Results of trials and registry studies have shown lower long-term mortality after coronary-artery bypass grafting (CABG) than after percutaneous coronary intervention (PCI) among patients with multivessel disease. These previous analyses did not evaluate PCI with second-generation drug-eluting stents.

Chatterjee et al. (2011) established the consistency of the maximum likelihood estimator in the β-model for undirected random graphs when the number of vertices goes to infinity. By approximating the inverse of the Fisher information matrix, we prove asymptotic normality of the maximum likelihood estimator under mild conditions. Simulation studies and a data example illustrate the theoretical results.

Background—: National and international guidelines recommend fasting lipid panel measurement for risk stratification of patients for prevention of cardiovascular events. However, the prognostic value of fasting versus nonfasting low-density lipoprotein cholesterol (LDL-C) is uncertain. Methods and Results—: Patients enrolled in the National Health and Nutrition Examination Survey III (NHANES-III), a nationally representative cross-sectional survey performed from 1988 to 1994, were stratified on the basis of fasting status (≥8 or <8 hours) and followed for a mean of 14.0 (±0.22) years. Propensity score matching was used to assemble fasting and nonfasting cohorts with similar baseline characteristics. The risk of outcomes as a function of LDL-C and fasting status was assessed with the use of receiver operating characteristic curves and bootstrapping methods. The interaction between fasting status and LDL-C was assessed with Cox proportional hazards modeling. Primary outcome was all-cause mortality. Secondary outcome was cardiovascular mortality. One-to-one matching based on propensity score yielded 4299 pairs of fasting and nonfasting individuals. For the primary outcome, fasting LDL-C yielded prognostic value similar to that for nonfasting LDL-C (C statistic=0.59 [95% confidence interval, 0.57–0.61] versus 0.58 [95% confidence interval, 0.56–0.60]; P =0.73), and LDL-C by fasting status interaction term in the Cox proportional hazards model was not significant ( P _interaction =0.11). Similar results were seen for the secondary outcome (fasting versus nonfasting C statistic=0.62 [95% confidence interval, 0.60–0.66] versus 0.62 [95% confidence interval, 0.60–0.66]; P =0.96; P _interaction =0.34). Conclusions—: Nonfasting LDL-C has prognostic value similar to that of fasting LDL-C. National and international agencies should consider reevaluating the recommendation that patients fast before obtaining a lipid panel.