Background: The aim of this post hoc analysis of a large cohort study was to evaluate the association between night-time surgery and the occurrence of intraoperative adverse events (AEs) and postoperative pulmonary complications (PPCs). Methods: LAS VEGAS (Local Assessment of Ventilatory Management During General Anesthesia for Surgery) was a prospective international 1-week study that enrolled adult patients undergoing surgical procedures with general anaesthesia and mechanical ventilation in 146 hospitals across 29 countries. Surgeries were defined as occurring during 'daytime' when induction of anaesthesia was between 8:00 AM and 7:59 PM, and as 'night-time' when induction was between 8:00 PM and 7:59 AM. Results: Of 9861 included patients, 555 (5.6%) underwent surgery during night-time. The proportion of patients who developed intraoperative AEs was higher during night-time surgery in unmatched (43.6% vs 34.1%; P<0.001) and propensity-matched analyses (43.7% vs 36.8%; P¼0.029). PPCs also occurred more often in patients who underwent night-time surgery (14% vs 10%; P¼0.004) in an unmatched cohort analysis, although not in a propensity-matched analysis (13.8% vs 11.8%; P¼0.39). In a multivariable regression model, including patient characteristics and types of surgery and anaesthesia, night-time surgery was independently associated with a higher incidence of intraoperative AEs (odds ratio: 1.44; 95% confidence interval: 1.09e1.90; P¼0.01), but not with a higher incidence of PPCs (odds ratio: 1.32; 95% confidence interval: 0.89e1.90; P¼0.15). Conclusions: Intraoperative adverse events and postoperative pulmonary complications occurred more often in patients undergoing night-time surgery. Imbalances in patients' clinical characteristics, types of surgery, and intraoperative management at night-time partially explained the higher incidence of postoperative pulmonary complications, but not the higher incidence of adverse events. Clinical trial registration: NCT01601223.

For a given permutation π n in S n , a random permutation graph is formed by including an edge between two vertices i and j if and only if (i − j)(π n (i) − π n (j)) < 0. In this paper, we study various statistics of random permutation graphs. In particular, we prove central limit theorems for the number m-cliques and cycles of size at least m. Other problems of interest are on the number of isolated vertices, the distribution of a given node (the mid-node as a special case) and extremal degree statistics. Besides, we introduce a directed version of random permutation graphs, and provide two distinct paths that provide variations/generalizations of the model discussed in this paper.