key: cord-0819161-3ss41iea authors: Xu, K.; Shang, N.; Levitman, A.; Corker, A.; Kudose, S.; Yaeh, A.; Neupane, U.; Stevens, J.; Mohan, S.; Sampogna, R.; D'Agati, V.; Kiryluk, K.; Barasch, J. title: Urine test predicts kidney injury and death in COVID-19 date: 2021-06-14 journal: nan DOI: 10.1101/2021.06.10.21258638 sha: 8e6b28a078595bc59b19c93349b0ad8ac1f68b09 doc_id: 819161 cord_uid: 3ss41iea Background: Kidney injury is common in COVID-19 infection, but serum creatinine (SCr) is not a sensitive or specific marker of kidney injury. We hypothesized that molecular markers of tubular injury could diagnose COVID-19 associated kidney damage and predict its clinical course. Methods: This is a prospective cohort study of 444 consecutive COVID-19 patients (43.9% females, 20.5% African American, 54.1% Latinx) in Columbia University's Emergency Department at the peak of the New York pandemic (March-April 2020). Urine and blood were collected simultaneously at admission (median time of day 0, IQR 0-2 days) and within 1 day of a positive SARS-CoV-2 test in 70% of patients. Biomarker assays were blinded to clinical data. Results: Urinary NGAL (uNGAL) was strongly associated with AKI diagnosis (267{+/-}301 vs. 96{+/-}139 ng/mL, P=1.6x10-10). uNGAL >150ng/mL had 80% specificity and 75% sensitivity to diagnose AKIN stage 2 or higher. uNGAL quantitatively predicted the duration of AKI and outcomes, including death, dialysis, shock, and longer hospital stay. The risk of death increased 73% per standard deviation of uNGAL [OR (95%CI): 1.73 (1.29-2.33), P=2.8x10-4] and was independent of baseline SCr, co-morbidities, and proteinuria [adjusted OR (95%CI): 1.51 (1.10-2.11), P=1.2x10-2]. Proteinuria and uKIM-1 also indicated tubular injury, but were not diagnostic of AKI. Typically, distal nephron segments transcribe NGAL, but in COVID-19 biopsies with widespread acute tubular injury (ATI), NGAL expression overlapped KIM-1 in proximal tubules. Conclusion: uNGAL predicted the diagnosis, duration, and severity of AKI and ATI, as well as hospital stay, dialysis, shock, and death in patients with acute COVID-19. Acute loss of kidney function, measured by a rise in serum creatinine (SCr), is common in the setting of acute SARS-CoV-2 infection [1] [2] [3] [4] . Elevated SCr is present in one-third of patients hospitalized with COVID-19 [5] [6] [7] [8] [9] [10] [11] . Yet, SCr fails to represent the full burden of COVID-19 kidney injury, because it cannot diagnose early stages or subtotal kidney damage 12, 13 . Moreover, a rise in SCr does not reveal the anatomical or physiological basis of kidney dysfunction. For example, an isolated SCr level cannot distinguish volume depletion from tubular injury, the two common entities that contribute to renal manifestations of COVID-19 and which must be distinguished for appropriate triage and fluid management at hospital admission 10 . Therefore, the well-established association between COVID-19 and loss of kidney function begs the question as to the best noninvasive method to detect COVID-19 associated tubular injury, to identify the mechanisms of functional failure, and to prognosticate the outcome of the illness. Prior research in human and mouse models has demonstrated that two molecular markers of tubular injury, urinary NGAL (uNGAL) and urinary KIM-1 (uKIM-1) derive from different segments of the kidney and allow for sensitive detection and real time distinction between volume depletion and acute tubular injury (ATI) 12, [14] [15] [16] . Consequently, these biomarkers may elucidate the pathogenesis of COVID-19 associated kidney injury, but have not been tested in this setting. Here, we assessed their performance in the diagnosis and stratification of kidney injury in acute COVID-19 patients presenting to Columbia University's Emergency Department at the peak of the New York City pandemic (March-April, 2020). We tested uNGAL and uKIM-1 for association with the diagnosis, duration, and severity of Acute Kidney Injury (AKI) defined by SCr-based AKIN criteria (primary outcomes), as well as with in-hospital death, dialysis, shock, respiratory failure, and length of hospital stay (secondary outcomes). All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 14, 2021. ; https://doi.org/10.1101/2021.06. 