A meta-analysis of tunica vaginalis fascia versus dartos fascia to prevent postoperative urethrocutaneous fistula in hypospadias and fistula repair

Background: Urethrocutaneous fistula (UCF) is the commonest postoperative complication in hypospadias and fistula repair. Several flap procedures have been recommended to decrease this complication rate, but no single flap procedure is ideal. The aim of this study was to compare the outcome of tunica vaginalis fascia (TVF) and dartos fascia (DF) as intermediate layers in prevention of the formation of UCF. Methods: We searched PubMed, EMBASE, the Cochrane Library and Web of Science for comparative studies up to July 1st, 2019. Studies were selected by the predesigned inclusion criterias. The primary outcomes was UCF incidence. Results: The pooled RR with 95% CI were calculated. We extracted the relevant informations from the included studie. 9 comparative studies were included The RR of UCF rate for TVF was 0.21 (95% CI: 0.09-0.51, P=0.0005) compared with DF in hypospadias and fistula repair. For other postoperative complications, the RR was 0.87 (0.28-2.70, P=0.80), 1.33 (0.41-4.35, P=0.64) and 0.17 (0.03-0.88, P=0.04) for meatal stenosis/ urethral stricture, glans dehiscence/ wound dehiscence and skin necrosis, respectively. Conclusions: This meta-analysis reveals that TVF is a better option in hypospadias and fistula repair as compared to DF in terms of decreasing the incidence of the UCF and skin necrosis. respectively. According whether magnification technique was used intraoperatively, we classified all studies as “Yes” and “No or NA”, and the RR were 0.17 (0.05–0.60, P = 0.006) and 0.28 (0.08–0.91, P = 0.03) respectively. These results showed that in TIP urethroplasty, one-stage urethroplasty, with or without magnification technique, TVF had lower postoperative UCF rate compared with DF as a soft tissue coverage in the hypospadias and fistula repair. But staged urethroplasty had no significant difference between TVF and DF.


Introduction
Hypospadias, which is caused by incomplete development of the urethra, is one of the most common congenital anomalies in male infants, with an estimated prevalence of 20.9 cases per 10 000 male live births around the world. The international total prevalence increased 1.6 times during 1980-2010, by 0.25 cases per 10 000 births per year [1]. The current purpose of hypospadias surgery is to improve the functional and cosmetic outcomes while minimizing the incidence of postoperative complications and avoiding reoperations. Urethrocutaneous fistula (UCF) is the commonest postoperative complication in hypospadias and fistula repair (0-35%) [2] and is also a major cause of reoperation. Additional urethral coverage is now routinely useed by most pediatric urologists to reduce the risk of UCF formation [3]. Dartos fascia (DF) and tunica vaginalis fascia (TVF) are the two most widely used urethral coverages with good surgical results while results varies reported by different studies [4][5][6]. At present, there has been no consensus on the better choice between the DF and TVF flap techniques, as well as on the short-and long-term outcomes of both techniques [3]. The aim of this meta-analysis was to compare the outcome of TVF and DF as intermediate layers in prevention of the formation of UCF in hypospadias and fistula repair.

Methods
This study is registered with PROSPERO, number CRD42019148554.

Inclusion criteria
The studies were selected according to the following criteria: (1) participants were diagnosed as hypospadias or fistula, (2) reported the incidence of postoperative UCF with or without the incidence of other complications, (3) comparative studies with control groups will be included in the review, (4) the article was written in English.
COMPLICATIONS WERE DEFINED AS UCF, meatal stenosis, urethral stricture, glans dehiscence, wound dehiscence or skin necrosis. The primary outcomes was fistula incidence (the proportion of patients who formed UCF postoperatively). The secondary outcomes were incidences of other complications (the proportion of patients who formed meatal stenosis, urethral stricture, glans dehiscence, wound dehiscence and skin necrosis postoperatively).
Data extraction and quality assessment Two independent authors (HY, QS) screened all retrieved titles and abstracts according to the pre-described inclusion criteria to identify potentially eligible studies. After screening, we accessed the full text to determine the final included studies independently. The following data from each included study were extracted: study characteristics (first author, published year, study design, hypospadias type, surgery type, and follow up time), patient characteristics (patients numbers, patients age, incidence of fistula, incidence of other complications). For the whole process, discrepancies were resolved by discussion between the two reviewers and reevaluation with a third author (XG). The level of evidence and publication type was classified according to the Oxford CEBM [7].

