Introduction

Cystometry is widely used for the evaluation of lower urinary tract (LUT) function in many indications. Video urodynamics (VUD), which is the combination of cystometry with contrast fluid filling and X-ray imaging, is considered the optimum procedure for urodynamic investigation in neuro-urological disorders [1] including spinal cord injury (SCI) [2]. Radiographical observations during cystometry may affect the interpretation of pressure measurements. For example, the pop-off of bladder pressure by vesicoureteral reflux may lead to underestimation of bladder compliance. Concomitant radiography also allows diagnosis and follow-up over time of aberrant LUT morphology or function, such as bladder diverticula or bladder neck insufficiency. The variables and instructions for data collection of imaging in patients with SCI have been published in the International Urinary Tract Imaging Basic SCI Data set [3]. Despite the strong support in expert opinion and guidelines, and the accepted benefit it provides to the interpretation of cystometry, determining the real-life clinical value of VUD today could benefit from further research.

In this study, we retrospectively evaluated the radiographical observations during VUD in a cohort of SCI patients with neurogenic bladder. We categorized the images and investigated their presence in different types of SCI and in different clinical and urodynamic diagnoses. Finally, we evaluated the change in radiographical findings over time. Our aim was to gather real-life data and to describe the yield of radiography during urodynamics in a cohort of SCI patients as indicated by current guidelines.

Methods

VUD data from consecutive patients with neurogenic bladder due to SCI were evaluated retrospectively. Patient age and sex, time since SCI, etiology of SCI, level of SCI, the neurological status determined according to the ASIA/ISCoS International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) and American Spinal Injury Association Impairment Scale (AIS) score, method of bladder emptying, and treatment with bladder relaxing drugs were collected from the records.

VUD were performed according to a standardized protocol following the ICS guidelines on urodynamic equipment performance and ICS guidelines on good clinical practice [4, 5]. The standard evaluation for neuro-urological patients in our center also includes strict decubitus prevention, clinical neurological examination, renal ultrasound, and flexible cystoscopy. The patients were in the supine or half-sitting position between a C-arm X-ray unit (Ziehm 8000 Germany). Five-channel urodynamic equipment was used (Medical Measurement Systems, Gladbeck, Germany). The bladder was filled with a radiocontrast solution (Urografin 30%-Schering AG, Berlin, Germany) at room temperature through a transurethral 8Fr water-filled double lumen catheter connected to an external transducer, and at a continuous inflow rate of 30 ml/min (infusion pump), with the filling bag hanging at a load cell. The filling was stopped at 500 ml or earlier if neurogenic detrusor overactivity (NDO) occurred with leakage, if persistent leakage was seen without detrusor contraction, if detrusor pressure consistently rose above 40 cm H2O, if the patient reported a strong desire to void, or if pain or autonomic dysreflexia occurred. Video images were taken before starting with an empty bladder, when sensations of bladder filling occurred, at each consecutive 200 ml of filling if no sensation was reported, at the end of the test, when sustained leakage and/ or detrusor overactivity occurred, and during voiding (if applicable). After VUD, all patients received one-time antibiotic prophylaxis with Fosfomycin 3 g.

The radiation exposure was recorded in every patient and every clinician, and data were reviewed by the radiation committee of the hospital following the National Law of 30 August 2001 on protection against ionizing radiation for education, information, equipment, and control. The aim was to keep the total radiation exposure below 0.5 mSv and the fluoroscopic exposure time under 100 s. In patients with known radiocontrast allergy, VUD was not performed and the patient was not included in this study. Further exclusion criteria were symptomatic urinary tract infection, difficulties to control autonomic dysreflexia, insufficient patient cooperation and or cognitive function, hematuria, pain, and inflammation in the pelvic region (n = 26) [6].

Radiographical observations of the bladder were categorized based on expert opinion as normal/standard, tonic (pinecone/Christmas tree), or flaccid. Cartoon drawings for clarification are presented in Fig. 1. The nomenclature and categories were based on (our) expert opinion as there is currently no ICS terminology describing cystographic observations during urodynamics. Other findings such as vesicoureteral reflux, bladder diverticula, cystocele, the image of enterocystoplasty, an open bladder neck during filling, and the presence of foreign material were recorded. If voiding was possible or if sustained leakage filled the urethra, urethrography images were also included in the evaluation.

