Dynamic graciloplasty for treatment of faecal incontinence

1163

REFERENCES 1. Ludlam CA, Tucker J, Steel CM, et al. Human T lymphotrophic virus type III (HTLV III) infection in seronegative haemophiliacs after transfusion of factor VIII. Lancet 1985; ii: 233-36. 2. Cuthbert RJG, Ludlam CA, Steel CM, et al. Five year prospective study of HIV infection in the Edinburgh haemophiliac cohort. Br Med J 1990; 301: 956-60. 3. Steel CM, Ludlam CA, Beatson D, et al. HLA haplotype A 1 B8 DR3 as a risk factor for HIV-related disease. Lancet 1988; i: 1185-88. 4. Kaslow RA, Duquesnoy R, Van Raden M, et al. A1, Cw7, B8, DR3 HLA antigen combination associated with rapid decline of T-helper lymphocytes in HIV-1 infection. Lancet 1990; 335: 927-30. 5. Mallal S, Cameron PU, French MAM, Dawkins RL. MHC genes and HIV infection. Lancet 1990; 335: 1591-92. 6. Peutherer JF, Rebus S, Barr P, et al. Confirmation of non-infection in persistently HIV-seronegative receipients of contaminated factor VIII. Lancet 1990; 336: 1008. 7. Carr R, Veitch SE, Edmund E, et al. Abnormalities of circulating lymphocyte subsets in haemophiliacs in an AIDS-free population. Lancet 1984; i: 1431-34. 8. Balfe P, Simmonds P, Ludlam CA, Bishop JO, Leigh Brown AJ. Concurrent evolution of human immunodeficiency virus type 1 in patients infected from the same source: rate of sequence change and low frequency of inactivating mutations. J Virol 1990; 64: 6221-33. 9. Parry JV. An immunoglobulin G capture assay (GACRIA) for anti-HTLV-III/LAV and its use as a confirmatory test. J Med Virol

1986; 19: 387-97. 10. Parry JV, Mortimer PP. Place of IgM antibody testing in HIV serology. Lancet 1986; ii: 979-80. 11. Simmonds P, Lainson FAC, Cuthbert RJG, et al. HIV antigen and antibody detection: variable responses to infection in the Edinburgh haemophiliac cohort. Br Med J 1988; 296: 593-98. 12. Lang JM, Coumaros G, Levy S, et al. Elevated serum levels of soluble interleukin 2 receptors in HIV infection: correlation studies with markers of cell activation. Immunol Lett 1988; 19: 99-102. 13. Cheingsong-Popov R, Panagiotidi C, Bowcock S, Aronstam A, Wadsworth J, Weber J. Relation between humoral response to HIV

gag and env proteins at seroconversion and clinical outcome of HIV infection. Br Med J 1991; 302: 23-26. 14. Lange JMA, de Wolf F, Goudsmit J. Markers for progression in HIV infection. AIDS 1989; 3 (suppl 1): S153-60. 15. Teitel JM, Freedman JJ, Garvey MB, Kardish M. Two-year evaluation of clinical and laboratory variables of immune function in 117

haemophiliacs seropositive

or

seronegative

1989; 32: 262-72. 16. Edelman AS, Zolla-Pazner S. AIDS:

dysregulation, dysfunction

and

a

for HIV. Am

J Hematol

syndrome of immune FASEB J 1989; 3:

deficiency.

