Comparison of modified MMK Operation vs Kelly’s Plication for female stress incontinence

Chander Chawla*, Shripad Hebbar**     

* Dr. Chander Chawla, Former Prof & Head, Dept of O.B.G., Manipal Teaching Hospital, Pokhara, Nepal

** Dr. Shripad  Hebbar, Associate Professor, Dept of O.B.G., Kasturba Medical College, Manipal (Currently on deputation to Melaka Manipal Medical College, Malaysia)

Corresponding Author

Dr. Shripad Hebbar, MD,Shrigandha, 1-71-C, Budnar,Opposite to MGM College, Kunjibettu,UDUPI-576102, KARNATAKA STATE, Phone: 0820-2531228

Email: drshripadhebbar@yahoo.co.in

Abstract: 

OBJECTIVES: To compare the results of modified MMK operation with vaginal repair in female stress incontinence and to study the prognostic variables that determine the success. 

MATERIALS AND METHODS: Urinary stress incontinence was treated by modified MMK operation in 59 and by Kelly’s vaginal repair in 42 women. All were assessed preoperatively by detailed history, clinical examination. No sophisticated urodynamic studies were done. Follow up was done initially at 3 months, then at 6 and 12 months.   

RESULTS: Better results were obtained with modified MMK operation (88.1%) compared to vaginal repair (69%). Results were excellent in type I incontinence and in the absence of pelvic floor relaxation. Hysterectomy did not seem to improve the results. Complication rates were acceptable with modified MMK operation. 

CONCLUSION: Though laparoscopic Burch colposuspension is the preferred management in female urinary stress incontinence, this modality is not available at all centers. For institutions which provide basic gynaecological services, we recommend modified MMK operation as a primary surgery for female urinary stress incontinence to obtain higher cure rates and to minimize the incidence of recurrent urinary stress incontinence following vaginal repair.

 KEY WORDS:   Stress incontinence, Kelly’s repair, modified MMK operation

 Introduction:

 Since Marshall, Marchetti, and Krantz [1] introduced the application of retropubic urethrovesical suspension in 1949, transabdominal procedures have been used with increasing frequency and higher success rates to correct urinary incontinence. [2]  Although numerous technologies and variation of retropubic repairs have been described, the basic concept remains the same; to suspend and to stabilize the bladder neck and proximal urethra in a retropubic position to prevent their descent out side the sphere of abdominal pressure transmission during raised intra-abdominal pressure.

 In a recent systematic review of 39 trials involving 3301 women, Lapitan M et al (2005) assessed effects of open retropubic colposuspension for the treatment of urinary incontinence. [3]  Overall cure rates were 68.9% to 88.0% for open retropubic colposuspension. Two small studies suggested lower failure rates after open retropubic colposuspension than conservative treatment. Similarly, one trial suggested lower failure rates after open retropubic colposuspension compared to anticholinergic treatment. Evidence from six trials showed a lower failure rate for subjective cure after open retropubic colposuspension than after anterior colporrhaphy. Such benefit was maintained over time. In comparison with needle suspensions there was a lower failure rate after colposuspension in the first year after surgery. Evidence from twelve trials in comparison with suburethral slings found no significant difference in failure rates in all time periods assessed. Patient-reported failure rates in short-, medium- and long-term follow-up showed no significant difference between open and laparoscopic retropubic colposuspension, but with wide confidence intervals. In general, the evidence available does not show a higher morbidity or complication rate with open retropubic colposuspension, compared to the other open surgical techniques, although pelvic organ prolapse is more common than after anterior colporrhaphy and sling procedures. 

The evidence available indicates that open retropubic colposuspension is an effective treatment modality for stress urinary incontinence especially in the long term. Within the first year of treatment, the overall continence rate is approximately 85-90%. After five years, approximately 70% patients can expect to be dry. Newer minimal access procedures like tension free vaginal tape look promising in comparison with open colposuspension but their long-term performance is not known and closer monitoring of its adverse event profile must be done. Laparoscopic and open approaches for performing retropubic surgery have similar short-term cure indexes, as well as similar occurrence of complications and satisfaction level of patients. [4]   Laparoscopic colposuspension should allow speedier recovery but its relative safety and effectiveness is not known yet.

In Manipal Teaching Hospital, Pokhara, Nepal, we have been using modified MMK operation for female stress incontinence, whenever indicated. We present an analysis of our results and discussion of our experience.

