The Importance of Urine Concentration on the Diagnostic Performance of the Urinalysis for Pediatric Urinary Tract Infection

Predictive performance of urinalysis for urine culture results according to causative microorganisms: an integrated analysis with artificial intelligence

Urinary tract infections (UTIs) are pervasive and prevalent in both community and hospital settings. Recent trends in the changes of the causative microorganisms in these infections could affect the effectiveness of urinalysis (UA). We aimed to evaluate the predictive performance of UA for urinary culture test results according to the causative microorganisms. In addition, UA results were integrated with artificial intelligence (AI) methods to improve the predictive power. A total of 360,376 suspected UTI patients were enrolled from two university hospitals and one commercial laboratory. To ensure broad model applicability, only a limited range of clinical data available from commercial laboratories was used in the analyses. Overall, 53,408 (14.8%) patients were identified as having a positive urine culture. Among the UA tests, the combination of leukocyte esterase and nitrite tests showed the highest area under the curve (AUROC, 0.766; 95% CI, 0.764-0.768) for predicting urine culture positivity but performed poorly for Gram-positive bacteriuria (0.642; 0.637-0.647). The application of an AI model improved the predictive power of the model for urine culture results to an AUROC of 0.872 (0.870-0.875), and the model showed superior performance metrics not only for Gram-negative bacteriuria (0.901; 0.899-0.902) but also for Gram-positive bacteriuria (0.745; 0.740-0.749) and funguria (0.872; 0.865-0.879). As the prevalence of non-Escherichia coli-caused UTIs increases, the performance of UA in predicting UTIs could be compromised. The addition of AI technologies has shown potential for improving the predictive performance of UA for urine culture results.IMPORTANCEUA had good performance in predicting urine culture results caused by Gram-negative bacteria, especially for Escherichia coli and Pseudomonas aeruginosa bacteriuria, but had limitations in predicting urine culture results caused by Gram-positive bacteria, including Streptococcus agalactiae and Enterococcus faecalis. We developed and externally validated an AI model incorporating minimal demographic information of patients (age and sex) and laboratory data for UA, complete blood count, and serum creatinine concentrations. The AI model exhibited improved performance in predicting urine culture results across all the causative microorganisms, including Gram-positive bacteria, Gram-negative bacteria, and fungi.

Prevalence of Urinary Tract Infections and Antibiogram of Bacteria Isolated From Children With Sickle Cell Disease in Tanzania

Introduction Individuals with sickle cell disease (SCD) are particularly vulnerable to urinary tract infections (UTIs) due to immunological deficits and renal abnormalities associated with the disorder. These infections can exacerbate underlying health issues and lead to severe complications if not managed promptly and effectively. Due to the heightened risk and potential consequences of UTIs in this population, this study aimed to determine their prevalence and explore the resistance patterns of causative pathogens among children attending the SCD Clinic at Muhimbili National Hospital (MNH), Dar es Salaam, Tanzania. Focusing on this demographic group, we sought to provide targeted insights to inform better clinical protocols and intervention strategies in regions heavily affected by SCD. Materials and methods This prospective cross-sectional study was conducted at the MNH, Dar es Salaam, Tanzania, with an enrollment over two months from 19th March to 21st May 2015. We diagnosed UTIs in children with SCD using dipstick and culture methods. Antibiotic susceptibility was assessed using the Kirby-Bauer disc diffusion method, evaluating resistance patterns to antibiotics such as ampicillin, cloxacillin, erythromycin, chloramphenicol, ceftriaxone, and trimethoprim-sulfamethoxazole. The diagnostic accuracy of the dipstick and culture methods was validated to ensure reliability in detecting UTIs. Statistical analysis was conducted using Statistical Product and Service Solutions (SPSS) software (Released 2019; IBM Corp., Armonk, New York, United States). Results Among the 250 children, 56 (22.4%) were UTI-positive according to the culture method and 62 (24.8%) were UTI-positive according to the dipstick test. Girls were more likely to be UTI-positive than boys (29.1% and 13.6%, respectively; p-value = 0.011). Escherichia coli was the most common uropathogen, followed by Klebsiella, Staphylococcus, Proteus, and Pseudomonas (44.2%, 26.9%, 21.2%, 3.8%, and 1.9%, respectively). All isolates were resistant to ampiclox. Resistance rates to ampicillin, erythromycin, cotrimoxazole, chloramphenicol, and ceftriaxone were 94.2%, 76.9%, 59.6%, 46.2%, and 21.2%, respectively. Conclusion This study indicated that dipsticks diagnosed more UTIs. The prevalence was higher in girls than in boys. Escherichia coli was the most commonly isolated antibiotic-resistant organism. High resistance levels were observed against the combination of ampicillin and cloxacillin. However, the isolates were less resistant to ceftriaxone. These results call for increased surveillance of resistant uropathogens in the pediatric population with SCD.