Fig. 1
figure 1

Cartoon drawings of the types of bladder morphology during video urodynamics in patients with Spinal Cord Injury.

Full size image

The time since SCI was categorized into: the period of spinal shock and of first rehabilitation (till 26 weeks in our management schedule), the period of continued rehabilitation until discharge (between 26 and mean 78 weeks), and the period of chronic follow-up (after discharge). The etiology of SCI was classified as traumatic or non-traumatic. The findings on renal ultrasound and during flexible cystoscopy were also described.

Ethical approval was obtained from the local review board (Edge number 001176).

Statistical analysis was done with SPSS26. Continuous variables are expressed as mean ± standard deviation. Chi-Square was used for categorical data and nonparametric Kruskal–Wallis for continuous data. Statistical significance was set at p < 0.05.

Results

Two hundred thirty-one patients with SCI (159 men and 72 women), with a mean age of 49 years (±17 years, Range 11-88 years) were included in the study. Mean time since injury was 437 weeks (±599 weeks). The etiology of SCI was traumatic in 191 (83%) patients and non-traumatic in 40 (17%). AIS scores and SCI level are presented in Table 1: 72% had cervical or thoracic lesion above T10, and 61% of patients had a complete injury (AIS A).

Table 1 AIS score of the cohort of SCI patients.
Full size table

The radiographical observations during VUD are presented in Table 2: in 34% of patients, the radiography was not categorized as normal/standard. The bladder neck was clearly open in only one patient, a 21-year-old man with a complete C4 lesion after a gunshot wound. His bladder image was categorized as normal/standard and he had no NDO on VUD.

Table 2 Radiographical observations during VUD in patients with SCI.
Full size table

The categories of radiographical observations did not differ by AIS score (p = 0.759). Vesicoureteral reflux was found in two patients at 6 and 9 weeks after SCI, but of low grade and disappearing on a subsequent test after bladder management was changed from an indwelling catheter to CIC. A tonic bladder image was seen as early as eight weeks after SCI. Diverticula on the other hand were only seen in patients on long-term follow-up.

There was no significant association between bladder compliance and radiographical category (p = 0.259). Maximum cystometric capacity was significantly lower in bladder images categorized as tonic and with diverticula compared to normal/standard and flaccid classified bladders (tonic 284 ml ± 126, diverticula 229 ml ± 148, normal/standard mean 367 ml ± 132, flaccid 495 ml ± 115). P < 0.05. The presence of NDO differed significantly between patients with different radiographical bladder categories (p < 0.001): it was present in 80% (40/50) of patients with a tonic categorized bladder, but also in 30% (3/10) with a flaccid categorized bladder and in 48% (73/152) with a normal/standard categorized bladder. NDO was also found in 5/5 (100%) patients with diverticula, while in 1/4 (25%) patients with enterocystoplasty undulations with high pressure were seen. NDO was not seen in patients with cystocele, reflux, or the patient with a ureteric catheter. Maximum detrusor pressure during filling above 40 cm H2O was also associated with radiographical bladder category (p = 0.016): it was present in 24% (12/50) of patients categorized as having a tonic bladder image, 10% (1/10) with flaccid, 11% (16/152) with normal/standard categorized bladder, and in 80% (4/5) of patients with diverticula.

Table 3 shows that 78% of patients had catheter-assisted bladder emptying, either by performing clean intermittent catheterization (CIC–43%) or through an indwelling transurethral (27%) or suprapubic catheter (9%). These patients were often categorized as having a normal/standardized radiographical bladder image (69%, 71%, 68%, and 65%, respectively), and as having a tonic bladder in 19%, 17%, 26%, and 26%, respectively. Three out of four patients with diverticula used tapping as the means to empty the bladder, whereas only one out of five patients with a cystocele emptied the bladder using Crede/Valsalva. Eleven percent of patients used standard voiding. Vesicoureteral reflux was found in four patients, of which three had a transurethral catheter. As mentioned above, the finding disappeared in two patients after switching to CIC.

