Assessing the associations between immunosuppressant use and COVID-19 severity in patients with autoimmune hepatitis

Table 1 Autoimmune hepatitis cohort characteristics, n (%)

Characteristic	Total (n = 15187)
	Individuals not on IS (n = 10081)	Individuals on IS (n = 5106)
Age
18-29	780 (7.7)	572 (11.2)
30-39	1003 (9.9)	620 (12.1)
40-49	1392 (13.8)	685 (13.4)
50-59	1922 (19.1)	869 (17.0)
60 +	4976 (49.4)	2358 (46.2)
Sex
Female	7176 (71.2)	3609 (70.7)
Male	2681 (26.6)	1275 (25.0)
Unknown	224 (2.2)	222 (4.3)
Race/ethnicity
White	6898 (68.4)	3344 (65.5)
Black	1479 (14.7)	921 (18.0)
Hispanic	640 (6.3)	396 (7.8)
Asian	453 (4.5)	142 (2.8)
Other/unknown	611 (6.1)	303 (5.9)
Height (in.)	65.5	65.8
Weight (lbs.)	180.51	177.4
BMI	30.24	29.5
Region
Northeast	432 (4.3)	252 (4.9)
Midwest	1558 (15.5)	1034 (20.3)
South	2026 (20.1)	1248 (24.4)
West	1049 (10.4)	779 (15.3)
Other/unlisted	5016 (49.8)	1793 (35.1)
Comorbid conditions
CCI	5.16	5.17
Diabetes	1072 (10.6)	635 (12.4)
MI	369 (3.7)	181 (3.5)
CHF	533 (5.3)	307 (6.0)
PVD	568 (5.6)	302 (5.9)
Chronic renal disease	512 (5.1)	461 (9.0)
Chronic pulmonary disease	1406 (13.9)	883 (17.3)
Cirrhosis
Compensated cirrhosis	284 (2.8)	159 (3.1)
Decompensated cirrhosis	143 (1.4)	74 (1.4)
Complications of cirrhosis
Hepatic encephalopathy	< 201	< 201
Esophageal varices	33 (0.3)	< 201 (0.5)
Ascites/pleural effusion	76 (0.8)	27 (0.5)
Hepatorenal syndrome	< 201	< 201
Hepatocellular carcinoma	361 (3.6)	312 (6.1)
IS used
Azathioprine		2112 (41.36)
Prednisone		2977 (58.3)
Budesonide		720 (14.1)
Mycophenolate mofetil		208 (4.1)
Tacrolimus		940 (18.4)
Cyclosporine		200 (3.9)
Laboratory tests
MELD 3.0 score	13.8	13.8
Sodium (mmol/L)	135.8	135.8
Creatinine (mg/dL)	1.2	1.2
Bilirubin (mg/dL)	12	2.2
INR	1.3	1.3
Albumin (g/dL)	3.2	3.3
AST (U/L)	112.7	95.8
ALT (U/L)	109.9	87.0
Alkaline phosphatase (U/L)	139.3	157.6
IgG (mg/dL)	1662.3	1595.2
WBC (1000/μL)2	4.7	6.2
Neutrophils (1000/μL)	8.8	6.9
Hemoglobin (g/dL)	12.6	12.1
Platelet (1000/ μL)	214.1	195.7
Number of IS
0	10081 (100.0)
1		3326 (65.1)
2		1526 (29.9)
3+		254 (5.0)

¹Per the N3C’s data download policy, all variables with ‘under 20’ individual counts have been marked < 20.

²Limited sample size for mean calculation. Continuous variables are listed as means. Categorical variables are counts (percentages).

IS: Immunosuppression; BMI: Body mass index; CCI: Charlson comorbidity index; MI: Myocardial infarction; CHF: Congestive heart failure; PVD: Peripheral vascular disease; MELD 3.0: Model for end-stage liver disease 3.0 score; INR: International normalized ratio; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase; IgG: Immunoglobulin G; WBC: White blood cell count.

Table 2 Adjusted logistic regression for associations between immunosuppression type and severe acute respiratory syndrome coronavirus 2 infection

Immunosuppressant	Odds ratio	P value	95%CI
Prednisone	0.81	0.04	0.65-0.99
Azathioprine	0.98	0.84	0.78-1.21
Budesonide	0.77	0.22	0.50-1.15
Tacrolimus	0.66	0.06	0.41-0.99
Mycophenolate mofetil	1.32	0.47	0.59-2.61
Cyclosporine	0.81	0.44	0.44-1.35
Azathioprine × budesonide	1.88	0.04	1.02-3.44
Prednisone × azathioprine	1.06	0.78	0.72-1.54
Prednisone × budesonide	1.25	0.50	0.65-2.32
Budesonide × mycophenolate mofetil	0.84	0.87	0.04-5.23
Prednisone × mycophenolate mofetil	0.64	0.42	0.21-1.88
Budesonide × tacrolimus	0.80	0.73	0.18-2.47
Prednisone × tacrolimus	1.99	0.02	1.14-3.53

Table 3 Adjusted logistic regression for associations between immunosuppression type and severe acute respiratory syndrome coronavirus 2 infection: Sensitivity analysis for sub-population with full model for end stage liver disease 3.0 score

Immunosuppressant	Odds ratio	P value	95%CI
Prednisone	0.93	0.74	0.58-1.43
Azathioprine	1.07	0.79	0.63-1.74
Budesonide	0.60	0.36	0.18-1.57
Tacrolimus	0.76	0.57	0.25-1.81
Mycophenolate mofetil	1.56	0.68	0.08-8.97
Cyclosporine	0.93	0.89	0.27-2.38
Azathioprine × budesonide	4.44	0.02	1.29-16.72
Prednisone × azathioprine	0.93	0.86	0.41-2.10
Prednisone × budesonide	1.46	0.56	0.39-5.16
Budesonide × mycophenolate mofetil	1.92e-06	0.99	NA, infinity
Prednisone × mycophenolate mofetil	0.29	0.41	0.01-8.19
Budesonide × tacrolimus	3.21	0.23	0.38-20.00
Prednisone × tacrolimus	1.85	0.32	0.58-6.69

NA: Not available.

Table 4 Adjusted Cox regression for associations between immunosuppression type and all-cause mortality amongst autoimmune hepatitis patients with coronavirus disease 2019

Immunosuppressant	Hazard ratio	P value	95%CI
Prednisone	0.77	0.40	0.42-1.41
Azathioprine	0.66	0.41	0.24-1.78
Budesonide	2.03	0.21	0.67-6.14
Tacrolimus	3.46	0.10	0.80-14.92
Mycophenolate mofetil	3.47	0.24	0.44-27.59
Cyclosporine	1.18	0.83	0.25-5.53

Table 5 Adjusted Cox regression for associations between immunosuppression type and all-cause mortality amongst autoimmune hepatitis patients with coronavirus disease 2019: Sensitivity analysis for sub-population with full model for end stage liver disease 3.0 score

Immunosuppressant	Hazard ratio	P value	95%CI
Prednisone	0.25	0.11	0.05-1.35
Azathioprine	0.06	0.25	0.00-7.24
Budesonide	12.2	0.30	0.11-1321.74
Tacrolimus	1.51	0.74	0.14-16.50
Mycophenolate mofetil	NA	NA	NA, NA
Cyclosporine	1.75e-09	1.00	0.00 to infinity

NA: Not available.