Development and validation of artificial intelligence models to predict urinary tract infections and secondary bloodstream infections in adult patients

BACKGROUND

Traditional culture methods are time-consuming, making it difficult to utilize the results in the early stage of urinary tract infection (UTI) management, and automated urinalyses alone show insufficient performance for diagnosing UTIs. Several models have been proposed to predict urine culture positivity based on urinalysis. However, most of them have not been externally validated or consisted solely of urinalysis data obtained using one specific commercial analyzer.

METHODS

A total of 259,187 patients were enrolled to develop artificial intelligence (AI) models. AI models were developed and validated for the diagnosis of UTI and urinary tract related-bloodstream infection (UT-BSI). The predictive performance of conventional urinalysis and AI algorithms were assessed by the areas under the receiver operating characteristic curve (AUROC). We also visualized feature importance rankings as Shapley additive explanation bar plots.

RESULTS

In the two cohorts, the positive rates of urine culture tests were 25.2% and 30.4%, and the proportions of cases classified as UT-BSI were 1.8% and 1.6%. As a result of predicting UTI from the automated urinalysis, the AUROC were 0.745 (0.743-0.746) and 0.740 (0.737-0.743), and most AI algorithms presented excellent discriminant performance (AUROC > 0.9). In the external validation dataset, the XGBoost model achieved the best values in predicting both UTI (AUROC 0.967 [0.966-0.968]) and UT-BSI (AUROC 0.955 [0.951-0.959]). A reduced model using ten parameters was also derived.

CONCLUSIONS

We found that AI models can improve the early prediction of urine culture positivity and UT-BSI by combining automated urinalysis with other clinical information. Clinical utilization of the model can reduce the risk of delayed antimicrobial therapy in patients with nonspecific symptoms of UTI and classify patients with UT-BSI who require further treatment and close monitoring.

Urine specific gravity, pyuria, and neutrophil gelatinase-associated lipocalin for identifying urinary tract infection in young children

BACKGROUND

To determine whether urine neutrophil gelatinase-associated lipocalin (uNGAL) might be superior to pyuria for detecting urinary tract infection (UTI) regardless of urine specific gravity (SG) in young children.

METHODS

We conducted a retrospective analysis of children aged < 3 years who were evaluated for UTI with urinalysis, urine culture, and uNGAL measurements during a 5-year period. Sensitivity, specificity, likelihood ratios (LRs), predictive values (PVs), area under the curves (AUCs) of uNGAL cut-off levels, and various microscopic pyuria thresholds for detecting UTI were calculated for dilute (SG < 1.015) and concentrated urine (SG ≥ 1.015).

RESULTS

Of 456 children included, 218 had UTI. The diagnostic value of urine white blood cell (WBC) concentration to define UTI changed with urine SG. For detecting UTI, uNGAL cut-off of 68.4 ng/mL had higher AUC values than pyuria ≥ 5 WBCs/high power field (HPF) for dilute and concentrated urine samples (both P < 0.05). Positive LR and PV and specificity of uNGAL were all greater than those of pyuria ≥ 5 WBCs/HPF regardless of urine SG, although the sensitivity of pyuria ≥ 5 WBCs/HPF was higher than that of uNGAL cut-off for dilute urine (93.8% vs. 83.5%) (P < 0.05). At uNGAL ≥ 68.4 ng/mL and ≥ 5 WBCs/HPF, posttest probabilities of UTI were 68.8% and 57.5% for dilute urine and 73.4% and 57.3% for concentrated urine, respectively.

CONCLUSIONS

Urine SG can affect the diagnostic performance of pyuria for detecting UTI and uNGAL might be helpful for identifying UTI regardless of urine SG in young children. A higher resolution version of the Graphical abstract is available as Supplementary information.

Decreasing Misdiagnoses of Urinary Tract Infections in a Pediatric Emergency Department

BACKGROUND AND OBJECTIVES

Urinary tract infection (UTI) is a common diagnosis in the emergency department (ED), often resulting in empirical antibiotic treatment before culture results. Diagnosis of a UTI, particularly in children, can be challenging and misdiagnosis is common. The aim of this initiative was to decrease the misdiagnosis of uncomplicated pediatric UTIs by 50% while improving antimicrobial stewardship in the ED over 4 years.

METHODS

By using the Model for Improvement, 3 interventions were developed: (1) an electronic UTI diagnostic algorithm, (2) a callback system, and (3) a standardized discharge antibiotic prescription. Outcome measures included the percentage of patients with UTI misdiagnosis (prescribed antibiotics, but urine culture results negative) and antibiotic days saved. As a balancing measure, positive urine culture results without a UTI diagnosis were reviewed for ED return visits or hospitalization. Statistical process control and run charts were used for analysis.

RESULTS

From 2017 to 2021, the mean UTI misdiagnosis decreased from 54.6% to 26.4%. The adherence to the standardized antibiotic duration improved from 45.1% to 84.6%. With the callback system, 2128 antibiotic days were saved with a median of 89% of patients with negative culture results contacted to discontinue antibiotics. Of 186 patients with positive urine culture results with an unremarkable urinalysis, 14 returned to the ED, and 2 were hospitalized for multiresistant organism UTI treatment.

CONCLUSIONS

A UTI diagnostic algorithm coupled with a callback system safely reduced UTI misdiagnoses and antibiotic usage. Embedding these interventions electronically as a decision support tool, targeted audit and feedback, reminders, and education all supported long-term sustainability.

Association of Pyuria with Uropathogens in Young Children

OBJECTIVE

To examine the association between uropathogens and pyuria in children <24 months of age.

STUDY DESIGN

A retrospective study of children <24 months of age evaluated in the emergency department for suspected urinary tract infection (UTI) with paired urinalysis and urine culture during a 6-year period. Bagged urine specimens or urine culture growing mixed/multiple urogenital organisms were excluded. Analysis was limited to children with positive urine culture as defined by the American Academy of Pediatrics clinical practice guideline culture thresholds.

RESULTS

Of 30 462 children, 1916 had microscopic urinalysis and positive urine culture. Urine was obtained by transurethral in-and-out catheterization in 98.3% of cases. Pyuria (≥5 white blood cells per high-powered field) and positive leukocyte esterase (small or more) on the urine dipstick were present in 1690 (88.2%) and 1692 (88.3%) of the children respectively. Children with non-Escherichia coli species were less likely to exhibit microscopic pyuria than children with E coli (OR 0.24, 95% CI 0.17-0.34) with more pronounced effect on Enterococcus and Klebsiella (OR 0.08, 95% CI 0.03-0.18 and OR 0.18, 95% CI 0.11-0.27 respectively). Similarly, positive leukocyte esterase was less frequently seen in non-E coli uropathogens compared with E coli.

CONCLUSIONS

Pyuria and leukocyte esterase are not sensitive markers to identify non-E coli UTI in young children. More sensitive screening biomarkers are needed to identify UTI with these uropathogens.

Clinical Features for the Diagnosis of Pediatric Urinary Tract Infections: Systematic Review and Meta-Analysis

PURPOSE

Accurate diagnosis of urinary tract infection in children is essential because children left untreated can experience permanent renal injury. We aimed to assess the diagnostic value of clinical features of pediatric urinary tract infection.

