Science, Medicine, and the Anesthesiologist

Although the use of intraoperative antibiotic prophylaxis has been established to reduced surgical site infections, the association of additional antibiotic prophylaxis in the postoperative period in children undergoing surgery is not. This multicenter cohort study used 30-day postoperative infection data from the American College of Surgeons’ Pediatric National Surgical Quality Improvement Program linked to antibiotic-use data from medical record review at 93 hospitals in children (younger than 18 yr of age) undergoing nonemergent surgery. The primary outcome was 30-day postoperative rate of incisional or organ space infection adjusted for differences in procedure mix, patient characteristics, and comorbidity profiles and procedure groups. A total of 40,611 patients (47% female; median age, 7 yr) were studied; 42% received postoperative prophylaxis (hospital range, 0 to 71%). Postoperative prophylaxis use ranged 190-fold across hospitals (odds ratio, 0.10 to 19.30) and infection rates ranged 4-fold (odds ratio, 0.55 to 1.90). There were no correlations between postoperative prophylaxis and overall infection rates (r = 0.13; P = 0.20), when stratified by either infection type (incisional, r = 0.08; P = 0.43 and organ space, r = 0.13; P = 0.23) or surgical specialty.

Take home message: In this observational cohort analysis using American College of Surgeons’ Pediatric National Surgical Quality Improvement Program database, postoperative surgical antimicrobial prophylaxis was not correlated with infection rates at the hospital level after adjustment for differences in procedure mix and patient characteristics.

Although extracorporeal cardiopulmonary resuscitation (CPR) is increasingly used in the in-hospital setting, its impact on survival with a favorable neurologic outcome in out-of-hospital cardiac arrest is unclear. This multicenter study of 10 Dutch centers randomized 160 patients with out-of-hospital cardiac arrest to either extracorporeal or conventional CPR (standard advanced cardiac life support). Patients had received bystander CPR with an initial ventricular arrhythmia and did not have a return of spontaneous circulation within 15 min after initiation. The primary outcome was survival with a favorable neurologic outcome, defined as a Cerebral Performance Category score of 1 or 2 (range, 1 to 5, higher scores with more severe disability) at 30 days. Of the 134 subjects analyzed, no difference in the primary outcome was noted between groups (20% in the extracorporeal CPR group vs. 16% in conventional CPR group, odds ratio, 1.4; 95% CI, 0.5 to 3.5; P = 0.52). The number of serious adverse events per patient was not different between groups.

Take home message: In a randomized trial of extracorporeal versus standard advanced cardiac life support for subjects sustaining refractory out-of-hospital cardiac arrest, no differences were noted in survival with a favorable neurologic outcome by 30 days.

Sodium-glucose co-transporter-2 (SGLT-2) inhibitors are antihyperglycemic agents that block resorption of filtered glucose in the proximal tubule. These medications lower the risk of heart failure and progression of kidney disease in patients with type 2 diabetes, actions spurring interest in utility beyond diabetes. This multicenter trial (241 centers in eight countries) randomized 6,609 patients with race-adjusted glomerular filtration rates of 20 to 45 ml · min^–1 · 1.73 m^–2 or 45 to 90 ml · min^–1 · 1.73 m^–2 plus urinary albumin-to-creatinine ratio greater than 200 mg/g to empagliflozin (10 mg daily) or placebo and followed for a median of 2 yr. The primary outcome, a composite of first occurrence of progression of kidney disease or death from cardiovascular causes, was lower in patients treated with empagliflozin compared to placebo (13% vs. 17%; hazard ratio, 0.72; 95% CI, 0.64 to 0.82, P < 0.001). Secondary outcomes, including hospitalization rate or death from any cause, and adverse event rates did not differ between treatment groups.

Take home message: In a randomized multicenter study, empagliflozin (10 mg daily) lessened progression of kidney disease and risk of death from cardiovascular events in patients with chronic kidney disease compared to placebo over a median of 2 yr.

