Validation of a Natural Language Processing Algorithm using National Reporting Data to Improve Identification of Anesthesia-related ADVerse evENTs: The "ADVENTURE" Study.

BACKGROUND: Reporting and analysis of adverse events (AE) is associated with improved healthcare learning, quality outcomes, and patient safety. Manual text analysis is time-consuming, costly, and prone to human errors. We aimed to demonstrate the feasibility of machine learning and natural language processing (NLP) approaches for early predictions of adverse events and provide input to direct quality improvement and patient safety initiatives. METHODS: We used machine learning to analyze 9,559 continuously reported AE by clinicians and healthcare systems to the French National Health accreditor (HAS) between January 1, 2009, and December 31, 2020, for a total of 135,000 unique de-identified AE reports. We validated the labeling and determined the associations between different root causes and patient consequences. The model was validated by independent expert anesthesiologists. RESULTS: The machine learning and Artificial Intelligence (AI) model trained on 9,559 AE datasets accurately categorized 8800 (88%) of reported AE. The three most frequent AE types were "difficult orotracheal intubation" (16.9% of AE reports), "medication error" (10.5%), and "post-induction hypotension" (6.9%). The accuracy of the AI model reached 70.9% sensitivity, 96.6% specificity for "difficult intubation", 43.2% sensitivity, and 98.9% specificity for "medication error." CONCLUSIONS: This unsupervised method provides an accurate, automated, AI-supported search algorithm that ranks and helps to understand complex patterns of risky patient situations and has greater speed, precision, and clarity when compared to manual human data extraction. Machine learning (ML) and natural language processing models can effectively be used to process natural language AE reports and augment expert clinician input. This model can support clinical applications and methodological standards of implementations and be used to better inform and enhance decision-making for improved risk management and patient safety. TRIAL REGISTRATION: The study was approved by the ethics committee of the French Society of Anesthesiology (IRB 00010254-2020-20) and the CNIL (CNIL: 118 58 95) and the study was registered with ClinicalTrials.gov (NCT: NCT05185479).

Epidemiology of perioperative anaphylaxis in France in 2017-2018: the 11th GERAP survey.

BACKGROUND: Perioperative anaphylaxis is rare but is associated with significant morbidity. This complication has been well described in France by the GERAP (Groupe d'Etude des Réactions Anaphylactiques Périopératoires), a network focused on its study. The epidemiology of perioperative anaphylaxis is evolving, influenced by environmental factors and clinical practice. The aim of this study was to update the epidemiology of perioperative anaphylaxis in France. METHODS: This multicentre retrospective study was performed in 26 allergy clinics of the GERAP network in 2017-8. RESULTS: There were 765 patients with perioperative anaphylaxis included. Most cases were severe, with 428 (56%) reactions graded as 3 or 4 according to the Ring and Messmer classification. Skin test results were available for 676 patients, with a culprit agent identified in 471 cases (70%). Neuromuscular blocking agents were the main cause of perioperative anaphylaxis (n=281; 60%), followed by antibiotics (n=118; 25%) and patent blue dye (n=11; 2%). Cefazolin was the main antibiotic responsible for perioperative anaphylaxis (52% of antibiotic-related reactions). Suxamethonium and rocuronium were the main neuromuscular blocking agents responsible for perioperative anaphylaxis with 7.1 (6.1-8.4) and 5.6 (4.2-7.4) reactions per 100,000 vials sold, respectively, whereas cefazolin-related cases were estimated at 0.7 (0.5-0.9) reactions per 100,000 vials sold. CONCLUSIONS: Our results confirm that most commonly identified triggering agents remain neuromuscular blocking agents. Reactions to antibiotics, particularly cefazolin, are becoming increasingly frequent. The origin of sensitisation to cefazolin is unknown, as no cross-sensitisation has been described, and it should be the subject of further study. Perioperative anaphylaxis should be followed over the years and understood given the changing triggers. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov (NCT04654923).

Hypersensitivity transfusion reactions to platelet concentrate: a retrospective analysis of the French hemovigilance network.