Table 3 Techniques used for bladder emptying in SCI patients with different radiographical bladder categories during VUD.
Full size table

Renal ultrasound revealed a pathological finding in only two patients: one had limited hydronephrosis and another a shrunken kidney on the right side. Cystoscopy findings are presented in Table 4 (161 patients). Forty-six patients with radiographically classified normal/standard bladders had aberrant findings on cystoscopy. “Eggshell” bladder stones were found on cystoscopy in 20, whereas none had been seen on plain X-ray. Diverticula were found in twice as many patients on cystoscopy than seen during VUD, and trabeculation is also underestimated when evaluated by cystography alone.

Table 4 Findings on X-ray and on cystoscopy.
Full size table

Antimuscarinics were taken at the time of examination by 46% with tonic, 40% with flaccid, and 40% with normal/standard categorized bladders. Injections of botulinum toxin A (300 IE) had been done once or several times in 17% of patients categorized as tonic, in 20% with flaccid and 12% with normal/standard bladder image, but not during the ten months preceding VUD.

Eighty-six patients had repeated tests with 72 ± 143 weeks (range 3–1007 weeks) intervals. In 59 (70%), the radiographic bladder category was unchanged. The changes in the other 27 patients were from normal/standard to tonic (N = 9), tonic to normal/standard (N = 11), flaccid to normal/standard (N = 3) and flaccid to tonic (N = 2). Vesicoureteral reflux appeared in 1 and disappeared in 2. The double J catheter had been removed in one patient.

In this cohort, the influx into the prostate, urethral signs of detrusor sphincter dyssynergia, or other urethral pathology were not seen during sustained leakage or during voiding in those who could void (14 men and 9 women).

Discussion

Deterioration of the urinary tract is no longer the leading cause of death after SCI [7]. This great improvement is partly due to significant progress in urological diagnosis and management in the last decades. Cystography can help to document anatomical pathology in the LUT. VUD presents a combination of standard urodynamic techniques (cystometry, pressure flow studies) and cystography [2]. Examples of the radiographic features which can be studied are bladder shape, the configuration of the bladder wall, trabeculation, sacculation, diverticula, and vesicoureteral reflux [8]. There are no absolute indications for VUD, but patients with neurogenic bladder are considered the most important target group [9, 10]. Adding video to the urodynamic investigation is advocated by several authors and guidelines. Wöllner and Pannek made a strong statement in favor of adding imaging in patients with SCI [11,12,13]. Suskind et al. [14] found VUD to be a clinically useful tool that altered the clinical impression and treatment plan in 29.5% of carefully selected patients including some with neurogenic bladder. In our setting, VUD was done in all SCI patients.

However, critical notes have recently been expressed by world experts taking part in two ICI-RS (International Consultation on Incontinence Research Society) Think Tanks (working groups) on the added value of video to urodynamics in children and adults with LUT dysfunction. The body of evidence for the addition of cystography to filling and voiding cystometry was considered scant and insufficient. Standards were judged to be poor and variable, and practice to be dominated by uncontrolled expert opinion. The evidence was considered low level and improved standardization of the tests and better evaluation of the added value were strongly advocated [15, 16]. This was the main reason we decided to make an evaluation of the added value of X-ray during urodynamics in a cohort of our SCI patients. In this study, we described and categorized the radiographical bladder images found during VUD. We evaluated the presence of these radiographical observations in patients with different types of SCI (level, AIS score, and time since SCI) and we examined whether the findings were different in SCI patients with different urodynamic findings and different methods of bladder emptying. Finally, an assessment of changes in the findings over time was made.

VUD were performed according to the ICS good urodynamic practice guidelines [5]: the contrast medium was specified, X-ray images were made at relevant moments during filling, and the protection against and the exposure to radiation are reported, showing strict adherence to the ALARA (As Low As Reasonably Achievable) principle [17].

One-third of the SCI patients in our cohort were found to have an aberrant (not normal/standard) bladder appearance during VUD (Table 2). Other than we expected, the categories of radiographical observations did not differ by AIS score. A tonic bladder appearance was the most frequent finding in 63% (Table 2—N = 50/72). These latter patients had a significantly lower maximum cystometric capacity, showed more NDO (80%), and ¼ was found to have pressures during NDO above 40 cm H2O. The category of the tonic bladder was found as early as eight weeks after SCI. Continuous bladder drainage can have played a role: one-third of patients with a tonic bladder image during VUD had an indwelling transurethral or suprapubic catheter.

