Keywords urinary incontinence - graciloplasty - pudendal nerve - microsurgery
Introduction
Male urinary incontinence is a common complication of radical prostatectomy (RP) with
a significant impact on the patient's quality of life. For total urinary incontinence,
an artificial urinary sphincter (AUS) is implanted in the bulbous urethra, and patients
obtain urinary continence by manually adjusting the pressure.[1 ]
[2 ] Although effective, the device causes several complications, such as infections,
erosion, mechanical failure, and urethral atrophy. The reoperation rate is 26.0% (15.0–45.0%).[2 ] Furthermore, the device is not physiological, because it takes several minutes to
operate and there is no indication that the device will operate when cognitive function
declines. The male sling procedure, synthetic mesh-like surgical tape is positioned
around the bulbous urethra, is an alternative to AUS. This procedure is minimally
invasive and has few complications, but its long-term data is not sufficient to prove
whether it is as effective as the AUS.[3 ]
The possibility of using the patient's own muscle flaps has been considered to reconstruct
a neosphincter to overcome the limitations of the AUS procedure. In 1926, Deming[4 ] first reported graciloplasty to treat urinary incontinence. In 1956, Pickrell et
al[5 ] reported six patients in whom the distal part of the gracilis muscle was transposed
around the bladder neck and urethra; however, this procedure did not gain widespread
clinical acceptance. Williams et al[6 ] and Janknegt et al[7 ]
[8 ] demonstrated dynamic graciloplasty, in which an electrical stimulating device was
implanted to transform the gracilis muscle from a fatigue-prone to a fatigue-resistant
muscle. However, clinical outcomes were disappointing because the urethra was wrapped
around with the distal portion of the gracilis muscle, which was ischemic with a narrow
width, without abundant contractile force, and sufficient volume.[8 ]
[9 ] Previously, we demonstrated the anatomic basis of anal sphincter reconstruction
via dynamic graciloplasty with pudendal nerve anastomosis and verified the feasibility
of this procedure.[10 ] Additionally, wrapping the urethra with a muscle flap increased both the maximum
urethral closure pressure (MUCP) at rest, and electrical stimulation further increased
the MUCP in a male rabbit model.[11 ] Guo et al[12 ] reported the efficacy of adynamic gracilis urethral myoplasty with a pedicled gracilis
muscle flap wrapped around bulbar urethra for male patients with mild to moderate
urinary incontinence. However, like the male sling, it is a static reconstruction
and is ineffective for patients with severe incontinence. In this study, we proposed
a new, more physiological procedure in which the well-vascularized belly at the middle
part of the gracilis muscle is wrapped around the bulbous urethra and its motor nerve
is sutured to the pudendal nerve, which originally controls the external urethral
sphincter (EUS). We performed a simulation surgery on three fresh cadavers and a clinical
study of a patient with severe postprostatectomy urinary incontinence in order to
determine the feasibility of this procedure.
Idea
The surgical procedure was simulated on three male cadavers at the Clinical Anatomy
Laboratory at the Keio University School of Medicine, Tokyo, Japan, complying with
the Guidelines for Cadaver Dissection in Education and Research of Clinical Medicine.
The procedure was approved by the Japan Surgical Society and the Japanese Association
of Anatomists.
Surgical Procedure
The cadavers were positioned in the lithotomy position. A skin incision was made at
the medial thigh, parallel to the long axis of the gracilis muscle. The origin of
the muscle at the pubis was detached, and the dominant pedicle arising from the profunda
femoris artery into its deep surface at the junction of the upper one-third with the
lower two-thirds was preserved for the vascular pedicle ([Fig. 1A ]). The motor nerve of the muscle was dissected proximally up to the bisection with
the obturator nerve, and the proximal end was severed. If the nerve length was inadequate
for subsequent nerve coaptation to the pudendal nerve without tension, it was further
lengthened by dissecting distally into the muscle belly.[10 ] Both ends of the gracilis muscle were isolated only on its vascular pedicle, and
the total muscle length, muscle width, distance from the point of the vascular pedicle
into the gracilis muscle to the pubic symphysis, obturator nerve length, and diameter
were measured. Through a perineal mid-incision, periurethral dissection was circumferentially
performed at the level of the bulbous urethra, allowing urethral mobilization and
muscle wrapping. After the excess parts at both the ends of the gracilis muscle were
trimmed, its proximal end was subcutaneously tunneled without tension into the perineum,
passed under the bulbous urethra, folded back, and sutured to the ipsilateral inferior
ramus of the pubis ([Figs. 1B ], [2A ], and [2B ]). Next, the distal end of the gracilis muscle was folded over its proximal side
and sutured to the contralateral inferior ramus of the pubis ([Figs. 1C ], [2A ], and [2C ]). Consequently, urethral graciloplasty was achieved in a γ-loop configuration, which
could be used with the thicker and well-vascularized belly in the middle part of the
gracilis muscle.