 Our technique of MMK operation

 We have modified the original MMK operation to our convenience. After the usual retropubic dissection, we place three sutures on either sides of urethrovesical junction using No. 1 Vicryl. The most medial stitch is placed 1 cm lateral to urethrovesical junction (as delineated by inflated Foley’s bulb), the intermittent stitch 2cm and lateral most 3 cm lateral to the bladder neck ensuring that the good thickness of paraurethral tissue is included in the bite excluding vaginal mucosa. After taking each bite on paracolpos, the needle is passed through the periosteum of pubic symphysis just below the cooper’s ligament at similar distances. Sutures are tied only after taking all bites successively on either side.

 Figure 1.  Anatomy of retropubic space

Figure II. Our modification of MMK Operation

Our modification differs from the original MMK in the sense that we avoid exclusive midline sutures on cartilage of symphysis pubis (hence avoiding possibilities of urethral occlusion) and it differs from Burch colposuspension because cooper ligaments are avoided, as we feel that sutures are placed relatively high in Burch colposuspension and the extra pull while tying the knots may cut through the paravaginal tissues. 

Another potential concern with Burch colposuspension is extreme anteriorization of anterior vaginal wall resulting in alteration of intra abdominal pressure distribution which subsequently leads to pelvic organ prolapse. [5, 6]  This is because of high lift that occurs during knot tying in Burch procedure because the cooper’s ligament is anatomically at a higher place compared to cartilage of pubic symphysis. Our modification involves taking bites at a lower level there by averting too much pull on anterior vagina. 

Patients and Methods:

 Between 1998 and 2003, a total of 93 cases were operated for stress incontinence. Selection of the patients for the appropriate surgery was done with careful history taking, physical examination which included Bonney’s test and Q-tip test. All the patients were screened for urinary tract infection and if bacilluria was present, suitable antibiotics were given for minimum of 3 weeks before taking up for surgery. Similarly, the medico-surgical diseases which may increase intra-abdominal pressure were treated before hand. Diabetes was not a contraindication once it was well controlled. We have not performed sophisticated urodynamic tests because of their non-availability and limited utility. We performed Q tip test using lubricated cotton tipped applicator in the urethra with the cotton tip at the urethrovesical junction with the patient lying in the supine position. The angle formed by the applicator stick and an imaginary line drawn parallel to the floor was measured at rest and while patient performed the Valsalva maneuver. The difference of more than 30⁰ was an indication of poor anatomical support and rotational descent of urethra. 

The mean age of the patients at the time of admission was 52 years (range 27-68 years) and 62 patients were menopausal. The mean weight was 56 kgs and 19 patients weighed more than 70 kgs. 

The modified MMK operation was done as a primary procedure in 51 women (86%) and as secondary for recurrence of incontinence following Kelly’s plication in 8 (14%). In 19 women, it was done along with total abdominal hysterectomy with Chawla’s sling operation. Following the surgery, the patients were followed up in OPD at 3 months, at 6 months and at the end of one year. Follow up was possible in all cases up to 3 months, in 68 cases up to 6 months and in 52 subjects at the end of one year.  

Study design (Fig 3):

We divided the study population into three groups according to preexisting pelvic floor prolapse, previous abdominal or pelvic surgery for the purpose of evaluating influence of these factors on cure rates.

Group I: Consisted of patients who had coexistent pelvic floor relaxation. They were further divided into

Group I a: Includes those with 1^st and 2^nd degree uterovaginal descent who underwent abdominal hysterectomy with Chawla’s sling operation and modified MMK operation. (n=19)

Group I b: Those with 3^rd and 4^th degree uterovaginal descent who underwent vaginal hysterectomy with pelvic floor repair and Kelly’s plication. (n=42)

Group II: Those with no genital prolapse and no previous pelvic surgery. (n=23)

Group III:  Included patients who had previous hysterectomy for benign pelvic pathology (n=9).

We had 19 patients of stress incontinence (Group I a) along with mild degree of pelvic descent (Grade I and Grade II uterine descent according to Shaw’s classification). [7]   They underwent Chawla’s sling operation (where in vaginal vault is anchored to anterior superior iliac spine on either side using mersilene tape). [8]   17 of them were successfully treated. There were 42 cases (Group I b) of significant uterovaginal prolapse (Grade III and IV) and they were offered Ward Mayo surgery (vaginal hysterectomy with anterior colporrhaphy and posterior colpoperineorrhaphy) along with Kelly’s bladder neck plication. Stress incontinence recurred in 13 patients and only 8 of them further agreed for abdominal retropubic surgery. We could achieve success in 6 of those who underwent second surgery.

