato da fístula interesfincteriana com um total de 147 pacientes. Tanto o LIFT original quanto suas modificações têm resultados promissores e desfechos potencialmente semelhantes; cicatrização primária na ligadura original do trato da fístula interesfincteriana de 73,95% (IC 95% 60,3-85,6) menos realizada que as modificações de 82,3% (IC 95% 64,8-94,7). Cicatrização geral na ligadura original do trato da fístula interesfincteriana de 78,9% (IC 95% 58,5-93,7) realizada relativamente menos do que as modificações (93,6%, IC 95% 81,4-99,6). A falha na ligadura original do trato da fístula interesfincteriana (17,9%; IC 95% 4,9-36,5) realizada quase tanto quanto as modificações (17,7%; IC 95% 5,3-35,2). Recidiva na ligadura original do trato da fístula interesfincteriana em 9,7% (IC 95% 1,7-23,2). No entanto, não houve recorrência nas modificações.

Conclusão:

A ligadura do trato da fístula interesfincteriana e suas modificações são procedimentos eficazes e simples no tratamento de fístulas anais simples, especialmente as transesfincterianas altas. No entanto, mais estudos devem ser realizados para avaliar sua eficácia em relação às fístulas complexas.

Palavras-chave:
Ligação do trato da fístula interesfincteriana; Fístula; Fístula no ânus; Fístula anal; Revisão

Introduction

Anal fistula, fistula-in-ano, or perianal fistula is a hollow tract lined with granulation tissue, connecting a primary opening inside the anal canal or rectum to a secondary opening in the perianal skin.¹

Over the centuries, the probing of the fistula tract has been the procedure of choice for final identification of its anatomy and planning the treatment. Laying open a fistulous tract (fistulotomy) is the treatment of choice.² However, this treatment can become challenging when a larger portion of the sphincter muscle is involved. Use of Seton is a traditionally favored method for treating high fistulae to minimize the incontinence problem. More surgical procedures in the form of fibrin glue, Anal Fistula Plug (AFP), Anorectal Advancement Flap (ARAF), and Ligation of the Intersphincteric Tract (LIFT) have been introduced and have met a wide recognition.²

In 2007, Rojanasakul, a Thai colorectal surgeon, and his coworkers presented a new novel approach called “ligation of Intersphincteric Fistula Tract” or LIFT procedure. The procedure included identification and exposure of the intersphincteric tract; then, the tract was ligated and divided. The main idea of LIFT is that ligation and excision of the intersphincteric tract could block the entrance for fecal particles into the tract, therefore, eliminating the intersphincteric sepsis.³

Although its simplicity and high success rates in its initial results, many recent studies had recorded lower healing rates and higher recurrences; this was mainly due to nonstandardized operative techniques and nonstandardized inclusive criteria of patients.

Many researchers tried to add modification to the LIFT technique targeting better results, insertion of a bioprosthetic in the intersphincteric plane to reinforce the closure of the fistula tract BioLIFT,⁴ and adding a transanal advancement flap to evaluate the effect of an additional ligation of the fistula tract on the outcome of transanal advancement flap repair.⁵ LIFT with partial coreout fistulectomy (LIFT plus)⁶ and insertion of bioprosthetic plug (LIFT Plug).⁷

The main aim of this review is to outline healing rates, primary and overall of the original LIFT procedure and compare them with the healing rates of LIFT modification described in literatures. Failure, recurrence rates, incontinence status, and other main complications were also outlined. Complications included wound dehiscence, hematoma, secondary bleed, and purulent discharge at the intersphincteric wound.

Methods

This study was a systematic review and meta-analysis of ligation of intersphincteric fistula track or LIFT procedure as a management of perianal fistula. This review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.⁸

Search strategy

Two independent reviewers searched PubMed (from January 2007 to June 2017), the Cochrane database and Ovid (January 2007 to June 2017). The keywords used were “fistula, fistulas, fistula-in-ano, perianal fistula, anal fistula, ligation of intersphincteric fistula tract, LIFT procedure and LIFT”. Searches were limited to English articles.

The abstracts of all potentially relevant studies were examined to identify suitable studies for inclusion. The full texts of all eligible articles were obtained. Additional searching of the cited references of selected studies for unidentified articles was also done.

Inclusion criteria

All randomized/non-randomized, controlled/non-controlled clinical trials, retrospective and prospective studies involving the procedure, and its modification were eligible. Full original peer-reviewed studies and fully published in English were eligible. The trials had to have a minimal median follow-up of one year.