10 .21258638 doi: medRxiv preprint The Columbia University COVID-19 Biobank was established in response to the surge of SARS-CoV-2 in New York City in March 2020. The Biobank recruited consecutive COVID-19 cases, regardless of age, sex, or race/ethnicity, diagnosed and treated at Columbia University Irving Medical Center (CUIMC). All included cases were diagnosed by positive SARS-CoV-2 PCR test based on nasopharyngeal sampling. The biobank collected residual blood and urine samples after clinical testing from every COVID-19 patient diagnosed at CUIMC, providing us with the largest cohort of COVID-19 patients for urinary biomarker analysis to date 17, 18 . The cohort was recruited from 3/24/2020 to 4/27/2020 with exclusion only for end stage renal disease on dialysis (Supplemental Figure S1 ) and linked to the Electronic Health Records for patient characteristics and outcomes (Supplemental Table S1 ). Shock was defined by the need for vasopressors, and respiratory failure was defined by the need for either invasive or noninvasive positive pressure ventilation. Kidney biopsies were accessioned by the CUIMC Renal Pathology Laboratory from 3/13/2020 to 6/1/2020, including 13 kidney biopsies from COVID-19 cases 19 and 4 non-COVID-19 specimens. Baseline SCr was determined, as described by Stevens et al. 20 , as follows in order of preference, • Median SCr from 365 to 31 days before urine collection. If not available, then: • Minimum SCr from 30 days before urine collection to the day of collection. If not available, then: • Minimum SCr from urine collection to 7 days in hospital. The patient was classified as "unknown" status if none of the above criteria were met. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The loss of kidney excretory function was determined by SCr kinetics, according to Acute Kidney Injury Network (AKIN) definitions 21 , and interpreted as an absolute increase in SCr of ³0.3mg/dL or a ³ 50% increase from baseline. AKIN stages were classified as follows: • AKIN Stage 1: ≥0.3mg/dL increase in sCr within a 48-hour window OR 1.5 to 2-fold increase in sCr compared to baseline. • AKIN Stage 2: >2 to 3-fold increase in SCr compared to baseline. • AKIN Stage 3: ≥0.5mg/dL increase in SCr within a 48-hour window when SCr ≥4.0mg/dL OR >3-fold increase in SCr compared to baseline. The first SCr in days 0-2 after urine collection was used to diagnose AKIN stage using the above criteria. In select cases, the day 1 AKIN score was imputed when the preceding and subsequent AKIN scores were identical. Further categorization of elevated SCr was based on the duration of SCr elevation above baseline: • No AKI (AKIN=0) -not meeting AKIN criteria within 2 days of presentation (must have SCr values for both days). • Transient AKI (tAKI) -met AKIN criteria on day 0 or 1 of presentation but normalized below AKIN detection thresholds within 2 days after first detection (total AKI duration <72 hours). • Sustained AKI (sAKI) -met AKIN criteria within 2 days of presentation but normalized below the AKIN detection thresholds only after 2 days from the first detection (total AKI duration >72 hours). • Unknown -missing baseline SCr, or insufficient SCr measurements to determine SCr kinetics, or missing measurements on day 0 or 1 that could not be imputed due to discrepant AKIN scores. Diagnosis and staging of chronic kidney disease (CKD) utilized KDIGO criteria 22 . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Urine was centrifuged (12,000rpm; 10min) and 10uL applied to NGAL gRAD dipsticks (BioPorto). Color development over 15min was compared to the manufacturer's test line by two independent, blinded readers and associated with semi-quantitative bins for risk of kidney damage. Extent of acute tubular injury (ATI) was assigned based on the following semi-quantitative scale: none (<5% of tubules involved), mild (<25%), moderate (25-50%), severe (>50%). Histologic features of ATI included loss of brush border, epithelial simplification, intracytoplasmic vacuolization, overt necrosis, apoptosis and cell shedding. Male and female wild-type C57Bl/6 mice, aged 8-10 weeks (Jackson Labs) were anesthetized with isoflurane and placed on a warming table to maintain a rectal temperature of 37°C. The left All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 14, 2021. ; https://doi.org/10.1101/2021.06.10.21258638 doi: medRxiv preprint renal pedicle was clamped for 10, 20, 30, or 40 minutes using microvascular clamps (Fine Science Tools). After the clamps were removed, reperfusion of the kidneys was visually confirmed. The kidneys were harvested at 24hrs. In Situ Hybridization (ISH) on formalin-fixed paraffin-embedded (FFPE) human kidney tissues was performed using the chromogenic RNAscope® 2. Continuous variables were compared using a two-sample