Everted saphenous vein graft for long anterior urethral strictures in men with tobacco-exposed oral mucosa: A prospective nonrandomized study.

INTRODUCTION: Oral mucosal graft (OMG) is the gold standard for urethral substitution but has poor results in long anterior urethral strictures and chronic tobacco-exposed oral mucosa. Saphenous vein has been recently described for long-segment anterior urethral stricture with successful initial results. Our objective was to compare the early outcomes of everted saphenous vein graft (eSVG) substitution urethroplasty in patients with tobacco-exposed oral mucosa and OMG urethroplasty in patients with nontobacco-exposed oral mucosa for long anterior urethral strictures.
MATERIALS AND METHODS: 30 patients with long anterior urethral strictures underwent substitution urethroplasty using Dorsolateral onlay approach. Fifteen patients with healthy oral mucosa underwent OMG urethroplasty (Group 1) and 15 patients who had unhealthy oral mucosa due to chronic tobacco exposure underwent eSVG urethroplasty (Group 2). Outcomes were assessed with the International Prostate Symptom Score (IPSS); uroflowmetry; donor and recipient site complications at 1, 3, and 6 months; and symptomatic assessment thereafter. Retrograde urethrogram was done at 3 months in both the groups. Successful urethroplasty was defined as satisfactory voiding (Qmax>15 ml/s) and no need for endoscopic dilatation/direct vision internal urethrotomy in follow-up.
RESULTS: Mean stricture and harvested graft length were 10.8 cm and 12.33 cm in Group 1, while in Group 2 were 13.6 cm and 15.73 cm, respectively. Nine of 13 patients in Group 1 (69.2%) and 11 of 14 in Group 2 (78.5%) with a minimum follow-up till 18 months had successful outcome at an average follow-up of 23.13 months. Donor and recipient site complications were comparable in both the groups. At 18 months, mean IPSS and Qmax in successful patients were 7.9 and 25.6 ml/s in Group 1, while in Group 2 were 8.0 and 22.6 ml/s.
CONCLUSIONS: Outcomes of great saphenous vein graft urethroplasty are comparable to OMG, and it is an acceptable option in long-segment anterior urethral stricture patients with chronic tobacco-exposed oral mucosa.

INTRODUCTION

Oral mucosa graft (OMG) is the graft of choice for all urethral substitutions and has been considered as the gold standard graft. Recently, concerns have been raised about the factors responsible for poorer outcomes after buccal graft anterior urethroplasty which include long-segment urethral strictures[12] and tobacco-exposed oral mucosa.[3] There is thus a need to search for alternate grafts for substitution in such scenario.

After a report of 10 urethroplasties by el-Morsi[4] in 1972, saphenous vein as a graft in stricture has been revived after a few favorable reports in hypospadias.[567] Kim et al.[8] used autologous saphenous vein graft (SVG) in surgically induced stricture in a rabbit model with good outcomes at 3 months. Similarly, in another study on rabbit model, Xu et al.[9] demonstrate good outcome of survival of graft as dorsal onlay patch with everted SVG (eSVG). In a prospective human study, Akhtar et al.[10] described a successful initial experience with eSVG urethroplasty (eSVGU) in 17 patients with long anterior strictures and tobacco-exposed oral mucosa.

In this study, we compared the outcomes of eSVG substitution urethroplasty in patients with tobacco-exposed oral mucosa and OMG urethroplasty in patients with nontobacco-exposed oral mucosa for long anterior urethral strictures.

MATERIALS AND METHODS

Patients with long anterior urethral stricture (6 cm or more) were assigned to either of the two groups based on their tobacco exposure in this prospective nonrandomized comparative study between November 1, 2015 and January 31, 2017. All patients without a history of tobacco abuse and good oral hygiene/healthy oral mucosa were subjected to OMG urethroplasty (Group 1), whereas those with a history of tobacco abuse and poor oral hygiene/unhealthy oral mucosa were subjected to eSVGU (Group 2). Patients with near-obliterative strictures (urethral caliber <6 Fr), age more than 75 years, presence of two separate strictures requiring two separate grafts, and any previous history of urethroplasty were excluded from the study. In addition, patients with a history of deep vein thrombosis or venous insufficiency in both the limbs were excluded for saphenous vein use. The recruitment stopped when each group completed 15 patients.