22-30. 17. Ascher MS, Sheppard HW. AIDS as immune system activation II. The panergic imnesia hypothesis. J Acquir Immune Defic Syndr 1990; 3: 177-90. 18. Jones P, Proctor S, Dickinson A, George S. Altered immunology in haemophilia. Lancet 1983; i: 120-21. 19. Brieva JA, Sequi J, Zabay JM, et al. Abnormal B cell function in haemophiliacs and their relationship with factor concentrates administration. Clin Exp Immunol 1985; 59: 491-98. 20. Simmonds P, Zhang LQ, Watson HG, et al. Hepatitis C quantification and sequencing in blood products, haemophiliacs, and drug users. Lancet 1990; 336: 1469-72. 21. Manca F, Habeshaw I, Dalgleish A. The naive repertoire of human T helper cells specific for gp120, the envelope glycoprotein of HIV. J Immunol 1991; 146: 1964-71. 22. Geha RS, Hyslop N, Alani S, Farah F, Schneeberger EE, Rosen FS. Hyperimmunoglobulin M immunodeficiency (dysgammaglobulin aemia): presence of immunoglobulin M-secreting plasmacytoid cells in peripheral blood and failure of immunoglobulin M-immunoglobulin G switch in B cell differentiation. J Clin Invest 1979; 64: 385-91. 23. Mayer L, Kwan SP, Thompson C, Rosen FS. Correction of a defect in immunoglobulin class switching in patients with immunodeficiency and hyper-IgM by "switch" T cells. In: Eibl MM, Rosen FS, eds. Primary immunodeficiency disease. Amsterdam: Elsevier, 1986: 181-86. 24. European Collaborative Study. Children born to women with HIV-1 infection: natural history and risk of infection. Lancet 1991; 337: 253-60.

Dynamic graciloplasty for treatment of faecal incontinence

Serious faecal incontinence due to anal sphincter damage should be treated by surgery. Graciloplasty has had limited success because the gracilis is a fast-twitch muscle and fatigues quickly. A favourable outcome in a patient who had dynamic

(electrically stimulated) graciloplasty encouraged us procedure. Gracilis muscle transposition was done in ten patients with complete anal incontinence due to anal atresia, sphincter damage, or neurogenic causes, and

to further assess this

who had had several other unsuccessful treatments. 6 weeks after muscle transposition, intramuscular leads were implanted and connected to an implantable electric stimulator. Eight patients became continent, one patient still has a diverting colostomy, and a fistula developed in the other patient. Anal sphincter pressure improved from 35 mm Hg without stimulation to 62 mm Hg with stimulation at 8 weeks (mean increase 28 mm Hg [95% confidence interval 18, 36], p<0·01). Retention time of a phosphate enema increased from 22 to 281 s (mean increase 259 s [82, 436],

Defaecography showed that the new functioning. Defaecation was possible sphincter

p<0·01).

was

when the stimulator was turned "off" with a magnet. Dynamic graciloplasty can restore continence and it improves quality of life in faecally incontinent patients for whom other treatments have been

unsuccessful. Introduction Faecal incontinence is a distressing disorder and its prevalence is substantial but difficult to estimated Serious incontinence due to anal sphincter damage should be treated by surgery. When no functional sphincter remnant is ADDRESSES: Departments of Surgery (C. G. M. I. Baeten, MD, J. Konsten, MD, Prof P. B. Soeters, MD), Clinical Neurophysiology (Prof F. Spaans, MD), and Pathology (R. Visser, MD), Maastricht University Hospital; Bakken Research Centre (A. M. M. C. Habets, PhD, I. M. Bourgeois, MS) Maastricht; and Department of Human Biology, University of Limburg (A. J. M. Wagenmakers, PhD), Maastricht, Netherlands. Correspondence to Dr C. G. M. I. Baeten, Maastricht University Hospital, PO Box 5800, 6202 AZ

Maastricht, Netherlands.

1164

available, anal sphincter function can be replaced by a Thiersch wire, by an inflatable cuff, or with autologous material, such as a transposed gracilis muscle around the

CAUSE OF INCONTINENCE AND PREVIOUS THERAPY

anal canal (graciloplasty), in which the transposed gracilis muscle acts as a sling and is stretched open by bowel contents.2 However, graciloplasty may not improve continence3 because gracilis is a voluntary muscle,and its "fast-twitch" fibres are easily fatiguable. To sustain a prolonged contraction, electrical stimulation can be.used to induce a transition in muscle composition,4which has been shown to improve sphincter function of transposed skeletal muscle in animals.5 A favourable outcome in a patient’ with electrically

stimulated (dynamic) graciloplasty encouraged us to further assess the technique. We here report the effect of electrical stimulation on continence and muscle fibre composition in patients who had been completely incontinent, despite conventional treatment.

