256-OR: Hypoglycemia Effect on Relationship of Renal and/or Retinal Disease with CVD and Vice Versa

Baseline microvascular complications have been reported to affect the relationship of severe hypoglycemia (SH: needing assistance) with cardiovascular disease (CVD) in DCCT/EDIC type 1 diabetes (T1D) cohort followed for ~ 30 years. We investigated if this finding holds true considering any presence/absence of renal/retinal disease (any RR: sustained AER >30 mg/24 h and/or eGFR <60 mL/min/1.73 m2, and/or PDR and/or CSME) and whether it is bidirectional during longer follow up of >35 DCCT/EDIC years. DCCT/EDIC data provided by NIDDK CR were evaluated applying time dependent Cox models. Outcome was composite of CV death, myocardial infarct (MI), silent MI, revascularization, congestive heart failure, and angina. Subset SH (sSH) is SH with seizures and or loss of consciousness. Interaction between RR and SH (sSH) was evaluated and probed. During follow up, of 1441 patients (mean baseline age 27 ± 7 years, 47% women) 975 (660) had a SH (sSH) event, 741 experienced RR, and 222 had a CV event. All unadjusted (u) / adjusted (a) interactions (I) were significant: I(SH, RR) with p<0.05 (u) and p~0.01 (a), and I(sSH, RR) with p<0.01 (u) and <0.001 (a). Probing interactions showed: 1) In absence of SH or sSH, hazard ratio (HR) for RR was nonsignificant (ns): p>0.5, p>0.6, respectively. 2) In presence of SH or sSH, HR for RR was 1.74 [1.19, 2.53] p<0.01, 2.40 [1.46, 3.96] p<0.001, respectively. 3) In absence of RR, HR for SH and sSH was ns: p>0.4, p~0.08, respectively. 4) In presence of RR, HR for SH and sSH was 1.78 [1.22, 2.60] p<0.01 and 1.81 [1.28, 2.56] p<0.001, respectively. Subanalyses (stratification by SH (sSH) or RR absence / presence) were consistent. Our analyses provide additional evidence for multifactorial nature of relationship between SH and CVD. We demonstrate that SH increases strengths of association between any RR and CVD during > 35 years of follow up, and vice versa. This bidirectionality in a young cohort should be realized and considered in T1D management to identify individuals at higher CVD risk.