The optimal volume of IV fluid required to treat sepsis-induced hypotension remains controversial with conflicting observational and randomized data. This multicenter unblinded superiority trial (60 U.S. intensive care units) randomized patients within 4 h of meeting criteria for sepsis-induced hypotension refractory to initial treatment with 1 to 3 l of intravenous fluid, to either a restrictive (favoring vasopressors and lower fluid volumes) or a liberal fluid strategy (favoring higher fluid volume prior to vasopressor use) for a 24-h protocol period. The primary outcome was all-cause mortality before discharge home by day 90. There were 1,563 patients enrolled, with 51% received restrictive fluids versus 49% liberal fluids. The restrictive group received less fluid (difference of medians, −2,134 ml; 95% CI, −2,318 to −1,949) with earlier, more frequent, and longer duration of vasopressor therapy. Death from any cause before discharge home by day 90 occurred in 109 patients (14%) in the restrictive fluid group and in 116 patients (15%) in the liberal fluid group (estimated difference, –0.9 percentage points; 95% CI, –4.4 to 2.6; P = 0.61). No differences in serious adverse events were noted between the two groups.

Take home message: In a multicenter randomized trial, use of either a restrictive or a liberal fluid replacement protocol after prior administration of guideline-recommended initial fluid resuscitation did not result in a significant difference in mortality before discharge home by day 90.

Epidemiologic evidence suggests that excessive artificial light may be a risk factor for metabolic disorders, but links between light and metabolism are poorly understood. In addition to light activation of photoreceptors for vision, light also activates photosensitive retinal ganglion cells that innervate multiple brain areas for controlling the pupillary light reflex, circadian rhythms, sleep, mood, and cognitive functions. The present study utilized genetically modified mice and neuronal retrograde labeling to identify a neural pathway linking sunlight, white LED, and blue LED lights to activation of intrinsically photosensitive retinal ganglion cells. These ganglion cells modulate the hypothalamic supraoptic nucleus that excites the paraventricular nucleus projecting to the solitary tract nucleus, which in turns projects to the rostral raphe pallidus modulating sympathetic outflow. Sympathetic innervation of brown adipose tissue via β3-adrenergic receptors block thermogenesis resulting in decreased glucose tolerance. Human volunteers demonstrated light-induced decreases in glucose metabolism at 19°C but not 29°C independent of circadian rhythm consistent with this functional neural circuit regulating brown adipose tissue in humans. Glucose tolerance during nighttime was worse than during daytime, and nighttime light further impaired glucose tolerance, suggesting that artificial light at nighttime may be an environmental factor that contributes to glucose intolerance in humans.

Take home message: Natural and artificial light lessened glucose tolerance mediated by brown adipose tissue in both mice and humans. A neuronal-to-metabolic pathway was identified from photosensitive retinal ganglion cells through several brain nuclei ultimately modulating sympathetic outflow to brown adipose tissue. This fundamental link of environmental light to glucose metabolism may partially explain the epidemiologic link of artificial light to metabolic disarray in diabetes and obesity.

Electroencephalogram (EEG) monitoring is useful in assessing recovery from birth asphyxia as indicated by the onset of continuous background cortical activity and sleep-wake cycling. However, EEG interpretation requires considerable specialist expertise, which is not widely available. Thus, machine learning methods may be useful in interpreting the raw EEG. A training dataset of 2,561 h of EEG data were collected from 39 infants (tertiary Finnish neonatal intensive care unit) recovering from birth asphyxia or stroke. An unsupervised classifier was applied to these data to recognize background activity. This was then combined with a classifier of sleep-wake cycling to produce a novel index of cortical function, the Brain State of the Newborn. For EEG background activity, the index had a 92% agreement with human scoring consensus (95% CI, 91 to 96; range, 85 to 100% for individual infants), similar to the between-expert rating accuracy. When validated on a separate, publicly available database of 105 h EEG data (from 31 newborn infants with hypoxic-ischemic encephalopathy), accuracy was 88% (range, 61% to 100% for individual infants). Temporal evolution of the Brain State of the Newborn showed early divergence between infants who recovered well and those with abnormal outcomes.

Take home message: A novel machine learning–derived index, the Brain State of the Newborn, is capable of deployment at the bedside in patient monitors and shows similar accuracy to human EEG assessment with regard to onset of continuous background cortical activity, sleep-wake cycling, and clinical trajectories.

