Can urinary albumin-creatinine ratio and protein-creatinine ratio be used interchangeably in diagnosis and management of pediatric steroid sensitive nephrotic syndrome?

Background: In pediatric nephrotic syndrome (NS), conventionally, spot urinary protein-creatinine ratio (PCR) is used for diagnosis, and to define relapse and remission. This study evaluates whether urine albumin-creatinine ratio (ACR) can reliably predict proteinuria status to manage NS according to existing PCR guidelines Methods: Children between 1 to 18 years of age with steroid sensitive NS were included. Blood samples for albumin, creatinine and morning urine samples for albumin, total protein and creatinine were taken, PCR and ACR were calculated in mg/mg. Data obtained was randomly split into training (70%) and test (30%) cohorts. Data was bootstrapped 100,000 times, a linear regression model predicting ACR from PCR was fit to the training cohort, and the ACR cut-offs for NS relapse and remission were derived from conventional PCR thresholds. Finally, the model's accuracy was validated using the test cohort. Results: In 202 patients (median age 7.4 years, 63% male, normal creatinine for age), ACR and PCR were highly correlated with r _s =0.97. In the training cohort, the final regression equation was estimated as ACR = 0.76(PCR)–0.06. Using this, PCR values of 2 mg/mg and 0.2 mg/mg yielded ACR cut-offs of 1.46 mg/mg (relapse) and 0.09 mg/mg (remission) respectively. In the test cohort, these cut-offs showed high diagnostic accuracy, with sensitivities of 97–100% and specificity of 96% Conclusion: Our data shows very strong correlation between spot urinary ACR and PCR in children with steroid sensitive NS. The ACR cut-offs obtained can be useful in diagnosis and follow-up of SSNS, or in evaluating retrospective study data, when ACR is available rather than PCR.