Fig. 1 Cadaveric simulation of urethral gracioplasty reinnervated by the pudendal nerve.
(A ) In the lithotomy position, both origin and insertion of the gracilis muscle were
severed, with preservation of the dominant vascular pedicle. The motor nerve of the
muscle was dissected, and the proximal end was severed. (B , C ) The proximal end of the gracilis muscle was subcutaneously tunneled into the perineum,
passed under the bulbous urethra, folded back, and sutured to the ipsilateral inferior
ramus of the pubis. Next, the distal end of the muscle was folded over its proximal
side, achieving urethral graciloplasty in a γ-loop configuration. (D ) Through an ipsilateral perineal incision, extrapelvic branches of the pudendal nerve
with pudendal vessels were exposed and preserved for the donor nerve.
Fig. 2 Operative schema. (A ) The proximal end of the motor nerve into the gracilis muscle was severed, and the
proximal end of the muscle was folded back under the bulbous urethra. (B ) After the excess part at both ends of the gracilis muscle was trimmed, its proximal
end was sutured to the ipsilateral inferior ramus of the pubis. (C ) The distal end of the gracilis muscle was folded over its proximal side and sutured
to the contralateral inferior ramus of the pubis.
Somatic motor innervation of the EUS arises from the cell bodies of sacral spinal
cord segments S2–S4 and is organized in Onuf's nucleus. The nerves to the EUS can
be classified as (1) extrapelvic branches of the pudendal nerve, (2) intrapelvic branches
of the pudendal nerve, or (3) branches of the inferior hypogastric plexus.[13 ] The extrapelvic branches travel through the greater sciatic notch to course anteriorly
within Alcock's canal, then follow a constant course ventromedially, and several centimeters
distally, give branches to the EUS and the superficial transverse perineal, bulbospongiosus,
and ischiocavernosus muscles and to the scrotal and ventral penile skin.[13 ]
[14 ] Through an ipsilateral perineal incision, these extrapelvic branches with pudendal
vessels were exposed at their entrance to Alcock's canal and preserved for the donor
nerve ([Fig. 1D ]). The gracilis muscle flap was subcutaneously tunneled into the perineal region,
and we confirmed whether it could be sutured with the pudendal nerve, tension-free
in an end-to-side fashion.
Results
The gracilis muscle was easily transferred to the periurethral region by detaching
both its origin and insertion and leaving only the vascular pedicle. The average length
of the muscle belly was 291.7 (range 260–320) mm, and the average muscle width was
350 (range 330–370) mm ([Table 1 ]). The vascular pedicle entered the muscle 76.7 (range 70–85) cm distal to the symphysis.
After dissection and transection of the proximal end of the pudendal nerve at its
bisection from the obturator nerve, its average length was 107.7 (range 95–123) mm
and average diameter 2.5 (range 2–3) mm. The extrapelvic branch of the pudendal nerve
was constantly confirmed, and its mean diameter was 2.7 (range 2.5–3.0) mm. In all
simulations, the gracilis muscle was subcutaneously tunneled without tension into
the perineum and could be wrapped around the bulbous urethra in a γ-loop configuration.
The motor nerve of the muscles was sufficient for direct neurorrhaphy with the pudendal
nerve, without requiring graft interposition.