Exclusion criteria

Case reports, reviews, abstracts, letters, short communication, and comments were excluded. Incomplete studies or studies with primary results were excluded. Studies with median follow-up less than 12 months were excluded. Studies which include fistulas of noncryptogenic origin were excluded. Studies which did not have a clear view of not including fistulas in HIV patients, tuberculosis, inflammatory bowl diseases, and recto-vaginal origin were excluded. Studies with unclear results or results favoring the procedure were excluded to eliminate the fall in the trap of authors' bias.

Study selection and study quality

Two independent reviewers identified the relevant studies. All of the abstracts were initially reviewed for potential inclusion; selected articles were then reviewed, categorized as meeting or not meeting the inclusion criteria, and scored according to the level of evidence using the Oxford Centre for Evidence-Based Medicine 2011 Levels of Evidence system.⁹

Data extraction

Data were extracted from the included articles by an independent reviewer. Data included; authors, period of study, year of publication, study design, operative technique, patient criteria and number, types and complexity of fistula, and all data related to the defined in the outcomes.

Data analysis

Meta-analysis was done using the open source software “OpenMetaAnalyst”.¹⁰ The outcome studies are proportions of primary healing, overall healing, failure and recurrence in both original LIFT and modified LIFT. The last group included more than one modification of the operation. The program output included the pooled estimate with the 95% confidence interval for both the fixed effect and random effect models. A test of homogeneity of the outcomes was performed and when proved statistically significant the random effect model estimates should be used, otherwise the fixed effect model is satisfactory.

Search outcome

The literature search identified 208 results for further reviewing. After careful reviewing of articles, a total of 10 studies were included, according to the review criteria for analysis (Fig. 1).

Fig. 1
Flow diagram of the search strategy.

Results

After reviewing of operative details of the selected 10 studies (Table 1), we concluded original LIFT in 5 studies and 4 modifications in the remaining 5 studies; two studies included LIFT with partial coreout fistulectomy, and the remaining 3 studies included LIFT with partial coreout with intraoperative Seton, LIFT with partial coreout with advancement flap, and the LIFT-plug.

Table 1
Original LIFT data by author, type of study, procedure, type of fistula, etc.

Author	Period of study	Year published	Type of study	Procedure (n)	Type of fistula (%)	1ry healing (%)	Follow-up, months (n)	Continence evaluation	OC-EBM level
Tan11	April 2006-July 2011	2012	R	Original LIFT (24) vs. ERAF (31)	Transsphincteric (100%)	62.5%	13.0	Clinical	4
Mushaya12	December 2007-February 2011	2012	RCT	Original LIFT (25) vs. ARAF (14)	Transsphincteric (100%)	68.0%	19.2	CCF-FI	2
Madbouly13	July 2011-February 2013	2014	P	Original LIFT (33) vs. MAF (35)	Transsphincteric (100%)	94.2%	12	WIS	3
Wallin14	March 2007-September 2011	2012	R	Original LIFT (93)	Transsphincteric (77%), complex (16%)	66%	19	CCF-FI	4
Dalbem15	May 2012-September 2013	2014	P	Original LIFT (22)	Transsphincteric (100%)	77%	14	Clinical	4

• R, retrospective; RCT, randomized controlled; P, prospective; LIFT, ligation of intersphincteric tract; ERAF, endorectal advancement flap; ARAF, anorectal advancement flap; MAF, mucosal advancement flap; CCF-FI, Cleveland Clinic Florida Fecal Incontinence score; NR, not reported; WIS, Wexner Incontinence Score; OCEBM, Oxford Centre for Evidence-based.

Original LIFT outcomes

Regarding original LIFT, a total of 5 studies were included with a population of 199 patients (Fig. 2 and Table 1),^11-15 all of cryptoglandular origin and mostly transsphincteric.

Fig. 2
Population of original LIFT: (A) type of fistula; (B) gender.

The pooled data of primary healing were 73.95% (95% CI 60.3-85.6) (Table 2), median healing time was 3.8 to 4 weeks when reported, and median follow-up period varied from 12 months to 19.2 months.

Table 2
Meta-analysis: proportion of primary healing in original lift.

