Risk of urinary tract infection after detrusor botulinum toxin A injections for refractory neurogenic detrusor overactivity in patients with no antibiotic treatment


Xavier Gamé, Service d’Urologie, Transplantation Rénale et Andrologie, CHU Rangueil, TSA 50032, 31059 Toulouse, France.
e-mail: xaviergame@hotmail.com


Study Type – Harm/Safety (case series)
Level of Evidence 4


To determine if antibiotic prophylaxis is required for intradetrusor botulinum toxin A (BoNT-A) injections.


This was a prospective non-randomized study. Inclusion criteria were adult patients with neurogenic detrusor overactivity who failed anticholinergic-based first-line treatment and had sterile urine culture 7 days before injections, a negative dipstick test on the day of the injections, and were performing clean intermittent self-catheterization. Injections were performed using a rigid cystoscope, under local anaesthetic. Patients received BoNT-A as either Botox® (Allergan Inc., Irvine, CA, USA) 300 U in 30 sites or Dysport® (Ipsen Ltd, Ipsen, Paris, France) 750 U in 20 sites. Urine culture was done 6 days and 6 weeks after the injections. If urinary tract infection (UTI) was suspected, a physician performed a complete physical examination. The primary outcome criterion was the occurrence of UTI during the first week after injections.


In all, 42 patients (22 women, 20 men) with a mean (sd) age 45.3 (16.4) years were included. A symptomatic UTI occurred during the first week after the injections in three patients (7.1%). The urinary colonization rate was 31% and 26% at 6 days and 6 weeks after injection, respectively. The most common bacterium was Escherichia coli (62.5%). There were no differences between 20 and 30 injection sites or between the sexes.


The rate of UTI after intradetrusor injections of BoNT-A was 7.1%. The most frequent bacterium involved was Escherichia coli. These results indicate that antibiotic prophylaxis for intradetrusor BoNT-A injections seems necessary.


botulinum toxin


neurogenic detrusor overactivity


Today, intradetrusor injections of botulinum toxin (BoNT-A) is offered as a second-line treatment for neurogenic detrusor overactivity (NDO) [1–3]. Furthermore, several studies have reported the efficacy of BoNT-A in the management of non-NDO [2]. It is used after failure of treatment with anticholinergic drugs or if intolerance or contraindications to these medications occur.

Although, guidelines on patients’ selection and on the follow-up of these neurological patients after injections are available, no recommendations are made about antibiotic prophylaxis with intradetrusor injections of BoNT-A [2,4,5]. Furthermore, no trial assessing intradetrusor injections of BoNT-A efficacy has made any specifications about antibiotic prophylaxis [1,2,6]. To the best of our knowledge, there has been no study assessing the incidence of symptomatic UTIs during the first week after injections, although studies report 2–32%[1,2] incidence of UTIs during the follow-up which are usually of 6 months on average.

Guidelines claim antibiotic prophylaxis should be done when intravesical lithotrity or TURP are performed [7–9] but it is not recommended for flexible or rigid cystoscopy and transurethral bladder resection or biopsy. There are no recommendations for antibiotic prophylaxis during subureteric Deflux® or Macroplastic® injections.

The aim of the present study was to evaluate the rate of symptomatic UTIs and bacterial colonization during the first week after intradetrusor injections of BoNT-A in patients with NDO and so to determine if antibiotic prophylaxis is required.


Between June 2007 and July 2008, a prospective non-randomized open-labelled cohort single-centre study was performed in neurological patients needing intradetrusor injections of BoNT-A.

The inclusion criteria for patients were: patients with NDO who failed to respond to oral antimuscarinics, aged >18 years, able and willing to do clean intermittent self-catheterization (CISC) and a sterile urine culture 7 days before the injections. Patients were excluded if they were unable or unwilling to do CISC, were aged <18 years, had a UTI or urinary bacterial colonization on urine culture performed 7 days before the injections, on antibiotic or cranberries therapy, or with any abnormality on the dipstick test carried out just before the injections, a history of myasthenia, expectant women and those with coagulation disorders.

Before inclusion, patients kept a bladder diary for 3 days, and underwent urodynamic investigation (Medtronic Duet®, Minneapolis, MN, USA) according to the ICS [10,11], with renal ultrasound and 24-h creatinine clearance as recommended [4]. A urine culture was performed 7 days before the injections.