Table 1
Cadaveric study: patient characteristics and surgical parameters
No.
Age (y)
Sex
Size of the gracilis muscle (mm)
Muscle length from the symphysis to vascular pedicle (mm)
Length of the obturator nerve (mm)
Diameter of the obturator nerve (mm)
Diameter of the pudendal nerve (mm)
The possibility of nerve coaptation
1
85
Male
37 × 260
70
95
3
2.5
Possible
2
88
Male
35 × 320
85
105
2.5
3
Possible
3
97
Male
33 × 295
75
123
2
2.5
Possible
Case
A 71-year-old male patient presented with severe postprostatectomy urinary incontinence
defined as a positive 24-hour pad test, which was 670 g (> 400 g). Physical therapy
was ineffective, so surgery was indicated. The patient did not meet any of the following
exclusion criteria: previous pelvic radiation therapy, incomplete bladder emptying,
maximum flow urinary rate < 12 mL/s, detrusor overactivity, bladder capacity < 300 mL,
and urethral stricture.
Urodynamic examinations were performed before surgery. The intraurethral pressure
was measured three times using a pressure sensor unit (GMMS-600; Star Medical Co.,
Ltd., Tokyo, Japan) and a 14-French (Fr) pressure transducer (Unitip catheter; Unisensor
AG, Attikon, Switzerland). The bladder was emptied. A catheter connected to the pressure
sensor system was placed in the bladder 18 cm deep from the external meatus. The sensor
was set to 0, and the catheter was withdrawn at 1 mm/s while recording the urethral
pressure. The MUCP was defined as intrabladder pressure–urethral pressure and was
set at 0 cm H2 O. Preoperative urodynamic examination revealed an MUCP of 40.7 cm H2 O and a functional profile length (FPL) of 40.1 mm ([Table 2 ]).
Table 2
Comparison of pre- and postoperative urodynamic values
Variables
Preoperative values
Postoperative values
MUCP, mean (range)
46.1 ± 4.0 (42.5 − 50.5)
70 ± 2.29 (68 − 72.5)
FPL, mean (range)
40.1 ± 1.67 (38.8 − 42)
45.3 ± 1.59 (43.5 − 46.5)
Abbreviations: FPL, functional profile length (mm); MUCP, maximum urethral closure
pressure (cm H2 O).
The surgical procedure was performed under general anesthesia. Two skin incisions
were made at the medial thigh, parallel to the long axis of the gracilis muscle in
the lithotomy position ([Fig. 3A ]). The origin of the muscle at the pubis was detached with a monopolar electric scalpel,
and the proximal dominant vascular pedicle was preserved ([Fig. 3B ]). The motor nerve of the muscle was dissected, and the proximal end was severed
with a length of 12.5 cm. Next, the distal end of the muscle was detached, and minor
vascular pedicles were ligated. The length of the muscle belly of the harvested muscle
was 250 mm, and the maximal muscle width was 55 mm.
Fig. 3 Surgical procedure used in the clinical case. (A ) A preoperative incision line was made at the medial thigh and perineal region in
the lithotomy position. (B ) The origin and insertion of the gracilis muscle were detached, with preservation
of the proximal dominant vascular pedicle. The motor nerve of the muscle was dissected,
and the proximal end was severed. (C ) The bulbous urethra was isolated and exposed circumferentially through a perineal
mid-incision. (D ) The gracilis muscle was subcutaneously tunneled into the perineum and trimmed to
a width of 45 mm. (E ) The gracilis muscle was wrapped around the bulbous urethra in a γ-loop configuration.
(F ) The pudendal nerve was bluntly dissected through the ipsilateral perineal incision.
(G ) The motor nerve of the gracilis muscle was subcutaneously tunneled into the perineal
region and directly sutured with the pudendal nerve in an end-to-side fashion using
a surgical microscope. (H ) Image after completion of surgery.
A 14 Fr Foley catheter was introduced into the bladder, and a perineal mid-incision
was made. The bulbous urethra was isolated and exposed circumferentially ([Fig. 3C ]), and the gracilis muscle was subcutaneously tunneled into the perineum, trimmed
to a 45-mm width ([Fig. 3D ]) and wrapped around the bulbous urethra in a γ-loop configuration ([Fig. 3E ]).