23 patients (Group II) had neither pelvic prolapse nor previous history of pelvic surgery and presented only with stress incontinence. They were operated abdominally using modified MMK technique. Stress incontinence recurred in two of them at the end of one year of follow-up.

Nine had undergone hysterectomy for benign pelvic disease (dysfunctional uterine bleeding 4, fibroid uterus 3, chronic pelvic inflammatory disease 1, endometriosis 1). All of them underwent modified MMK operation.  8 of them were continent at one year of follow-up.

Results: 

It can be observed from Table I that most of our patients were multiparae. 

In 61 patients the urinary incontinence was associated with pelvic floor dysfunction.

Table II shows the incidence of different grades of uterovaginal prolapse seen in these women (Shaw’s classification is followed).

It is evident from table III that many of the patients had rotational descent of the urethra (type II) abnormality). This may be due to the fact that majority had genital prolapse.

Table IV shows that whenever there is type II GSI, the results will be on lower side whatever the procedure adapted. However modified MMK operation gives better results even in type II abnormality. The overall result with modified MMK operation is 21% higher than Kelly’s plication and this difference is statistically significant (chi square test, c2=6.59, p<0.01).

The average operating time exclusively for modified MMK operation was 34 minutes. Whenever we opened anterior abdominal wall for the first time, we could limit our incision to 6 cm in majority of the cases. However, in the presence of previous scar we extended the incision to 8-10 cm, anticipating bladder injury due to fibrosis.

The overall results of modified MMK operation are shown in table V. When the modified MMK operation was done as a primary procedure, we could achieve 90% success rate, however it dropped to 75% when done as a secondary procedure following failed Kelly’s plication. Combining hysterectomy did not improve the overall success rate (89.5% vs 91.3%, p value >0.05). Similarly we did not observe any difference in cure rates in Type I and Type II incontinence (91.7% vs 88.9%, p>0.05).

We observed hematuria during surgery in 4 cases. We presumed it to be due to sutures accidentally running through bladder wall and we took the help of our surgical wing to remove the offending sutures under cystoscopic guidance. However we had no cases of bladder tear requiring repair. Other complications noted were postoperative urinary retention (10), urinary tract infection (2), wound hematoma (1), wound infection (3) and thrombophlebitis (1). Two developed osteitis pubis in 3^rd week. We did not have any cases of ureteric injuries or bladder fistula.  

Discussion:

In our series, 69% of those who had Kelly’s repair and 88.1% of those who had modified MMK operation were cured. Rotational descent of urethra (Type II incontinence) and pelvic floor relaxation were adverse prognostic factors. [9]  Though there are contradicting reports regarding success rate according to the type of urinary stress incontinence, [10]   in our series better results were obtained in patients with type I incontinence. Combining hysterectomy did not seem to improve the overall results. Similar observations have been made when Burch retropubic urethropexy has been combined with abdominal hysterectomy. [11]   

Mainprize and Drutz summarized 56 articles reporting the results of MMK procedures. [12]  Only few of these researches used preoperative diagnostic urodynamic tests. Of 2712 cases, 2334(86.1%) succeeded, 73 (2.7%) improved and 305 (11.2%) failed. The success rate of primary surgery was 92.1%; the success rate was 84.5% when used for recurrent urinary incontinence following failed vaginal procedure. Table VI shows previous comparative studies between abdominal approach and vaginal route indicating continence rates at the end of one year and it is evident that our series support the superiority of modified MMK operation over Kelly’s plication in accordance with published reports.

The complication rates were acceptable. All our patients who underwent modified MMK operation were on catheter drainage for minimum of seven days. The residual urine volume ranged between 65-150 ml (mean 78 ml) on the day of catheter removal. In 10 patients recatheterization was done as 6 of them could not void at all and in 4 it was considered because the residual urine exceeded 100 ml. However all of them could void spontaneously after some days of catheterization and bladder drill.

Whenever the space of Retzius had the tendency to ooze at the end of procedure, we used romavac suction drain for retropubic space. 5 patients were transfused in post operative ward as drainage volume exceeded 400 ml. None of them needed more than one pint of blood.

