MUSCULAR URINARY SPHINCTER: ELECTRICALLY STIMULATED MYOPLASTY FOR FUNCTIONAL SPHINCTER RECONSTRUCTION

0022-5347/98/1605- 1867$03.00/0 T H E JOURNAL OF UROLOGY Copyright 0 1998 by AMERICM

Vol. 160, 1867-1871, November 1998 Printed i n U.S.A.

UROLOGIC.& ASSOCIATION, INC

MUSCULAR URINARY SPHINCTER: ELECTRICALLY STIMULATED MYOPLASTY FOR FUNCTIONAL SPHINCTER RECONSTRUCTION M. M. PALACIO, V. C. VAN AALST, G. A. PEREZ ABADIA, R. W. STREMEL, P. M. N. WERKER, X. REN, G. D. PETTY, S. J. HEILMAN, J. G. VAN SAVAGE, A. GARCIA FERNANDEZ, M. KON, G. R. TOBIN AND J. H. BARKER* From the Divisions of Plastic and Reconstructive Surgery and Urology, Departments of Surgery and Physiology and Biophysics, School of Medicine, University of Louisville, Louisville, Kentucky, the Division of Plastic, Reconstructive and Hand Surgery, University Hospital Utrecht, The Netherlands, and the Private Center of Pediatric Urology, the Division of Urology, and the Department of Pediatric Surgery, Children's Hospital, Cordoba, Argentina

ABSTRACT

Purpose: To reconstruct an electrically stimulated muscular urinary sphincter (MUS) using a tailored gracilis muscle free flap with intact nerve. Materials a n d Methods: Unilateral surgically tailored gracilis muscle free flaps were transferred into the pelvis in eight dogs, leaving the obturator nerve intact. The muscle's pedicle vessels were anastomosed to the inferior epigastric artery and vein in the pelvis and the muscle w a s wrapped around the bladder neck. Electrodes were inserted into the MUS and connected to a programmable pulse generator. After 8 weeks of training the MUS, the pulse generator was programmed to be "on" for 4 hours and "off' for 15 minutes in a continuous cycle. Urodynamic studies were performed periodically, and at the end of the experiment the MUS and proximal urethra were harvested for histology. Three control dogs h a d s h a m operations. Results: All MUSS functioned well following the procedure. Histology of the MUShrethra complex showed no evidence of stricture. Except for one dog, all urethras were easily catheterized. Conclusions: This electrically stimulated innervated free-flap MUS technique effectively increases bladder outlet resistance without producing urethral obstruction. KEY WORDS:urinary incontinence, muscle free-flap, vascular anastomosis Urinary incontinence has a tremendous impact on quality of life and is one of the primary contributing reasons that people enter long term care facilities. Intrinsic urinary incontinence caused by sphincter incompetence has a variety of congenital, traumatic and iatrogenic etiologies. Traumatic causes are mainly due to blunt external forces to the pelvis while the iatrogenic is seen after radical prostatectomy. The most frequent congenital causes of sphincter incompetence are bladder extrophy, epispadias and neurogenic bladder secondary to myelodysplasia. Treatment options in severe urinary sphincter incompetence include procedures that increase outlet resistance, like the Young,' Dees? and Leadbetter3 procedures, slings: and periurethral injection techniques;" procedures that create a valve mechanism like the Kropp,' Pipi Salle' and de Badiola' procedures; and the artificial urinary ~ p h i n c t e r None .~ of these treatment modalities is ideal and all are associated with complications. The most widely accepted treatment is the artificial urinary sphincter (AUS). Though the AUS has many advantages, the fact that it is a foreign body can lead to complications such as erosion of the bladder neck and/or infection. Furthermore, Once in place, the AUS cuff pressures cannot be modified without another surgical intervention. An alternative treatment is dynamic graciloplasty. In spite of poor preliminary clinical outcomes using this procedure," recent experimental work in our laboratory indicates that

with a variation in the surgical technique, dynamic graciloplasty could have promising potential." The main problems in the currently used graciloplasty procedure appear to be associated with limited length of the gracilis muscle. When lifted on its proximal pedicle the distal part of the gracilis flap is at risk of becoming ischemic and fibrotic. Additionally, the spiral orientation of the gracilis flap around the urinary outlet may cause twisting and pulling rather than circumferential squeezing of the urethra. Fibrosis, limited length and spiral orientation of the gracilis may potentially lead to functional failure of the neo-sphincter and stricture of the urethra." To address the above problems, we developed a new procedure which enables us to transfer the proximal, wellperfused part of the gracilis muscle as an innervated free-flap into the pelvis and tailor it into a sphincter which is anatomically and functionally similar to a true sphincter. This study describes the application of this new dynamic gracilis free flap myoplasty procedure to create a MUS, and the resulting changes in histology, static urethral and leak point pressures, over a period of 16 weeks. MATERIALS AND METHODS