Study	Sample size	Proportion (%)	95% CI	Weight (%)
				Fixed	Random
Tan et al. (2012)	24	62.5	40.594-81.201	12.25	18.46
Mushaya et al. (2012)	25	68	46.500-85.050	12.75	18.7
Wallin et al. (2012)	93	65.591	55.021-75.139	46.08	24.37
Dalbem et al. (2014)	22	77.273	54.630-92.179	11.27	17.94
Madbouly et al. (2014)	35	94.286	80.843-99.300	17.65	20.53
Total (fixed effects)	199	72.414	65.739-78.426	100	100
Total (random effects)	199	73.95	60.255-85.605	100	100
Test for heterogeneity
Q					15.7797
DF					4
Significance level					p = 0.0033
I2 (inconsistency)					74.65%
95% CI for I2					37.33-89.75

On the other hand, overall healing rates of the pooled data were 78.9% (95% CI 58.5-93.7) (Table 3). Overall healing rates increased in 2 studies: 92% (95% CI 73.969-99.016) and 100% (95% CI 84.56-100.0).^13,15 However, overall healing rates decreased in only 2 studies: 56% (95% CI 45.238-66.20) and 74.3% (95% CI 56.744-87.511).^14,15 Rates did not have any change in one study regarding overall healing (62.5%) (95% CI 40.594-81.201).¹¹

Table 3
Meta-analysis: proportion of overall healing in original LIFT.

Study	Sample size	Proportion (%)	95% CI	Weight (%)
				Fixed	Random
Tan et al. (2012)	24	62.5	40.594-81.201	12.25	19.36
Mushaya et al. (2012)	25	92	73.969-99.016	12.75	19.47
Wallin et al. (2012)	93	55.914	45.238-66.203	46.08	21.76
Dalbem et al. (2014)	22	100	84.563-100.000	11.27	19.11
Madbouly et al. (2014)	35	74.286	56.744-87.511	17.65	20.29
Total (fixed effects)	199	71.489	64.766-77.575	100	100
Total (random effects)	199	78.893	58.498-93.652	100	100
Test for heterogeneity
Q					36.9388
DF					4
Significance level					p = 0.0001
I2 (inconsistency)					89.17%
95% CI for I2					77.48-94.79

Failure rates from the pooled data were 17.9% (95% CI 4.9-36.5) (Table 4). Causes and management of failure were only mentioned in 3 studies^11,13,15 (Table 6). However, it is noticeable that 2 studies^14,15 with 37 failures showed that 14 failed transsphincteric fistulas were converted into intersphinteric fistulas and were treated successfully by simple fistulectomy and one of these studies mentioned that all its failure and management were of similar pattern.¹⁵ Recurrences data from the pooled studies were 9.7% (95% CI 1.7-23.2). Recurrences occurred in only 3 studies with 8% (95% CI 0.984-26.031),¹² 20% (95% CI 8.441-36.938),¹³ (and 25.8% (95% CI 17.287-35.923)¹⁴ with median time to recurrence of 3.5, 4, and 7 months respectively (Table 5).

Table 4
Meta-analysis: proportion of failure in original LIFT.

Study	Sample size	Proportion (%)	95% CI	Weight (%)
				Fixed	Random
Tan et al. (2012)	24	37.5	18.799-59.406	12.25	19.29
Mushaya et al. (2012)	25	0	0.000-13.719	12.75	19.42
Wallin et al. (2012)	93	34.409	24.861-44.979	46.08	21.95
Dalbem et al. (2014)	22	22.727	7.821-45.370	11.27	19.02
Madbouly et al. (2014)	35	5.714	0.700-19.157	17.65	20.32
Total (fixed effects)	199	21.983	16.500-28.299	100	100
Total (random effects)	199	17.858	4.943-36.472	100	100
Test for heterogeneity
Q					33.4762
DF					4
Significance level					p < 0.0001
I2 (inconsistency)					88.05%
95% CI for I2					74.65-94.37

Table 5
Meta-analysis: proportion of recurrence in original LIFT.

Study	Sample size	Proportion (%)	95% CI	Weight (%)
				Fixed	Random
Tan et al. (2012)	24	0	0.000-14.247	12.25	19.06
Mushaya et al. (2012)	25	8	0.984-26.031	12.75	19.22
Wallin et al. (2012)	93	25.806	17.287-35.923	46.08	22.61
Dalbem et al. (2014)	22	0	0.000-15.437	11.27	18.72
Madbouly et al. (2014)	35	20	8.441-36.938	17.65	20.39
Total (fixed effects)	199	14.784	10.213-20.410	100	100
Total (random effects)	199	9.711	1.723-23.165	100	100
Test for heterogeneity
Q					25.429
DF					4
Significance level					p < 0.0001
I2 (inconsistency)					84.27%
95% CI for I2					64.71-92.99

Table 6
Showing anatomy of failure recurrence of original LIFT.

