Tubularized incised plate urethroplasty and grafted tubularized incised plate urethroplasty: systematic review, meta-analysis and trial sequential analysis

Introduction

Hypospadias is defined by an ectopic opening of the urethral meatus on the ventral aspect of the penis rather than the tip. Embryologically, this occurs because of the arrest in the normal development of the penis.1 The incidence of hypospadias in Europe is around 18.6 per thousand, with the highest prevalence in North America and the lowest in Asia.2 Hypospadias correction is usually recommended between the ages of 6 and 18 months. There are around 300 methods for the surgical correction of hypospadias which evolved over a thousand years from after Christ to the modern era. Galen from the second century was the first to use the term ‘hypospadias’.3 Sophisticated urethral surgery was only possible after the introduction of anesthesia by Morton in 1946.3 In 1880, Duplay first described the tubularization of local skin over a tube,3 and over the next century, various techniques were introduced using local flaps and free grafts for reconstruction of the neourethra.

In 1994, Warren Snodgrass published his tubularized incised plate (TIP) urethroplasty technique.4 He used this technique to correct distal hypospadias with minimal chordee. This TIP technique involves a midline incision over the urethral plate from the anomalous meatus to the glans tip, which allows mobilization of the plate for tubularization. It creates a functional neourethra with a vertically oriented slit-like meatus. Soon thereafter, the Snodgrass technique became very popular as an alternative to meatal-based and onlay island flaps for distal hypospadias. However, over the years, many surgeons have noticed that the results of this technique may be compromised, especially in patients with a narrow or shallow urethral plate. TIP repair has also been reported to have complications such as meatal stenosis (MS) and urethrocutaneous fistula (UCF). In 2000, Kolon and Gonzales5 reported a technique using a free graft of inner prepuce to bridge the gap created by the Snodgrass incision. This showed promising results with none of the patients in their series of 32 patients developing MS or UCF. Hayes and Malone6 used free buccal mucosa graft instead of inner preputial skin in their patients with satisfying results; however, preputial skin is usually preferred and widely used for hypospadias repair. Later, a few authors reported encouraging results using the technique described by Kolon and Gonzales.5 Mouravas and Sfoungaris7 were the first to publish a randomized trial comparing the results of TIP with grafted TIP (GTIP) urethroplasty. They concluded that the GTIP technique had a considerably lower rate of complications than TIP. The results of TIP and GTIP repair for primary distal hypospadias correction were comparable in a prospective randomized study by Helmy et al.8 They preferred the Snodgrass technique as the procedure of choice for primary distal hypospadias correction. A prospective randomized study by Eldeeb et al9 also reported equivalent results for both TIP and GTIP groups with a shorter operating time in the TIP group. In their prospective comparative study, Ahmed et al10 concluded that despite being a statistically insignificant result, GTIP repair showed better clinical outcomes. The HOSE (Hypospadias Objective Scoring Evaluation) score as a measure of cosmetic outcomes was also comparable in both groups in their study.

Further objective evaluation of urethral function by urinary flow measurement after hypospadias correction by these techniques was performed by Helmy et al and González and Ludwikowski.8 11 Both techniques have proven their results in hypospadias repair. Published literature shows the relatively recent GTIP technique to be equivalent or superior to TIP in terms of operative complications, cosmesis, and functional outcomes. This quantitative analysis aimed to systematically compare the reported outcomes of TIP and GTIP in children undergoing primary hypospadias repair.

Methods

This meta-analysis was conducted as per Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.12 In the included studies, patients included children under the age of 18 years undergoing primary hypospadias repair through dorsal inlay GTIP compared with classic TIP. Primary outcomes that were assessed included UCF, MS, glans dehiscence (GD), operative time, wound infection, and success rate. Secondary outcomes included uroflow, cosmetic scores, and urethral diverticulum. We only included randomized controlled trials (RCTs) for our meta-analysis.

Results

Study characteristics

Our initial literature search was as per Boolean search results in 1002 articles. After removing the duplicates and excluding non-eligible studies, 10 randomized studies including 603 children7–10 16–21 were found to be eligible. The details of the excluded studies are also depicted in the PRISMA flow diagram (figure 1).

Request permission

Figure 1

PRISMA Flow Diagram.