Eleven adult female mongrel dogs (20 ? 3 kg. weight), were allocated into experimental or MUS (n = 8) and control (n = 3) groups. Prior to surgery, all dogs underwent a physical examination, blood test, urine culture, and a base line urodynamic study. All urodynamics studies were performed under xylazine sedation (IV,2 to 3 m1./20 kg. body weight). All dogs were housed in individual cages, fed standard dog chow and provided water ad libitum. The animals were euthanized with pentobarbital sodium and phenytoin sodium (6

Accepted for publication May 1, 1998. * Requests for reprints: Division of Plastic and Reconstructive Surgery, 320 MDR-Building, 51 1 South Floyd Street, Louisville, Kentucky 40292. Supported by grants from the Alliant Community Trust Fund, the Jewish Hospital Foundation, Louisville, Kentucky. Presented at the American Academy of Pediatrics 1997 Annual Meeting, New Orleans, LA,November 2nd,1997. 1867

1868

ELECTRICALLY STIMULATED MYOPLASTY FOR SPHINCTER RECONSTRUCTION

mlJ20 kg.body weight). Studies were performed in the American Association of Laboratory Animal Care-approved Research and Resource Center at the University of Louisville Health Science Center. Surgical procedure. Dogs were given penicillin G procaine (IM, 300,000 UU20 kg. body weightf24 hours) and gentamycin (i.m., 50 mg./20 kg. body weighUl2 hours) 30 minutes before and daily for 3 days after surgery. Prior to surgery all dogs received atropine sulfate (s.c., 0.1 ml./kg. body weight) and intravenous thiopental sodium (6 to 12 mg./kg. body weight) anesthesia. They were intubated and mechanically ventilated with a gas mixture of 2% halothane/94% oxygen/4%nitrous oxide 0.2 l./kg. body weight. In all dogs, a total bilateral pudendal neurotomy was performed to promote low outlet resistance. Using sterile surgical technique, with the dog supine and both hind limbs abducted, a longitudinal incision was made on the medial side of the left thigh and the gracilis muscle was exposed along its entire length. So the muscle could be extended to its original length after detaching it from its proximal and distal insertions, marking sutures were placed along its entire length 5 cm. apart. The muscle was then tailored into a rectangular shape (6 cm. long and 2 cm. wide) with the main neurovascular pedicle protruding laterally (fig. 1). To transfer the gracilis into the pelvis, the muscle was isolated and its distal blood supply was severed leaving it entirely dependent upon its main neurovascular pedicle for its blood supply. Microsurgical technique was used to dissect free and prepare the artery and vein for anastomosis to their recipient vessels (deep inferior epigastric artery (DIEA) and vein (DIEV))in the pelvis. The gracilis’ main innervation, a branch of the obturator nerve, was also dissected free but was left intact. A midline abdominal incision was made and the left anterior rectus abdominis fascia was opened and retracted laterally to expose the deep inferior epigastric vessels. After preparing the recipient vessels in the pelvis, the g r a d i s “innervated” free flap was passed through a subcutaneous tunnel in the inguinal region into the pelvis and the muscle’s main artery and vein were anastomosed end-to-end (10-0 nylon suture) to the DIEA and DIEV (fig. 2). Following completion of the anastomosis, blood flow to the flap was reestablished. The total ischemic time during the transfer was 60 to 110 minutes. Two electrodes (Model 4300, Medtronic Inc.) were inserted and fixed in the muscle (2 to 3 cm. apart) near the nerve entry. The peritoneum was opened and the urethra and bladder neck were gently dissected free. To create the MUS, the tailored gracilis free flap was wrapped around the bladder neck like a ring and sutured in place using 3-0 silk. The electrode leads were tunneled to the lower abdomen and connected to a pulse generator (Itrel 3,

Intad Nerve

Stimulator

Bladder DIEA a DIEV

MUS Ufethra

\w

Microanastomosis Neuro4ascular Pedlcle

FIG.2. Muscular Urinary Sphincter (MUS) procedure. Following completionof anastomosis and reestablishment of flow, tailored gracilis free flap is passed around bladder necWurethra circumferentially like a true sphincter and sutured in place. Two electrodes are inserted and fixed in muscle and connected to subcutaneously placed pulse generator.

