Urinary angiopoietin-2 is associated with adverse kidney outcomes in the ICU

Acute kidney injury (AKI) is a heterogeneous syndrome encompassing a range of pathophysiology. Past studies have demonstrated the function of the angiopoietin-Tie2 signaling pathway in maintaining the endothelium and its role in AKI. Angiopoietin-2 (Ang-2) is an antagonist in the Tie2 pathway promoting endothelial dysfunction by increasing vascular permeability [1]. Elevated plasma Ang-2 is strongly associated with the development and progression of AKI. Although urinary Ang-2 (uAng-2) offers a non-invasive biomarker, its relationship with kidney-related outcomes remains unknown. We examined whether uAng-2 is associated with AKI and renal replacement therapy (RRT) in two independent, prospectively enrolled intensive care unit (ICU) cohorts.

The Covid-19 Host Response and Clinical Outcomes (CHROME: n = 187) study enrolled patients admitted to the ICU with suspicion for acute respiratory infection at the University of Washington [2]. The Critical Illness Translational Research Cohort (CITRC: n = 259) study enrolled ICU patients at University of Washington without COVID-19 [3]. Both cohorts, collected urine within 24 h of admission. We measured uAng-2 using immunoassays (MesoScale Discovery, Rockville, MD). Approximately one-third of patients had undetectable uAng-2 levels, that were imputed to 5 pg/mL. To account for variations in urine dilution, uAng-2 levels were divided by urine creatinine to calculate a normalized uAng-2.

AKI was defined as an increase in serum creatinine (SCr) of ≥ 0.3 mg/dL or 50% from a baseline SCr consistent with the KDIGO guidelines. In CHROME, baseline SCr was defined as the SCr on study enrollment. In CITRC, baseline SCr was determined by a prior outpatient SCr within 1 year of ICU admission (25%). If unavailable, then the SCr on day of study enrollment (75%) was used as baseline.

Log-linear regression was used to estimate the association of uAng-2 with the outcomes of AKI and RRT, adjusted for the potential confounders of age, sex, body mass index, invasive mechanical ventilation (IMV), and COVID status (where applicable). Huber-White robust standard errors were used to provide 95% confidence intervals.

In CHROME, the mean age was 55 (± 16) years old, 65% were male and 64% had a primary diagnosis of COVID-19. In CITRC, the mean age was 51 (± 14) years old, 66% were male and 0% of patients had COVID-19. The median baseline SCr, in CHROME was 0.9 mg/dL (IQR: 0.7–1.5) and in CITRC was 1.0 mg/dL (IQR: 0.8–1.4). On study enrollment, 67% (CHROME) and 63% (CITRC) of patients received IMV. The median normalized uAng-2 was 49 pg/mg (IQR, 10-181pg/mg) in CHROME and 73 pg/mg (IQR, 14–370 pg/mg) in CITRC. Across cohorts, patients in the highest uAng-2 tertile had greater illness severity, higher rates of diabetes, IMV and vasopressor use (Table S1 and S2).

In CHROME, 30% developed AKI within 7-days after study enrollment and 42% developed AKI by 28-days. In CITRC, 31% developed AKI within 7-days and 36% within 28-days. In multivariable adjusted models, patients in tertile 3 had a greater risk of AKI within 7-days compared to tertile 1 (CHROME: adjusted relative risk [aRR], 2.14 [95% CI, 1.28–3.59]; CITRC: aRR, 2.11 [95% CI, 1.29–3.45]) (Table 1). In both cohorts, elevated uAng-2 was similarly associated with AKI at 28-days (Table S3). We found no significantly higher risk for AKI within 7 or 28-days while comparing tertile 2 to 1. Modeling uAng-2 continuously, a doubling of uAng-2 was significantly associated with AKI within 7 and 28-days in both cohorts (Table 1). Lastly, the rates of RRT during hospitalization were 12% and 5% in CHROME and CITRC, respectively and across cohorts uAng-2 was significantly associated with RRT. Finally, to account for circulating levels of Ang-2, we found that uAng-2 continued to be associated with 7 and 28-day AKI adjusting for plasma Ang-2 (Tables S5 and S6).

