Characteristics of Symptomatic and Asymptomatic Bacteriuria in Patients with Neurogenic Bladder under Neurorehabilitation

Background: The definitions of urinary tract infections (UTI) and asymptomatic bacteriuria (ABU) are problematic to apply in patients with neurogenic bladder (NB). Here, we carried out a comparative analysis of the main clinical and laboratory data of NB patients with UTI and ABU. Methods: One hundred ninety five patients with neurogenic bladder were evaluated in the Urology Sector at a neurorehabilitation hospital. Patients were divided into either ABU or UTI group based on clinical and laboratory data. The sociodemographic data, clinical history, and laboratory test results were collected and used in the comparative analysis. Results: Of the patients evaluated, 161 (82.6%) had ABU. Patients of different age groups were affected, predominantly young adults (20-39 years). The median time of bladder involvement was 8.9 years (0-35). Neurogenic bowel was observed in 97.5% of cases and renal lithiasis in 11.3%. The main underlying pathologies leading to urologic involvement were spinal cord injury, myelomeningocele, stroke, and neoplasms. Only 16.4% of patients were not on intermittent catheterization, in which the risk of recurrent infection was higher than in patients who were under for intermittent bladder catheterization (p = 0.016, OR 2.65). Infection rates were significantly different between patients with histories of recurrent urinary tract infections (asymptomatic bacteriuria 29.8% vs 52.9% infection, p = 0.016). Leukocyturia was frequent in both groups, however, our data suggested that only values ≤ 30 cells/high power field excluded infection. Conclusions: In summary, intermittent catheterization was observed to be essential in the prevention of recurrent UTI, as well as the need to adjust the reference values for leukocyturia in the definition of the infectious condition.

Escherichia coli is the most frequent pathogen to cause symptomatic urinary tract infection (UTI), but can also lead to asymptomatic bacteriuria (ABU) [3]. Clinically, while UTI is characterized by the presence of host inflammatory responses against microorganisms triggering relevant signs and symptoms [3], asymptomatic bacteriuria (ABU) is a clinical condition in which bacteria are isolated from a properly collected urine sample and the patient has no signs or symptoms related to the urinary tract [4,5]. Such clinical definitions are problematic to apply, particularly in patients in whom ABU originates from predisposing situations, as the use of a bladder catheter. Another important aspect is that some hospitalized and/or institutionalized patients may be unable to verbalize their symptoms and, consequently, can be mistakenly diagnosed as having ABU or UTI [6].
The clinical heterogeneity among the NB patients and the scarcity of studies make it difficult to establish specific and reliable guidelines. There is a significant variation in the diagnosis, management, and treatment of UTI in these patients between urological rehabilitation centers.
Management of this population seems to be largely based on the particularities of each clinical center rather than on existing evidence and guidelines [7,8]. Thus, we compared the main clinical and laboratory data of the NB patients from a neurorehabilitation hospital with symptomatic and asymptomatic bacteriuria by E. coli.

Study, sampling and hospital setting
This is a descriptive, analytical, and cross-sectional study with patients with neurogenic bladder who underwent urine culture. The study consisted of two groups of patients clinically defined as UTI or ABU carriers. The participants were selected after receiving conventional care by the hospital team, with the purpose of admission, monitoring, and vesical education. Bacterial strains were isolated from urine cultures requested according to the context of clinical follow-up. Other laboratory tests were also requested, according to the clinical condition of each patient.
In 2013, the Sarah Network of Rehabilitation Hospitals located in the city of São Luís -Maranhão, Northeast Brazil, carried out a total of 1,730 urine culture exams. During this period, 354 (20.5%) patients with neurogenic bladder had bacteriuria with the growth of only E. coli. Based on these data, the minimum number required for to perform this study, with a confidence level of 95% and a sample error of 5%, was 169 patients. The group for analysis included 195 patients selected among those who had growth of only E. coli and with colony counts greater than or equal to 100,000 colony forming units per milliliter (≥ 10 5 CFU/mL). Clinical samples were collected from July 2014 to February 2015. Patients who did not consent to participate in the study urine and cultures with polymicrobial growth were excluded from the study.

Epidemiological and clinical data
The sociodemographic data, clinical history, and laboratory test results were collected. The following clinical and demographic characteristics were investigated: gender, age, type of lesion, time of evolution of the lesion that led to bladder involvement, type and need for catheterization, number of UTIs per year, presence of neurogenic bowel, use of diapers, and occurrence of renal lithiasis. Despite the difficulty in clinically classifying the type of urinary infection [7], we used the previously established criteria as described below [8,9]: -ABU: patients with counts greater than or equal to 10 5 CFU/mL without relevant clinical manifestations in the urinary tract.
-UTI: patients with the same characteristics as for ABU accompanied by the following manifestations: fever (temperature greater than or equal to 38 °C), dysuria, bacteremia, significant leukocyturia (≥ 10 leukocytes/high power field -HPF), tissue invasion, abdominal pain, sweating, hypotension, and dysreflexia or spasticity.
The laboratory tests used were: white blood cell (WBC) counting, C-reactive protein, procalcitonin, cystatin C, and urinalysis (automated counting using the iQ 200 Iris device and evaluation by using bright field microscopy (x 400) for confirming the presence of leukocyturia and hematuria).

Statistical analysis
Descriptive statistical analysis was performed for all variables using the NCSS 11 Statistical Software -2016 (NCSS, LLC. Kaysville, Utah, USA). The Lilliefors test was applied to evaluate the data distribution, and as the distribution was not normal (p < 0.05), the time of bladder involvement was evaluated using the Mann-Whitney test. The screening test was used to determine sensitivity, specificity, positive predictive value, negative predictive value and accuracy, and the ROC curve was used to determine a cut-off point. The Chi-square test (c 2 ) was used for independent samples and to evaluate the similarity of the groups regarding demographic variables.  Considering only the 163 patients who were on intermittent catheterization (136 with ABU and 27 with UTI), the most suggested cut-off point (ROC curve) for the evaluation of leukocyturia as a parameter to distinguish between UTI and ABU was 30 leukocytes/HPF with the highest area under the curve (AUC = 0.875 ± 0.035, p = 0.0001 CI 95% = 0.806 -0.943) (Fig. 1). This value corresponded to a sensitivity of 81.5%, specificity of 77.2%, positive predictive value of 41.5%, negative predictive value of 95.5%, and accuracy of 77.9% (Table 2). The main clinical and laboratory findings identified in the 34 patients with UTI are described in Table   3. Renal function impairment with changes in proteinuria and cystatin C tests was observed in 20.6% Abdominal and/or lumbar pain 16 (47.1) for UTI in NB patients, which was not observed among the patients of this study.

Results
Patients who were not on intermittent catheterization faced a relatively higher risk of having recurrent UTI, despite the fact that this procedure is recommended in the neurological practice of patients with neurogenic bladder and is indicated as the main factor in reducing the occurrence of UTI [9]. Patients who were not on intermittent catheterization showed resistance to the completion of the procedure or had not yet started urological rehabilitation at the time of data collection in this study.

Conclusions
Knowledge of the sociodemographic conditions of patients with neurogenic bladder increases the understanding of risk factors for the development of the infection. The findings of this study suggest that follow-up should be permanent, as low adherence to intermittent catheterization is a significant risk factor for recurrent UTI. It is also important to note that, in cases of bladder dysfunction, the assessment of intestinal function should always be considered and, if the neurogenic intestine is identified, monitoring and control measures must be implemented.
Although the confirmation of infection should not be based solely on pyuria, the presence of less than