Medtronic Inc.) which was placed in a subcutaneous pocket. All wounds were closed in layers. Three control animals underwent bilateral pudendal neurotomy without manipulation of their gracilis muscles. The peritoneum was opened and their urethra and bladder neck were dissected. All wounds were closed in layers and the animals were treated in a like manner to the MUS group. Electrical stimulation protocol. To transform the gracilis MUS from a fatigue prone, fast-twitch muscle into a fatigue resistant, slow-twitch muscle, an electrical stimulation protocol was initiated in the experimental group two weeks after surgery. The protocol used was that specified by the FDA for Clinical Dynamic Graciloplasty” (table). Following this 8-week training regimen, all MUSS were programmed to be “On” for 4 hours and “Off” for 15 minutes in continuous cycles. Urodynamic studies. Static urethral pressure (SUP) and intravesical leak point pressure (LPP) were performed in all dogs before surgery (baseline) and every 2 weeks thereafter for 16 weeks. In the experimental group, this was done with the MUS both “On” and “Off”. Two 6 Fr. double-lumen catheters were used to measure bladder pressure and urethral pressure. The catheters were connected to pressure transducers ( a u l d , P23 ID) and the signals were recorded with a computer-based data acquisition system (CED 1401p’”8).The bladder was infused (10 ml. per minute) with sterile saline (25C) solution. The urethral pressure port was maintained patent by a bias infusion pump (Model 600-900 v, Harvard Apparatus) at 2 ml. per minute with sterile saline solution. Bladder and urethral pressures were measured simultaneously during bladder filling and voiding. Sterile urine samples were collected during each study and cultured for bacterial growth. Histology. Biopsies were taken from the bladder, MUS/ urethra complex, urethra and ureters at the end of the experiment. All samples were fixed in formalin and embedded in paraffin for Hematoxylin & Eosin and Masson’s trichrome staining. Statistical analysis. Paired Student’s t test and one-way analysis of variance (ANOVA)were performed. Multiple comFIG. 1. Muscular Urinary Sphincter (MUS). Innervated free-flap is tailored from gracilis muscle into rectangular shape with lateral parisons and the Bonferroni post-hoc test were used. All data neurovascular pedicle. A artery, V vein, N: nerve. Proximal inser- are presented as mean -+ SEM. A p value less than 0.05 was tion is to right in drawing and distal is to left. considered significant.

ELECTRICALLY STIMULATED MYOPLASTY FOR SPHINCTER RECONSTRUCTION

1869

RESULTS

The urethras in all dogs, except one, were easily catheterized with the pulse generator in the “On” position. In this dog, catheterization was easy with the stimulator in the “Off position. All animals presented seromas at the gracilis donor site, which were reabsorbed during the first postoperative week. Because of repeated urethral catheterizations, all but one dog presented with urinary infections, which were successfully treated, based on antibiotic sensitivities. Urodynamic studies. The baseline (before surgery, mean 2 SEM) static urethral pressures (SUP) in the MUS group were 16.3 2 3.6 cm. H,O and 16.2 5 2.7 cm. H,O in the Control group. The mean SUP did not change significantly over the 16 weeks study in either the Control or MUS group with the pulse generator “Off”. From 10 weeks until termination, the MUS was considered to be fully trained (fatigue resistant). The MUS group, with pulse generator turned U O ~ ,at , , 12 weeks (30.7 5 8.7 cm. H,O, p <0.01); at 14 weeks (33.9 5 6.2 cm.H,O, p <0.001); and 16 weeks post-surgery (40.0 5 13.5 an. H,O, p <0.001), achieved significantly greater SUP when compared with the baseline and Control group measurements (fig. 3). Measurements of SUP taken a t 10 weeks after surgery, during which the MUS was in a transformed and trained state, demonstrated that contraction of the MUS elicited an increase of 10.8 cm. H,O. The baseline (mean 5 SEMI leak point pressure (LPP) in the MUS group was 20.0 ? 5.6 cm. H,O and 21.9 5 3.3 cm. H,O in the Control group. The mean LPP did not change significantly over the course of the study in either the Control or MUS group with the pulse generator “Off. The MUS, with pulse generator “On” at 12 weeks (61.4 5 40.5 cm. H,O, p <0.001) and 14 weeks post surgery (57.1 5 20.8 cm. H,O, p <0.01), achieved significantly greater LPP compared with baselines and control animals (fig. 4). Histology. Histological examination of the bladder of animals with the MUS revealed that acute inflammation of the serosa and muscle hypertrophy occurred in one dog. The MUShrethral complex appeared histologically viable with integration of the gracilis muscle into smooth muscle of the urethral wall (fig. 5). Four dogs demonstrated a partial area of smooth muscle (that is native sphincter) loss. Focal skeletal muscle atrophy and fatty infiltration were also present.

