Skin collagen fluorophore LW-1 versus skin fluorescence as markers for the long-term progression of subclinical macrovascular disease in type 1 diabetes

Table 1 Logistic regression models of microvascular complications vs. age- and duration-adjusted LW-1 levels with and without adjustment for DCCT mean HbA1c up to the biopsy across treatment groups

Outcome	Total N	Event N	LW-1^a			LW-1 Adjusted for HbA1c ^b			DCCT Mean HbA1c^a		DCCT Mean HbA1c Adjusted for LW-1^b
Outcome	Total N	Event N	Odds ratio^c	Entropy R2^d	P value^e	Odds ratio^c	Entropy R2^d	P value^e	Entropy R2^d	P value^e	Entropy R2^d	P value^e
DCCT complications (at DCCT closeout)
Retinopathy/microaneurysms (sustained ≥3 MA ever in DCCT) (primary cohort)	123	31	1.97 (1.40, 3.27)	9.0	0.0004	1.66 (1.00, 2.33)	2.3	0.077 (NS)	14.9	<0.0001	8.1	0.0008
Nephropathy (AER closest to biopsy > 40 mg/24 h)	206	15	1.97 (1.19, 3.27)	7.8	0.0038	1.66 (1.00, 2.76)	4.1	0.0357	8.3	0.0028	4.6	0.0259
EDIC complications (EDIC years 13–16)
Retinopathy/progression (3 or more step progression)	213	91	1.37 (1.04, 1.81)	1.8	0.023	1.09 (0.80, 1.48)	0.1	0.58 (NS)	10.5	<0.0001	8.8	<0.0001

Abbreviations used: see Fig. 4. Analysis limited to those with no respective complication at DCCT baseline excluding 10 subjects with AER >40 mg/24 h and the secondary cohort for sustained ≥3 MA
^aUnadjusted models are from univariate logistic regression with LW-1 or DCCT mean HbA1c up to the biopsy time as the risk factor
^bAdjusted models are from logistic regressions with both LW-1 and the DCCT mean HbA1c up to the biopsy time as risk factors
^cOdds ratio is calculated based on one SD increase in LW-1; i.e., 170
^dEntropy R2 is calculated as the ratio of Chi square from a likelihood ratio test and −2*log transformation of the likelihood from the null model with intercept only
^eP value is from the Likelihood Ratio Test (LRT) between the respective reduced vs. full model

ISSN: 1475-2840