Author	Time to recurrence	Failure n (%)	Recurrence n (%)	Treatment of failure
Tan11	NA	9 (37.5%)	0	Incision and drainage1
				Fistulectomy 4
				Seton 4
				ERAF 2
				LIFT 1
Mushaya12	4 months	0	2 (8%)	NA
Madbouly13	3.5 months	2 (5.7%)	7 (20%)	NA
Wallin14	7.0 months	32 (34.4%)	24 (25.8%)	9 intersphincteric by fistulectomy
				2 transsphincteric by fistulectomy
				20 by Seton
				13 by LIFT
				2 by PLUG
				1 by advancement flap
				9 by drainage
Dalbem15	NA	5 (23%)	0	5 intersphincteric by fistulectomy

Other notable complications when mentioned were, secondary bleeding in 1 patient,¹¹ one patient had hematoma in a study with 93 patient,¹³ and finally wound dehiscence and infection in intersphinteric wound of 7 patients in 2 studies with 47 patient.^12,15

In all five studies, there was no change in continence function, except in one study which reported that 1 patient had incontinence for liquid and gas, and 4 patients reported incontinence for gas.¹⁵

Three of those 5 trials compared LIFT with other procedures.^11-13 One retrospective study by Tan¹¹ compared ligation of the intersphincteric fistula tract with endorectal advancement flap with healing rates of 62.5% and 93.5% respectively. Failure rates of the former (37.5%) were almost five times higher than the latter (6.5%).

A prospective study by Madbouly comparing LIFT with Mucosal Advancement Flap (MAF) showed that primary healing was 94.2% in LIFT vs. 91.4% in MAF,¹³ with median healing time in first group (26.6 days) significantly less than the second group (38.1 days). Failure rates were 5.8% in LIFT group and 8.5% in MAF group. After complete healing, 7 out of 35 (21.2%) patients showed recurrence after LIFT while 9 out of 35 (28.1%) patients after MAF. Overall, all healing were higher in LIFT (74.3%) in comparison to MAF (65.7%).

A notable part of this trial was the satisfactory scaling of patient stating that LIFT has the advantage of less postoperative pain. In addition, there was no change of continence levels after 16 weeks in patients who underwent lift, but 2 patients who underwent MAF showed insentience to (1) gas and (1) liquid.

A controlled randomized study by Mushaya¹² comparing LIFT with Anorectal Advancement Flap (ARAF) concluded that the major findings were that the LIFT technique can be performed more rapidly with less pain, a better satisfaction score, and faster resumption of normal activities than the ARAF technique. Although the overall recurrence rate was low in both procedures (7% and 8% respectively), there was no difference in recurrence rate between the 2 procedures. However, one patient in the ARAF group reported a slight incontinence (a CCF-FI score of 4) which had resolved in a few weeks.¹²

LIFT modifications results

Regarding LIFT modifications, 5 trials were included in this review, with population of 147 patients with demographic values according to Fig. 3. Four modifications were identified (Table 3).

Fig. 3
Population of LIFT modification group: (A) gender; (B) type of fistula.

First modification of original LIFT procedure included coring out of fistula tract from the external opening of the fistula tract to the lateral border of the external sphincter. It was described by Sirikurnpiboon “LIFT PLUS” in 2013.⁷ Two studies, included in this review, applied this technique, one prospective¹⁶ and another retrospective,¹⁷ with total numbers of 61 patients with 22 low transsphincteric fistulas and 39 with transsphincteric fistula respectively (Table 7).

Table 7
LIFT modifications data by author, type of study, procedure, type of fistula, etc.

Author	Period of study	Year published	Type of study	Procedure (n)	Type of fistula (%)	1ry healing (%)	Follow-up, months (n)	Continence evaluation	OCEBM level
Onkelen16	June 2009-March 2012	2012	P	LIFT with partial coreout fistulectomy (22)	Low transsphincteric fistula	82%	19.5	RFISI	4
Feng Ye17	June 2012-March 2013	2014	R	LIFT with partial coreout fistulectomy (39)	High transsphincteric	87.2%	15	WIS	4
Onkelen6	June 2009-December 2010	2012	p	LIFT with partial coreout fistulectomy and TAFR (41)	High transsphincteric fistula	51.0%	15	RFISI	4
Han4	December 2010-March 2011	2012	P	LIFT-plug (21)	Transsphincteric	95.0%	14	WIS	4
Tsunoda18	March 2010-August 2012	2012	p	LIFT with partial coreout fistulectomy and Seton (20)	Low transsphincteric (60%), complex (40%).	95.0%	18.0	Manometry, RFISI, clinical	4

• R, retrospective; P, prospective; LIFT, ligation of intersphincteric tract; TARA, transanal advancement flap; RFISI, Rockwood Fecal Incontinence Severity Index; WIS, Wexner Incontinence Score; OCEBM, Oxford Centre for Evidence-Based.