BACKGROUND: Among labile blood products, platelet concentrates (PCs) are the leading cause of hypersensitivity transfusion reactions (HTRs). These reactions often lead to interruption of PC transfusion and can result in a prolonged transfusion process leading to significant morbidity and use of premedication and close monitoring for patients with a history of allergic transfusion reactions. The French hemovigilance database is one of the largest standardized databases providing information on HTRs following administration of labile blood products. In this study, we analyzed this database to assess the relative risk of HTR for each type of PC. STUDY DESIGN AND METHODS: HTRs following PC transfusion were retrospectively extracted from the e-Fit Hemovigilance database of the French National Agency for Medicines and Health Products Safety (ANSM). Frequencies were calculated using the number of specific PCs transfused. RESULTS: Between 2008 and 2014, the overall estimated incidence of HTRs following PC administration was calculated at 232 HTRs per 100,000 PCs transfused. The rate of HTRs was significantly higher with apheresis PC (337/100,000) than with buffy-coat PC (94/100,000). Platelets in additive solutions (PAS) were associated with a significantly lower frequency of HTRs when compared with PCs in native plasma. Amotosalen/UVA- PCs (APCs and BCPCs) which are always in PAS in France, exhibited the lowest frequency of HTRs when compared with their corresponding PCs in native plasma or in PAS (p < 10-7 in all comparisons). CONCLUSION: Our results showed that the type of PC and its processing may have an impact on the risk of HTR.

Treatment with a platelet-activating factor receptor antagonist improves hemodynamics and reduces epinephrine requirements, in a lethal rodent model of anaphylactic shock.

BACKGROUND: In some cases, anaphylactic shock (AS) is still lethal, despite rapid use of epinephrine. High doses of epinephrine are associated with severe complications. Platelet-activating factor (PAF) is secreted in massive amounts during AS, and a high plasma level is correlated with increased AS severity. OBJECTIVE: To assess the effect of ABT-491, a PAF-receptor antagonist and possible adjunct treatment, alone or in combination with epinephrine during AS. METHODS: AS was induced by intravenous injection of 1 mg ovalbumin into ovalbumin-sensitized rats. Rats were then randomly assigned to 5 groups (n = 10 per group): SHAM (vehicle only), SHOCK (no treatment), ABT (ABT-491 1 mg/kg), EPI (epinephrine 5 µg as a bolus then 10 µg kg-1  min-1 by continuous infusion, followed by a reducing protocol) and EPI-ABT (both treatments). RESULTS: Ovalbumin injection resulted in a severe decrease in mean arterial pressure, left ventricular inotropy (max dP/dt) and left ventricular shortening fraction (LVSF). All rats from the ABT group survived until the end of the experiment. ABT-491 prevented the LVSF decrease observed in the SHOCK group (at T15: ABT 50% ± 11% vs SHOCK 36% ± 9%, P = .01), significantly reduced the dose of epinephrine needed to treat anaphylactic shock (EPI-ABT 314 ± 67 µg/kg vs EPI 475 ± 69 µg/kg, P < .001) and reduced the time to restore basal MAP (ABT 23 ± 7 minutes vs EPI-ABT 13 ± 5 minutes, P < .01). CONCLUSIONS AND CLINICAL RELEVANCE: AS was characterized by early cardiac dysfunction in our model. Treatment with ABT-491 allowed survival until the end of the experiment and reduced cardiac dysfunction. Use of the PAF-R antagonist had a synergistic effect with epinephrine and allowed a significant reduction in epinephrine consumption. Use of PAF-R antagonists during AS could reduce epinephrine-related complications and improve the treatment of epinephrine refractory cases.

Paired acute-baseline serum tryptase levels in perioperative anaphylaxis: An observational study.

BACKGROUND: Anaphylaxis is recognized mainly through clinical criteria, which may lack specificity or relevance in the perioperative setting. The transient increase in serum tryptase has been proposed since 1989 as a diagnostic tool. Sampling for well-defined acute and baseline determinations has been recommended. We assessed the performance of four proposed algorithms with tightly controlled time frames for tryptase sampling, their robustness with inadequate sampling times, and the possible use of mature tryptase determination. METHODS: A retrospective study was performed on 102 adult patients from the Aix-Marseille University Hospitals who had experienced a perioperative hypersensitivity reaction clinically suggesting anaphylaxis. EAACI and ICON criteria were used to diagnose anaphylaxis. Mature and total serum tryptase levels were measured. RESULTS: Based on EAACI guidelines, clinical diagnostic criteria for anaphylaxis were found in 76 patients and lacking in 26. The most effective algorithm was the international consensus recommendation of 2012 that acute total tryptase levels should be greater than ([1.2×baseline tryptase] + 2] µg/L to be considered a clinically significant rise. In our cohort, this algorithm achieved 94% positive predictive value (PPV), 53% negative predictive value (NPV), 75% sensitivity, 86% specificity, and a Youden's index value of 0.61. A detectable acute mature tryptase level showed lower sensitivity, particularly in patients with acute total tryptase levels lower than 16 µg/L. Acute tryptase levels varied as a function of the clinical severity of anaphylaxis. CONCLUSION: Total tryptase levels in serum discriminated between nonanaphylactic and anaphylactic events in a perioperative setting when acute and baseline levels were collected and analyzed by the consensus algorithm.