t-test and summarized as mean ± SD. Non-normally distributed continuous variables (i.e., uNGAL, uKIM-1) were natural logtransformed and standard-normalized before statistical testing. Categorical variables were compared using Chi-squared or Fisher's Exact test. For testing binary outcomes, we used logistic regression. Ordinal outcomes, such as AKIN stage, were tested using ordinal logistic regression. Ordinal predictors, such as urine dipstick category or proteinuria grade, were tested under the assumption of linear effects using a slope test within the framework of a generalized linear model tailored to the outcome of interest (e.g. logistic or ordinal logistic for binary or ordinal outcomes, respectively). We used Cox proportional hazards model for the time-to-event analyses of mortality. We used competing risks regression model for the analysis of the length of hospital stay, with death as a competing risk 23, 24 . The proportional hazards assumption was verified by All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 14, 2021. ; https://doi.org/10.1101/2021.06.10.21258638 doi: medRxiv preprint testing scaled Schoenfeld residuals for each predictor against observation time. Associations of urinary biomarker levels with clinical outcomes were adjusted for the following covariates: age, sex, race, and ethnicity (minimally-adjusted model), baseline serum creatinine and pre-existing obesity, diabetes, hypertension, transplant (any organ), cancer (hematologic cancers and any solid tumors), cardiovascular disease (coronary artery disease, heart failure, cerebral infarction), pulmonary disease (asthma, chronic obstructive pulmonary disease, interstitial pulmonary disease, primary pulmonary hypertension, idiopathic pulmonary fibrosis) (fully-adjusted model-1), as well as proteinuria (fully-adjusted model-2). In the analysis of primary outcomes, we considered two-sided P<0.05 as statistically significant. In the analysis of secondary outcomes, we considered P<0.01 as significant, corresponding to the Bonferroni-correction for 5 major independent outcomes tested. All statistical analyses were performed using R (CRAN The Columbia University Biobank COVID-19 studies were reviewed and approved by the All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. We analyzed urine samples from 444 COVID-19 patients collected prospectively by the Columbia University COVID-19 Biobank at a median time of day 0 (hospital admission, IQR 0-2 days), within 1 day of a positive SARS-CoV-2 test in 70% of patients (Supplemental Figure S1 ). The cohort was diverse in age, sex, race, ethnicity, and pre-existing comorbidities (Supplemental Table S1 ). Admission uNGAL levels were higher among patients reaching the diagnosis of AKI (uNGAL: 267±301 vs. 96±139 ng/mL, P=1.6x10 -10 ) or sustained AKI (lasting >72 hours; sAKI; uNGAL: 332±324 vs. 96±139 ng/mL, P=3.3x10 -11 ) compared to those without AKI (Figure 1a) . uNGAL levels were also associated with increasing AKIN stage (slope P=4.0x10 -21 ) (Figure 1b) . Accordingly, the area under the receiver operating characteristics (ROC) curve for uNGAL increased for higher AKIN stages (Figure 1c ). For example, uNGAL had 80% specificity and 75% sensitivity to diagnose AKIN stage 2 or 3 at a cutoff level of 150ng/mL (Table 1) . Notably, uNGAL levels were marginally elevated in patients who experienced transient AKI (tAKI) (187±257 vs. 96±139 ng/mL, P=0.021) or the AKIN-1 stage (162±219 vs. 96±139 ng/mL, P=6.8x10 -3 ) compared to those with no AKI. Measurement of uNGAL using a novel rapid point-of-care semi-quantitative dipstick 20 strongly correlated with ELISA measurements (Spearman's correlation ρ=0.84, P=1.5x10 -119 ), reproducing the association of uNGAL with AKI, sAKI, and AKIN stages at the same level as ELISA uNGAL (Supplemental Figure S2 ). The association of uNGAL with the primary outcomes was independent of age, sex, race and ethnicity (minimally-adjusted model), baseline creatinine and other comorbidities (fully-adjusted model-1) and was also independent of proteinuria measured in the same urine sample (fully adjusted model-2; Supplemental Table S2 ). In addition, uNGAL was quantitatively associated with four out of five secondary outcomes of COVID-19, including in-hospital death, acute dialysis, All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. shock, and length of hospital stay (Figure 2a) , independent of demographics, comorbidities, baseline renal function, and proteinuria (Supplemental Table S2 Table S2 ). To determine if our findings