Dataset Information

Prediction of urine culture results by automated urinalysis with digital flow morphology analysis.

ABSTRACT: To investigate the association between the results of urinalysis and those of concurrent urine cultures, and to construct a prediction model for the results of urine culture. A total of 42,713 patients were included in this study. Patients were divided into two independent groups including training and test datasets. A novel prediction algorithm, designated the UTOPIA value, was constructed with the training dataset, based on an association between the results of urinalysis and those of concurrent urine culture. The diagnostic performance of the UTOPIA value was validated with the test dataset. Six variables were selected for the equation of the UTOPIA value: age of higher UTI risk [odds ratio (OR), 2.069125], female (OR, 1.400648), nitrite (per 1 grade; OR, 3.765457), leukocyte esterase (per 1 grade; OR, 1.701586), the number of WBCs (per 1 × 10⁶/L; OR, 1.000121), and the number of bacteria (per 1 × 10⁶/L; OR, 1.004195). The UTOPIA value exhibited an area under the curve value of 0.837 when validated with the independent test dataset. The UTOPIA value displayed good diagnostic performance for predicting urine culture results, which would help to reduce unnecessary culture. Different cutoffs can be used according to the clinical indication.

SUBMITTER: Kim D

PROVIDER: S-EPMC7966378 | biostudies-literature | 2021 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of urine culture results by automated urinalysis with digital flow morphology analysis.

Kim Dokyun D Oh Seoung Chul SC Liu Changseung C Kim Yoonjung Y Park Yongjung Y Jeong Seok Hoon SH

Scientific reports 20210316 1

To investigate the association between the results of urinalysis and those of concurrent urine cultures, and to construct a prediction model for the results of urine culture. A total of 42,713 patients were included in this study. Patients were divided into two independent groups including training and test datasets. A novel prediction algorithm, designated the UTOPIA value, was constructed with the training dataset, based on an association between the results of urinalysis and those of concurre ...[more]

PMID: 33727643

Similar Datasets

Project description: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.

Project description:BackgroundCurrent evidence does not recommend screening urine culture and curing asymptomatic bacteriuria (ASB) before joint arthroplasty. The bacteriuria count on pre-operative urinalysis is a more common clinical parameter. We aimed to investigate whether the bacteriuria count on preoperative urinalysis can increase postoperative wound complications in primary total hip arthroplasty (THA).MethodsWe conducted a retrospective study that included patients who underwent primary THA in our institution from 2012 to 2018. We obtained preoperative urinalysis results before THA during the same hospitalization and identified patients with abnormal urinalysis. Receiver operating characteristic (ROC) curves were first generated to evaluate the predicted value of leukocyte esterase (LE), nitrite, bacteriuria, and pyuria in the urinalysis for superficial wound infection. Then, all included patients were divided into two groups according to the preoperative urinalysis: a bacteriuria-positive group and a bacteriuria-negative group. The primary outcome was the superficial wound infection rate within 3 months postoperatively, and the secondary outcomes included wound leakage, prosthetic joint infection (PJI), pulmonary infection, urinary tract infection (UTI), readmission rate within 3 months postoperatively, and length of stay (LOS) during hospitalization. We utilized univariable analyses to compare the outcomes between the two groups. A multivariable logistic regression model was generated to explore the potential association between bacteriuria and the risk of superficial wound infection, wound leakage, and readmission rate controlling for baseline values.ResultsA total of 963 patients were included in the study. One hundred sixty patients had abnormal urinalysis. The AUCs for LE, nitrite, bacteriuria, and pyuria were 0.507 (95% confidence interval (CI), 0.315 to 0.698), 0.551 (0.347 to 0.756), 0.675 (0.467 to 0.882), and 0.529 (0.331 to 0.728), respectively. Bacteriuria was diagnostically superior to LE, nitrite, and pyuria. Among the 963 patients, 95 had a positive bacteriuria on preoperative urinalysis, and only 9 (9.5%) had a positive urine culture. Compared with the bacteriuria-negative group, the bacteriuria-positive group had a higher superficial wound infection rate (4.2% vs. 0.6%, P = 0.008), higher wound leakage rate (11.6% vs. 4.5%, P = 0.007), higher readmission rate (5.3% vs. 1.3%, P = 0.015) within 3 months postoperatively and longer LOS (6.19 ± 2.89 days vs. 5.58 ± 2.14 days, P = 0.011). After adjustment, the bacteriuria-positive group had a significantly increased risk of superficial wound infection (OR = 7.587, 95%CI: 2.002 to 28.755, P = 0.003), wound leakage (OR = 3.044, 95%CI: 1.461 to 6.342, P = 0.003), and readmission (OR = 4.410, 95%CI: 1.485 to 13.097, P = 0.008).ConclusionPreoperative bacteriuria positivity on urinalysis significantly increased the risk of postoperative wound complications, readmission, and LOS in primary THA regardless of the result of the urine culture. Urinalysis is a fast and cost-acceptable test whose advantages have been underestimated.Level of evidenceLevel III, observational study.

Dataset Information

Prediction of urine culture results by automated urinalysis with digital flow morphology analysis.

Publications

Prediction of urine culture results by automated urinalysis with digital flow morphology analysis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets