Wim Van Gemert

The optimal treatment of desmoid tumours is controversial. We evaluated desmoid management in Dutch familial adenomatous polyposis (FAP) patients. Seventy-eight FAP patients with desmoids were identified from the Dutch Polyposis Registry. Data on desmoid morphology, management, and outcome were analysed retrospectively. Progression-free survival (PFS) rates and final outcome were compared for surgical vs non-surgical treatment, for intra-abdominal and extra-abdominal desmoids separately. Also, pharmacological treatment was evaluated for all desmoids. Median follow-up was 8 years. For intra-abdominal desmoids (n=62), PFS rates at 10 years of follow-up were comparable after surgical and non-surgical treatment (33% and 49%, respectively, P=0.163). None of these desmoids could be removed entirely. Eventually, one fifth died from desmoid disease. Most extra-abdominal and abdominal wall desmoids were treated surgically with a PFS rate of 63% and no deaths from desmoid disease. Comparison ...

The VBG was originally performed with a Marlex band and characterized by a satisfactory weight loss and low morbidity. The effect of the material used for the banding procedure (Marlex vs Dacron) in vertical banded gastroplasty (VBG) is evaluated. In 49 consecutive obese patients treated with a VBG, a Marlex band was used in 17 patients and a Dacron band in 32 patients. Data were analyzed retrospectively with regard to the type of band, weight loss and complications. A significant difference was found in the percentage excess weight 5 years postoperatively in favor of the Dacron group (59.2% vs 39.2%; p < 0.05) because of more band-related complications in the Marlex group. The difference in percentage excess weight disappeared 8 years postoperatively (43.3% vs 46.8%), due to the renewed weight loss of the Marlex group following reoperation. The Dacron band is superior to the Marlex band in VBG because sustained weight loss is satisfactory and morbidity is low.

An increasing number of patients with a failed primary bariatric procedure present themselves for secondary treatment. Only a few studies have investigated critically the success of revisional surgery. In the present study, the effectiveness of revisional surgery for failed vertical banded gastroplasty (VBG) is analyzed: restoration of the VBG (reVBG) is compared to conversion to a Roux-en-Y gastric bypass (RYGB). From 1980 to 1996, 136 consecutive morbidly obese patients underwent primary RYGB (n = 20) or VBG (n = 16). Weight loss, indications and complications after revisional surgery were registered. The rate of revisional surgery after primary and secondary bariatric procedures was estimated by means of a Kaplan-Meier analysis. Kaplan-Meier analysis revealed that 56% of the patients will eventually require revisional surgery after initial VBG over a 12-year period compared to 12% after initial RYGB (P<0.01). After reVBG 68% will need revisional surgery over a 5-year period, while no further revisional surgery was required after conversion to a RYGB (P<0.05). Body mass index dropped significantly after reVBG or conversion to RYGB for insufficient weight loss (P<0.05), however, more revisional surgery was necessary after reVBG to achieve this result. The complication rate was comparable between reVBG and conversion to RYGB (33%). Conversion of a failed VBG to a RYGB is more effective than a reVBG, because conversion to RYGB provides satisfactory weight loss without requiring further revisional surgery.

Surgical treatment of morbid obesity is gaining in popularity, because conservative treatment is ineffective. However, a cost-effectiveness analysis has never been performed and is the main goal of the present study. 21 consecutive morbidly obese patients were tested before and after vertical banded gastroplasty (VBG). Quality of life was assessed with the Nottingham Health Profile and a visual analogue scale. A prevalence-based cost-of-illness analysis of morbid obesity was performed and the cost-effectiveness of VBG assessed. VBG resulted in a significant weight loss and an improved quality of life. The improved quality of life combined with 3.6 life-years gained after VBG resulted in 12 quality adjusted life-years (QALY) gained in a lifelong scenario. Lifelong costs of illness of morbidly obese persons ranged from $8,304 to $9,367. Total direct costs of VBG were $5,865. The percentage of patients performing paid labor increased from 19% before VBG to 48% after VBG, resulting in an average productivity gain of $2,765 per year. In summary, the cost-effectiveness analysis revealed that surgical treatment of morbid obesity by means of VBG saves $4,004 to $3,928 per QALY (overall dominance). Because treatment of morbid obesity with VBG results in QALYs gained and less costs, there is no doubt that this procedure should be introduced or continued from a societal point of view.