FIG. 4. Leak Point Pressure (LLp). ComPafisonof(mean + SEMI LLP with MUS turned “Off (light bars) and turned “On”(dark bars). MUS with pulse generator turned “On” achieved significantly greater LLP values than Controls (open bars) (*p <0.001 at 12 weeks, and p <0.01 at 14 weeks).

DISCUSSION

The use of the gracilis muscle in the surgical treatment of urinary incontinence was first reported by Deming in 1926 in one patient.” In 1956 Pickrell reported 6 patients in whom the distal part of the gracilis was transposed around the bladder neck and urethra. Both Deming and Pickrell relied

FIG. 5. A, histological cross section of the muscular urinary sphincter (MUSYurethra complex, obtained 16 weeks after construction of the MUS (magnification6 X). B, Masson’s trichrome staining. Arrows: interface between MUS and urethral wall; GM: gracilis muscle; U. urethral wall; L: urethral lumen (magnification, X 16). FIG. 3. Static urethral pressure (SUP). Comparison of (mean + SEMI SUP with the (MUS) ‘‘off” (light bars), and ‘%”(dark bars). MUS with pulse generator turned “On” achieved significantly greater SUP values than Controls (open bars) and MUS turned “OfP (*p <0.001 at 12 weeks, and p <0.001 at 14 and 16 weeks).

upon volitional abduction of the leg to contract the gracilis. pickrel113 reported SUCCeSS in all his patients, but long term follow up was not provided. In spite of this report, the use of