Apart from routine assessment, all patients had an examination of the oral cavity, International Prostate Symptom Score (IPSS), uroflowmetry, and postvoided residual urine (PVR) estimation. Retrograde urethrogram (RGU) was done to delineate the extent and size of stricture segment. Color Doppler ultrasound of both lower limbs was done in each patient planned for SVG urethroplasty to check for patency and competence of superficial and deep venous system.

Urethroplasties were performed under general anesthesia through endotracheal intubation in OMG group and under regional anesthesia in eSVGU group in the lithotomy position. Penile invagination with one-sided urethral mobilization (Kulkarni technique)[11] was employed from the ventral midline to beyond the midline on the dorsal aspect. Urethra was opened in the midline dorsally, and length of stricture segment including the caliber of urethra was measured. A two-team approach was used for both the groups. For oral mucosa harvesting, a suitable sized oral mucosa graft was harvested from one or both the cheeks if needed. Extension toward lip was done to gain the length if required staying away from the angle of the mouth. Wound was left raw for spontaneous healing after adequate hemostasis. Lingual graft was not used in any of the patients. For saphenous vein harvesting, multiple island incisions over premarked course of saphenous vein starting from saphenofemoral junction were given[10] in the same lithotomy position as for urethroplasty [Figure 1a]. Using “no touch” technique, a suitable sized saphenous venous graft was retrieved using tunneling between the small island incisions in the upper thigh with ligation of even small tributaries avoiding cautery use [Figures 1b, c and 2a]. After harvesting, the vein was submerged in a bowl of 100 ml of saline with 30 mg of papaverine. The vein was occluded from one end and hydrodistended using the same solution before detubularization along its entire length [Figure 2b and c]. It was then used as an everted graft (endothelium facing outside the lumen).

Figure 1

(a) Course of saphenous vein premarked with color Doppler with planned sites of incisions (b) Vein dissected at the saphenofemoral junction. Any tributaries are tied and divided (not divided by energy sources) to facilitate further hydrodistention (c) The vein is divided cranially and tunneled into caudal incisions depending on the length required

Figure 2

(a) Long segment of the saphenous vein is retrieved through three island incisions and is ready to be divided. All tributaries are ligated. The penis has been inverted into the perineum, urethra dissected on one side and opened dorsally, (b) the vein is hydrodistended, and (c) the vein is then opened all along its length to be used as graft with endothelium kept on the outside

Respective graft was used as dorsolateral onlay patch. Graft was sutured proximally to urethral end by few interrupted sutures at the apex, and the medial margin of the urethra using 4/0 polyglactin 910. One edge of the harvested vein graft was sutured to urethral mucosal margin using continuous sutures [Figure 3]. A 16 Fr Foley catheter was placed. The graft was fixed to the underlying corpora with a running suture in the middle along the whole length of the graft, while the other edge of the graft was sutured to the margin of the spongiosum. In patients with meatal involvement, a generous dorsal meatotomy was given and one end of the graft was sutured at the meatus dorsally, to the raw area thus created, before being fed in the perineal wound through the meatus to be continued as dorsolateral onlay. A perineal drain (vacuum suction drain, 14 Fr) was kept in all patients and the wound was closed in layers using interrupted 3/0 Polyglactin 910.