Primary healing rates of this variation were reported as 81.8% (95% CI 59.715-94.813) and 87.17% (95% CI 72.570-95.703). Overall, all healing rates reached 100% (95% CI 84.563-100.000) in both studies. Median follow-ups were 19.5 and 15 months. There were no changes in continence levels in both studies.

There was no recurrence; however, an interesting fact was that both studies reported 9 failures only in the form of intersphincteric fistulas rather than the original transsphincteric and they were treated successfully with simple fistulectomy.

Other complications also were wound dehiscence (6 out of 39 patients) mentioned in one study.¹⁷

A second modification included in this review was LIFT with partial coring out fistulectomy and an additional transanal advancement flap. Only one study was included in this review with primary healing rates of 51.2% (95% CI 35.134-67.122) in 41 patients with 73.1% (95% CI 57.056-85.779) overall healing rates.⁶

There was no recurrence after median follow-up of 15 months. 20 patients had failure of procedure; however, in 8 (40%) patients, transsphincteric fistulas were converted into intersphincteric treated with simple fistulectomy.

There was no change in continence levels. Other complications also were purulent discharge at the intersphincteric wound resolved spontaneously in 4 patients.

LIFT Plug, a third alternation, which involve insertion of a plug in the external portion of the track. One prospective study was included, and primary healing rates were 95.2% (95% CI 76.184-99.880) in total of 21 patients, with failure in one patient and no recurrence. Only 1 (5%) patient reported rare incontinence for gas postoperatively.⁴

A final modification was included in this review which was left with partial coring out fistulectomy and insertion of Seton during operation. Only one study was included with 20 patients, and 12 of them were with low transsphincteric fistulas and 8 were complex transsphincteric. Healing rates were 95% (95% CI 75.127-99.873). With no recurrence and one reported failure, however, there was no explanation of the failure. No change in the continence levels was observed. Additionally, other complications included were 1 patient with wound dehiscence.¹⁸

Overall, pooled data from LIFT modifications included primary healing rates of 82.3% (95% CI 64.8-94.7) (Table 8), overall healing rates of 93.6% (95% CI 81.4-99.6) (Table 9), failure rates of 17.7% (95% CI 5.3-35.2) (Table 10), and 0 recurrences.

Table 8
Meta-analysis: proportion of primary healing in Modifications of LIFT.

Study	Sample size	Proportion (%)	95% CI	Weight (%)
				Fixed	Random
Onkelen et al. (2012)	22	81.818	59.715-94.813	15.54	19.36
Onkelen et al. (2012)	41	51.22	35.134-67.122	28.38	21.29
Han et al. (2012)	21	95.238	76.184-99.880	14.86	19.19
Tsunoda et al. (2012)	20	95	75.127-99.873	14.19	19
Feng Ye et al. (2014)	39	87.179	72.570-95.703	27.03	21.16
Total (fixed effects)	143	79.506	72.100-85.693	100	100
Total (random effects)	143	82.297	64.764-94.668	100	100
Test for heterogeneity
Q					24.1237
DF					4
Significance level					p = 0.0001
I2 (inconsistency)					83.42%
95% CI for I2					62.39-92.69

Table 9
Meta-analysis: proportion of overall healing in modifications of LIFT.

Study	Sample size	Proportion (%)	95% CI	Weight (%)
				Fixed	Random
Onkelen et al. (2012)	22	100	84.563-100.000	15.54	19.28
Onkelen et al. (2012)	41	73.171	57.056-85.779	28.38	21.44
Han et al. (2012)	21	95.238	76.184-99.880	14.86	19.09
Tsunoda et al. (2012)	20	95	75.127-99.873	14.19	18.88
Feng Ye et al. (2014)	39	100	90.975-100.000	27.03	21.3
Total (fixed effects)	143	92.853	87.442-96.436	100	100
Total (random effects)	143	93.62	81.420-99.577	100	100
Test for heterogeneity
Q					21.5523
DF					4
Significance level					p = 0.0002
I2 (inconsistency)					81.44%
95% CI for I2					56.92-92.01

Table 10
Meta-analysis: proportion of failure in modifications of LIFT.