The study estimated the risks of myocardial infarction (MI), ischemic stroke, and all-cause and cardiovascular-related mortality after Roux-en-Y gastric bypass for weight loss compared to obese patients without surgery and matched nonobese controls. Obese patients 20 to 65 yr old in the nationwide Swedish Patient Registry from 2001 to 2013 were divided into those who had surgery within 2 yr of a diagnosis of obesity (n = 28,204) and those without (n = 40,827) and matched with nonobese population controls. Participants were followed for the primary outcome or end of follow-up. Gastric bypass conferred a lower risk of MI (hazard ratio, 0.44 [95% CI, 0.28 to 0.63]), although risk of ischemic stroke (hazard ratio, 0.79 [95% CI, 0.54 to 1.14]) was similar to nonoperated obese patients. Cardiovascular-related (hazard ratio, 0.47 [95% CI, 0.35 to 0.65]) and all-cause mortality (hazard ratio, 0.66 [95% CI, 0.54 to 0.81]) was lower in the gastric bypass patients versus obese patients without surgery during the first 3 yr of follow-up, but not later. In comparison with nonobese controls, gastric bypass patients had higher risks of ischemic stroke (hazard ratio, 1.57 [95% CI, 1.08 to 2.29]), cardiovascular-related mortality (hazard ratio, 1.82 [95% CI, 1.29 to 2.60]), and all-cause mortality (hazard ratio, 1.42 [95% CI, 1.16 to 1.74]), but not of MI (hazard ratio, 1.02 [95% CI, 0.72 to 1.46]).

Take home message: Roux-en-Y gastric bypass for weight loss is associated with a lesser risk of MI compared to that of nonoperated obese individuals, but not with risk of ischemic stroke.

The human opioid system consists of four receptors—µ opioid, δ opioid, κ opioid, and the nociception receptor—as well as a set of respected endogenous opioid peptides that interact variably with them. Despite their analgesic effectiveness, the clinical use of opioids is limited by their serious side effects, including addiction and respiratory depression. These side effects of opioids are responsible in part for the ongoing opioid crisis in the United States. An alternative strategy is to develop novel pain therapeutics based on endogenous opioid peptides, which may produce fewer side effects. Although the binding of these peptides to opioid receptors has been extensively studied, how they recognize and activate their receptors is not well characterized. This study systematically characterized the binding of several peptides to the four human opioid receptors and further determined the Gi complex structures associated with these opioid receptors. The structures demonstrated a universal activation mechanism for all four opioid receptors revealing a conserved pocket for the YGGF motif of these peptides and unique structural features of the extracellular loop 2/3 for the selectivity of the opioid peptides.

Take home message: This study may provide a structural framework for rational design of safer opioid drugs for pain relief.

Common knowledge holds that the central nervous system is lined by three meninges, namely dura, arachnoid, and pia mater. Using advanced in vivo neuroimaging and molecular biology, a previously unknown fourth meningeal layer, termed the subarachnoidal lymphatic-like membrane, has been reported in this study. This anatomic structure divides the subarachnoid space of the brain into two compartments and helps control the flow of cerebrospinal fluid, which is transporting and removing waste from the brain. The subarachnoidal lymphatic-like membrane (14.2 ± 0.5 µm, in mice) is thinner than dura (21.8 ± 1.3 µm) but acts as a tight barrier (at least 3 kDa) that limits the exchange of most peptides and proteins between the upper and lower subarachnoid space. It also expresses podoplanin, a specific glycoprotein marker of the mesothelium lining all body cavities, and thus may represent the brain’s mesothelium reducing friction between brain and skull during movements. As with all mesothelia, it harbors immune cells important for immunosurveillance, and acts as an immune barrier, as the central nervous system maintains its own native population of immune cells, preventing outside immune cells from entering the brain.

Take home message: Discovery of the subarachnoidal lymphatic-like membrane opens the door for research investigating its mechanistic role in the context of brain injury, neurodegenerative disorders, and aging. As it also affects the delivery of drugs to the brain, its function needs to be considered when developing new biologic therapies.