Table 1.

Association of urinary Ang-2 with 7-day AKI

Cohorts	Exposure of Urine Ang-2	N at risk (N events)*	Unadjusted RR (95% CI)	Unadjusted p-value	Adjusted RR (95% CI)	M1 adjusted p-value
CHROME	Tertile 1	63 (14)	1.0 (Ref.)		1.0 (Ref.)
	Tertile 2	62 (10)	0.73 (0.35, 1.51)	0.39	0.72 (0.35, 1.50)	0.38
	Tertile 3	62 (32)	2.32 (1.38, 3.91)	0.002	2.14 (1.28, 3.59)	0.004
	Per Doubling		1.16 (1.08, 1.24)	< 0.0001	1.14 (1.07, 1.22)	0.0001
CITRC	Tertile 1	87 (17)	1.0 (Ref.)		1.0 (Ref.)
	Tertile 2	86 (25)	1.49 (0.87, 2.55)	0.15	1.36 (0.79, 2.34)	0.26
	Tertile 3	86 (38)	2.26 (1.39, 3.68)	0.001	2.11 (1.29, 3.45)	0.003
	Per Doubling		1.12 (1.06, 1.17)	< 0.0001	1.11 (1.06, 1.17)	< 0.0001

Adjusted for age, sex, body mass index, invasive mechanical ventilation, and COVID-19 status (COVID-19 adjustment in CHROME only)

*Among patients with AKI in CHROME, 2 patients were started on RRT in tertile 1, 3 in tertile 2 and 8 in tertile 3 by 7 days. In CITRC, 1 patient had RRT in tertile 1, 2 in tertile 2, and 7 in tertile 3 by 7 days

Plasma Ang-2 is a well-established marker of endothelial dysfunction in critical illness, and associated with AKI [4]. Previous studies demonstrated an association between uAng-2 and albuminuria in patients with diabetes but less is known in hospitalized patients [5]. Our results extend these findings by demonstrating uAng-2 is detectable and associated with subsequent AKI and RRT, indicating it may serve as a marker of kidney-specific endothelial injury in critically ill patients.

Strengths include evaluating two independent ICU cohorts and use of a non-invasive urinary biomarker. Limitations include that a baseline outpatient SCr was unavailable in most patients in our study, which is not uncommon in ICU studies. To ensure that AKI occurred after measurement of uAng-2, we used the SCr value at study enrollment as the baseline and determined AKI rates at 7 days. Additionally, we cannot infer the causality of uAng-2 with kidney-related outcomes and thus, the biological role of uAng-2 warrants additional study.

Taken together, uAng-2 is a promising non-invasive biomarker for detection of AKI and need for RRT in ICU patients. Further investigation is needed to identify opportunities for uAng-2 to improve risk stratification or as a target for therapeutic intervention.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AKI

Acute kidney injury

Ang-2

Angiopoietin-2

CHROME

Covid-19 host response and clinical outcomes

CITRC

Critical illness translational research cohort

ICU

Intensive care unit

IMV

Invasive mechanical ventilation

RRT

Renal replacement therapy

Scr

Serum creatinine

uAng-2

Urine angiopoietin-2

Author contributions

Conception and design: ZB, MW, EM, IS, LZ, PBEnrollment of subjects, collection of data and completion of measurements: JH, SS, MW, EM, PBAnalysis and interpretation: All authorsDrafting the manuscript for important intellectual content: All authorsAll authors read and approved this manuscript.

Funding

Funding was received from the NIDDK R01DK133177 (PB).

Data availability

The datasets generated during and/or analyzed during the current study are not publicly available due to currently ongoing research studies, but the data are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The University of Washington institutional review board approved this study.

Consent for publication

No individual personal data is included in the study. The IRB provided waiver of consent to participate in this study.

Competing interests

The authors declare no competing interests.