were specific to COVID-19, we examined a second cohort of comparable size (426 patients) admitted through the same Emergency Department (6/2017-1/2019) 20 and analyzed using identical methods (Supplemental Table S1 ). The COVID-19 cohort was older and enriched in Latinx patients, but the burden of chronic kidney disease was similar in both cohorts. Notably, COVID-19 patients were 2.6-times more likely to present with AKI (35.2% vs. 13.6%, P=2.8x10 -13 ), 3.9-times more likely to have sustained AKI (17.5% vs. 4.5%, P=2x10 -9 ) and 1.8-times more likely to have more severe disease (AKIN 2-3, 12.5% vs. 6.8%, P=6.6x10 -3 ) compared to our historical cohort (Supplemental Table S1 ), and similar to published data 26 . Urinary findings also differed; both KIM-1 and proteinuria were elevated in all COVID-19 cases compared to the historical cohort, even in patients without AKI, or AKIN=0 (uKIM-1: 2.57±2.44 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Figures S3 and S4) . Consistently, urinary shedding of proximal tubule cells specifically marked by LRP2+ at AKIN=0 was more prominent in COVID-19 than in non-COVID-19 patients (n=40; 2.61-fold increase in urinary LRP2+ cells, P=5.5x10 -3 ) while UMOD+ cells were present regardless of COVID-19 status (Supplemental Figure S5) . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. A simple urine test for NGAL -a protein first discovered in mouse models 13 -was quantitatively associated with the diagnosis, the severity, and the duration of COVID-19 associated functional kidney failure. Increasing the severity of injury in COVID-19 human kidneys and in mouse models of ischemia broadened NGAL RNA expression. In this light, we demonstrate that the level of uNGAL mirrors the degree of histopathological ATI. The diagnostic potential of uNGAL to reflect ATI and subsequent functional failure differs from SCr, which increases only as a marker of functional failure 12 . Unlike uNGAL, an acute rise of SCr is delayed after the onset of injury 28, 29 and blunted by mechanisms that enhance its excretion (the renal reserve) or limit its production 30 . In fact, in this study SCr measurements underestimated the impact of COVID-19 since both uKIM-1 and proteinuria were markedly elevated even in the absence of functional kidney failure by SCr criteria (AKIN=0). The shedding of proximal tubule cells specifically marked by LRP2 even at AKIN=0 further demonstrated the insensitivity of SCr to appropriately diagnose proximal tubular injury. These results provide the possibility of a sensitive diagnostic strategy that bypasses the delays and insensitivity of SCr. Increased proteinuria and uKIM-1 in patients with acute COVID-19 are indicative of proximal tubular injury, while uNGAL, but not uKIM-1, predicts subsequent stages of nephron injury resulting in functional failure and associated clinical outcomes. We suggest that together these biomarkers offer sensitive, timely, and comprehensive evaluation of both tubular injury and functional loss in COVID-19. Moreover, accurate testing is possible at the bedside using a rapid point-of-care dipstick 20,31 , which mitigates the risk of handling infectious body fluids and may be particularly helpful in the setting of high patient volumes witnessed in Emergency Departments during recent COVID-19 surges. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In summary, uNGAL is diagnostic of COVID-19-associated AKI, correlates with the severity and duration of AKI, and predicts key clinical outcomes including length of stay, dialysis, shock, and death. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. collecting or analyzing samples, interpretation of data or the preparation and revision of the manuscript and datasets. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or Columbia University. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 14, 2021. ; https://doi.org/10.1101/2021.06.10.21258638 doi: medRxiv preprint Figure 2 . Higher urinary NGAL levels are associated with critical illness and death in patients with COVID-19: (a) Urinary NGAL levels were associated with AKI and sustained AKI (>72hrs) after adjustment for age, sex, race, and ethnicity (minimally-adjusted model, blue) and baseline SCr and pre-existing comorbidities (fully-adjusted model, red; N=371). Urinary NGAL levels were also associated with secondary outcomes of death, dialysis, shock, and respiratory failure in both minimally-and fully-adjusted models, N=440. In contrast, uKIM-1 was not associated with AKI or any secondary