Statistical analysis
We used RevMan 5.3 (Cochrane Library, Oxford, UK) and STATA 13.1 (Corp LP, College Station, TX, USA) softwares to perform this meta-analysis and relevant subgroup analysis.
All categorical data were analyzed by estimating the pooled risk ratios (RR) and 95% confidence intervals (CI). The Mantel-Haenszel method was used to calculate pooled RR [8]. Heterogeneity was evaluated by the Cochrane's Q-statistic and the I 2 statistic [9]. A random-effects model would be used if heterogeneity was significant (the Q statistic was significant or I 2 values > 50%). Otherwise, a fixed-effects model would be used. Sensitivity analysis and subgroup analysis were performed to assess the robustness of the results and find the possible sources of heterogeneity. Several subgroup analysis were conducted, including different surgery type, study design and other available groups. The symmetry of funnel plot was used to judge the existence of publication bias subjectively. And Begg and Egger tests [10,11] were conducted to measure the potential publication bias through STATA software. A P values < 0.05 was considered to be statistically significant.

Literature selection and study characteristics
Of 1928 databases articles titles and 57 trial register titles screened, 1269 abstracts were reviewed, 31 full text were reviewed, and 9 articles [12-20] met our inclusion criteria and were included in our meta-analysis (Fig. 1). The 9 comparative studies involving a total of 564 hypospadias or fistula patients (193 patients in the TVF group and 371 in DF group).
In the subgroup analysis (Table 3)   the robustness of the results in this meta-analysis. Reoperation for failed hypospadias or fistula repair has been considered to be a seriously problem because the dense fibrotic tissue causes difficulties in wound healing and increases the rate of complications [21].
Various surgical methods and modifications have been developed for repairing the hypospadias and fistula while UCF are still one of the most common complications of these techniques [22]. The total UCF rate is 7.71% in our study, consistent with the 7.5% incidence in a systematic review [2]. Patients age, glans size, urethral defect length, urethral operation history, surgical procedure, type of surgical repair, chordee degree, magification technique, caudal anesthesia, preoperative hormonal stimulation and other many factors are may related to the development of UCF and other complications postoperatively [21][22][23][24][25]. Additional soft coverages on the neourethra are also introduced to avoid these complications especially for decreasing the incidence of postoperative UCF. DF is a layer of connective tissue found in the penile dorsal or ventral, foreskin, and scrotum and can be used in hypospadias or fistula repair in different techniques [33,34]. Excellent vascularity, easy availability and adequate source are advantages of DF, making this flap technique more popular for many pediatric urologists especially for young surgeons. Penile rotation and preputial skin necrosis were common reported relevant complications and could be avoided by careful operation and technical improvement.
Snodgrass described additional interposition of vascularized tissues between the tubularized plate and glans closure dissected from dorsal prepucial and shaft skin [35].
Duckett has described that when dartos is separated from skin, it compromises the vascularity of the overlying skin [36]. So the dissection of DF may compromise the vascularity of the preputial skin covering and result in subsequent skin necrosis. This is consistent with the conclusion of our secondary outcomes. TVF can be harvested through penile incision by degloving till the root of penis [26] or additional scrotal incision which reach and cover the neourethra through a subcutaneous scrotal tunnel [37] which however may damage to the vas or vessels of the testicular therefore resulting in scrotal abscess or scrotal hematoma. The blood supply of the neourethra tissue may be affected due to the dissection and utilization of DF, which mainly comes from the shortage of dartos or preputial skin necrosis. However, this is rarely affected in TVF technique for its ventral skin covering is rarely compromised. All these can theoretically explain the advantage of TVF over DF. And we performed this meta-analysis to prove the advantages of TVF over DF with the data.
The results of our meta-analysis were partially consistent with the result of a systematic review by Fahmy et al [30]. However, there were several differences among the two studies.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this published article.