Patients and methods We studied the clinical outcome of all patients (seven women, three men) with complete faecal incontinence (median duration of incontinence 20 years [range 2-40]) who were treated with electrically stimulated graciloplasty (dynamic graciloplasty) from May, 1986, to January, 1991. Median age was 38 years (range 20-65). Follow-up ranged from 26 weeks to 5 years (median 13 months). Eight patients had had previous anal surgery (table). In two patients (A and F) unsuccessful gracilis transposition had been done 10 years previously. Four patients had tried to achieve pseudocontinence with enemas. In patient I a colostomy had been constructed elsewhere. No patient wanted to have a permanent colostomy, despite their severe incontinence. In the first operation, the gracilis muscle was transposed according to the method of Pickrellwith minor modifications. No protective colostomies were done. An interval of at least 6 weeks was chosen before the second stage of the procedure so that the gracilis tendon could be anchored tightly to the ischial tuberosity and the gracilis muscle could recover from manipulation and dissection of the peripheral vessels. At the second operation, the site of the lowest threshold of the transposed gracilis muscle was assessed with temporary testing needles connected to an external stimulator; at this site intramuscular electrodes (’SP 5528’, Medtronic, Kerkrade, Netherlands) were implanted. The electrodes were then tunnelled to the lower abdomen and connected with an implantable stimulator (’Itrel 7420’, Medtronic, Minnesota, USA), which was placed in a subcutaneous pocket. The muscle was activated immediately after implantation, with intermittent mode electrical stimulation for 8 weeks. The pulse width was 210 s, frequency 25 Hz, and duty cycle ("on" time) 6%. The duty cycle was increased every 2 weeks (6,11,80,100%) for 8 weeks with telemetry. This procedure can be regarded as an "in-service training" of the new sphincter. The output of the electrical stimulator and thus the tonic contractions of the new sphincter could be switched "off’ with a magnet to allow defaecation at a convenient time. We evaluated patients before transportation and during followup at regular intervals (0, 2, 4, 6, 8, 12 weeks) after implantation of the stimulator. A questionnaire was used to record improvement of continence with respect to the use of incontinence pads and frequency of defaecation. Anal pressure was recorded before transposition with a microtransducer (Gaeltec Ltd, Dunvegan, Isle of Skye, UK) in eight patients (the other two had had a graciloplasty previously). Anal manometry was repeated at the previously indicated intervals in all patients. Electromyography of the external sphincter was done before transposition, and repeated after graciloplasty to assess the vitality of the muscle. Retention times of a 250 ml phosphate enema were measured before and after 8 weeks of stimulation. Defaecography was done before transposition and 16 weeks after stimulation was started. Muscle fibre changes were studied by immunoMstochernistry8 in four patients in samples

taken from open biopsies of the transposed gracilis muscle at implantation and 16 weeks later; a monoclonal antibody R11D10,’ specific for type I skeletal myosin, was used because type I muscle fibres are capable of prolonged contractions.’*’" The results were expressed as mean and 95 % confidence interval (CI). Statistical analysis was done with the paired Student’s t test, with the computer program SPSS/PC + (version 3.1) (Microsoft, Chicago, Illinois, USA). The study was approved by the medical ethical committee of Maastricht University Hospital.

Results There were no technical difficulties during the operative procedures. Continence did not improve in any patient who had transposition alone. Eight patients achieved continence

after 8 weeks’ electrical stimulation, of whom six do not wear incontinence pads any longer and two sometimes wear pads. After implantation of the stimulator, the frequency of defaecation decreased from a mean of 8/day (95% CI 5,11) before stimulation to 3 (1, 5) after stimulation (mean The ability to postpone decrease 5 [2, 9], n=9, p<0-01). defaecation increased from 12 s ( - 7, 31) before stimulation to 40 min (-11, 89) after stimulation (geometric mean increase, 4 min [16 s, 49 min], n 9, p 0108). One patient (H) achieved continence after 8 weeks, but a fistula developed after 16 weeks when the transposed gracilis muscle perforated the anal canal. Another patient (I) still has a diverting colostomy and is scheduled to undergo restoration of bowel continuity. Two other patients had complications; a partly atrophic gracilis developed in patient D after 9 weeks, but he =

=

80ï

Efficacy of gracilis muscle stimulation Closed columns= mean anal pressure at rest; hatched columns= anal pressure due to electrical stimulation. Time after implantation of the stimulation equipment is shown.

