Complications and functional outcome following artificial anal sphincter implantation (Br J Surg 2002; 89: 877-881)

Abstract

SirWe commend the authors for their objectivity. Infection, migration and erosion are well-documented serious complications of artificial anal sphincter implantation, yet may be relatively under-reported. It has long been our belief that any artificial devices or material brought into direct contact and completely encircling the bowel wall would eventually fail as a result of displacement and/or subsequent erosion. In this long-term study over three-quarters of the patients experienced postoperative complications and nearly a third had to have definitive explantation of the device. Similar complications led to the failure and subsequent abandonment of the Angelchik prosthesis for gastro-oesophageal reflux disease 1 . In the light of the evidence in this study, we feel that the electrically stimulated gracilis neoanal sphincter perhaps offers a more suitable alternative than the artificial bowel sphincter for the treatment of severe end-stage faecal incontinence, since it utilises autologous tissue 2 . Whilst it appears a more complex procedure, with advances in technology and appropriate training it can now be performed with greater consistency and lower morbidity than during its initial development. Results from our centre show a significant decrease in sepsis and mechanical failure and an overall success rate of approximately 70 per cent in the longer-term. This improvement has been considerably aided by developing a supra-regionally funded multidisciplinary team with reasonable throughput. Whilst we reserve judgement on the artificial bowel sphincter, we suspect that with longer follow-up, like the Angelchik prosthesis, it will be confined to the annals of surgical history.J SirSeow Choen is right to point out that poor surgical performance should not be allowed to hide behind a mask of adjuvant radiotherapy. But there is more than a hint of surgical arrogance in his suggestion that master surgeons only need to do a good operation to ensure a favourable cancer outcome. Sure, there is a minority of surgeons with high local recurrence rates who should be discouraged from continuing to practice on these challenging patients. On the other hand there are probably not enough master surgeons, however defined, to operate on the 9000 new cases of rectal cancer in the United Kingdom each year. There will always be a range of ability; most of us will be around the mean and the aim must be to drive down the average local recurrence rate, by whatever means. This will of course include tutelage under a specialist teacher, but at what point does the newly appointed consultant surgeon become expert? It may take 5-10 years for mature data on local recurrence to become available and it will then be too late for some patients. The master surgeon will still be challenged by the anterior lower rectal third tumour with a whisp of mesorectum separating if from the prostate and an MRI scan predicting tumour within 1mm of the radial margin. More trials are needed to define neo-adjuvant treatment protocols but there is already some e...

The Colorectal Development Unit: impact on functional outcome for the electrically stimulated gracilis neoanal sphincter

“…This discrepancy can be accounted for by the fact that many patients underwent an additional procedure to an ESGN and the complication rates included those due to these additional procedures. Sacral nerve stimulation has far less complications but the patients in this study were far too impaired for sacral nerve stimulation and we feel that an artificial bowel sphincter is not an option owing to sepsis [16]. Although ESGN has a high morbidity this is still preferable for our well‐counselled patients desperate to avoid a stoma at all costs.…”

“…Longer‐term results have been disappointing [14] whilst a systematic review of an artificial bowel sphincter concluded that it is of uncertain benefit and may actually harm many patients [15]. It remains to be seen how long‐term function will be with such a device [16].…”