Dynamic graciloplasty (DGP) improves anal continence and quality of life for most patients. However, in some patients, DGP fails and fecal incontinence is unsolved or only partially improved. Constipation is also a significant problem after DGP, occurring in 13-90%. Colonic irrigation can be considered as an additional or salvage treatment for defecation disorders after unsuccessful or partially successful DGP. In this study, the effectiveness of colonic irrigation for the treatment of persistent fecal incontinence and/or constipation after DGP is investigated. Patients with defecation disorders after DGP visiting the outpatient clinic of the University Hospital Maastricht were selected for colonic irrigation as additional therapy or salvage therapy in the period between January 1999 and June 2003. The Biotrol(R) Irrimatic pump or the irrigation bag was used for colonic irrigation. Relevant physical and medical history was collected. The patients were asked to fill out a detailed questionnaire about colonic irrigation. Forty-six patients were included in the study with a mean age of 59.3 +/- 12.4 years (80% female). On average, the patients started the irrigation 21.39 +/- 38.77 months after the DGP. Eight patients started irrigation before the DGP. Fifty-two percent of the patients used the irrigation as additional therapy for fecal incontinence, 24% for constipation, and 24% for both. Irrigation was usually performed in the morning. The mean frequency of irrigation was 0.90 +/- 0.40 times per day. The mean amount of water used for the irrigation was 2.27 +/- 1.75 l with a mean duration of 39 +/- 23 min. Four patients performed antegrade irrigation through a colostomy or appendicostomy, with good results. Overall, 81% of the patients were satisfied with the irrigation. Thirty-seven percent of the patients with fecal incontinence reached (pseudo-)continence, and in 30% of the patients, the constipation completely resolved. Side effects of the irrigation were reported in 61% of the patients: leakage of water after irrigation, abdominal cramps, and distended abdomen. Seven (16%) patients stopped the rectal irrigation. Colonic irrigation is an effective alternative for the treatment of persistent fecal incontinence after DGP and/or recurrent or onset constipation additional to unsuccessful or (partially) successful DGP.

The aim of this study was to compare long-term results for patients born with an anorectal malformation and fecal incontinence treated with a dynamic graciloplasty with those for the total group of patients undergoing dynamic graciloplasty. Consecutive patients with fecal incontinence after surgical treatment of anorectal malformation and treated with dynamic graciloplasty were included in this study. Preoperative assessment was performed. Postoperative follow-up consisted of anorectal manometry and registration of defecation frequency, continence scores, and postponement time of defecation. Twenty-eight patients with a median age of 25.5 years were included in the study. The median follow-up was 4 years. A high anorectal malformation was present in 89.3 percent of patients. Conventional graciloplasty had been previously performed in 36 percent. All patients were incontinent for stools. Median frequency of defecation was four times/day. Median postponement time of defecation was 0 minutes. Rectoanal inhibition reflex was present in 17 percent of patients. Median preoperative sensory threshold during balloon distention was 30 ml and median maximum urge threshold was 165 ml. Satisfactory continence was reached in 35 percent of patients, however, 7.1 percent of patients gained this continence score by additional bowel irrigation. Twenty-nine percent of patients were incontinent for loose stool, 36 percent were incontinent for formed stool. Satisfactory continence was achieved in only 18 percent of patients with a high anorectal malformation, compared with 100 percent in patients with a low anorectal malformation. In the total group of patients with dynamic graciloplasty, satisfactory continence was obtained in 76 percent. The sensitivity threshold in patients with a successful dynamic graciloplasty was lower than that in patients with a failing dynamic graciloplasty (45 vs. 24 ml, P = 0,06). When we compare median preoperative rectal sensitivity threshold in our study group with that in the total patient group with dynamic graciloplasty, statistical difference was established (P = 0.008). Postponement time (0 to 20 minutes) and anal squeeze pressure (81 to 120 mmHg) increased significantly after surgery. Patients with an anorectal malformation had significantly lower resting and stimulation pressure than that of the total group of patients, but the difference between resting and stimulation pressure in both groups was not significantly different (P = 0.33). The difference between resting and stimulation pressure was not significantly different between anorectal malformation patients with a failing dynamic graciloplasty and patients with a successful dynamic graciloplasty. Complications were noted in 57 percent of patients. Explantation of the dynamic graciloplasty was necessary in 32 percent of patients, mainly because of infection of the implant. Results of dynamic graciloplasty for fecal incontinence are reasonable for this specific group of patients with limited treatment options. Despite functional dynamic graciloplasty, the results are worse than those for the total group of patients with dynamic graciloplasty. Rectal sensitivity and type of malformation are prognostic factors for outcome and can be used to select patients for treatment with dynamic graciloplasty, thereby improving treatment outcome.