Study	Sample size	Proportion (%)	95% CI	Weight (%)
				Fixed	Random
Onkelen et al. (2012)	22	18.182	5.187-40.285	15.54	19.36
Onkelen et al. (2012)	41	48.78	32.878-64.866	28.38	21.29
Han et al. (2012)	21	4.762	0.120-23.816	14.86	19.19
Tsunoda et al. (2012)	20	5	0.127-24.873	14.19	19
Feng Ye et al. (2014)	39	12.821	4.297-27.430	27.03	21.16
Total (fixed effects)	143	20.494	14.307-27.900	100	100
Total (random effects)	143	17.703	5.332-35.236	100	100
Test for heterogeneity
Q					24.1237
DF					4
Significance level					p = 0.0001
I2 (inconsistency)					83.42%
95% CI for I2					62.39-92.69

In conclusion, when comparing the original LIFT and its modification, it can be seen that primary healing rates in the original LIFT were 73.95% (95% CI 60.3-85.6), which performed less than that in the modified LIFT with 82.3% (95% CI 64.8-94.7). Overall healing rates in the original LIFT were 78.9% (95% CI 58.5-93.7), which performed less than that in the modified LIFT with 93.6% (95% CI 81.4-99.6). Failure rates in the original LIFT were 17.9% (95% CI 4.9-36.5), which performed almost the same as the modified LIFT with 17.7% (95% CI 5.3-35.2). Recurrence rates in the original LIFT were 9.7% (95% CI 1.7-23.2), which performed less than the modified LIFT with zero recurrence.

Funnel plots for all estimates of original LIFT showed equal balanced distribution of the estimates of different studies around the pooled estimate. However, Funnel plots of the modified LIFT showed some bias toward higher values around the estimates (Figs. 4-17).

Fig. 4
Meta-analysis; original LIFT, primary healing forest plot.

Fig. 5
Meta-analysis; LIFT modifications, primary healing forest plot.

Fig. 6
Meta-analysis; original LIFT, overall healing forest plot.

Fig. 7
Meta-analysis; original LIFT, overall healing forest plot.

Fig. 8
Meta-analysis; original LIFT, failure forest plot.

Fig. 9
Meta-analysis; LIFT modification, failure forest plot.

Fig. 10
Meta-analysis; original LIFT, recurrence forest plot.

Fig. 11
Meta-analysis; original LIFT, primary healing, funnel plot.

Fig. 12
Meta-analysis; LIFT modification, primary healing, funnel plot.

Fig. 13
Meta-analysis; original LIFT, overall healing, funnel plot.

Fig. 14
Meta-analysis; LIFT modification, overall healing, funnel plot.

Fig. 15
Meta-analysis; original LIFT, failure, funnel plot.

Fig. 16
Meta-analysis; LIFT modification, failure, funnel plot.

Fig. 17
Meta-analysis; original LIFT, recurrence, funnel plot.

Discussion

This review had concluded that LIFT procedure and its modification have an effective healing in over two-third of the population included. It showed that this procedure had minimal or no changes in continence levels. Failed cases usually transformed transsphincteric fistulae into intersphincteric ones which were treated simply by fistulectomy (Tables 6 and 11). Recurrence rates were low and even absent in most of the studies. Other complications were mainly limited to the intersphincteric wound either with hematoma or mostly wound dehiscence which resolved spontaneously without any surgical intervention. LIFT procedure was introduced in 2007 to find a better outcome in anal surgery regarding saving sphincters unlike other procedure, also to provide better healing and lower recurrence rate. And since then it has been popularized among many surgeons' worldwide, thanks to its simplicity and its promising outcomes, comparing its result to other procedures.

Table 11
Anatomy of failure and recurrence in LIFT modifications.

Author	Time to recurrence	Failure	Recurrence	Treatment of failure
Onkelen16	NA	4 (18%)	No recurrence	4 intersphincteric by fistulectomy
Feng Ye17	NA	5 (1.2%)	No recurrence	5 intersphincteric fistula and fistulectomy
Onkelen6	NA	49% 20 pts	No recurrence	8 intersphincteric fistula by fistulectomy
				4 by TARF
				8 by Seton
Han4	NA	1(5%)	No recurrence	NA
Tsunoda18	NA	1 (5%)	No recurrence	NA

• NA, not given.