A urine dipstick test was performed on the day of the procedure and the injections were performed using a rigid cystoscope, under local anaesthesia in the operating theatre. During the procedure 300 U of Botox® (Allergan Inc., Irvine, CA, USA) at 30 sites or 750 U of Dysport® (Dysport®, Ipsen, Paris, France) at 20 sites, were injected into the detrusor sparing the bladder trigone. Cardiac and respiratory functions were systematically monitored. The patients were discharged from the hospital on the evening of the same day.

Patients were reviewed in clinics 7 days, 6 weeks and 6 months after the procedure and a clinical examination and urine culture were performed at each follow-up visit. If there was a medical concern between the follow-up visits, patients were seen by a physical medicine and rehabilitation physician or neuro-urologist and an urine culture carried out together with a detailed physical examination.

Urine cultures were interpreted according to the Kass criteria (bacterial count in culture over 105 colony-forming units per millilitre and >10 leukocytes per high-power field (×400) [12].

Symptomatic UTI was defined by the association of bacteriological criteria and symptoms such as fever, intensification of spasticity, intensification of autonomic hyperreflexia, pain and worsening of the neurological status.

The primary outcome was the number of symptomatic UTIs during the first week after BoNT-A injections. The secondary outcomes were the urinary bacterial colonization rate at 6 days and 6 weeks after injection, type of bacteria, number of UTIs after the first week after injections, gender, the type of BoNT-A and the number of injections’ impact on UTIs.

Data are presented as the mean (sd). Quantitative values were compared with the Student’s t-test and qualitative values were compared with the chi-square test or Fisher’s exact test; P < 0.05 was considered to indicate statistical significance.


Of the 153 patients treated, 42 patients, 22 women and 20 men, with a mean (sd) age of 45.3 (16.4) years were enrolled in the present study. The patients’ characteristics are summarized Table 1. The neurological disease was spinal cord injury in 17 patients, multiple sclerosis in 14, spina bifida with myelomeningocele in six, stroke in three and myelitis in two. In the patients with multiple sclerosis, the median (range) Expanded Disability Status Scale score was 4 (2–6). The mean time between the beginning of the neurological disease and the enrolment was 15.1 (11.7) years. All patients performed CISC. It was the 2.4 (1.6)th injection course on average. Dysport was used in 26 and Botox in 16 patients.

Table 1.  Characteristics of the enrolled patients
  1. ASIA, American Spinal Injury Association classification.

 Neurological disease: 
  Spinal cord injury:17
   ASIA A6
   ASIA B3
   ASIA C5
   ASIA D3
  Multiple sclerosis14
Mean (sd): 
 Age, years45.3 (16.4)
 Duration of the disease, years15.1 (11.7)
 Number of BoNT-A injections2.4 (1.6)
 Type of BoNT-A 
 Number of sites of injections 
  20 sites26
  30 sites16

During the first week after the injections, three patients (7.1%) had symptomatic UTIs. The symptoms of the UTIs were fever in two patients and worsening of neurological status in one. Escherichia coli (E.coli) was the causative organism in all cases. At 6 days after injection, the urine culture was sterile in 26 patients (62%) but there was bacterial colonization in 13 patients (31%) and the bacteria involved are shown Table 2. The most frequent bacterium was E.coli (10/16 cases). No extended-spectrum β-lactamases or methicillin-resistant staphylococcus aureus were found.

Table 2.  Bacteria involved in UTIs and urinary colonization during the first week after the BoNT-A injections
E. coli10/16
Enterococcus faecalis1/16
Serratia marcescens1/16
Staphylococcus epidermidis1/16
Klebsiella oxytoca1/16
Proteus mirabilis1/16

There was urinary bacterial colonization in 11 patients (26%) at 6 weeks after BoNT-A injection. Urine culture remained sterile in 31 patients. Involved bacteria were E.coli in eight patients, Proteus mirabilis in one, Klebsiella pneumonia in one and Staphylococcus haemolyticus in one. No patient reported any symptomatic UTI between the first week and the sixth week after the BoNT-A injections.

Between the sixth week and the sixth month, seven patients had eight symptomatic UTIs (16.7%). During the 6 months after intradetrusor injections of BoNT-A, the mean number of symptomatic UTIs was 0.28 (0.51) per patient.

There were no statistical differences on the number of symptomatic UTIs during the first week after intradetrusor injections of BoNT-A according to gender, age, neurological disease, duration of the neurological disease, or number of injections’ sites and type of BoNT-A (Table 3).