Hypersensitivity reactions in the anesthesia setting/allergic reactions to anesthetics.

PURPOSE OF REVIEW: Immediate hypersensitivity reactions are an important cause for mortality and morbidity in anesthesia. The present review considers reports covering epidemiology, diagnosis, and treatment of these reactions. RECENT FINDINGS: Immediate hypersensitivity reactions are largely under-reported, adult women being at significantly higher risk than men. The role of sex hormones associated with increased risk in adult women has been demonstrated. Neuromuscular blocking agents (NMBAs) remain the most frequently incriminated drugs. Reactions involving antibiotics, dyes, or nonsteroidal anti-inflammatory agents are reported with increasing frequency, in parallel with changing trends in clinical practice. A recent hypothesis concerning a link between pholcodine exposure and allergic reactions to NMBAs is under investigation. Detailed guidelines concerning skin testing have been provided. The use of several inotropes or vasopressor such as vasopressin is proposed in case of reactions refractory to epinephrine and volume expansion. The use of cyclodextrin to mitigate severe allergic reactions to rocuronium, by specific drug encapsulation, has been proposed. SUMMARY: Hypersensitivity reactions remain a major cause of concern in the perioperative setting. Although largely under-reported, their incidence is higher than previously reported. NMBAs remain the most frequently incriminated drug, followed by latex and antibiotics. The number of reactions involving new allergens like vital dyes or nonsteroidal anti-inflammatory drugs is rapidly increasing. The mechanism of sensitization to NMBAs could be influenced by as yet unidentified environmental factors. The possible role of pholcodine is under investigation. Several guidelines concerning the diagnosis and management of immediate hypersensitivity reactions in anesthesia are now available.

Refractory anaphylactic cardiac arrest after succinylcholine administration.

Refractory shock from anaphylaxis can occur after induction of general anesthesia. We report two cases of fatal cardiac arrest with increased blood tryptase and immunoglobulin E values after succinylcholine administration. Tryptase and immunoglobulin E assays may help to identify anaphylactic reactions when cardiac arrest occurs at induction of anesthesia.

Exercise release of cardiac natriuretic peptides is markedly enhanced when patients with coronary artery disease are treated medically by beta-blockers.

OBJECTIVES: This study sought to identify determinants of the exercise rise in plasma levels of cardiac natriuretic peptides (NPs) in patients with coronary artery disease (CAD). BACKGROUND: During stress, there is a variable rise in the plasma level of NPs, but this rise frequently reaches levels that are known to lower the cardiac load and that thus might be beneficial to CAD patients. METHODS: Plasma venous concentrations of atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) were determined at rest and peak exercise in 104 patients with chronic CAD who were referred to exercise thallium-201 ((201)Tl) single-photon emission computed tomography (SPECT) and radionuclide angiography. RESULTS: The extent of scarred myocardium by (201)Tl-SPECT and patient age were the best independent predictors of NP concentrations at rest, but also of increases in NP concentration during exercise (all p < 0.001). Moreover, beta-blocking treatment was an additional and strong independent predictor of the increase in NP concentrations at exercise (p < 0.001 for ANP; p = 0.001 for BNP). On average, exercise increases in NP concentrations were more than twice as high in patients with (n = 55) than in those without (n = 49) beta-blocker treatment (ANP: +49 +/- 63 vs. +22 +/- 25 ng/l, p = 0.01; BNP: +24 +/- 5 vs. +11 +/- 15 ng/l, p = 0.04), whereas NP concentrations at rest were equivalent in the two groups (ANP: 34 +/- 34 vs. 30 +/- 33 ng/l, p = NS; BNP: 85 +/- 152 vs. 57 +/- 101 ng/l, p = NS). CONCLUSIONS: Patients with chronic CAD exhibit much higher exercise releases of ANP and BNP when they are treated with beta-blockers. This enhanced secretion of potent vasodilating and natriuretic agents constitutes an original therapeutic mechanism for further protecting diseased hearts against stress.