1870

ELECTRICALLY STIMULATED MyOPLASTY FOR SPHINCTER RECONSTRUCTION

the gracilis muscle for urinary sphincter reconstruction did not become widespread. I t was not until thirty years later that this procedure reappeared when Williams’* and Janknegt” reintroduced it, adding the use of implantable pulse generators to make the gracilis muscle contract. Using electrical stimulation, the gracilis was transformed from its usual fatigue prone, fast-twitch state with predominantly type I1 fibers into a fatigue resistant, slow-twitch,mainly type I muscle. This approach was reported to keep the muscle contracted for prolonged periods without fatiguing.I6 Using the original “graciloplasty” procedure together with implantable pulse generators, approximately 30 clinical cases have been reported to l7 The results in these cases have been disappointing, with an approximately 50% failure rate. These failures appear to be caused by stricture of the bladder necuurethra at the location of the gracilis neosphincter. We believe this stricture is due t o three factors related to the transfer of the gracilis as a pedicled flap: 1) as a pedicled flap with intact proximal insertion, the gracilis muscle is too short to reach around the bladder neck and thus it may be wrapped too tightly; 2) the distal part of the gracilis is often rendered ischemic when it is lifted on its proximal pedicle; and 3) the presently used wrapping technique twists rather than circumferentially squeezes the urinary outlet. Recently a modification of the wrap was report with preliminary results.” To address these problems, we developed a new procedure which consists of making the gracilis muscle into an “innervated” free flap rather than a pedicled flap. This variation frees the gracilis from its proximal insertion, allowing surplus length, so that the proximal well-perfused part of the muscle, rather than its distal tip, can be wrapped circumferentially around the bladder necWurethra like a true sphincter. In a previous anatomical study in human cadavers, we found that using the gracilis as an “innervated” free flap allowed us to transfer the entire muscle into the pelvis through the obturator foramen. We also found that the proximal part of the muscle could easily be wrapped circumferentially around the bladder neck.” In a n acute animal study we demonstrated that this “innervated” gracilis free flap procedure did not compromise muscle function or perfusion. l1 In the present study, static urethral pressures measured at the site of the MUS wrap, with the pulse generator in the “On” position, were significantly higher than urethral pressures with the pulse generator “Off” or in the controls. Leak point pressures with the pulse generator #On” were significantly higher than controls. The LPP “Off was also higher (not significantly) than controls. This was probably due to the fact that the MUS modified the anatomy of the bladder neck and probably its function during micturition. In the control dog the bladder becomes funnel-shaped at micturition (fig. 6, A-B).In the dog with MUS, the bladder neck cannot become funnel-shaped, due to the wrapping of the new sphincter (fig. 6, C-D). In these experiments, muscle biopsies of the MUS demonstrated areas of skeletal muscle atrophy after 4 months. These morphological changes are most likely to be a consequence of transposition and tailoring. G u e l i n ~ k x ’reported ~ that transposed gluteus maximus muscle on a n intact neurovascular pedicle, with or without electrical stimulation, undergoes 25 to 30% atrophy of the muscle fibers. Pathological findings included hypertrophy of the bladder, which was likely due to the straining of the animal against the MUS while it was “On”. Despite the muscle atrophy and increased fibrosis, the functional measurements demonstrated that the MUS significantly increased urethral pressure and leak point pressure in every animal. Advantages of this procedure. 1) Adequate gracilis muscle length is provided to comfortably tailor and construct the

FIG.6. Voiding cystourethrogram. Control group (A-B),and muscular urinary sphincter (MUS) group (C-D). Micturition sequence from A to B and C to D and illustration of MUS modifications of cervicourethral angle (bladder neck).

MUS. 2) The proximal, well-perfused muscle part of the gracilis muscle can be used to construct the MUS, thus avoiding ischemia and fibrosis. 3) The MUS is tailored and wrapped circumferentially around the bladder necwurethra, thus squeezing like the true sphincter. 4) Autologous tissue in contact with urethral wall does not pose the risk of bladder necWurethra erosion as seen in techniques using foreign materials. 5 ) The well-perfused MUS is less prone to infection than a foreign body sphincter. 6) Urethral pressure can be adjusted by regulating the stimulus intensity to the MUS. Disadvantages of this procedure. 1) The MUS procedure requires microsurgical instruments and skills. 2) The microsurgical procedure can add 1 t o 2 hours to the length of the procedure. 3) Donor site morbidity is associated with harvesting the gracilis muscle.20 In conclusion, this new “innervated gracilis free-flap urinary sphincter procedure provides increased urethral and leak point pressures which can provide continence for periods of at least 4 hours in the absence of urethral stricture. The new MUS procedure may be applicable in patients with severe urinary incontinence due to urinary sphincter incompetence. Stimulation protocol changes parameters during training weeks: pulse rate (Hz) on time (sec) off time (sec) duty circle

~ _ _ _ _(%I“ ___

1&2 210 25 0.1 1

6

3&4 210 25 0.2 1 11

5&6 210 25 0.4

0.7 25

7&8 210 25 2 0.5 80

~

’. ‘2 of time during which the muscle is actually stimulated.

8 210 14 continuous 0 100

ELECTRICALLY STIMULATED MYOPLASTY FOR SPHINCTER RECONSTRUCTION

Acknowledgments. The a u t h o r s would like t o thank I v a n Bourgeois and Don Erickson from Medtronic Inc. for their technical assistance; G. Randolph Schrodt, M.D., and S h e r o n Lear, ASCP, from the D e p a r t m e n t of Pathology and Special Procedure Laboratory, School of Medicine, University of Louisville for the histology procedures, and H i r a m Polk, chairman of the D e p a r t m e n t of Surgery of the University of Louisville, for his advice in p r e p a r i n g this manuscript.