Table 3.  Number of UTIs during the first week after intradetrusor injections of BoNT-A according to the gender, age, number of sites of injections, BoNT-A type and neurological disease
VariableUTINo UTIP
 Gender:  0.086
 Neurological disease:  0.09
  Spinal cord injury017 
  Multiple sclerosis113 
Mean (sd):   
 Age, years37.7 (12.7)45.9 (16.7)0.38
 Duration of the disease, years14.8 (12.6)15.1 (11.9)0.97
 Type of BoNT-A  0.29
 Number of sites of injections  0.29
  20 sites125 
  30 sites214 


Given the successful and sustained response to intradetrusor injections of BoNT-A, the number of patients receiving this treatment will significantly increase worldwide. It is therefore very important to have clear recommendations on the use of antibiotic prophylaxis during this procedure. To the best of our knowledge, this is the first study to show that the rate of symptomatic UTIs during the first week after intradetrusor injections of BoNT-A was 7.1% and that the urinary bacterial colonization rate was 31% 1 week after BoNT-A injection, which decreased to 26% at 6 weeks. The present study carried out in neurological patients with no history of recurrent UTIs, with sterile urine cultures 1 week before the BoNT-A injections and normal urine dipstick test on the day of the procedure argues for the use of antibiotic prophylaxis.

Except after TURP and cystoscopy, the incidence of UTIs after other endoscopic procedures has not been well studied [7]. This rate of symptomatic UTI during the first week is higher than reported after cystoscopy, which is between 0.7 and 2.7%[13–18]. Clark and Higgs [19] and Rane et al. [20] reported rates of 7.5% and 21%, respectively, but they included urinary bacterial colonization. The high level of colonization in that study is comparable with previous reports in patients using CISC and does not justify any antibiotic therapy [21].

The Pan-European study on nosocomially acquired UTIs identified the three most important factors for infectious complications after surgery: indwelling catheter, previous urogenital infection and long preoperative hospital stay [22]. These conditions occur often in the management of neurological patients emphasizing the need for antibiotic prophylaxis in this population. Furthermore, decreasing the occurrence of UTIs in neurological patients is an important issue because it is a major cause of morbidity, alteration of the neurological status and is one of the main reasons for hospitalization.

The diagnosis of UTI after detrusor injections of BoNT-A is difficult in neurological patients especially as they may experience an increase of incontinence episodes, urgency, frequency and haematuria for few days. That is why before prescribing any antibiotic treatment, a complete clinical examination and an urine culture should be performed.

No risks factors of symptomatic UTI after BoNT-A injections have been identified. However, the lack of a statistically significant difference according to the sex and the neurological disease could be related to the population size. All patients with UTI were women and were not spinal cord injured.

The aim of the present study was to determine the incidence of UTI after BoNT-A injection, despite the use of two different formulations of BoNT-A (Dysport and Botox) and number of sites injected. However, as it was on average the second injection, patients were not randomized. The dosage used was comparable with those previously suggested for each type of BoNT-A [3,23].

Although the present study argues for an antibiotic prophylaxis, its impact has to be assessed through a controlled randomized trial, as antibiotic prophylaxis may not always be associated with a decrease in symptoms. Wilson et al. [17] and Tsugawa et al. [18] did not find a decrease in either bacteriuria or in symptomatic UTIs after use of antibiotic prophylaxis after cystoscopy.

The used antibiotics should be adapted to local ecology but as shown, it has also to be active against E.coli. In endoscopic surgery, the main antibiotics assessed to prevent UTIs were fluoroquinolones, gentamicin and trimethoprim [7]. Aminoglycosides potentiate the neuromuscular activity of BoNT-A and is best avoided with BoNT-A. Bacteria involved in the present study were sensitive to both fluoroquinolones and trimethoprim. Interestingly, no specific antibiotic resistance was reported. However, because of its low cost and low selective power, trimethoprim could be the antibiotic of choice. However, of note, in cases of urinary colonization before the injections, an antibiotic therapy should be used and adapted to the patient ecology.

In conclusion, as the incidence of symptomatic UTIs during the first week after intradetrusor injections of BoNT-A was 7.1% and the urinary bacterial colonization rate was 31%, antibiotic prophylaxis is necessary in the neurological patients. The most frequent bacterium involved was E.coli and no risk factors of symptomatic UTIs were found.


None declared.