Paravertebral versus Pectoralis-II (Interpectoral and Pectoserratus) Nerve Blocks for Postoperative Analgesia after Non-Mastectomy Breast Surgery: A Randomized, Controlled, Observer-Masked Noninferiority Trial.

BACKGROUND: Pectoralis-II and paravertebral nerve blocks are both used to treat pain following breast surgery. Most previous studies involving mastectomy identified little difference of significance between the two approaches. Whether this is also accurate for non-mastectomy procedures remains unknown. METHODS: Participants undergoing uni- or bilateral non-mastectomy breast surgery anticipated to have at least moderate postoperative pain were randomized to a pectoralis-II or paravertebral block (90 mg ropivacaine/side for both). Surgeons and recovery room staff were masked to treatment group assignment, and participants were not informed of their treatment group. Injectate for pectoralis-II blocks was ropivacaine 0.3% (30 mL) per side. Injectate for paravertebral blocks was ropivacaine 0.5% (9 mL in each of 2 levels) per side. We hypothesized that pectoralis-II blocks would have noninferior (1) analgesia [Numeric Rating Scale] and (2) cumulative opioid consumption within the operating and recovery rooms combined (dual primary outcomes). The study was adequately powered with n=100, but the target enrollment was raised to n=150 to account for higher-than-anticipated variability. RESULTS: The trial was ended prematurely with 119 (79%) of the original target of 150 participants enrolled due to (masked) surgeon preference. Within the recovery room, pain scores were higher in participants with pectoralis-II (n=60) than paravertebral blocks (n=59): median [IQR] 3.3 [2.3, 4.8] vs 1.3 [0, 3.6]; 95% CI: 0.5 to 2.6; P < 0.001. Similarly, intravenous morphine equivalents were higher in the pectoralis-II group: 17.5 [12.5, 21.9] vs 10.0 mg [10, 20]; 95% CI: 0.1 to 7.5; P = 0.004. No block-related adverse events were identified in either group. CONCLUSIONS: Following non-mastectomy breast surgery, 2-level paravertebral blocks provided superior analgesia and opioid sparing compared with pectoralis-II blocks. This is a contrary finding to the majority of studies in patients having mastectomy in which little significant difference was identified between the two types of blocks.

A Narrative Review of Ultrasound-Guided and Landmark-based Percutaneous Cryoneurolysis for the Management of Acute and Chronic Pain.

PURPOSE OF REVIEW: Cryoneurolysis refers to the process of reversibly ablating peripheral nerves with extremely cold temperatures to provide analgesia for weeks to months. With ultrasound-guidance or landmark-based techniques, it is an effective modality for managing both acute and chronic pain. In this review, we summarize the reported literature behind its potential applications and efficacy. RECENT FINDINGS: Here, we summarize several studies (from case reports to clinical trials) describing the use of ultrasound-guided and landmark-based cryoneurolysis for acute and chronic pain. Acute pain indications included pain related to knee arthroplasty, limb amputations, mastectomies, shoulder surgery, rib fractures, and burn. Chronic pain indications included chronic knee pain (due to osteoarthritis), shoulder pain, painful neuropathies, postmastectomy pain syndrome, phantom limb pain, facial pain/headaches, foot/ankle pain, inguinal pain, and sacroiliac joint pain. For both acute and chronic pain indications, more high quality randomized controlled clinical trials are needed to definitively assess the efficacy of cryoneurolysis versus other standard therapies for a multitude of pain conditions.

Enhancing Laryngoscopy Mastery: The Impact of Autonomous Practice with Feedback-Providing Simulators.

Mastery learning with fixed end points and variable training time leads to more consistent expertise but is difficult to implement. Here we piloted mastery learning of laryngoscopy with independent practice. 35 learners participated in independent mastery learning on a manikin that provides automated performance feedback. A pre- and postpractice assessment of intubation skills was completed. After an average of 21 minutes of open practice, the percentage of subjects that met mastery criteria improved from 24% to 89% (P < .05). Independent intubation practice with manikin feedback facilitated mastery learning, enhanced procedural education, and may impact clinical care.

Association of amisulpride and recovery room length of stay among patients with postoperative nausea and vomiting following outpatient surgery.