1165

maintained satisfactory continence without stimulation. In another patient (E) the stimulator had to be removed a wound infection. The equipment was 2 months of total incontinence, with after reimplanted subsequent recovery of continence. Anal manometry before and after transposition of the gracilis muscle showed that voluntary contraction of the transposed gracilis muscle increased the anal pressure from 31 (19,43) to 78 mm Hg (59, 97) (mean increase 47 [29, 64], n = 10, p < 001). However, this contraction could not be sustained for longer than a minute. After implantation of the electrical stimulator, the resting pressure (figure) increased from 31 (19, 43) to 68 (57, 79) mm Hg with stimulation (mean increase 37 [27,47], n = 10, p < 0-01). After 8 weeks these values were 35 (25,44) without stimulation and 62 (47, 77) with stimulation (mean increase 28 [18, 36], n = 10 p < 001 ). There was no significant decrease in anal pressure with stimulation (initial induced contraction 68 mm Hg [57, 79], after 8 weeks 62 [47, 77]; mean decrease 6 [ 4, 16], n = 10). After 5 years, electrical stimulation still induces an increase of anal pressure of 20 mm Hg in patient A. At implantation, the lowest mean frequency of stimulation that produced a fused muscular contraction was 25 Hz. After 8 weeks of stimulation, it was decreased to 15 Hz. As judged by electromyography, all patients had severe anal sphincter damage before transposition, and the gracilis muscle innervation had not been damaged by transposition. Retention time of a 250 ml phosphate enema increased from 22 s (-19, 62) before stimulation to 281 s (107, 454) after stimulation (mean increase 259 s [82,436], n = 10 p < 0-01). Defaecography showed that when stimulation was discontinued, voluntary defaecation was possible. After a minimum of 16 weeks after the second operation, there was an increase in the percentage of type I fibres in muscle biopsy samples from 44% (38, 49) before stimulation to 63% (54,71) after stimulation (mean increase 19% [13,25],

because of

-

n=4, p <001). Discussion

findings show that treatment with dynamic graciloplasty is much more successful than graciloplasty alone because it offers the opportunity for a prolonged Our

muscular contraction for many hours. Several observations in our patients suggest that there is a type II (fast-twitch) to type I (slow-twitch) conversion in the stimulated gracilis muscle: firstly, the percentage of type I fibres increased; secondly, the muscle could function for a prolonged period; and lastly, the minimum stimulation frequency at which a fused contraction was obtained could be reduced from 25 to 15 Hz. The continuous stimulation of transposed gracilis muscle is needed to maintain a prolonged contraction, which is essential for continence. Incontinence might return when stimulation is discontinued because there is a reverse transformation from slow-twitch, fatigue-resistant fibres to fast-twitch, fatiguable fibres Cavina et al,14 using temporary stimulation of transposed gracilis muscle, found that they had to further rely on voluntary contraction of the gracilis. By contrast, in our series the prolonged contraction of the gracilis was produced by the implanted stimulator, and voluntary contraction was no longer necessary. The aim of stimulation is to obtain faecal continence and not a maximum contraction; thus, the mean anal pressure with stimulation was less than the mean pressure of a maximum voluntary contraction after transposition. Diarrhoea can

serious difficulties, which may explain the moderate in patient E who had diarrhoea and also lacked anal canal sensitivity. The best results might be obtained in patients in whom some anal sensitivity is preserved.1s We used a stimulation pattern modified from that used for the latissimus dorsi muscle in cardiomyoplasty.12,13 However, whether stimulating patterns are critical to induce fibre-type changes is not certain 14,15 since other investigators have used low-frequency stimulation in graciloplasty with equally good results,16,17 although they did not show the type II to type I fibre transition. Further research is needed to define the best stimulus pattern for a lifelong stimulated cause

success

neosphincter.14 Our findings suggest that dynamic graciloplasty is a promising treatment for faecal incontinence. The "automatic" and prolonged contraction allows the transposed gracilis to function effectively as a sphincter. This study was supported by a grant of the Netherlands Ministry of Trade and Industry (STIPT).