Demirci F et al (1999) reviewed intraoperative and early complications of retropubic colposuspension in 360 patients. [18]   Ten patients had massive haemorrhage and 8 of them had a blood transfusion. Three patients had retropubic haematoma. Bladder injuries were noticed in 10 patients, 3 of whom were diagnosed postoperatively. One patient had unilateral ureteral kinking. Urinary retention occurred in 20 patients for more than 10 days and 2 required catheterization for 26 and 32 days respectively. Eighteen patients had a wound infection and 4 had wound abscess. Twenty nine patients had urinary infection. Urinary tract injury, haemorrhage and blood transfusion were significantly more common in women having secondary surgery than those having primary surgery. Deep venous thrombosis was diagnosed in 3 patients who had a Burch colposuspension with concomitant abdominal hysterectomy. The complication rates in our series were comparable and acceptable.

The requirement for blood transfusion in our series was 8.5 % (5/59), slightly on higher side compared to Burch colposuspension. This is because of the fact that lateral stitches we place in our modification need wider exposure. Anatomical knowledge of retropubic dissection, careful dissection, using cautery and gel foam to control bleeding and passive retropubic drainage using low pressure redivac system are the key factors in controlling peri-operative haemorrhagic complications.

Another potential problem with original MMK operation is postoperative voiding dysfunction because of urethral obstruction. [19]   This may be due to exclusive midline stitches on pubic symphysis. Our modification involves relatively lateral stitches and we did not experience any permanent voiding complications in our patients.

During follow-up period none of our patient had recurrence of pelvic floor relaxation. Fuat Demircia et al (2001) reviewed incidence of pelvic floor dysfunction in 220 women who had undergone Burch colposuspension. [20]   They observed cystocoele in 18; rectocoele in 32; enterocoele in 35. We presume our technique is superior to Burch colposuspension as too much anteriorization of vagina is avoided. However long term follow-up is needed to arrive at a definitive conclusion.

In our series of 59 patients who underwent modified MMK operation 2 had osteitis pubis (3.4%), as evidenced by severe suprapubic pain radiating to thighs, marked tenderness and swelling over the pubic symphysis developing in 3^rd week of surgery. None of them had radiological features of bone destruction. Both responded to antibiotics and NSAIDs. Kammerer-Doak et al (1998) reviewed 2030 cases of MMK operation and reported 15 cases of osteitis pubis (0.74%) and 7 of them required surgical intervention.[21]  

Few surgeons suggest that the patient should complete her childbearing before surgical correction of incontinence is attempted. Only eight pregnancies have been reported after MMK operation in literature, 7 of them delivered vaginally. We had 4 (one nullipara and 3 primiparae) young patients who underwent modified MMK operation with uterine conservation. No pregnancies were reported in follow up period so far. 

Conclusions:

From this study it is evident that modified MMK operation is an effective treatment modality for stress urinary incontinence compared to Kelly’s plication. Within first year of treatment, the overall continence rate is 90% similar to other open retropubic approaches. Though success rates slightly decline when done for recurrent stress incontinence, it is still the method of choice as cure rates up to 85% can be achieved. Concomitant hysterectomy does not seem to improve overall success rates and hence should be done only when an indication for hysterectomy co-exists. However it should be born in mind that complications such as osteitis pubis and retropubic haematoma are exclusive to MMK operation due to retropubic approach.  

We continue to favour the less complex Kelly’s plication combined with vaginal hysterectomy with pelvic floor repair (Mayo-Ward) for the usual patient with minimal moderate stress incontinence and associated pelvic floor relaxation. However, abdominal procedure should be considered for those with severe urinary incontinence (perhaps aggravated by a medical disease such as COPD or obesity); those with apparently good vesical support (e.g., a snug vagina with minimal anterior rotational descent of urethrovesical junction) and those associated with intra-abdominal disease that requires suprapubic incision. The primary retropubic suspension, which relies more on substantial tissues, would appear to provide more permanent urethral support and everlasting control of symptoms in patients with complex and severe degree of urinary incontinence. It will also eliminate the distressing incidence of recurrent stress incontinence associated with vaginal repair.

REFERENCES:

1. Marshal VF, Marchetti AA, Krantz KE: The correction of stress incontinence by simple vesicourethral suspension. Surg. Gynecol. Obst., 88:509, 1949.

2. Parnel JP, Marshal VF, Vaughan ED Jr: Primary management of urinary stress incontinence by the Marshall-Marchetti-Krantz vesicourethropexy. J Urol 1982;127:679.