The characteristics of all the included studies are mentioned in table 1. Nine studies were available in English literature, and one was in Chinese with abstracts in English. In the studies, there were 300 patients in the TIP group and 303 patients in the GTIP group. The primary and secondary outcomes of all studies are mentioned in table 2.

Table 1

View inline|View popup

Study characteristics

Table 2

View inline|View popup

Study outcomes

Risk of bias assessment

Risk of bias assessment (RoB 2) was performed across all the studies using the robvis tool and robvis web app (figure 2). Seven studies out of the 10 had an overall low risk of bias. However, for the remaining three studies, there were some concern in the domain of randomization7 16 and concern over missing outcome data.10

Request permission

Figure 2

Risk of Bias Domains.

Data synthesis and analysis of outcome

Urethrocutaneous fistula

This outcome was reported by nine studies included in this meta-analysis. We calculated the incidence of UCF at the maximum follow-up period mentioned in the respective studies. There were nine UCFs in the GTIP group (2.97%) and 18 in the TIP group (6%). The study by Helmy et al8 did not report any UCF events among the comparison groups. Pooled analysis of this outcome for the included studies showed that there was no statistically significant difference in the occurrence of UCF between the groups (RR=0.52, 95% CI 0.25 to 1.10), with no heterogeneity among the studies (I²=0%) (figure 3A).

Request permission

Figure 3

(A) Forest Plot (UCF) (B) Forest Plot (Meatal Stenosis) (C) Forest Plot (Operative Time). GTIP, grafted TIP; TIP, tubularized incised plate.

Meatal stenosis

This outcome was reported by all 10 studies included in the meta-analysis. There were 6 events of MS in the GTIP group (2%) and 24 events in the TIP group (8%). Pooled analysis of the included studies for this outcome has shown a lower incidence of MS in the GTIP group (RR 0.32, 95% CI 0.15 to 0.67). No heterogeneity was observed for this pooled analysis (I²=0%) (figure 3B).

Glans dehiscence

This outcome was reported by seven studies included in the meta-analysis. Studies by Patel et al,21 Mohammed et al,18 and Zeina et al19 did not report any event of GD in their reported series. There were eight events of GD in the GTIP group (2.64%) and three in the TIP group (1%). Pooled analysis of the included studies for this outcome showed that there was no significant difference in the occurrence of GD between the groups without heterogeneity (RR=1.89, 95% CI 0.68 to 5.24, I²=0%).

Operative time

All studies have reported this outcome except for Patel et al.21 Mouravas and Sfoungaris7 reported operative time as a mean with range, and Eldeeb et al9 reported this outcome as a median with range. Mohammed et al18 reported this outcome as a mean without range or SD. Ahmed et al10 reported that a few of their patients failed to report for follow-up. For the meta-analysis of the outcome of operative time, we considered the total number of patients operated on by surgeons; however, for the analysis of the remaining outcome measures, we calculated the total number of patients who had completed follow-up. A pooled analysis of the remaining six studies was performed. Quantitative synthesis of this outcome showed that operative time was shorter in the TIP group than in the GTIP group, with significant heterogeneity (MD=19.09, 95% CI 10.24 to 27.95, I²=97%) (figure 3C).

Maximum flow rate

Only three studies reported this outcome.8 16 20 Pooled analysis of the three studies showed that the maximum flow rate was better in the TIP group than that in the GTIP group, with significant heterogeneity (MD=3.78, 95% CI 0.30 to 7.27, I²=95%). Although the maximum flow rate was in favor of TIP, as there were only a small number of participants in the three studies and high heterogeneity, the results need to be interpreted cautiously.

HOSE score

Only three studies reported this outcome. El Shazly et al17 mentioned this outcome but did not provide the mean and SD values. Pooled analysis of the other two studies showed no significant difference in HOSE score among both groups, with significant heterogeneity (MD=0.52, 95% CI −0.92 to 1.96, I²=95%).

Success rate

Three of the included studies9 10 16 reported this outcome as a separate parameter. The study by Eldeeb et al9 quoted success as having no complications. In the other two studies, the criteria for success were unclear. In the studies that did not provide the success rate as a separate outcome, we derived it after subtracting the complications from the total number of participants. In a study by Patel et al,21 we could not obtain the total number of complications, so the study was removed from the final analysis. Pooled analysis of the remaining nine studies for the outcome of success rate showed no significant difference between the groups, with heterogeneity (RR=1.02, 95% CI 0.97 to 1.08, I²=0%).

Other outcomes, such as diverticulum, wound infection and urethral stricture, have been reported by only one study each, so pooled analysis was not applicable. Nonetheless, these outcomes were not statistically significant in the described respective studies.

Trial sequential analysis

We performed TSA for the two primary outcomes, UCF and MS. The analysis of UCF shows that the information size of 603 was inadequate for the evidence to be conclusive, and it has not crossed conventional borders. The RIS for this outcome is 1504 (figure 4A). Analysis of MS showed that the cumulative Z-score crossed conventional boundaries and alpha-spending boundaries in favor of the TIP group. After reaching the RIS, the Z value remained greater than 1.96, making the meta-analysis of this outcome statistically significant (figure 4B).

Request permission

Figure 4

(A) Trial sequential analysis for UCF; (B) trial sequential analysis for MS. GTIP, grafted TIP; TIP, tubularized incised plate.

Discussion

Our literature search yielded 10 studies satisfying the eligibility criteria. There were 603 patients across the 10 studies used for our quantitative analysis. As per the quantitative analysis, supplemented by TSA, MS had a lower incidence in the GTIP group than in the TIP group. Our analysis also concludes that operative time is less in the TIP group than in the GTIP group. The other outcomes, such as UCF, GD, HOSE score, and success rate, were comparable in both groups.

The first case of TIP was performed by Warren Snodgrass in 1990 in a 9-month-old child who was planned for Mathieu repair.22 The rounded meatal appearance after Mathieu repair and the possible presence of hair follicles in the proximal margin of the outlined flap for Mathieu repair have incited the idea of TIP based on Rich et al’s23 principle of hinging of the urethral plate. In their own review of TIP, Warren Snodgrass and Nicol Bush mentioned UCF and GD are the most common complications encountered after TIP repair.22 They encountered only one case of MS among a series of 426 patients.22 A systematic review of complications rate of TIP in 3261 patients of distal hypospadias by Pfistermuller et al mentioned an overall incidence of 3.6% (1.7–7.4%) for MS, 5.7% (4.0–8.2%) for UCF and 1.3% (0.8–2.2%) for urethral stricture.24 The mean incidence of UCF (6.33%) in the TIP group in our pooled studies also falls within the range of the study by Pfistermuller et al,24 but the mean value of MS (8%) in our pooled data exceeds the range mentioned in that study. As mentioned earlier, Kolon and Gonzales5 modified the Snodgrass procedure by placing the graft at the bed of the midline incision of the TIP. In TIP, this midline incision of TIP heals through granulation and subsequent fibrous tissue, and the graft is covered with epithelium, minimizing fibrotic reaction, which is a possible explanation for many modifying TIP by placing the midline graft.7 25 The TIP procedure is frequently criticized for being unable to extend the midline incision to the tip of the glans beyond the urethral plate because of the possibility of scarring and consequent MS. According to proponents of the GTIP procedure, grafting distal incised glans and urethral plate enables the meatus to be appropriately positioned with a lower chance of MS. A case series of GTIP of 230 patients by Ahmed and Alsaid26 reported a UCF incidence of 3.91%, with no patients developing MS and urethral diverticulum in the postoperative period. A similar result was reported by Gupta et al27 for 263 cases of GTIP with a 3.7% incidence of UCF with no reported cases of MS. Our meta-analysis findings regarding MS corroborate the two studies mentioned above; as there is a significantly reduced incidence of MS in the GTIP group supported also by TSA.

Our meta-analysis also yields a non-significant result for the outcome measure UCF between the TIP and GTIP groups. Similar results were also reported by retrospective evaluation by Shuzhu et al28 and almost all other studies included in our quantitative analysis. This meta-analysis finding has been seconded by our TSA with the Z curve, although in the area of GTIP, it has not crossed conventional boundaries and has not reached the RIS. In our quantitative analysis, the outcome variable of operative time was less in the TIP group than in the GTIP group. The extra time taken for graft harvesting and fixing at the site of TIP usually accounts for an increase in the total duration of the procedure, which is the possible reason behind the prolonged operative time in the GTIP group. Although GTIP is a more time-consuming procedure, there was no difference in the cosmetic HOSE score between the groups. GD is also one of the complications that has been reported by a few authors included in our study and by Snodgrass and Bush.22 The difference between the groups for this outcome was not significant.