Figure 3

The vein has been placed as a dorsal graft and has been sutured to one edge of the strictured urethra

Immediate postoperative complications, if any, along with wound complications in perineum and donor site were noted. Visual analog scale (VAS) for pain was measured in each group postoperatively on day 1, 2, and 3. Catheter was removed at 3 weeks in all patients without doing a periurethral contrast study. Symptom assessment using IPSS and uroflowmetry with PVR was done in all patients at 1, 3, and 6 months and with 6-monthly symptoms assessment thereafter. Response to quality of life (QOL) question of the IPSS was also recorded on each occasion. RGU was done in all patients at 3 months. Endoscopic dilatation or direct vision internal urethrotomy (DVIU) was done for symptomatic narrowing on follow-up. Successful urethroplasty was defined as satisfactory voiding (Qmax>15 ml/s) and no need for endoscopic dilatation/DVIU or any other auxiliary procedures in follow-up.

Data from the patients were analyzed using Statistical Package for the Social Sciences software version 21.0 Categorical variables were presented in number and percentage (%), and continuous variables were presented as mean ± standard deviation and median. Normality of data was tested using the Kolmogorov–Smirnov test. If the normality was rejected, nonparametric test was used. Quantitative variables were compared using the unpaired t-test/Mann–Whitney test (when the data sets were not normally distributed) between the two groups. Qualitative variables were correlated using the Chi-square test/Fisher's exact test. P < 0.05 was considered statistically significant.

RESULTS

Thirty patients were taken up for urethroplasty in this period. Demography and preoperative characteristics of patients in both the groups are summarized in Table 1, and intraoperative details are summarized in Table 2. The etiologies were infectious urethritis related – 10 (5 in each group), Lichen Sclerosus (LS) related – 4 (2 in each group), iatrogenic catheter related – 9 (5 in OMG Group 1 and 4 in SVG Group 2), iatrogenic post-TUR – 2 (both in OMG Group 1), and idiopathic – 5 (1 in Group 1 and 4 in Group 2).

Table 1

Demography and preoperative characteristics of patients in both the groups

Variables	Group 1 (OMG) (n=15)		Group 2 (eSVG) n=15			P
	Mean	Range	Mean	Range
Age (years)	37.6±17	16-76	46.73±15.82	20-66	0.949
Duration of symptoms (years)	5.63±4.9	1-20	5.43±5.81	1-25	0.769
Serum creatinine	0.84±0.28	0.5-1.6	0.96±0.36	0.7-2.1	0.171
IPSS	24±2.83	20-28	24.91±2.43	20-28	0.646
QOL measurement	5.67±0.62	4-6	5.6±0.63	4-6	0.718
Qmax (ml/s)	6.44±3.37 (n=9; 6 were on SPC)	2.5-13.3	7.22±3.21 (n=11; 4 were on SPC)	2.9-12.5	0.797
PVR (ml)	100.56±83.01	15-260	82.18±69.21	10-230	0.879
Narrow meatus (n)	3		4		1.00
Previous dilatation (n)	5		5		0.435
Previous DVIU (n)	4		4		0.766
History of catheterization (n)	12		11		1.000
Urethral discharge (n)	5		5		1.000
Trauma (n)	2		0		0.483

OMG: Oral mucosal graft, eSVG: Everted saphenous vein graft, PVR: Postvoided residual urine, DVIU: Direct vision internal urethrotomy, SPC: Suprapubic Cystosctomy, IPSS: International Prostate Symptom Score , QOL: Quality of life

Table 2

Intraoperative details

Variables	Group 1 (OMG)			Group 2 (eSVG)			P
	Maximum	Minimum	Mean±SD	Maximum	Minimum	Mean±SD
Length of stricture (cm)	18	6	10.8±3.21	18	9	13.6±2.59	0.015
Caliber of urethra in stricture segment (Fr)	10	6	6.8±1.26	8	6	6.8±1.01	0.825
Length of graft harvested (cm)	19	7	12.33±3.31	20	11	15.73±2.6	0.007
Operative time (min)	210	150	192±18.21	240	150	194.6±21.34	0.697
Graft harvesting time (min)	40	32	35.73±2.58	40	30	35.67±2.85	0.340

OMG: Oral mucosal graft, eSVG: Everted saphenous vein graft, SD: Standard deviation

Fifteen patients underwent OMG urethroplasty (Group 1) and equal number underwent eSVGU (Group 2).