outcomes except for respiratory failure. Odds ratios (OR) and hazard ratios (HR) are expressed per one unit of standard deviation of biomarker distribution; 95% CI: 95% confidence intervals. (b) Kaplan-Meier survival analysis demonstrates survival differences by tertile of urinary NGAL levels measured by ELISA or (c) by three levels of urinary NGAL dipstick test (unadjusted P-values provided for both b and c; N=440). All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 14, 2021. ; https://doi.org/10.1101/2021.06.10.21258638 doi: medRxiv preprint Figure 3 . The expression level of Ngal (red-brown) and Kim-1 (blue-purple) depended on the dose of arterial ischemia in mouse: (a-e) Ngal expression was found at the cortico-medullary junction after 10min of ischemia, but throughout the medulla and papilla after 30-40min of ischemia. Kim-1 expression was found in the cortex and throughout the cortico-medullary junction. (e-f) Prolonged ischemia (40min) broadened the expression domain of Ngal to include the proximal tubule marked by Kim-1. In contrast to Ngal, Kim-1 expression remained localized to the cortex and cortico-medullary junction. Bars a-e: 500μm; Bars f: 20μm. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Table 1 . Diagnostic properties of urinary NGAL test at test cutoffs corresponding to specificity of 80%, 85%, and 90%. N=371. PPV, positive predictive value; NPV, negative predictive value; +LR, positive likelihood ratio; -LR, negative likelihood ratio. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted June 14, 2021. ; https://doi.org/10.1101/2021.06.10.21258638 doi: medRxiv preprint Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study Characterization and clinical course of 1000 patients with coronavirus disease 2019 in New York: retrospective case series Extrapulmonary manifestations of COVID-19 Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area Kidney disease is associated with in-hospital death of patients with COVID-19 Outcomes for Patients With COVID-19 and Acute Kidney Injury: A Systematic Review and Meta-Analysis High rate of renal recovery in survivors of COVID-19 associated acute renal failure requiring renal replacement therapy Acute kidney injury in patients hospitalized with COVID-19 Acute Kidney Injury in COVID-19: Emerging Evidence of a Distinct Pathophysiology COVID-19-associated acute kidney injury: consensus report of the 25th Acute Disease Quality Initiative (ADQI) Workgroup Prevalence and impact of acute renal impairment on COVID-19: a systematic review and meta-analysis Diagnostic and prognostic stratification in the emergency department using urinary biomarkers of nephron damage: a multicenter prospective cohort study The Ngal reporter mouse detects the response of the kidney to injury in real time Unique Transcriptional Programs Identify Subtypes of AKI Kidney Injury Molecule-1 (KIM-1): a novel biomarker for human renal proximal tubule injury Kidney injury molecule-1 (KIM-1): a urinary biomarker and much more Urinary Neutrophil Gelatinase-Associated Lipocalin in Critically Ill Patients With Coronavirus Disease Incorporation of Urinary Neutrophil Gelatinase-Associated Lipocalin and Computed Tomography Quantification to Predict Acute Kidney Injury and In-Hospital Death in COVID-19 Patients Kidney Biopsy Findings in Patients with COVID-19 Rule Out Acute Kidney Injury in the Emergency Department With a Urinary Dipstick Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury Nomenclature for kidney function and disease: report of a Kidney Disease: Improving Global Outcomes (KDIGO) Consensus Conference A Proportional Hazards Model for the Subdistribution of a Competing Risk How to handle mortality when investigating length of hospital stay and time to clinical stability The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies AKI in Hospitalized Patients with and without COVID-19: A Comparison Study Single-Cell Profiling of AKI in a Murine Model Reveals Novel Transcriptional Signatures No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Combining functional and tubular damage biomarkers improves diagnostic precision for acute kidney injury after cardiac surgery Identification of neutrophil gelatinase-associated lipocalin as a novel early urinary biomarker for ischemic renal injury A rapid urine test for early detection of kidney injury Assessment of Acute Kidney Injury and Longitudinal Kidney Function After Hospital Discharge Among Patients With and Without COVID-19 No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.