Effects of intravesical onabotulinumtoxinA on bladder dysfunction and autonomic dysreflexia after spinal cord injury: role of nerve growth factor


Mohamed S. Elkelini MD MSc, 399 Bathurst Street, MP 8-306, Toronto, Ontario, M5T 2S8, Canada. e-mail: m.elkelini@utoronto.ca


Study Type – Aetiology (case control)

Level of Evidence 3c

What's known on the subject? and What does the study add?

Neurogenic detrusor overactivity (NDO) and autonomic dysreflexia (AD) are common outcomes following spinal cord injury (SCI).

In this study, we showed that onabotulinumtoxinA controlled NDO and AD in rats with T4-SCI, and also provided a mechanism for the control of AS.


  • • To assess the significance of onabotulinumtoxinA (onabotA) intravesical administration in blocking autonomic dysreflexia (AD) response induced by cystometrogram (CMG) after T4 spinal cord transection (SCT).


  • • Female rats were stratified into three groups: a sham group; a SCT-only group; and a SCT with onabotA treatment group. Each group was further subdivided into two subgroups: AD assessment, or nerve growth factor (NGF) assessment via enzyme-linked immunosorbent assay (ELISA).
  • • Three weeks after T4-SCT, all groups were assessed. Arterial pressure and heart rate were measured during and after CMG.
  • • NGF was also extracted from the bladder and the dorsal root ganglia (DRG) of the T4 root and quantified by ELISA. In the onabotA-treated group, 48 h before assessment, onabotA (1 mL, 20 U/mL in saline) was given using a urethral tube and was left indwelling for 30 min.
  • • Univariate anova was used to analyse the data and statistical significance was set at P < 0.05.


  • • The maximum voiding pressure and the number of uninhibited contractions were significantly lower in the group treated with intravesical onabotA than in the SCT-only group.
  • • Intravesical onabotA significantly blocked the dysreflexia response (high arterial pressure with bradycardia) induced by CMG after SCT.
  • • Intravesical onabotA also significantly lowered NGF concentrations in the bladder and the T4 DRG segment.


  • • The results of the present study showed that intravesical onabotA controls neurogenic detrusor overactivity and AD after SCT.
  • • The findings shed light on the potential benefits of intravesical onabotA treatment in patients with spinal cord injury, and also provide a novel mechanism for the control of AD via a minimally invasive treatment modality.

autonomic dysreflexia




dorsal root ganglia


neurogenic detrusor overactivity


nerve growth factor


onabotulinumtoxinA onabotA, onabotulinumtoxinA


spinal cord injury


spinal cord transection.


Spinal cord injury (SCI) is a significant cause of morbidity and mortality in developed nations, with a global annual incidence of 1:25 000 [1,2]. Bladder dysfunction and autonomic dysreflexia (AD) are common outcomes of cervical and high thoracic SCI. Bladder dysfunction after SCI includes a brief spinal shock phase, followed by neurogenic detrusor overactivity (NDO) 2–12 weeks later. Bladder distension can cause initiation of AD, which involves life-threatening episodes of paroxysmal hypertension and bradycardia [3].

Autonomic dysreflexia and NDO after SCI develop in a time-dependent manner, suggesting that neuroplasticity contributes significantly to both conditions. After SCI, bladder afferent pathways reform and induce the growth of C-fibre afferents, containing calcitonin gene-related peptide. Nerve growth factor (NGF) is a member of the neurotrophin family and an important regulator of neural survival, development, function and plasticity. NGF plays a major role in the enlargement of the afferent arbour after SCI [4], and intrathecally delivered anti-NGF has reduced AD in rats with SCI [5]. Hence, NGF and its receptors in the bladder and spinal cord might offer potential targets for new therapies to control AD and NDO after SCI.