REFERENCES

Leigh RJ, Turnberg LA. Faecal incontinence: the unvoiced symptom. Lancet 1982; i: 1349-51. 2. Christiansen J, Sorenson M, Ramussen OO. Gracilis muscle transposition for faecal incontinence. Br J Surg 1990; 77: 1039-40. 3. Cormann ML. Management for fecal incontinence by gracilis muscle transposition. Dis Colon Rectum 1979; 22: 290-92. 4. Pette D. Activity induced fast to slow transition in mammalian muscle. Med Sci Sports Exercise 1984; 6: 517-28. 5. Hallan RI, Williams NS, Hutton MRE, et al. Electrically stimulated sartorius neosphincter: canine model of activation and skeletal muscle transformation. Br J Surg 1990; 77: 208-13. 6. Baeten C, Spaans F, Fluks A. An implanted neuromuscular stimulator for fecal continence following previously implanted gracilis muscle: report 1.

of a

Dis Colon Rectum 1988; 31: 134-37. Broadbent TR, Masters FW, Metzger JT. Construction of a rectal sphincter and restoration of anal incontinence by transplanting the gracilis muscle: a report of four cases in children. Ann Surg 1952; 135: 853-63. 8. Havenith MG, Visser R, Schrijvers van Schendel JMC, Bosman FT. Muscle fiber typing in routinely processed skeletal muscle with monoclonal antibodies. Histochemistry 1990; 93: 497-99. 9. Salmons S, Hendriksson J. The adaptive response of skeletal muscle to increased use. Muscle Nerve 1981; 4: 94-105. 10. Eisenberg BR, Brown JMC, Salmons S. Restoration of fast muscle characteristics following cessation of chronic stimulation. Cell Tissue Res 1984; 238: 221-30. 11. Cavina E, Evangelista G, Chiarugi M, Buccianti P, Tortora A, Chirico A. Perineal colostomy and electrical stimulated gracilis neosphincter after abdomino-perineal resection of the colon and anorectum: a surgical experience and follow-up study in 47 cases. Int J Colorect Dis 1990; 5: 6-11. 12. Chachques JC, Grandjean PA, Carpentier A. Dynamic cardiomyoplasty: experimental cardiac wall replacement with a stimulated skeletal muscle. In: Chiu RCJ, ed. Biomechanical cardiac assist, cardiomyoplasty and muscle-powered devices. New York: Futura Publishing, 1986: 59-84. 13. Carpentier A, Chachques JC. The use of skeletal muscle to replace diseases human heart muscle. In: Chiu RJC, ed. Biomechanical cardiac assist, cardiomyoplasty and muscle-powered devices. New York: Futura Publishing, 1986: 85-102. 14. Baeten CGMI, Spaans F. Construction of neorectum and neoanal sphincter. Br J Surg 1990; 77: 473-74. 15. Edström L, Grimby L. Effect of exercise on the motor unit. Muscle Nerve case.

7. Pickrell KL,

1986; 9: 104-26. 16. Williams NS, Hallan RI, Koeze TH, Watkins ES. Construction of a neorectum and neoanal sphincter following previous proctocolectomy. Br J Surg 1989; 76: 1191-94. 17. Williams NS, Hallan RI, Koeze TH, Pilot M-A, Watkins S. Construction of a neoanal sphincter by transposition of the gracilis muscle and prolonged neuromuscular stimulation for the treatment of faecal incontinence. Ann R Coll Surg Eng 1990; 72: 108-13.