The original LIFT procedure appeared most frequently in the literature. Rojanasakul initially reported 94% healing rates, and healing rates from pooled data were 73.95% (95% CI 60.3-85.6) for primary healing and 78.9% (95% CI 58.5-93.7) for overall healing in original LIFT in this review. Other reviews reported healing rates in 759 patients that ranged from 51% to 94%.¹⁹ Another review showed success rates from 40% to 95%, in 352 of 495 patients.²⁰

Additionally, healing rates from the pooled data of LIFT modification were 82.3% (95% CI 64.8-94.7) for primary healing and 93.6% (95% CI 81.4-99.6) for overall healing. Overall, healing rates of LIFT and its modifications showed better results than other sphincter-preserving procedures, for example, 10-67% for fibrin glue,²¹ <50% for fistula plug²² and 25-95% for advancement flap.²² All of the studies in our review included a standard definition of healing whether primary healing or overall healing.^11,14 However, a clear definition of failure or recurrence has not been universally described yet.

What is interesting is that time to heal after surgery in given studies ranged from 2 weeks¹⁷ to 7 weeks,¹⁸ however, time for healed fistulae to reoccur were around 3 to 7 months in reported studies.^12-14 This means that long follow up is essential to determine the overall success of this procedure and its modifications. Nevertheless, all reviews which discussed this types of anal surgery did not put consideration on follow up period, and what makes this review more reliable is that it only included studies with follow up more than or equal to 12 months. Also, most of the reviews regarding this procedure did not compare original results with LIFT modifications results. However, more studies are needed to compare the techniques to fulfill better understanding of a superior approach.

Systemic risk factors did not play any major role in studies. In addition, preservation of continence levels play a key in favoring LIFT and its modulations over other sphincter preserving procedure and other anal procedures like fistulectomy; as seen is this review, continence was rarely damaged in all patients (1 patient mild incontinence to liquid and 5 to gas in total of 340 patients).

On the other hand, other sphincter-preserving procedures turned out to have greater impact on it. For example, various studies of advancement flap repairs have reported rates of incontinence ranging from 0% to 35%.²² The impact on continence appears to be minimal; also, most of studies included had a formal assessment of incontinence (8 out of 10 studies)^{4,6,12-14,16-18} (Tables 1 and 7). Therefore, we may conclude complete results of the rates and nature of continence changes after this procedure from this review.

This review also highlighted other minor complications following surgery, the main and most notable one is wound dehiscence and surgical site infection or discharge in intersphinteric incision, 18 patients of 340; however, all patients resolved spontaneously with proper dressing, wound care and sometime antibiotics, with no further intervention.^6,4,12,15,17

Regarding preoperative Seton insertion in original and modified LIFT, its result was irrelevant and did not promote better healing. Nevertheless, one study included preoperative Seton to enforce or enhance fibrosis in the tracts to make them well defined.¹⁵ Therefore, further future studies should elaborate the factor of preoperative Seton insertion and its outcomes on healing and recurrence. Similarly, preoperative imagining, with endoanal ultrasound or MRI, in included studies did not play a major role in pre-evaluating the fistula and most surgeons relied heavily on intraoperative examination to outline the anatomy of the fistula. Results from studies which used MRI and endoanal ultrasound were similar to those that discarded imaging. Therefore, there is no need for routine preoperative Seton placement and imaging if not needed, because it could provide patients and the health system with significant cost and time reduction to achieve the targeted outcomes. Regarding fistula complexity, most of the studies on LIFT and its modifications were performed on high transsphincteric fistulae with limited numbers on complex branching ones which could limit the use of these procedures on straight forward fistulae, and therefore more trials should be conducted to access their effectiveness on various types of fistulae.

Conclusion

Ligation of the intersphincteric fistula tract is a feasible, minimally invasive, cheap, and relatively easy procedure, which is safe and effective at the same time. LIFT and its modifications can be ideal for treating straightforward high anal fistulas in patient with no previous intervention. Additionally, more studies should be conducted to compare results regarding different approaches of this procedure with longer follow-up and randomization of patients. Finally, a universal explanation of failure and recurrence should be applied with a proposition that failure should be explained as an unsuccessful achievement of primary healing and recurrence as a failure after successful primary healing.