Current management protocols for AD and NDO are mostly symptomatic and ineffective. [6]. OnabotulinumtoxinA (onabotA) has been used in SCI patients to reduce detrusor overactivity via inhibition of acetylcholine release from efferent nerve endings [7]. There is increasing evidence that onabotA might also affect sensory nerve fibres and afferent signalling mechanisms [7]. OnabotA also reduces NGF concentrations in bladder tissue of patients with detrusor overactivity [8,9]. Intravesical onabotA administration might block AD by acting on primary afferent fibres and putative spinal neurones.

In the present study, our goal was to assess the effect of onabotA on AD and NDO after SCI in a clinically relevant animal model.


Female Sprague–Dawley rats (n= 44), weighing 200–250 g were used in the present study. Rats were stratified into three groups: sham control (laminectomy with intact spinal cord), SCT, and spinal cord transection (SCT) + onabotA. Each group was subdivided for AD and NGF assessment (six to eight rats per subgroup). Animals were maintained for 3 weeks after surgery. The protocol was approved by the Animal Care Committee of University Health Network/University of Toronto in accordance with the policies established in the Guide to Care and Use of Experimental Animals prepared by the Canadian Council on Animal Care.

Spinal cord transection was performed as described previously [10]. Briefly, under general anaesthesia, a limited laminectomy to the fourth and fifth vertebrae was performed. Using a sharp microscissor, a complete transection of the spinal cord at T4 level was performed under direct visual control then aided by an operating stereomicroscope (Spencer, American Optical Company, NY, USA). To ensure a complete transection of the spinal cord, the tip of 16-G needle was passed several times around the inner surface of the exposed vertebra. After surgery, the rat bladders were evacuated by manual expression.

Intravesical instillation of onabotA was performed as follows. Under general anaesthesia using a combination of xylazine (5 mg/kg) and ketamine (50 mg/kg), PE-50 tubing (Clay-Adams, Parsippany, NJ, USA) was inserted into the bladder through the urethra. The bladder was emptied of urine and slowly filled with onabotA (1 mL, 20 U/mL in saline; Allergan, Irvine, CA, USA), which was left indwelling for 30 min. Rats were allowed to recover, and 48 h later they either had a suprapubic catheter implanted for cystometrogram (CMG; AD group) or were killed for dorsal root ganglia (DRG) retrieval (NGF group). We measured the effect of onabotA 48 h after administration because previous reports using the same animal model showed improved bladder dysfunction after onabotA treatment [11,12].

Suprapubic catheter implantation and AD assessment during CMG were carried out as follows. Under general anaesthesia, a silicon tube was implanted in the bladder via laparotomy based on a method described previously [10]. CMG was conducted on conscious rats held within a restrainer (IITC Life Science, Woodland Hills, CA, USA). Rats were kept in a warming chamber at 30 °C to maintain optimal ambient temperature for blood pressure reading. A tail cuff (7/16 inch) with photoelectric sensors (IITC Life Science) measured blood pressure and heart rate. The system also includes an automated sphygmomanometer, amplifier and scanner (IITC Life Science) attached to a computer interface.

Rats were acclimatized for at least 30 min. Blood pressure measurements were initiated through a computer interface and recorded with the help of IITC BpMon software. Measurements were taken at baseline and during bladder dilatation via CMG. During CMG, the bladder was filled with sterile saline at a rate of 0.2 mL/min using an infusion pump (Model 2620, Harvard Apparatus Holliston, HA, USA). At least four micturition cycles were monitored in each rat. The ladder pressure was recorded with a Grass Polygraph (Model 7D). Immediately before maximum voiding pressure (which could be predicted after a few micturation cycles), blood pressure measurements were initiated and recorded; the heart rate was calculated by the software. Measurements were taken in triplicate and the mean calculated.

