Intravesical administration of drugs has been used commonly in spinal cord injury patients to suppress detrusor hyperreflexia (eg oxybutynin, verapamil, terodiline) or, to initiate a micturition reflex (eg 15S 15-methyl prostaglandin F2 alpha, protaglandin E2); however, the response has been variable and sometimes, unpredictable. This prompted us to study the presence of nerve fibres in the vesical urothelium and submucosa in mucosal biopsies taken from the dome and trigone (obtained prior to performing a therapeutic procedure eg, vesical lithotripsy, or a diagnostic cystoscopy) in 47 consecutive, unselected paraplegic/tetraplegic patients with a neuropathic urinary bladder. Nerve fibres were demonstrated by routine immunohistochemical technique using commercially available monoclonal and polyvalent antibodies against S-100 (DAKO A/S, Glostrup, Denmark) and Neurofilament (MILAB, Malmo, Sweden). Biopsy specimens were graded for the presence of nerve fibres on a 0-3 scale for urothelium, and superficial/deep submucosa separately in a blind and randomised manner. Virtually no fibre presence was found in one paraplegic patient and no superficial or single fibres were noted in a tetraplegic patient. Absence of C-fibre hyperplasia (Grade 0) was found in nine cases (paraplegic: 4; tetraplegic: 5); Grade 1 hyperplasia was observed in 17 cases (paraplegic: 4; tetraplegic: 13); Grade 2 hyperplasia was seen in 11 cases (paraplegic: 7; tetraplegic 4); and Grade 3 hyperplasia was noticed in eight cases (paraplegic 3: tetraplegic: 5). The magnitude of C-fibre hyperplasia was not significantly different between paraplegic and tetraplegic patients (chi(2) = 4.64; P = 0.3262). The relationship, if any, between the degree of C-fibre hyperplasia and duration of paralysis was studied by categorising patients as5 years, and5 years of paralysis. No evidence of single fibre or fibre bundle hyperplasia (Grade 0) was seen in five and six cases, grade 1 hyperplasia in six and 11 cases, grade 2 hyperplasia in two and nine cases, and grade 3 hyperplasia in three and five cases respectively in these two categories of patients. (chi(2) = 1.92; P = 0.58). The possible relationship between C-fibre hyperplasia in the vesical mucosa/submucosa and (i) the vesical response to intravesical drug administration; (ii) the vesical urothelial proliferation arrest; (iii) the electrical stimulation of urinary bladder by implanted electrodes (sacral anterior root stimulator); and (iv) long-term indwelling urethral Foley catheter drainage, are discussed with illustrative case reports. In conclusion, mucosal biopsy and study of nerve fibres in urothelium and submucosa of neuropathic bladder has helped to generate hypotheses on the association between C-fibre hyperplasia and response to intravesical pharmacotherapy and the predictive value of such a study in identifying those patients who are likely to respond to intravesical pharmacotherapy.