References

• ¹
  Oxforddictionaries.com. Fistula - definition of fistula in English from the Oxford dictionary; 2015. Available at: http://www.oxforddictionaries.com/definition/english/fistula
  » http://www.oxforddictionaries.com/definition/english/fistula
• ²
  Mangual RN, Tudu DN, Pattnaik SP, Mohanty SS, Prusty KP. The sphincter preserving perianal fistulectomy: a better alternative. Ind J Surg. 2004;66:31-5.
• ³
  Rojanasakul A, Pattanaarun J, Sahakitrungruang C, Tantiphlachiva K. Total anal sphincter saving technique for fistula-in-ano; the ligation of intersphincteric fistula tract. J Med Assoc Thai. 2007;90:581-6.
• ⁴
  Han JG, Yi BQ, Wang ZJ, Zheng Y, Cui JJ, Yu XQ, et al. Ligation of the intersphincteric fistula tract plus a bioprosthetic anal fistula plug (LIFT-Plug): a new technique for fistula-in-ano. Colorectal Dis. 2013;15:582-6.
• ⁵
  Song KH. New techniques for treating an anal fistula. J Korean Soc Coloproctol. 2012;28:7-12.
• ⁶
  van Onkelen RS, Gosselink MP, Schouten WR. Is it possible to improve the outcome of transanal advancement flap repair for high transsphincteric fistulas by additional ligation of the intersphincteric fistula tract? Dis Colon Rectum. 2012;55:163-6.
• ⁷
  Sirikurnpiboon S, Awapittaya B, Jivapaisarnpong P. Ligation of intersphincteric fistula tract and its modification: results from treatment of complex fistula. World J Gastrointest Surg. 2013;5:123-8.
• ⁸
  Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009;6:e1000100.
• ⁹
  Howick J, Chalmers I, Glasziou P, Greenhalgh P, Heneghan C, Liberati A, et al. Explanation of the 2011 Oxford Centre for Evidence-Based Medicine (OCEBM) levels of evidence. http://www.cebm.net/index.aspx?o=5653 [accessed 13.02.15].
  » http://www.cebm.net/index.aspx?o=5653
• ¹⁰
  Wallace BC, Dahabreh IJ, Trikalinos TA, Lau J, Trow P, Schmid CH. Closing the gap between methodologists and end-users: R as a computational back-end. J Stat Soft. 2012;49:5.
• ¹¹
  Tan KK, Alsuwaigh R, Tan AM, Tan IJ, Liu X, Koh DC, et al. To LIFT or to flap? Which surgery to perform following Seton insertion for high anal fistula? Dis Colon Rectum. 2012;55:1273-7.
• ¹²
  Mushaya C, Bartlett L, Schulze B, Ho YH. Ligation of intersphincteric fistula tract compared with advancement flap for complex anorectal fistulas requiring initial Seton drainage. Am J Surg. 2012;204:283-9.
• ¹³
  Madbouly KM, El Shazly W, Abbas KS, Hussein AM. Ligation of intersphincteric fistula tract versus mucosal advancement flap in patients with high transsphincteric fistula-in-ano: a prospective randomized trial. Dis Colon Rectum. 2014;57:1202-8.
• ¹⁴
  Wallin UG, Mellgren AF, Madoff RD, Goldberg SM. Does ligation of the intersphincteric fistula tract raise the bar in fistula surgery? Dis Colon Rectum. 2012;55:1173-8.
• ¹⁵
  Dalbem CS, Danilo S, Tomiyoshi T, Henrique C. Assessment of LIFT (ligation of the intersphincteric fistula tract) technique in patients with perianal transsphincteric fistulas. J Coloproctol. 2014;34:250-3.
• ¹⁶
  van Onkelen RS, Gosselink MP, Schouten WR. Ligation of the intersphincteric fistula tract in low transsphincteric fistulae: a new technique to avoid fistulotomy. Colorectal Dis. 2013;15:587-91.
• ¹⁷
  Ye F, Tang C, Wang D, Zheng S. Early Experience with the modificated approach of ligation of the intersphincteric fistula tract for high transsphincteric fistula. World J Surg. 2015;39:1059-65.
• ¹⁸
  Tsunoda A, Sada H, Sugimoto T, Nagata H, Kano N. Anal function after ligation of the intersphincteric fistula tract. Dis Colon Rectum. 2013;56:898-902.
• ¹⁹
  Kontovounisios C, Tekkis P, Tan E, Rasheed S, Darzi A, Wexner SD. Adoption and success rates of perineal procedures for fistula-in-ano: a systematic review. Colorectal Dis. 2016;18:441-58.
• ²⁰
  Yassin NA, Hammond TM, Lunniss PJ, Phillips RKS. Ligation of the intersphincteric fistula tract in the management of anal fistula. A systematic review. Colorectal Dis. 2013;15:527-35.
• ²¹
  Steele SR, Kumar R, Feingold DL, Rafferty JL, Buie WD. Practice parameters for the management of perianal abscess and fistula-in-ano. Dis Colon Rectum. 2011;54:1465-74.
• ²²
  Muhlmann MD, Hayes JL, Merrie AE, Parry BR, Bissett IP. Complex anal fistulas: plug or flap? ANZ J Surg. 2011;81:720-4.

Publication Dates

• Publication in this collection
  Oct-Dec 2018

History

• Published
  24 May 2018