Clinical prediction models for acute kidney injury. / Modelos de predição clínica para lesão renal aguda na unidade de terapia intensiva: uma revisão sistemática.

OBJECTIVE: To report on the currently available prediction models for the development of acute kidney injury in heterogeneous adult intensive care units. METHODS: A systematic review of clinical prediction models for acute kidney injury in adult intensive care unit populations was carried out. PubMed® was searched for publications reporting on the development of a novel prediction model, validation of an established model, or impact of an existing prediction model for early acute kidney injury diagnosis in intensive care units. RESULTS: We screened 583 potentially relevant articles. Among the 32 remaining articles in the first selection, only 5 met the inclusion criteria. The nonstandardized adaptations that were made to define baseline serum creatinine when the preadmission value was missing led to heterogeneous definitions of the outcome. Commonly included predictors were sepsis, age, and serum creatinine level. The final models included between 5 and 19 risk factors. The areas under the Receiver Operating Characteristic curves to predict acute kidney injury development in the internal validation cohorts ranged from 0.78 to 0.88. Only two studies were externally validated. CONCLUSION: Clinical prediction models for acute kidney injury can help in applying more timely preventive interventions to the right patients. However, in intensive care unit populations, few models have been externally validated. In addition, heterogeneous definitions for acute kidney injury and evaluation criteria and the lack of impact analysis hamper a thorough comparison of existing models. Future research is needed to validate the established models and to analyze their clinical impact before they can be applied in clinical practice.

Modelos de predição clínica para lesão renal aguda na unidade de terapia intensiva: uma revisão sistemática / Clinical prediction models for acute kidney injury

RESUMO Objetivo: Relatar os modelos atualmente disponíveis de predição para o desenvolvimento de lesão renal aguda em unidades de terapia para adultos heterogêneas. Métodos: Foi realizada revisão sistemática dos modelos de predição de lesão renal aguda em unidades de terapia intensiva para pacientes adultos. Fizemos busca na base PubMed® quanto a publicações que relatassem desenvolvimento de um novo modelo de predição, validação de um modelo estabelecido, ou impacto de um modelo de predição existente para diagnóstico precoce de lesão renal aguda em unidades de terapia intensiva. Resultados: Triamos 583 artigos potencialmente relevantes. Dentre os 32 artigos que restaram após a primeira seleção, apenas 5 cumpriram os critérios para inclusão. As adaptações não padronizadas feitas para definir o valor inicial de creatinina sérica, quando o valor desse exame antes da admissão estava faltando, levaram a definições heterogêneas do resultado. Preditores comumente incluídos foram sepse, idade e nível de creatinina sérica. Os modelos finais incluíram entre 5 e 19 fatores de risco. As áreas sob a curva Característica de Operação do Receptor para predição do desenvolvimento de lesão renal aguda nas coortes de validação interna variaram entre 0,78 e 0,88. Apenas dois estudos tiveram validação externa. Conclusão: Os modelos de predição clínica de lesão renal aguda podem ajudar na aplicação em tempo apropriado de intervenções preventivas para os pacientes adequados. Contudo, poucos modelos foram externamente validados para as populações da unidade de terapia intensiva. Além disto, definições heterogêneas para lesão renal aguda e critérios de avaliação, e a falta de análise do impacto comprometem uma comparação abrangente dos modelos existentes. São necessárias novas pesquisas para validar os modelos estabelecidos e analisar seu impacto clínico, antes que estes possam ser aplicados na prática clínica.

ABSTRACT Objective: To report on the currently available prediction models for the development of acute kidney injury in heterogeneous adult intensive care units. Methods: A systematic review of clinical prediction models for acute kidney injury in adult intensive care unit populations was carried out. PubMed® was searched for publications reporting on the development of a novel prediction model, validation of an established model, or impact of an existing prediction model for early acute kidney injury diagnosis in intensive care units. Results: We screened 583 potentially relevant articles. Among the 32 remaining articles in the first selection, only 5 met the inclusion criteria. The nonstandardized adaptations that were made to define baseline serum creatinine when the preadmission value was missing led to heterogeneous definitions of the outcome. Commonly included predictors were sepsis, age, and serum creatinine level. The final models included between 5 and 19 risk factors. The areas under the Receiver Operating Characteristic curves to predict acute kidney injury development in the internal validation cohorts ranged from 0.78 to 0.88. Only two studies were externally validated. Conclusion: Clinical prediction models for acute kidney injury can help in applying more timely preventive interventions to the right patients. However, in intensive care unit populations, few models have been externally validated. In addition, heterogeneous definitions for acute kidney injury and evaluation criteria and the lack of impact analysis hamper a thorough comparison of existing models. Future research is needed to validate the established models and to analyze their clinical impact before they can be applied in clinical practice.