METHODS

We performed a systematic review and meta-analysis of diagnostic test accuracy studies in ambulatory care. We searched the PubMed, Embase, Web of Science, Cumulative Index to Nursing and Allied Health Literature, Cochrane Central Register of Controlled Trials, Health Technology Assessment, and Database of Abstracts of Reviews of Effects databases from inception to January 27, 2020 for studies reporting 2 × 2 diagnostic accuracy data for clinical features compared with urine culture in children aged <18 years. For each clinical feature, we calculated likelihood ratios and posttest probabilities of urinary tract infection. To estimate summary parameters, we conducted a bivariate random effects meta-analysis and hierarchical summary receiver operating characteristic analysis.

RESULTS

A total of 35 studies (N = 78,427 patients) of moderate to high quality were included, providing information on 58 clinical features and 6 prediction rules. Only circumcision (negative likelihood ratio [LR-] 0.24; 95% CI, 0.08-0.72; n = 8), stridor (LR- 0.20; 95% CI, 0.05-0.81; n = 1), and diaper rash (LR- 0.13; 95% CI, 0.02-0.92; n = 1) were useful for ruling out urinary tract infection. Body temperature or fever duration showed limited diagnostic value (area under the receiver operating characteristic curve 0.61; 95% CI, 0.47-0.73; n = 16). The Diagnosis of Urinary Tract Infection in Young Children score, Gorelick Scale score, and UTIcalc (https://uticalc.pitt.edu) might be useful to identify children eligible for urine sampling.

CONCLUSIONS

Few clinical signs and symptoms are useful for diagnosing or ruling out urinary tract infection in children. Clinical prediction rules might be more accurate; however, they should be validated externally. Physicians should not restrict urine sampling to children with unexplained fever or other features suggestive of urinary tract infection.

Microscopic Bacteriuria Detected by Automated Urinalysis for the Diagnosis of Urinary Tract Infection

OBJECTIVE

To evaluate the test performance of microscopic bacteriuria by automated urinalysis for presumptive urinary tract infection (UTI) in young children.

STUDY DESIGN

This is a retrospective cross-sectional study of children aged <2 years evaluated for UTI in a single large emergency department with paired automated microscopic urinalysis and culture. Test characteristics were calculated for automated microscopic bacteriuria and pyuria, and a practical diagnostic threshold of bacteriuria was determined. Standard test performance measures and receiver operator characteristic curves were generated. The diagnostic performance of bacteriuria was compared with microscopic pyuria.

RESULTS

Two thousand five hundred fifty-four children with a median age of 6.1 months were studied, 19% of whom had a positive urine culture. Automated microscopic bacteriuria ≥1+ resulted in a positive likelihood ratio (LR+) of 4.5 (95% CI, 3.9-5.2) and negative LR (LR-) of 0.52 (95% CI, 0.47-0.57). Pyuria alone (≥5 WBC/high-power field) had a LR+ of 4.5 (95% CI, 4.1-5.0) and a LR- of 0.14 (95% CI, 0.11-0.18), whereas the addition of automated microscopic bacteriuria ≥1+ improved the LR+ to 16.3 (95% CI, 12.6-21.1) but raised the LR- to 0.51 (95% CI, 0.47-0.56). Test performance of automated microscopic bacteriuria measured by area under the curve analysis was lower (0.73; 95% CI, 0.70-0.76) than for pyuria (0.92; 95% CI, 0.90-0.93). Isolated automated microscopic bacteriuria without pyuria occurred in only 204 patients (8.0%), among whom only 20 (9.8%) had a positive urine culture.

CONCLUSIONS

Microscopic bacteriuria measured by automated urinalysis augments the diagnostic value of pyuria for identifying presumptive UTI in young children aged <2 years. Bacteriuria is diagnostically inferior to microscopic pyuria, and in children with bacteriuria without pyuria, presumptive UTI is unlikely.