After the excess parts at both the ends of the gracilis muscle were trimmed, the muscle
was secured to each pubic ligament with horizontal mattress sutures, the tightening
pressure of the muscle was adjusted to the extent that repeated urethral pressure
reached 80 to 90 cm H2 O, and 400 mL of saline injected into the bladder was leaked by manual compression
of the abdominal wall via intraoperative retrograde cystourethrography.
The pudendal nerve was bluntly dissected through the ipsilateral perineal incision
([Fig. 3F ]). The motor nerve of the gracilis muscle was subcutaneously tunneled into the perineal
region and directly sutured with the pudendal nerve, tension-free in an end-to-side
fashion using a surgical microscope ([Fig. 3G ]). Negative pressure drains were placed on the subcutaneous layer of the proximal
thigh and perineal region, and the wound was closed ([Fig. 3H ]).
The postoperative course was uneventful, without infection, muscle necrosis. The urinary
catheter was removed 5 days after surgery, and no obvious urinary retention was observed.
In order to convert to a new sphincter, rehabilitation was performed to tighten the
urethral or anal sphincter.
The 24-hour pad test improved to 285 g (58% decrease) at 12 months postoperatively.
On urodynamic examination, the mean MUCP and FPL were 70 cm H2 O and 45.3 mm, respectively ([Table 2 ]). The preoperative urethral pressure profile showed a monomodal wave ([Fig. 4A ]), consistent with the native EUS, while the postoperative profile showed a bimodal
wave, with the second wave at the wrapped gracilis muscle ([Fig. 4B ]). Although urinary incontinence was not completely cured, the patient could almost
maintain urinary continence at night, leading to a reduction in the number of pads
used.
Fig. 4 Pre- and postoperative urethral pressure profile. (A ) Before surgery and (B ) 12 months postoperatively.
Discussion
Dysfunction of the EUS or urethral sphincter muscle is a major cause of urinary incontinence
after RP.[15 ]
[16 ] Several studies showed a significant decrease of the MUCP and FPL related to urethral
sphincteric function after RP.[17 ] In these studies, mean MUCP at baseline was 73 cm H2 O (range 49–95 cm H2 O) and mean MUCP after RP was 56 cm H2 O (range 30–83 cm H2 O). Mean FPL at baseline was 5.0 cm (range 4.3–6.1) and mean FPL after RP was 2.6 cm
(range 1.6–3.1). Therefore, an accurate understanding of the urethral sphincter structure
and innervation is important for sphincter reconstruction.
The EUS is located at the lower part of the urethra and forms a muscular coat ventral
and lateral to the membranous urethra and prostate, the core of which is an omega-shaped
loop around the membranous urethra.[18 ] Murakami et al[19 ] reported that the thick, lateral portion of the EUS retracts the urethra backward
and upwards with the aid of the levator ani. In the elderly, sphincteric action is
weak or incomplete, so the EUS maintains urinary continence by retracting the urethra
backward and upwards with the aid of the levator sling, unlike the real sphincteric
action in younger men. Burnet and Mostwin[20 ] described that urethral closure is achieved by the urethral sphincteric complex
comprising the cylindrical EUS, ventral subpubic fascia, and medial fascia of the
levator ani muscle. The action of the EUS is to draw the complex upwards and laterally
along the inner undersurface of the pubis. Therefore, muscle fiber shortening at sphincteric
contraction allows urinary continence, whereas elongation at sphincteric relaxation
permits micturition.
In graciloplasty for fecal incontinence, various loops (e.g., γ, υ, c, α, and ε) can
be used to wrap the gracilis muscle around the anus.[21 ] The classical γ-loop is often used, but it raises the issue of insufficient contraction
because the distal part that is actually wrapped around the anus is thinner and has
poor vascularity. Our modified γ-loop not only achieves sphincteric contraction but
can also have the effect of lifting the urethra upwards by fixing both ends of the
gracilis muscle to the pubic ligament, thereby resulting in a more physiological function.