Out of 15 patients from Group 1 (OMG group), three patients failed at 3 months (symptomatic narrowing seen on RGU and all required DVIU) and one patient failed (required DVIU) after 12 months. Later on, out of these four failed urethroplasties, two patients required redo-urethroplasty using lingual and lower lip mucosa, respectively, while the other two are on self-intermittent calibration. Two patients were lost to follow-up after 12 months.

Out of 15 patients from Group 2 (eSVG group), two patients failed at 3 months (short stricture requiring DVIU in one and diffuse narrowing with complete failure in another requiring staged urethroplasty) and one patient failed after 12 months (required DVIU). Both the patients who required DVIU are on self-intermittent calibration. One patient stopped reporting for follow-up after 12 months.

No significant difference in VAS was noted in both the groups (P: 0.446, 0.150, and 0.31 on days 1, 2, and 3, respectively). In Group 1, two patients had slight decrease in mouth opening with decrease of sensation, the symptoms improved at 1 month and disappeared at 3 months. In Group 2, four patients developed donor-site wound infection; all patients were managed conservatively. Two patients in Group 1 developed complications in follow-up (one patient developed a small urethrocutaneous fistula after per urethral catheter removal, for which catheter was kept for further 2 weeks till fistula closed spontaneously and another patient developed a paraurethral abscess treated with incision and drainage with further 2 weeks per urethral catheter drainage), both recovered without any sequelae. Mean duration of follow-up was 21.31 ± 3.42 months (range: 18–30 months) in Group 1, while it was 24.93 ± 4.01 month (range: 20–33 months) in Group 2 patients, respectively. The IPSS, QOL score, Qmax, and PVR were not significantly different in the two groups [Table 3].

Table 3

Postoperative details

Variables	Group 1 (OMG)		Group 2 (eSVG)		P
	Mean	Range	Mean	Range
At 1 month (n)	15		15
IPSS	7.87±0.99	7-10	8.87±2	7-14	0.119
QOL	1.33±0.49	1-2	1.53±0.74	1-3	0.541
Qmax (ml/s)	25.33±5.09	18.5-34.4	22.11±5.19	12.2-28.8	0.940
PVR	29.87±20	10-78	30.07±19.26	10-85	0.933
At 3 month (n)	15		15
IPSS	9.07±2.96	7-18	10.13±4.55	7-22	0.378
QOL	1.67±0.72	1-3	1.87±0.99	1-4	0.700
Qmax	22.63±7.18	12.2-36.6	19.65±5.72	7.8-28.8	0.345
PVR	40.47±36.7	10-150	45.0±28.92	10-110	0.212
At 6 month (n)	12 (3 failure)		13 (2 failure)
IPSS	8.33±1.97	7-14	8.62±1.45	7-12	0.68
QOL	1.5±0.67	1-3	1.61±0.65	1-3	0.68
Qmax	24.80±5.62	16.2-32.6	21.32±3.22	16.4-27.2	0.07
PVR	26.50±11.61	10-50	27.92±10.95	15-60	0.76
At 12 months (n)	12 (3 failure)		13 (2 failure)
IPSS	8.75±3.02	7-18	8.76±2.86	7-18	0.99
QOL	1.83±0.72	1-3	1.53±0.88	1-4	0.36
Qmax	23.90±4.66	13.4-29.4	22.6±4.25	12.4-28.4	0.47
PVR	25.5±13.56	10-60	29.23±13.82	20-70	0.50
At 18 months (n)	9 (2 patients lost to follow-up and 4 failures)		11 (1 patient lost to follow-up and 3 failures)
IPSS	7.88±0.93	7-10	8.0±0.77	7-10	0.75
QOL	1.44±0.53	1-2	1.36±0.49	1-2	0.73
Qmax	25.69±3.30	20.8-30.6	22.65±3.35	17.8-28.2	0.06
PVR	20.44±6.77	10-30	24.53±6.88	15-35	0.20

n: Number of patients, OMG: Oral mucosal graft, eSVG: Everted saphenous vein graft, PVR: Postvoided residual urine, IPSS: International Prostate Symptom Score , QOL: Quality of life