This study was designed to assess the healing rate of complex perianal fistulas in Crohn's disease after a multistep strategy, including induction treatment with Infliximab in case of active proctitis, followed by definitive surgery. From 2000 to 2003, all consecutive patients with complex fistulas and Crohn's disease underwent pretreatment with noncutting setons and, in case of severe recurrent fistulas or abscesses, a diverting stoma. Infliximab was added in cases of active proctitis. After definitive surgical treatment, patients were examined. Seventeen patients were included (median age, 34 (range, 22-58) years). Seven patients were treated by surgery only, and in ten patients Infliximab was added. After a median follow-up of 19 (range, 8-40) months, fistula healing was observed in 17 patients (100 percent). One patient of the Infliximab group developed a recurrent fistula (10 percent) after 24 months, and in one patient (10 percent) soiling occurred. Two patients of the surgical group developed a recurrent fistula (29 percent) and soiling occurred in two patients (29 percent). A multistep strategy followed by definitive surgery for the treatment of complex perianal fistulas in patients with Crohn's disease is a promising treatment modality. The preliminary results of this study suggest that Infliximab treatment has a beneficial additive effect in the multistep treatment followed by definitive surgery of complex anal fistulas and active proctitis in Crohn's disease.

To assess the efficacy of a staged strategy for the treatment of complex perianal fistula. Between January 1999 and April 2003 all consecutive patients with complex perianal fistulas were treated according to a staged strategy. Fistula tracks originating from the middle third or upper part of the anal sphincter were included. Patients were examined for recurrent fistulas and complaints of incontinence and soiling. Initial treatment consisted of a noncutting seton with or without a diverting stoma. Definitive surgical treatment consisted of an advancement flap or fistulotomy. Thirty patients were included (median age; 42 years, range 22-68 years). Seven had Crohn's disease without signs of rectal and anal involvement other than the fistula. At a median follow up of 22 months (range 8-52 months) in 29 (97%) patients, the wounds had healed completely; 7 (22%) patients subsequently developed a recurrent fistula and minor soiling occurred in 7 (23%) patients. Initial treatment with a seton with and without a diverting stoma minimizing inflammatory activity at the fistula site before definitive surgical treatment gave good results in this difficult group of patients.

The aim of the study was to determine the therapeutic stimulation threshold in patients with successful sacral nerve modulation for faecal incontinence. Patients who had undergone successful permanent sacral nerve modulator implantation and had been followed up for a minimum of 3 months were included. The sensitivity threshold and motor threshold were determined and correlated with therapeutic response. Patients went home with the stimulator set at 0.6 V below the sensitivity threshold. Each week the voltage was increased by 0.2 V until the sensitivity threshold was reached. The effects on anorectal physiology and continence were recorded. Eight patients (seven women) with a median age of 58.5 years were included. The median follow-up was 6.3 months. The median sensibility threshold volume of rectal sensation was 50 ml, the median urge threshold volume was 140 ml and the median maximum tolerated rectal volume 240 ml. The median number of incontinence episodes and days per week affected by incontinence decreased from 5.0 and 3.8 before operation to 0.7 and 0.7 respectively after follow-up for 3 months. At anorectal manometry the median resting and stimulation anal canal pressures were 57 and 85 mmHg respectively, and remained constant over time. The therapeutic response threshold was significantly lower than the sensitivity threshold (median 1.6 versus 1.7 V; P = 0.042). The median motor threshold was 2.1 V, significantly higher than the sensitivity threshold (P = 0.009). The stimulation threshold for suboptimal therapeutic response was 1.4 V. In five of the eight patients the therapeutic response threshold was the same as the sensitivity threshold. Sacral nerve modulation can produce a therapeutic effect below the sensitivity threshold. A lower stimulation voltage increases the lifespan of the pulse generator.