Rohit Bhowmick1, Ramachandran Rameshkumar2, Madhusudhanan Ponnusamy3, Vishnukumar Rajaraman3, Muthu Chidambaram1,4, Abraar Sheriff1, Subramanian Mahadevan1. 1. Division of Pediatric Critical Care, Department of Pediatrics, Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Puducherry, 605 006, India. 2. Division of Pediatric Critical Care, Department of Pediatrics, Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Puducherry, 605 006, India. krramesh_iway@yahoo.co.in. 3. Department of Nuclear Medicine, Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Puducherry, 605 006, India. 4. Department of Pediatrics, Saveetha Medical College, Chennai, Tamil Nadu, 600004, India.
Abstract
BACKGROUND: Studies comparing the modified Schwartz formula with measured GFR (m-GFR) are lacking in critically ill children. METHODS: This prospective cohort study enrolled children aged 1 month to 12 years, within 24 h of admission. m-GFR measured by technetium-99m-labeled diethylenetriaminepentaacetic acid (99mTc-DTPA) and calculated by Russell's two-sample slope-intercept method. Serum creatinine was estimated by modified Jaffe method and estimated GFR (e-GFR) calculated by modified Schwartz formula. The primary outcome was to find agreement between the two methods. Bias, precision, and accuracy were calculated. Secondary outcomes were the incidence of AKI (by p-RIFLE criteria) and the difference between the two methods to diagnose AKI. RESULTS: A total of 208 pairs were analyzed. e-GFR showed good agreement with m-GFR with a mean bias of -4.37 ml/min/1.73 m2 and precision (SD of bias) of 33.07, 95% limit of agreement -69.18 to 60.45, and intraclass correlation of 74% (95%CI 66-80%, P < 0.001). e-GFR underestimated m-GFR by 19.8% (95% CI 7.9-31.7%). Accuracy of e-GFR values within 10%, 20%, and 30% of m-GFR were 68.3%, 72.6%, and 78.8%, respectively. Incidence of AKI within 24 h was 60.1% by e-GFR and 54.3% by m-GFR (kappa 0.569, P < 0.001; sensitivity of 85.8%, 95%CI (78-91.7%). CONCLUSIONS: The modified Schwartz formula shows good agreement with 99mTc-labeled DTPA double plasma sample clearance method for calculating GFR in critically ill children aged 1 month to 12 years. The underestimation of GFR should be kept in mind while applying the formula at the bedside in PICU. TRIAL REGISTRATION: Protocol accessible at Clinical Trial Registry of India (CTRI) www.ctri.nic.in . (Trial Registered Prospectively and Registration No. CTRI/2017/10/010014) ([Registered on: 06/10/2017] Trial Registered Prospectively.) (Title "Measured glomerular filtration rate using Diethylenetriaminepentaacetic acid (DTPA) renal scan versus estimated glomerular filtration rate using modified Schwartz formula in critically ill children: A prospective observational, analytical study."). A higher resolution version of the Graphical abstract is available as Supplementary information.
BACKGROUND: Studies comparing the modified Schwartz formula with measured GFR (m-GFR) are lacking in critically ill children. METHODS: This prospective cohort study enrolled children aged 1 month to 12 years, within 24 h of admission. m-GFR measured by technetium-99m-labeled diethylenetriaminepentaacetic acid (99mTc-DTPA) and calculated by Russell's two-sample slope-intercept method. Serum creatinine was estimated by modified Jaffe method and estimated GFR (e-GFR) calculated by modified Schwartz formula. The primary outcome was to find agreement between the two methods. Bias, precision, and accuracy were calculated. Secondary outcomes were the incidence of AKI (by p-RIFLE criteria) and the difference between the two methods to diagnose AKI. RESULTS: A total of 208 pairs were analyzed. e-GFR showed good agreement with m-GFR with a mean bias of -4.37 ml/min/1.73 m2 and precision (SD of bias) of 33.07, 95% limit of agreement -69.18 to 60.45, and intraclass correlation of 74% (95%CI 66-80%, P < 0.001). e-GFR underestimated m-GFR by 19.8% (95% CI 7.9-31.7%). Accuracy of e-GFR values within 10%, 20%, and 30% of m-GFR were 68.3%, 72.6%, and 78.8%, respectively. Incidence of AKI within 24 h was 60.1% by e-GFR and 54.3% by m-GFR (kappa 0.569, P < 0.001; sensitivity of 85.8%, 95%CI (78-91.7%). CONCLUSIONS: The modified Schwartz formula shows good agreement with 99mTc-labeled DTPA double plasma sample clearance method for calculating GFR in critically ill children aged 1 month to 12 years. The underestimation of GFR should be kept in mind while applying the formula at the bedside in PICU. TRIAL REGISTRATION: Protocol accessible at Clinical Trial Registry of India (CTRI) www.ctri.nic.in . (Trial Registered Prospectively and Registration No. CTRI/2017/10/010014) ([Registered on: 06/10/2017] Trial Registered Prospectively.) (Title "Measured glomerular filtration rate using Diethylenetriaminepentaacetic acid (DTPA) renal scan versus estimated glomerular filtration rate using modified Schwartz formula in critically ill children: A prospective observational, analytical study."). A higher resolution version of the Graphical abstract is available as Supplementary information.