Vesicoureteral reflux is a potential kidney- and life-threatening complication in SCI, and missing its presence during urodynamics may lead to an underestimation of bladder pressures during filling (compliance and/or presence and amplitude of NDO). The incidence of reflux in a previously published large cohort during long-term follow-up was 5% and mostly low grade [18, 19]. The prevalence in our cohort was 2%, the grade was low, and the finding disappeared in two of four patients after changing from an indwelling catheter to CIC. In these two patients, vesicoureteral reflux was found early after SCI. This is in line with previous reports [20]. Furthermore, only two patients in our cohort had aberrant findings on renal ultrasound. Bladder stones are the second most common urological complication in SCI patients [21]. Approximately 36% of those with an indwelling catheter develop bladder stones within eight years [22], but the 5-year cumulative incidence rate has declined during the past decades [23]. As previously reported, the detection by X-ray is 28.6% for struvite stones and 41.9% for calcium phosphate stones, and 78% of such bladder stones are not visible on plain X-ray. This may be explained by a combination of low stone contrast, a shell-like form, and by gas and feces in the bowel obstructing the view [24, 25]. In our cohort, no bladder stones were diagnosed through VUD. Nevertheless, 20 patients out of 161 who underwent cystoscopy were found to have bladder stones, which indeed had a shell-like form.

Nordling et al. [26] previously reported that trabeculation, diverticula, and a Christmas tree appearance were not related to the level of the lesion or the reflex pattern of the detrusor in a cohort of 57 patients with localized neurological lesions. In our cohort, the presence of NDO and increased amplitude of NDO were found more often in patients with a tonic bladder appearance and with bladder diverticula. Bladder compliance, on the other hand, did not differ between bladder appearance categories. Furthermore, diverticula were more often seen on cystoscopy than during cystography. This could be explained in several ways: superposition in a 2-D image, a narrow entrance of the diverticula, or insufficient contrast filling in large diverticula. Our study also illustrated that diverticula take long to develop as they were only seen in patients during long-term follow-up.

Finally, VUD could also be used to assess the urethra. However, voluntary voiding was rare and no major urethral pathology was seen in our SCI cohort.

Repeating VUD after ~1.5 years showed a change in bladder appearance category in 30% of patients. A normalization, from tonic or flaccid to normal/standard, was seen in 50% of these. In the other half, a new aberrant image was seen. These changes could have been caused by neurological changes, long-term bladder dysfunction, or a change in drug intake and/or bladder drainage method. However, we accept that this is pure speculation as pre- and post-event controls of treatment were not available.

The findings from our study have to be put into perspective. Our study was retrospective. Nevertheless, the data available in the files were standardized, reliable, and nearly complete. Our sample size was not small but from one single center. Evaluation in a much larger multicenter sample could provide further insights. Our findings reflect our way of management from the day of admission and may be different when other management rules are used. The data are from a Western country and may not represent the situation in other parts of the world. Our data are from a cohort of SCI patients, which is an acquired neurological disorder, and can thus not be extrapolated to congenital spinal cord deficiencies and progressive neurologic diseases. Finally, we cannot compare with alternative X-ray techniques or ultrasound as these data are not available.

This retrospective study presents a real-life example of the yield of VUD in a cohort of our patients with SCI. It is an illustration of diagnostic management according to current standards and therefore has not the aim nor the possibility to influence future guidelines. Nevertheless, it may serve as a background to stimulate large high-quality studies that evaluate the role of cystography added to urodynamics in patients with neurogenic bladder.

Taking all data into consideration, it is clear that concomitant X-ray during filling cystometry in SCI patients can reveal pathological changes in bladder appearance and complications. However, in our cohort, as in other series in the literature, such findings were rare. A high-quality evaluation of the added benefit of cystography during VUD in patients with SCI is needed to help define the role of X-ray in VUD in SCI. The potential danger of repeated radiographical testing, and the cost-benefit should be part of this investigation, as should potential phenotyping of which patients need X-rays to be done during urodynamics.