In addition, as the thicker and well-vascularized belly in the middle part of the
gracilis muscle can be wrapped around the bulbous urethra, a new sphincter with a
long FPL can be reconstructed.
The EUS predominantly consists of slow-twitch type I striated muscle, which is metabolically
designed to provide sustained tone and contract slowly, whereas the levator ani consists
of fast-twitch type II striated muscle, which contracts forcefully, rapidly, and for
a short duration,[22 ] indicating that the periurethral striated musculature provides tonic contraction
necessary for passive urinary continence, whereas the pelvic floor musculature provides
rapid contraction necessary for voluntary interruption of the urinary stream. Martinez
et al[23 ] reported that complete denervation of the EUS after bilateral pudendal nerve axotomy
decreases the MUCP, which may cause urinary incontinence. Moreover, the increase in
the MUCP via pudendal nerve end-to-end coaptation shows the primacy of the pudendal
nerve in EUS innervation.
Research on cross-reinnervation shows that motor neurons influence the regulation
of the mechanical and metabolic properties of skeletal muscles and mimic muscles.[22 ]
[23 ]
[24 ]
[25 ]
[26 ] A nerve to a slow-contracting muscle can convert a fast-contracting muscle almost
completely to a slow type, and vice versa. Vukovich et al[22 ] demonstrated that the vascularized gracilis muscle flap wrapped around the rabbit
urethra, cross-reinnervated by the pudendal nerve, has a reproducible response to
bulbocavernosus reflex testing. Histologic findings of reinnervated gracilis muscle
flaps suggest that reinnervation with the pudendal nerve facilitates a change in muscle
myofiber physiology from fast twitch to slow twitch. Sato et al[27 ] demonstrated the feasibility of reconstructing perineal colostomy following anorectal
resection using a transposed gluteus maximus muscle with pudendal nerve coaptation
to achieve the physiological and histological characteristics of the external anal
sphincter, such as voluntary contraction and relaxation in accordance with defecation
signals from the nervous system, tonic activity in the resting state, and a receptor
mechanism to sense the need to defecate, all of which are controlled by the pudendal
nerve.
A recent experimental and clinical study proved the feasibility of transposition of
skeletal muscle reinnervated by end-to-side neurorrhaphy with donor and recipient
nerves.[28 ]
[29 ] End-to-side neurorrhaphy can induce collateral sprouting to the target muscle without
losing donor nerve function. Pirro et al[30 ] demonstrated that anal sphincter reconstruction using the gracilis muscle transposed
around the anal canal with end-to-side pudendal neurorrhaphy is anatomically achievable
without losing the donor pudendal nerve function of the original anal sphincter and
allows to integrate intact sensory mechanisms and obtain spontaneous relaxation with
better muscular coordination during defecation. Sato et al[31 ] reported that the biceps femoris innervated by end-to-side pudendal nerve anastomosis
contracted with the evoked potential in 5 of the 7 dogs (71%) and demonstrated electric
activity at rest in 3 dogs (43%), which may be characteristics of the anal sphincter.
In our case, by wrapping the gracilis muscle around the bulbous urethra and switching
the obturator nerve to the pudendal nerve, both the MUCP and urinary incontinence
improved, so we believe that the gracilis muscle was reinnervated by the pudendal
nerve and that muscle fiber transformation occurred, resulting in tonic and voluntary
contraction.
However, our procedure has some limitations. The reasons why daytime continence was
not achieved despite the relatively high MUCP may include loosening of the transferred
muscle fixation and denervation muscle atrophy until reinnervation by the pudendal
nerve. The main differences between urinary and anal graciloplasty are that urine
leakage occurs more easily than stool leakage, and the spiral-shaped configuration
may cause twisting and pulling of the urethra rather than a circumferential closure
around the bulbous urethra.
The AUS has the disadvantage of being difficult for the elderly to operate due to
cognitive decline, whereas our procedure has the great advantage of restoring a more
physiological urinary function. Further technical improvements such as the method
of wrapping and securing the muscle are required to improve the treatment outcome.