1871

Baeten, C. G. M. I.: Electrically stimulated gracilis sphincter (dynamic graciloplasty) for treatment of intrinsic sphincter deficiency: a pilot study on feasibility and side effects. J . Urol., 154: 1830,1995. 11. Van Aalst, V. C. Werker, P. M. N., Stremel, R. W., Perez Abadia, G. A,, Petty, G. D., Heilman, S. J., Palacio, M. M., Kon, M., Tobin, G. R. and Barker, J . H.: Electrically stimulated freeflap graciloplasty for urinary sphincter reconstruction: a new surgical procedure. Plast. Reconstr. Surg., (in press). 12. Deming, C. L.: Transplantation of the gracilis muscle for inconREFERENCES tinence of urine. JAMA, 8 6 822,1926. 1. Young, H. H.: An operation for the cure of incontinence of urine. 13. Pickrell, K., Georglade, N., Grawford, H., Maguire, C. and Boone, A,: Gracilis muscle transplant for correction of urinary Surg. Gynecol. Obstet., 28: 84, 1919. incontinence in male children. Ann. Surg., 143: 764,1956. 2. Dees, J . E.: Congenital epispadias with incontinence. J. Urol., 14. Williams, N. S. Hallan, R. I., Koeze, T. H. and Watkins, E. S.: 62 513, 1949. 3. Leadbetter, J. R. Jr.: Surgical correction of total urinary inconConstruction of a neorectum and neoanal sphincter following tinence. J . Urol., 91: 261, 1964. previous proctocolectomy. Br. J . Surg., 7 6 1191,1989. 4. Blaivas, J . G. and Jacobs, B. Z.: Pubovaginal fascia1 sling for the 15. Janknegt, R. A,, Baeten, C. G. M. I., Weil, E. H. J . and Spaans, treatment of complicated stress urinary incontinence. J . Urol., F.: Electrically stimulated gracilis sphincter for treatment of 145: 1214,1991. bladder sphincter incontinence. Lancet, 340 1129,1992. 5. Perez, L. M., Smith, E. A,, Broecker, B. H., Massad, C. A. and 16. Salmons, S.and Hendriksson, J.: The adaptive response of skelWoodard, J . R.: Submucosal bladder neck injection of bovine etal muscle to increased use. Muscle Nerve, 4:94, 1981. dermal collagen for stress urinary incontinence in the pediat- 17. Chancellor, M. B., Hong, R. D., Rivas, D. A,, Watanabe, T., ric population. J . Urol., 156 633,1996. Crewalk, J . A. and Bourgeois, I.: Gracilis urethromyoplasty 6. Kropp, K.A. and Angwafo, F. F.: Urethral lengthening and a n autologous urinary sphincter for neurologically impaired reimplantation for neurogenic incontinence in children. patients with stress incontinence. Spinal Cord, 3 5 546, 1997. J. Urol., 135 533, 1986. 7. Pipi Salle, J., de Fraga, J. C. S., Amarante, A,, Silveira, M. L., 18. Chancellor, M. B., Watanabe, T., Rivas, D. A., Hong, R. D., Kumon, H., Ozawa, H. and Bourgeois, I.: Gracilis urethral Lambertz, M., Schmidt, M. and Rosito, N. C.: Urethral lengthmyoplasty: preliminary experience using an autologous uriening with anterior bladder wall flap for urinary incontinence: nary sphincter for post-prostatectomy incontinence. J. Urol., a new approach. J. Urol., 152: 803,1994. 158 1372,1997. 8. de Badiola, F., Ruiz, E., Denes, E. and Puigdevall, J.: Una nueva tecnica quinirgica para aumentar la resistencia cervico ure- 19. Guelinckx, P. J., Sinsel, N. K. and Gruwez, J. A.: Anal sphincter reconstruction with the gluteus maximus muscle: anatomic tral. Rev. de Cir. Infantil, 5 57, 1995. and physiologic considerations concerning conventional and 9. Levesque, P. E., Bauer, S. B., Atala,A., Zurakowski, D., Colodny, dynamic gluteoplasty. Plast. Reconstr. Surg., 98 293, 1996. A., Peters, C. and Retik, A. B.: Ten year experience with the artificial urinary sphincter in children. J . Urol., 156 625, 20. Whetzel, T.P. and Lechtman, A. N.: The gracilis myofasciocutaneous flap: vascular anatomy and clinical applications. Plast. 1996. Reconstr. Surg., 99.1642,1997. 10. Janknegt, R. A,, Heesakkers, J . P. F. A,, Weil, E. H. J. and