For NGF immunoassay, peptides were extracted from DRG and the total protein content was determined with a bicinchoninic acid assay (Pierce, Rockford, IL, USA) for standardization. A commercial rat NGF ELISA kit (Promega, Madison, WI, USA) was used. Samples were acid-treated to increase the amount of detectable NGF and stored at −20 °C. A Nunc MaxiSorp 96-well ELISA plate was coated with polyclonal antibody in carbonate coating buffer and incubated at 4 °C for 24 h. The plate was washed with Tris-buffered saline with Tween 20 wash buffer, blocked with buffer (Promega), and incubated at room temperature for 1 h. The plate was washed, monoclonal antibodies were added to the wells and the plate was incubated at 4 °C for 24h. The plate was washed, anti-rat immunoglobulin G horseradish peroxidase conjugate was added to the wells and the plate was incubated at room temperature for 2.5 h with shaking (220 rpm). The plate was washed, TMB One solution was added and the plate was incubated at room temperature for 10 min with shaking (220 rpm). The reaction was stopped with 1 N HCl and absorbance was read on a plate reader at 450 nm. Using a standard curve, sample NGF concentrations were derived.

For statistical analysis, an independent t-test and one-way analysis of variance (anova) were used to analyse the data. P < 0.05 was considered to indicate statistical significance.


Complete transection of the spinal cord at the T4 segment resulted in total flaccid paralysis of the lower limbs accompanied by bladder areflexia. Rats began regaining bladder contractility at ≈10 days after surgery as evidenced by smaller evacuated volumes of urine during daily bladder squeezing. By day 14, most rats were able to move their hips and knee joints; however, no weight-bearing ability was evident during the 3-week study period.

During CMG, the control group had a smooth filling phase, with no detrusor activity (Fig. 1). By contrast, the CMG of SCT rats showed a filling phase with uninhibited contractions (8.0 ± 0.7) and bladder pressure reached 31.24 ± 4.7 cmH2O. OnabotA treatment significantly reduced the number of uninhibited contractions (3.0 ± 0.4, P < 0.001); bladder pressure reached 24.8 ± 6.3 cmH2O. Uninhibited contractions were defined as intravesical pressure waves greater than 8 cmH2O; they were not accompanied by urethral fluid discharge. Large-amplitude voiding contractions occurred in all study groups; however, the voiding variables were markedly different. The resting (lowest) pressure was significantly higher in the SCT group (8.25 ± 0.85 cmH2O) than in the control (2.5 ± 0.28 cmH2O; P= 0.04). However, we found no significant difference in resting pressure between SCT and SCT + onabotA rats. The maximum voiding pressure was significantly higher in the SCT group (54.25 ± 2.1 cmH2O) than in the control (33.3 ± 2.4 cmH2O; P < 0.001), whereas it was significantly lower in the SCT + onabotA group (41.00 ± 1.2 cmH2O) than in the SCT group (54.25 ± 2.1cmH2O; P= 0.003) (Table 1).

Figure 1.

Bladder pressure changes (CMG) in control rats at baseline and during CMG. a, Normal rat; b, SCT rat; c, SCT + onabotA rat. Rats were tested 3 weeks after SCT. Asterisks indicate the voiding contractions. Note that uninhibited contractions occurred in the SCT rat but were not detected in the normal rat and were significantly reduced with onabotA treatment.

Table 1.  Cystometrogram data from control, SCT, SCT + onabotA groups
GroupsResting pressure, cmH2OMax. voiding pressure, cmH2ONo. of uninhibited contractions
  • *

    P= 0.003,

  • **

    P= 0.001.

Control2.5 ± 0.2833.3 ± 2.40
SCT8.25 ± 0.8554.25 ± 2.18.0 ± 0.7
SCT + onabotA8.75 ± 1.741.00 ± 1.2*3.0 ± 0.4**

