TABLE 3.

Association between posttransplant 25(OH)D and recurrence of specific GN

Primary GN		Sample size	Serum 25(OH)D	Subdistribution HR (95% CI)a
Focal segmental glomerulosclerosis	Primary analysesb	270 KTRs (20 events)	Per ng/mL lower	1.08 (1.03-1.12)
			Sufficiency (≥30 ng/mL)	Reference
			Insufficiency (21–29 ng/mL)	2.38 (0.72-7.89)
			Deficiency (≤20 ng/mL)	5.26 (1.63-16.95)
	Secondary analysesc	221 KTRs (16 events)	Per ng/mL lower	1.09 (1.02-1.16)
			Sufficiency (≥30 ng/mL)	Reference
			Insufficiency (21–29 ng/mL)	1.29 (0.19-8.84)
			Deficiency (≤20 ng/mL)	7.47 (2.18-25.65)
IgA nephropathy	Primary analysesb	315 KTRs (26 events)	Per ng/mL lower	1.01 (0.97-1.05)
			Sufficiency (≥30 ng/mL)	Reference
			Insufficiency (21–29 ng/mL)	0.85 (0.31-2.31)
			Deficiency (≤20 ng/mL)	2.16 (0.85-5.51)
	Secondary analysesc	239 KTRs (18 events)
			Per ng/mL lower	0.99 (0.96-1.03)
			Sufficiency (≥30 ng/mL)	Reference
			Insufficiency (21–29 ng/mL)	0.57 (0.17-2.00)
			Deficiency (≤20 ng/mL)	1.04 (0.32-3.36)

Subgroup analysis was infeasible in other GNs due to limited sample size. Competing risk regression models were fitted with allograft failure due to other causes (including death) as competing events. One year after each serum 25(OH)D measurement was regarded as a period at risk.

^aModels were adjusted for recipient age at transplant and measurement time of 25(OH)D (ie, months after transplant) in primary analyses; models were further adjusted for concurrent urine protein–creatinine ratio (UPCR), serum albumin, and estimated glomerular filtration rate (eGFR) in secondary analyses.

^bPrimary analyses were conducted in the total study population.

^cSecondary analyses were conducted in the study population where concurrent measurements of UPCR, serum albumin, and serum creatinine were available.

25(OH)D, 25-hydroxy vitamin D; GN, glomerulonephritis; HR, hazard ratio; CI, confidence interval; KTR, kidney transplant recipient.