A validation study of UCSD-Mayo risk score in predicting hospital-acquired acute kidney injury in COVID-19 patients

Figures & data

Figure 1. The flowchart of the enrolled patients.

Table 1. Demographic and outcome of patients from COVID-19 validation cohort and UCSD development cohort.

Variables	COVID-19 validation cohort	UCSD development cohort	p Value
	(n = 572)	(n = 573)
Age (years)	61 ± 14	54 ± 18	NA
Age > 70, n (%)	144 (25)	120 (21)	.09
Male, n (%)	300 (52)	367 (64)	<.01
Hypertension, n (%)	236 (41)	207 (36)	.07
CLD, n (%)	68 (12)	61 (11)	.51
CHF, n (%)	63 (11)	77 (13)	.21
CKD, n (%)	28 (5)	43 (8)	.07
ASCVD, n (%)	57 (10)	NA	NA
Anemia, n (%)	24 (4)	NA	NA
Acidosis, n (%)	62 (11)	79 (14)	.13
Severe infection, n (%)	48 (8)	93 (16)	<.01
Nephrotoxin exposure, n (%)	8 (1)	114 (20)	<.01
Mechanical ventilation, n (%)	14 (2)	238 (42)	<.01
SCr at admission, mg/dL	0.8 (0.6-0.9)	0.9 (0.7–1.1)	NA
Incidence of AKI, n (%)	44 (8)	127 (22)	<.01
Need for RRT, n (%)	18 (3)	30 (5)	.08
Mortality, n (%)	66 (12)	41 (7)	.01

COVID-19: coronavirus disease 2019; SCr: serum creatinine; CLD: chronic liver disease; CHF: congestive heart failure; CKD: chronic kidney disease; ASCVD: atherosclerotic coronary vascular disease; RRT: renal replacement therapy; AKI: acute kidney injury: an increase in serum creatinine by 0.3 mg/dL within 48 h or a 50% increase in serum creatinine from baseline within 7 d.

Figure 2. Validation of UCSD-Mayo Model in the COVID-19 cohort. (a) Area under curve of UCSD-Mayo risk score for prediction of AKI in COVID-19 validation cohort. (b) Kaplan–Meier curve with AKI-free survival of patients (n = 572) according to their risk group. (c) Calibration curve of UCSD-Mayo risk score for prediction of acute kidney injury in COVID-19 validation cohort.

Table 2. Cox regression analysis of predictors from the UCSD-Mayo model (n = 572 patients).

Predictors	AKI patients	AKI-free patients	HR [95%CI]
	(n = 44)	(n = 528)
Hypertension, n (%)	28 (64)	208 (39)	2.73 (1.48–5.04)*
CLD, n (%)	15 (34)	53 (10)	4.74 (2.54–8.86)*
CHF, n (%)	25 (57)	38 (7)	14.8 (8.13–26.96)*
CKD, n (%)	2 (5)	26 (5)	1.06 (0.26–4.37)
ASCVD, n (%)	12 (27)	45 (9)	4.03 (2.07–7.82)*
Anemia, n (%)	6 (14)	18 (3)	4.08 (1.72–9.65)*
Acidosis, n (%)	13 (30)	49 (9)	4.09 (2.14–7.81)*
Severe infection, n (%)	14 (32)	34 (6)	6.74 (3.57–12.73)*
Nephrotoxin exposure, n (%)	2 (5)	6 (1)	3.02 (0.73–12.49)
Mechanical ventilation, n (%)	9 (20)	5 (10)	23.89 (11.31–50.45)*

*With a p value < .01.

CLD: chronic liver disease; CHF: congestive heart failure; CKD: chronic kidney disease; ASCVD: atherosclerotic coronary vascular disease; AKI: acute kidney injury: an increase in serum creatinine by 0.3 mg/dL within 48 h or a 50% increase in serum creatinine from baseline within 7 d.

Figure 3. Subgroup analysis of UCSD-Mayo model in COVID-19 cohort.

Table 3. Clinical features of patients with ICU and non-ICU admission.

Variables	ICU admission	non-ICU admission	p Value
	(n = 45)	(n = 527)
Age, years, mean (SD)	70 ± 11	61 ± 14
Age >70, n (%)	23 (51)	121 (23)	<.01
Male, n (%)	27 (60)	273 (52)	.30
Hypertension, n (%)	31 (69)	205 (39)	<.01
CLD, n (%)	19 (42)	49 (9)	<.01
CHF, n (%)	27 (60)	36 (7)	<.01
CKD, n (%)	5 (11)	23 (4)	.10
ASCVD, n (%)	15 (33)	24 (5)	<.01
Anemia, n (%)	5 (11)	19 (4)	.04
Acidosis, n (%)	15 (33)	47 (9)	<.01
Severe infection, n (%)	13 (29)	35 (7)	<.01
Nephrotoxin exposure, n (%)	1 (0.02)	1 (0.002)	<.01
Mechanical ventilation, n (%)	14 (31)	0 (0)	<.01
SCr at admission, mg/dL	0.9 (0.6–1.2)	0.7 (0.6–0.9)
Incidence of AKI, n (%)	26 (58)	18 (3)	<.01
AKI stage 1, n (%)	11 (24)	6 (1)	<.01
AKI stage 2, n (%)	7 (16)	6 (1)	<.01
AKI stage 3, n (%)	8 (18)	6 (1)	<.01
Need for RRT, n (%)	10 (22)	8 (2)	<.01
Mortality, n (%)	39 (87)	27 (5)	<.01

ICU: intensive care unit; CLD: chronic liver disease; CHF: congestive heart failure; CKD: chronic kidney disease; ASCVD: atherosclerosis coronary vascular disease; RRT: renal replacement therapy; AKI: acute kidney injury: an increase in serum creatinine by 0.3 mg/dL within 48 h or a 50% increase in serum creatinine from baseline within 7 d.

Data availability statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.