Three weeks after SCT, the mean arterial pressures in the SCT (90.66 ± 1.4 mmHg) and SCT + onabotA (93.25 ± 6.7 mmHg) groups were not significantly different from the control (94.25 ± 5.0 mmHg). By contrast, the heart rate in the SCT group (520.0 ± 27.1 beats/min) was significantly higher than in the control (417.75 ± 11.5 beats/min; P= 0.006); the heart rate in the onabotA treatment group was not significantly different from the control (Table 2). In control rats, bladder distension via CMG did not cause a significantly different change in mean arterial pressure and heart rate. By contrast, bladder distension 3 weeks after SCT increased the mean arterial pressure by 35.67 ± 5.13 mmHg (P= 0.001), whereas in the SCT + onabotA group the mean arterial pressure increased by only 14.5 ± 2.0 mmHg (P= 0.011). Moreover, bladder distension 3 weeks after SCT decreased the heart rate by 104 ± 47.88 beats/min (P= 0.01) while, in the onabotA treatment group, the heart rate decreased by only 28.75 ± 9.1 beats/min (P= 0.024; Fig. 2).

Table 2.  Mean arterial pressure (MAP) and heart rate (HR) data at baseline and CMG for control, SCT, SCT + onabotA groups
MAP, mmHgHR, beats/minMAP, mmHgHR, beats/min
  • *

    P= 0.001,

  • **

    P= 0.01.

Control94.25 ± 5.0417.75 ± 11.593.00 ± 5.4426.25 ± 25
SCT90.66 ± 1.4520.0 ± 27.1126.33 ± 6.11*416.00 ± 43.7**
SCT + onabotA93.25 ± 6.7458.25 ± 11.98107.75 ± 4.1429.00 ± 9.9
Figure 2.

Arterial pressure (a) and heart rate (b) changes in response to CMG.

Three weeks after SCT, NGF concentration was significantly higher in the SCT T4-DRG group (557.66 ± 79.54 pg/mL) than in the control (105.50 ± 33.21 pg/mL; P= 0.002), whereas after onabotA treatment in rats with SCT, the NGF concentration in T4-DRG (152.66 ± 63.28 pg/mL) was significantly lower than in the SCT study group (P= 0.006). Furthermore, the bladder NGF concentration was significantly higher after SCT (610.33 ± 143.20 pg/mL) than in the sham control group (11.86 ± 1.97 pg/mL; P= 0.01); whereas, after onabotA treatment in rats with SCT, the NGF concentration in the bladders was significantly lower (136.00 ± 58.66 pg/mL) than in the SCT study group (P= 0.028; Fig. 3).

Figure 3.

NGF concentration at T4-DRG (a) and bladder (b) in control, SCT and SCT + onabotA groups.


The rat is a common SCI model because it provides an inexpensive and reliable method to characterize complex clinical problems. For instance, patients with SCI above T5 exhibit pressor response and bradycardia after bladder distension, and there are similar results in rats with SCI [13]. To assess AD and NDO, we developed an animal model with a complete SCT at the T4 segment. Bladder distension is the most common cause of AD [14], and usually occurs during urodynamics assessment [15]. Therefore, AD produced via bladder distension provides a consistent and clinically relevant assessment tool. In the present study, we measured the changes in blood pressure and heart rate in T4-SCT rats during urodynamics. We also correlated these measurements to a standard urodynamics variable (maximum voiding pressure). Anaesthesia markedly reduces voiding function in SCT rats; thus, we conducted CMG in conscious rats. Using non-invasive blood pressure monitoring we assessed AD in a situation that mimics clinical scenarios. To minimize bladder irritation we inserted the catheter only a few hours before urodynamics.

It is unlikely that intravesical onabotA instillation has an effect on normal animals, as previous studies have shown that onabotA is most effective in conditions of increased nerve activity [11]. Khera et al. also demonstrated no significant differences in the bladder contraction frequency or amplitude of contractions in any animal treated with instillation of saline alone, which shows that the effect of the intravesical instillation procedure is not great enough to affect bladder function. In the present study, there were also uninhibited contractions in conscious SCT rats, which agree with previous findings [16], and onabotA administration reduced the frequency of uninhibited contractions and significantly lowered the maximum voiding pressure. Thus, intravesical onabotA instillation improves bladder dysfunction. Furthermore, onabotA administered 3 weeks after SCT significantly reduced CMG-induced effects on blood pressure and heart rate, suggesting that AD improves with onabotA treatment. This was also associated with a decrease of NGF concentration in T4-DRG.

