Peter A McCullough1, Marlies Ostermann2, Lui G Forni3, Azra Bihorac4, Jay L Koyner5, Lakhmir S Chawla6, Jing Shi7, J Patrick Kampf8, Paul McPherson8, John A Kellum9. 1. Baylor University Medical Center, Baylor Heart and Vascular Hospital, Baylor Heart and Vascular Institute, Dallas, Texas, USA, peteramccullough@gmail.com. 2. Department of Critical Care, King's College London, Guy's and St Thomas' Hospital, London, United Kingdom. 3. Department of Clinical and Experimental Medicine, Faculty of Health Sciences, University of Surrey, Guildford, United Kingdom. 4. Precision and Intelligence in Medicine Partnership, Department of Medicine, University of Florida, Gainesville, Florida, USA. 5. Department of Internal Medicine, Section of Nephrology, Department of Medicine, University of Chicago, Chicago, Illinois, USA. 6. Veterans Affairs Medical Center, San Diego, California, USA. 7. Walker Biosciences, Carlsbad, California, USA. 8. Astute Medical Inc., San Diego, California, USA. 9. Center for Critical Care Nephrology, Department of Critical Care Medicine, School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, USA.
Abstract
INTRODUCTION: Over the course of critical illness, there is a risk of acute kidney injury (AKI), and when it occurs, it is associated with increased length of stay, morbidity, and mortality. The urinary cell-cycle arrest markers tissue inhibitor of metalloproteinase-2 (TIMP-2) and insulin-like growth factor binding protein 7 (IGFBP7) have been utilized to predict the risk of AKI over the next 12 h from the time of sampling. The aim of this analysis was to evaluate the utility of [TIMP-2] × [IGFBP7] measured serially to anticipate the occurrence of AKI over the first 7 days of critical illness. METHODS: This analysis is from a prospective, blinded, observational, international study of patients admitted to intensive care units. We designed the analysis to emulate a clinician-driven serial testing strategy. Urine samples collected every 12 h up to 3 days from 530 patients were considered for analysis. We evaluated [TIMP-2] × [IGFBP7] results for the first 3 measurements (baseline, 12 and 24 h) and continued to evaluate additional results if any of the first 3 were positive >0.3 (ng/mL)2/1,000. Patients were stratified by number of [TIMP-2] × [IGFBP7] results >0.3 (ng/mL)2/1,000 and number of results >2.0 (ng/mL)2/1,000. The primary endpoint was AKI stage 2-3 defined by the Kidney Disease: Improving Global Outcomes (KDIGO) criteria. RESULTS: The median (interquartile range) age was 64 (53-74) years, 61% were men, and 79% were Caucasian. The median APACHE III score was 71 (51-93), and 82% required mechanical ventilation. Baseline serum creatinine was 0.8 mg/dL and 164/530 (31%) developed the primary endpoint by day 7 with a median time from baseline to stage 2/3 AKI of 26 (8-56) h. In patients with negative values for the first 3 tests (≤0.3 (ng/mL)2/1,000), the cumulative incidence of the primary endpoint at 7 days was 13.0%. Conversely, for those with one, two, or three strongly positive values (>2.0 (ng/mL)2/1,000), the cumulative incidence for the primary endpoint at 7 days was 57.7, 75.0, and 94.4%, respectively, p < 0.001 for trend. There were 3.4% with test results between 0.3 and 2.0 (ng/mL)2/1,000 at all measurements; one third of those patients developed the primary endpoint. We observed a graded increase in the primary endpoint in Kaplan-Meier plots for successively positive test results over time. CONCLUSION: Serial urinary [TIMP-2] × [IGFBP7] at baseline, 12 and 24 h, and up through 3 days are prognostic for the occurrence of stage 2/3 AKI over the course of critical illness. Three consecutive negative values (≤0.3 (ng/mL)2/1,000) are associated with very low (13.0%) incidence of stage 2/3 AKI over the course of 7 days. Conversely, emerging or persistent, strongly positive results [>2.0 [ng/mL]2/1,000] predict very high incidence rates (up to 94.4%) of stage 2/3 AKI. There was a low rate of test results between 0.3 and 2.0 (ng/mL)2/1,000, where the primary endpoint was observed in a third of cases.