3. Lapitan M, Cody Dj, Grant A. Open retropubic colposuspension for urinary incontinence in women. Cochrane Database Syst Rev. 2005 Jul 20;(3):CD002912.

4. Bezerra CA, Rodrigues AO, Seo AL, Ruano JM, Borrelli M, Wroclawski ER. Laparoscopic Burch surgery: is there any advantage in relation to open approach? Int Braz J Urol. 2004 May-Jun;30(3):230-6.

5. Bezerra CA, Rodrigues AO, Seo AL, Ruano JM, Borrelli M, Wroclawski ER. Laparoscopic Burch surgery: is there any advantage in relation to open approach? Int Braz J Urol. 2004 May-Jun;30(3):230-6.

6. Kwon CH, Culligan PJ, Koduri S, Goldberg RP, Sand PK. The development of pelvic organ prolapse following isolated Burch retropubic urethropexy. Int Urogynecol J Pelvic Floor Dysfunct. 2003 Nov;14(5):321-5.

7. Demirci F, Yucel O, Eren S, Alkan A, Demirci E, Yildirim U. Long-term results of Burch colposuspension. Gynecol Obstet Invest. 2001;51(4):243-7.

8. Padubidri VG, Daftary SN. Shaw's Textbook of gynecology. New Delhi: BI-Churchill Livingstone; 2001

9. Chawla CD, Synopsis of Clinical Obstetrics, B.I. Churchill Living Stone Pvt Ltd, New Delhi, 2001,Chapter 20:179-182.

10. Park GS, Miller EJ. Surgical treatment of stress urinary incontinence: a comparison of the Kelly’s plication, Marshall-Marchetti-Krantz, and Pereyra procedures. Obstet Gynecol 1988,71:575-579.

11. Green TH, Development of a plan for diagnosis and treatment of urinary stress incontinence. Am J Obstet Gynecol 1962;83:632.

12. Meltomaa SS, Haarala MA, Taalikka MO, Kiilholma PJ, Alanen A, Makinen JI. Outcome of Burch retropubic urethropexy and the effect of concomitant abdominal hysterectomy: a prospective long-term follow-up study. Int Urogynecol J Pelvic Floor Dysfunct. 2001;12(1):3-8.

13. Mainprize TC, Drutz HP. The Marshall-Marchetti-Krantz procedure: a critical review. Obstet Gynecol Surv 1988:43:724-729.

14. Girao MJ, Sartori MG, Kobata SA, de Andrade FM, Juliano Y, Baracat EC, de Lima GR. Surgical treatment of stress urinary incontinence: analysis of the results by abdominal or vaginal pathways. Rev Assoc Med Bras. 1995 Nov-Dec;41(6):397-400.

15. Zhu L, Lang J, Liu Z. Comparison of different surgical procedures for urinary stress incontinence. Zhonghua Yi Xue Za Zhi. 1998 Aug;78(8):601-3.

16. Glazener CM, Cooper K. Anterior vaginal repair for urinary incontinence in women. Cochrane Database Syst Rev. 2000;(3):CD001755.

17. Lan Z, Jinghe L, Zhufeng L, Rongli H. Comparison the efficiency of different surgical procedures for urinary stress incontinence. Chin Med Sci J. 2000 Jun;15(2):107-10.

18. Demirci F, Yucel N, Ozden S, Delikara N, Yalti S, Demirci E. A retrospective review of perioperative complications in 360 patients who had Burch colposuspension. Aust N Z J Obstet Gynaecol. 1999 Nov;39(4):472-5.

19. Romero Maroto J, Prieto Chaparro L, Lopez Lopez C, Quilez Fenoll JM, Rodriguez Fernandez E. Obstruction following surgical repair of female stress urinary incontinence. Diagnosis and treatment. Arch Esp Urol. 2002 Nov;55(9):1107-14.

20. Fuat Demircia, Oguz Yucela, Sadiye Erenb, Akif Alkanb, Elif Demircib, Uzay Yldrma. Long-Term Results of Burch Colposuspension. Gynecologic and Obstetric Investigation 2001; 51:243-247.

21. Kammerer-Doak DN, Cornella JL, Magrina JF, Stanhope CR, Smilack J. Osteitis pubis after Marshall-Marchetti-Krantz urethropexy: a pubic osteomyelitis. Am J Obstet Gynecol. 1998 Sep;179(3 Pt 1):586-90.