Nerve growth factor regulates sensory and sympathetic neuronal growth and is known to increase at least fourfold within a week after SCI [17]. Exogenous anti-NGF also decreased dysreflexia in rats by 30% after SCI, suggesting that NGF plays a major role in the pathogenesis of AD [18]. Recent studies have shown a close relationship between NGF and changes in the afferent arbour, which can contribute to AD [19]. Intrathecal anti-NGF administration improves AD in rats [5], suggesting that drugs with anti-NGF properties such as onabotA might have a similar effect. In the present study, we demonstrated that intravesical onabotA treatment reduces NGF content in T4-DRG after SCT in rats and also blocked AD, possibly by lowering NGF content at the injury site.

Recent reports have found that increased NGF in the spinal cord after spinal cord injury is responsible for inducing hyperexcitability of C-fibre bladder afferent pathways [20], and that intrathecal application of NGF antibodies, which neutralized NGF in the spinal cord, suppressed detrusor hyperreflexia and detrusor sphincter dyssynergia in rats with SCI [21]. In addition, intravesical onabotA injection lowered NGF content in the bladder tissue of patients with neurogenic detrusor overactivity [8].Together, these reports support the findings of the present study that onabotA suppressed neurogenic bladder overactivity, lowered voiding pressure and blocked AD response during bladder distension via urodynamics in SCI rat. After SCI, increased NGF content in the bladder, dorsal root ganglia and spinal cord has been reported [22]. During development, NGF is released by the target tissue, taken up in responsive neurones by receptor-mediated endocytosis and transported retrogradely to the cell body where it exerts its trophic/differentiative effects [23], and intrathecal administration of NGF at the L6-S1 level of the spinal cord for 1 or 2 weeks caused bladder overactivity and hyperexcitability of bladder afferent neurones.

It has been proposed that detrusor sphincter dyssynergia is the initial insult after SCI [3] and it leads to bladder outlet obstruction and, subsequently, to bladder hypertrophy. Bladder smooth muscle and urothelium were also found to produce NGF [24]. Therefore, we believe that nascent NGF in the bladder arrives at DRG and the spinal cord via retrograde transport. At the site of injury, however, NGF concentration is highest because, in addition to retrograde transport from visceral end organs, NGF is also produced by surrounding neuroglia [4]. Furthermore, the urothelium is currently perceived as a sensing structure with signalling properties due to neurotrophin receptors (TrkA and p75NTR) and a number of Transient Receptor Potential (TRP) channels, including TRPV1 [25]. NGF has been shown to activate TRPV1 on small afferent nerves, which can promote the release of excitatory neuropeptides [25]. Thus, a reduction of bladder NGF could lead to afferent pathway desensitization. Very recently, one study using transgenic mice demonstrated that overexpression of NGF leads to urinary bladder enlargement characterized by marked nerve fibre hyperplasia in the submucosa and detrusor smooth muscle. They also found a marked increase in the density of calcitonin gene-related peptide and substance P-positive C-fibre sensory afferents, neurofilament 200-positive myelinated sensory afferents, and tyrosine hydroxylase-positive sympathetic nerve fibres in the sub-urothelial nerve plexus [24].

In conclusion, in the present study we showed that intravesical onabotA treatment blocks AD in rats with T4-SCT. This reduced dysreflexia response was associated with a decrease NGF concentrations at the bladder and dorsal root ganglia T4 segment, which suggest an afferent pathway modulation by intravesical onabotA treatment. The findings of the present study highlight the potential benefits of intravesical onabotA treatment in patients with SCI, and also provide a novel mechanism for the control of AD via a minimally invasive treatment modality.


Financial assistance (postdoctoral fellowship in the area of incontinence) was provided to M.S.E. through a partnership programme of the Canadian Foundation for Research on Incontinence and the Canadian Institutes of Health Research.


None declared.