Lack of efficacy of an intradural somatic-to-autonomic nerve anastomosis (Xiao procedure) for bladder control in children with myelomeningocele and lipomyelomeningocele: results of a prospective, randomized, double-blind study.

OBJECTIVE Xiao et al. and other investigators have studied an intradural somatic-to-autonomic (e.g., L-5 to S3-4) nerve transfer as a method to create a reflex arc to allow bladder emptying in response to cutaneous stimulation (the Xiao procedure). In previous clinical studies of patients with spinal dysraphism who underwent the Xiao procedure, high success rates (70%-85%) were reported for the establishment of a "skin-CNS-bladder" reflex arc that allows spontaneous, controlled voiding in children with neurogenic bladder dysfunction. However, many of these studies did not use blinded observers, did not have control groups, and/or featured only limited follow-up durations. METHODS A randomized, prospective, double-blind trial was initiated in March 2009, enrolling children with myelomeningocele (MM), lipomyelomeningocele (LMM), and neurogenic bladder dysfunction who were scheduled for spinal cord detethering (DT) for the usual indications. At the time of DT, patients were randomized between 2 arms of the study: half of the patients underwent a standard spinal cord DT procedure alone (DT group) and half underwent DT as well as the Xiao procedure (DT+X group). Patients, families, and study investigators, all of whom were blinded to the surgical details, analyzed the patients' strength, sensory function, mobility, voiding, and urodynamic bladder function before surgery and at regular intervals during the 3-year follow-up. RESULTS Twenty patients were enrolled in the study: 10 underwent only DT and the other 10 underwent DT+X. The addition of the Xiao procedure to spinal cord DT resulted in longer operative times (p = 0.024) and a greater chance of wound infection (p = 0.03). Patients in both treatment arms could intermittently void or dribble small amounts of urine (< 20% total bladder capacity) in response to scratching in dermatomes T-9 through S-2 using a standardized protocol, but the voiding was not reproducible and the volume voided was not clinically useful in any patient. Voiding in response to scratching was not more frequent in patients who underwent DT+X compared with those who underwent only DT. Bladder contractions in response to scratching occurred in both treatment arms at various intervals after surgery, but they were not more reproducible or more frequent in the patients who underwent the Xiao procedure than in the patients who did not. No patient in either treatment arm was continent of urine before, during, or after the study. CONCLUSIONS Patients with MM and LMM who underwent the Xiao procedure during spinal cord DT were no more likely to be able to void, to control their urination, to achieve continence, or to have a demonstrable urodynamic bladder contraction in response to cutaneous stimulation than patients who underwent only spinal cord DT. This study, in the context of disappointing results reported in other recent studies of the Xiao procedure, raises doubts about the clinical applicability of this procedure in humans until further basic science research is performed.