INTRODUCTION: Over the course of critical illness, there is a risk of acute kidney injury (AKI), and when it occurs, it is associated with increased length of stay, morbidity, and mortality. The urinary cell-cycle arrest markers tissue inhibitor of metalloproteinase-2 (TIMP-2) and insulin-like growth factor binding protein 7 (IGFBP7) have been utilized to predict the risk of AKI over the next 12 h from the time of sampling. The aim of this analysis was to evaluate the utility of [TIMP-2] × [IGFBP7] measured serially to anticipate the occurrence of AKI over the first 7 days of critical illness. METHODS: This analysis is from a prospective, blinded, observational, international study of patients admitted to intensive care units. We designed the analysis to emulate a clinician-driven serial testing strategy. Urine samples collected every 12 h up to 3 days from 530 patients were considered for analysis. We evaluated [TIMP-2] × [IGFBP7] results for the first 3 measurements (baseline, 12 and 24 h) and continued to evaluate additional results if any of the first 3 were positive >0.3 (ng/mL)2/1,000. Patients were stratified by number of [TIMP-2] × [IGFBP7] results >0.3 (ng/mL)2/1,000 and number of results >2.0 (ng/mL)2/1,000. The primary endpoint was AKI stage 2-3 defined by the Kidney Disease: Improving Global Outcomes (KDIGO) criteria. RESULTS: The median (interquartile range) age was 64 (53-74) years, 61% were men, and 79% were Caucasian. The median APACHE III score was 71 (51-93), and 82% required mechanical ventilation. Baseline serum creatinine was 0.8 mg/dL and 164/530 (31%) developed the primary endpoint by day 7 with a median time from baseline to stage 2/3 AKI of 26 (8-56) h. In patients with negative values for the first 3 tests (≤0.3 (ng/mL)2/1,000), the cumulative incidence of the primary endpoint at 7 days was 13.0%. Conversely, for those with one, two, or three strongly positive values (>2.0 (ng/mL)2/1,000), the cumulative incidence for the primary endpoint at 7 days was 57.7, 75.0, and 94.4%, respectively, p < 0.001 for trend. There were 3.4% with test results between 0.3 and 2.0 (ng/mL)2/1,000 at all measurements; one third of those patients developed the primary endpoint. We observed a graded increase in the primary endpoint in Kaplan-Meier plots for successively positive test results over time. CONCLUSION: Serial urinary [TIMP-2] × [IGFBP7] at baseline, 12 and 24 h, and up through 3 days are prognostic for the occurrence of stage 2/3 AKI over the course of critical illness. Three consecutive negative values (≤0.3 (ng/mL)2/1,000) are associated with very low (13.0%) incidence of stage 2/3 AKI over the course of 7 days. Conversely, emerging or persistent, strongly positive results [>2.0 [ng/mL]2/1,000] predict very high incidence rates (up to 94.4%) of stage 2/3 AKI. There was a low rate of test results between 0.3 and 2.0 (ng/mL)2/1,000, where the primary endpoint was observed in a third of cases.
Authors: Federico Nalesso; Leda Cattarin; Laura Gobbi; Antonio Fragasso; Francesco Garzotto; Lorenzo Arcangelo Calò Journal: Int J Nephrol Renovasc Dis Date: 2020-04-24
Authors: Luca Molinari; Fabienne Heskia; Sadudee Peerapornratana; Claudio Ronco; Louis Guzzi; Seth Toback; Robert Birch; Hadi Beyhaghi; Thomas Kwan; J Patrick Kampf; Donald M Yealy; John A Kellum Journal: Crit Care Med Date: 2021-10-01 Impact factor: 9.296