Spine pain is the leading cause of disability worldwide and the leading cause of healthcare spending in the United States, with no long-term reliable treatments. In a multicenter, three-group, pragmatic, open-label, cluster-randomized clinical trial, 2,971 patients with acute or subacute back or neck pain were randomized into three groups: (1) a biopsychosocially focused “identify, coordinate, and enhance care model” emphasizing physical therapy and healthcare coaches to address maladaptive coping behaviors (e.g., catastrophization), (2) individualized postural training, or (3) usual care for 6 to 8 weeks. The co–primary outcome measures were pain-related disability (measured by Oswestry Disability Index) up to 3 months and spine-related healthcare costs through 1 yr. Function was greater at 3 months (−5.8 [95% CI, −7.7 to −3.9], P < 0.001 for “identify, coordinate, and enhance care model”; −4.3 [95% CI, −5.9 to −2.6], P < 0.001 for individualized postural training) for the treatment groups, but also greater spending in the individualized postural training group ($1,448, $2,528, and $1,587 in the “identify, coordinate, and enhance care model,” individualized postural training, and usual care groups, respectively; P < 0.001 for individualized postural training vs. usual care group). One limitation in the study was the failure to perform a subgroup analysis in individuals with radicular pain, or those with back pain, neck pain, or a combination of the two.

Take home message: In patients with acute or subacute spine pain, a multidisciplinary biopsychosocial intervention and individualized postural training resulted in less pain-related disability at 3 months, but greater spine-related healthcare costs.

Hospitalized patients are monitored for signs of clinical deterioration (modified early warning scores), allowing earlier intervention by rapid response teams. The addition of routine postoperative visits by anesthesiologists to postsurgical patients as an adjunct to this approach has not been studied. This prospective, multicenter, stepped-wedge cluster-randomized interventional study of nine Netherlands hospitals evaluated the addition of standardized postoperative anesthesia visits on postoperative days 1 and 3 to routine use of modified early warning scores in 5,473 patients undergoing elective noncardiac surgery on 30-day all-cause mortality (primary outcome). Secondary outcomes included the incidence of postoperative complications, length of hospital stay, and intensive care unit admission. A total of 2,490 patients were randomized to control and 2,700 to the intervention group. There was no difference in the primary outcome, 0.56% control versus 0.44% intervention group (odds ratio, 0.74; 95% CI, 0.34 to 1.62). The overall incidence of postoperative complications was not different, although renal complications were higher in the control group (1.7% vs. 1.0%, P = 0.014). The median length of hospital stay was not different. Treatment recommendations were made in the intervention group in 16% and 11% of patients on days 1 and 3, respectively, of which 67% and 69% were followed up.

Take home message: The combination of use of modified early warning scores and postoperative anesthesia visits on postoperative days 1 and 3 did not affect 30-day mortality over use of modified early warning scores alone.

Although the U.S. Department of Veterans Affairs (VA) Veterans Choice Program has expanded health care access for eligible patients living 40 miles or more from their closest VA hospital (Choice Program), the effect on access to surgery and those postoperative outcomes has not been well studied. This retrospective study using VA administrative records evaluated its impact on surgical utilization and related outcomes using a nonrandomized controlled regression discontinuity design to reduce the impact of unmeasured confounders. Assessed variables included postoperative emergency department visits, inpatient readmissions, and mortality at 30 and 90 days. A total of 615,473 unique surgical procedures in 498,427 patients (mean ± SD age, 63 ± 13 yr; 90% male) were analyzed. Of these, 15% of procedures were performed outside the VA (VA-paid) and were more common in females than males (13% vs. 9%), white race (74% vs. 73%), and patients younger than age 65 yr (49% vs. 46%) with a lower comorbidity burden (1.8 ± 2.2 vs. 2.6 ± 2.7). No difference in postoperative mortality, readmissions, or emergency department visits were observed between services provided directly by the VA versus community based.

Take home message: In this retrospective, administrative analysis, expanded access to health care for U.S. veterans using community resources for providing surgical care had no measurable association with surgical outcomes.