Loading...
Publication
Infeções do trato urinário: uma análise estatística acerca do seu diagnóstico laboratorial
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 1.39 MB | Adobe PDF |
Abstract(s)
A infeção do trato urinário (ITU) é uma das infeções mais comuns no ambiente comunitário e hospitalar, representando desafios clínicos e económicos. O exame laboratorial desempenha um papel essencial no diagnóstico desta condição, mas enfrenta limitações como o tempo necessário para obter resultados e o impacto da crescente resistência antimicrobiana.
Este trabalho procura contribuir para a eficiência do diagnóstico laboratorial de ITU, avaliando os fatores associados a resultados positivos na cultura microbiológica de urina e propondo critérios que permitam reduzir a realização de culturas desnecessárias.
Foi realizado um estudo observacional retrospetivo a partir de uma amostra de 2528 pedidos de cultura de urina processados numa unidade hospitalar. A análise incluiu os resultados do sedimento urinário, como as contagens de bactérias e leucócitos e variáveis como a idade e sexo dos pacientes. Foram aplicados modelos de regressão logística multivariada e realizada uma análise ROC para identificar valores de cutoff com melhor capacidade preditiva.
Os resultados revelaram diferenças significativas associadas ao sexo e à idade dos pacientes e identificaram valores de cutoff para bactérias e leucócitos (40 bactérias/μL e 30 leucócitos/μL) com uma capacidade discriminativa moderada (AUC de 0,76). Estes valores de cutoff, quando ajustados à população de pacientes e aos objetivos específicos de cada laboratório, podem contribuir para a redução do número de culturas microbiológicas de urina desnecessárias, diminuir custos e contribuir para um uso mais racional de antibióticos, essencial no combate à resistência antimicrobiana.
Urinary tract infection (UTI) is one of the most common infections in both community and hospital settings, with significant clinical and economic challenges. Laboratory testing plays a crucial role in diagnosing this condition but faces limitations such as the time required to obtain results and the growing impact of antimicrobial resistance. This study aims to enhance the efficiency of laboratory diagnosis of UTI by evaluating factors associated with positive urine culture results and proposing criteria to reduce unnecessary urine cultures. A retrospective observational study was conducted using a sample of 2,528 urine culture requests processed in a hospital unit. The analysis included sediment urine test results, such as bacteria and leukocyte counts, as well as variables like patients’ age and sex. Multivariate logistic regression models were applied, and ROC analysis was performed to identify cutoff values with the best predictive capacity. The results revealed significant differences related to patients’ sex and age, identifying cutoff values for bacteria and leukocytes (40 bacteria/μL and 30 leukocytes/μL) with moderate discriminative capacity (AUC of 0.76). These cutoff values, when adjusted to the patient population and specific objectives of each laboratory, can help reduce unnecessary urine cultures, lower costs, and support a more rational use of antibiotics, which is essential in combating antimicrobial resistance.
Urinary tract infection (UTI) is one of the most common infections in both community and hospital settings, with significant clinical and economic challenges. Laboratory testing plays a crucial role in diagnosing this condition but faces limitations such as the time required to obtain results and the growing impact of antimicrobial resistance. This study aims to enhance the efficiency of laboratory diagnosis of UTI by evaluating factors associated with positive urine culture results and proposing criteria to reduce unnecessary urine cultures. A retrospective observational study was conducted using a sample of 2,528 urine culture requests processed in a hospital unit. The analysis included sediment urine test results, such as bacteria and leukocyte counts, as well as variables like patients’ age and sex. Multivariate logistic regression models were applied, and ROC analysis was performed to identify cutoff values with the best predictive capacity. The results revealed significant differences related to patients’ sex and age, identifying cutoff values for bacteria and leukocytes (40 bacteria/μL and 30 leukocytes/μL) with moderate discriminative capacity (AUC of 0.76). These cutoff values, when adjusted to the patient population and specific objectives of each laboratory, can help reduce unnecessary urine cultures, lower costs, and support a more rational use of antibiotics, which is essential in combating antimicrobial resistance.
Description
Tese de Mestrado em Bioestatística e Biometria, apresentada à Universidade Aberta
Keywords
Infeção do trato urinário Sedimento urinário Cultura microbiológica de urina Regressão logística multivariada Urinary tract infection Microbiologic urine culture Urinary sediment Multivariate logistic regression
Pedagogical Context
Citation
Publisher
CC License
Without CC licence