Five failures occurred within the first 6 months of follow-up (3 from Group 1 and 2 from Group 2) and two occurred after 12 months (one in each group).

DISCUSSION

This study aimed to compare the initial outcomes of eSVG and buccal mucosal graft (BMG) urethroplasty. We found that long length of saphenous vein had an advantage over OMG because a large saphenous graft with adequate caliber can be harvested from a single side as compared to OMG which needed to be retrieved from multiple sites (bilateral buccal and lower lip mucosal graft in three patients and bilateral buccal in nine patients). Another advantage of SVG over OMG is that it does not need multiple suture lines to join the two graft for single-staged urethroplasty in long-segment urethral stricture patients compared to OMG urethroplasty which needs multiple sutures line for the same. In this study, the maximum length of OMG harvested was 19 cm which needed multiple suture lines, while maximum length of harvested saphenous graft was 20 cm which needed a single suture line for urethroplasty. Suture lines joining the grafts are source of focal narrowing as are anastomotic ends.[12] It has been suggested that the length and the caliber of saphenous vein are very much adaptive and so a new urethra could be constructed easily as per the stretched length of the penis.[5]

We employed two-surgeon team techniques for urethroplasty; one team was harvesting OMG or SVG while another team was working for urethral mobilization. The mean duration of surgery in both the groups was comparable (192 and 194 min) and statistically insignificant (P = 0.697). The mean duration of OMG and SVG harvesting was also comparable (35.67 and 35.73 min) and statistically insignificant (P = 0.340) in both the groups. We employed dorsolateral onlay technique as it helps to preserve urethral vascularity. None of the patients developed penile lateral curvature or chordee in both the groups.

Four patients (26%) from SVG group (Group 2) developed wound infection, all patients were managed conservatively. Belczak et al.[13] in their study on clinical complications of limbs undergoing harvesting of great saphenous vein for coronary artery bypass grafting, wound infection (25%), edema (52,3%), paraesthesia (29.5%), erysipelas (9.1%), lymphorrhea (4.5%), and deep vein thrombosis (2.3%). None of our patients developed such complications.

In this study, eSVG group patients had worse strictures which were significantly longer with mean stricture length more than 10 cm as compared to Group 1 patients. At an average follow-up of 23.13 months, Group 1 had four failures with stricture lengths of 10, 13, 14, and 18 cm while Group 2 had three failures with stricture lengths of 12, 15, and 15 cm, respectively.

It has been studied that stricture length is an important factor to prognosticate results and to determine operative methods. Chen et al.[14] in their study in 40 patients with anterior urethral stricture treated with dorsal onlay BMG urethroplasty suggested that stricture length and localization are the most important variables for desirable success and also reported 82% success rate for patients with stricture length >4 cm and 76% for stricture length ≥6 cm. According to Yalcinkaya et al.,[2] length of stricture dramatically and statistically affects the outcomes of urethroplasty, and using BMG, they also reported 88% success rate for stricture length ≤7 cm and only 40% success rate for stricture length >7 cm.

SVG came as an alternative because of its sufficient length and easy availability. Hair is not a problem with vein graft, its multilayer structure provide robust graft, less chances of ballooning/fistula formation and rapid blood supply due to thin wall, it also provides elasticity through its muscle coat, and elastic fibers which help in preventing penile curvature during erection.[15] el-Morsi is credited with the first use of SVG for urethral reconstruction in stricture,[4] but the urethroplasty literature was later dominated by studies on OMG use. Later, successful outcome with saphenous vein in complex hypospadias failures[567] helped revival of this forgotten graft for strictures.

Due to valves leading to initial poor results with tubular substitution of the urethra with saphenous vein, Hübner et al.[16] first used vein in an everted fashion as a tube in an experimental study on dogs and found neourethral tube to be completely lined with urothelium. Other groups from Iran and Turkey showed good results from both noneverted and everted jugular venous tubes in experimental studies on rabbits but demonstrated that the graft stays longer when used in everted fashion.[1718] Akhtar et al.[10] used everted vein graft as dorsolateral onlay graft in 17 men with long strictures with 82% success at 6 months and biopsies of mucosal ingrowths during the study had confirmed ingrowth of uroepithelium. The present cohort was different from this study.

Recently, an experimental study on rabbit model by a Korean group has shown 1-year successful result with everted vein graft and also concluded that vein is more resistant to inflammation and prevents extravasation of urine and thus provides a framework for uroepithelialization.[9]

The mechanism for vein graft is different from that of a buccal graft, whereas the buccal graft “takes” and the junction between the graft and native urethral mucosa is well identifiable at 6 weeks in animals[19] and after many months in humans,[20] the vein graft on the other hand disintegrates except for a fibrous matrix which allows growth of highly organized urothelium over it by 3 months which becomes multilayered with well-formed collagen fibers at 1 year and becomes indistinguishable from the rest of the mucosa.[9]

About 50% of adult Indian men consume tobacco with the incidence of smokeless (chewed) being five times that of smoked.[21] Similar incidence reflects in preoperative characteristics of men undergoing urethroplasty in a study from Northern India (58.3% men were tobacco users),[3] and this may be representing the condition from most of the Asian population, making our study relevant.

Apart from tobacco exposure, the risk of inadvertent human papillomavirus transmission with OMG urethroplasty gained attention when a study from a dental department in Finland was published in a renowned urology journal[22] and was commented on in the editorial with a cry for searching alternatives to oral grafts. Although OMG has been considered an ideal substitute for urethroplasty for the last 25 years, few reported incidences of malignancy in the graft deserve attention.[2324]

From this study, we found that outcomes of everted saphenous vein graft urethroplasty are comparable to OMG urethroplasty, and it is a viable option in patients who are chronic tobacco chewer/smoker having unhealthy buccal mucosa in which the use of BMG is at risk. It is also an insightful alternative for lengthy urethral stricture. In BMG urethroplasty, multiple grafts require multiple anastomoses, leading to anastomotic site narrowing and more donor-site morbidity. It provides an option in patients with submucosal fibrosis and scarred or unhealthy mucosa due to chronic substance abuse. Harvesting SVG under regional anesthesia showed advantages that early oral intake for both solids and liquids could be allowed and that long grafts even for panurethral strictures could be harvested in regional anesthesia.

The limitations of our study are selection bias, small sample size, and a shorter duration of follow-up. Therefore, to establish this study, it needs further studies with a larger sample size for comparison and longer duration of follow-up as more recurrences are expected as the follow-up increases. A follow-up study of the same cohort with longer follow-up might bring more clarity as to how does the continuous exposure to urine affect the outcome. However, as the published data were not strong enough to encourage us to randomize patients to each arm, our encouraging and comparable results may embolden further researchers to use eSVG as a primary choice for long anterior strictures and conduct randomized studies to further test it as a substitution material. The study forms part of exploration phase (Phase 2b) of IDEAL recommendations for adoption of a surgical innovation.[25]

CONCLUSIONS

Everted saphenous vein is an acceptable alternative to OMG in patients with long-segment anterior urethral stricture with chronic tobacco-exposed oral mucosa. The early outcomes of eSVGU appear comparable to OMG urethroplasty with promising result in long anterior urethral strictures.