STUDY OBJECTIVE: Postoperative nausea and vomiting (PONV) is a common sequela of surgery in patients undergoing general anesthesia. Amisulpride has shown promise in its ability to treat PONV. The objective of this study was to determine if amisulpride is associated with significant changes in PACU efficiency within a fast-paced ambulatory surgery center. METHODS: This was a retrospective cohort study of 816 patients at a single ambulatory surgery center who experienced PONV between 2018 and 2023. The two cohorts analyzed were patients who did or did not have amisulpride among their anti-emetic regimens in the PACU during two distinct time periods (before and after amisulpride was introduced). The primary outcome of the study was PACU length of stay. Both unmatched analysis and a linear multivariable mixed-effects model fit by restricted maximum likelihood (random effect being surgical procedure) were used to analyze the association between amisulpride and PACU length of stay. We performed segmented regression to account for cohorts occurring during two time periods. RESULTS: Unmatched univariate analysis revealed no significant difference in PACU length of stay (minutes) between the amisulpride and no amisulpride cohorts (115 min vs 119 min, respectively; P = 0.07). However, when addressing confounders by means of the mixed-effects multivariable segmented regression, the amisulpride cohort was associated with a statistically significant reduction in PACU length of stay by 26.1 min (P < 0.001). CONCLUSIONS: This study demonstrated that amisulpride was associated with a significant decrease in PACU length of stay among patients with PONV in a single outpatient surgery center. The downstream cost-savings and operational efficiency gained from this drug's implementation may serve as a useful lens through which this drug's widespread implementation may further be rationalized.

Constructing synthetic datasets with generative artificial intelligence to train large language models to classify acute renal failure from clinical notes.

OBJECTIVES: To compare performances of a classifier that leverages language models when trained on synthetic versus authentic clinical notes. MATERIALS AND METHODS: A classifier using language models was developed to identify acute renal failure. Four types of training data were compared: (1) notes from MIMIC-III; and (2, 3, and 4) synthetic notes generated by ChatGPT of varied text lengths of 15 (GPT-15 sentences), 30 (GPT-30 sentences), and 45 (GPT-45 sentences) sentences, respectively. The area under the receiver operating characteristics curve (AUC) was calculated from a test set from MIMIC-III. RESULTS: With RoBERTa, the AUCs were 0.84, 0.80, 0.84, and 0.76 for the MIMIC-III, GPT-15, GPT-30- and GPT-45 sentences training sets, respectively. DISCUSSION: Training language models to detect acute renal failure from clinical notes resulted in similar performances when using synthetic versus authentic training data. CONCLUSION: The use of training data derived from protected health information may not be needed.

An Interdisciplinary, Comprehensive Recovery Pathway Improves Microsurgical Breast Reconstruction Delivery.

INTRODUCTION: Free-flap (autologous) breast reconstruction demonstrates superiority over alloplastic approaches but is offered infrequently. Enhanced recovery protocols can address postoperative challenges, but most literature is limited to inpatient interventions and outcomes. This study describes an adoptable, longitudinally comprehensive and multidisciplinary recovery pathway for autologous reconstruction which adds to the current guidelines. The authors aimed to allow perioperative outcomes comparable to alloplastic reconstructions. METHODS: All autologous Comprehensive Recovery Pathway (CRP) subjects from a single surgeon were retrospectively included. A comparator group of equal size was randomly selected from institutional subpectoral and dual-plane tissue expander patients having Enhanced Recovery After Surgery guideline-directed care. All subjects in both cohorts received preoperative paravertebral regional blocks. Operative detail, inpatient recovery, longitudinal morphine equivalents (MEs) required, and complications were compared. RESULTS: Each cohort included 71 cases (99 breasts). Despite longer operations, intraoperative MEs were fewer in autologous cases ( P = 0.02). Morphine equivalents during inpatient stay were similar between cohorts, with both being discharged on median day 2. Multivariate regression demonstrated a 0.8-day increased stay for autologous subjects with additional contribution from bilateral cases, body mass index, and age ( P < 0.05). Autologous subjects were regularly discharged postoperative day 1 (17%) and postoperative day 2 (39%), with trend toward earlier discharge ( P < 0.01). Outpatient MEs were significantly fewer in autologous subjects, corresponding to a 30- to 150-mg oxycodone difference ( P < 0.01). Major complication occurred in 12.7% of autologous and 22.5% of alloplastic subjects ( P = 0.11). Flap loss occurred in 1 autologous subject versus 11 alloplastic failures ( P < 0.01). CONCLUSIONS: This study details partnership between the plastic surgery service, regional and acute pain anesthesia services, and dedicated nursing with longitudinal optimizations allowing perioperative outcomes improved over current literature. Patients in the CRP used fewer opioids from operation through follow-up with comparable length of stay and significantly fewer reconstructive failures than alloplastic subjects. The pathway may be quickly adopted into academic practice patterns and mitigates traditional barriers, allowing extension of autologous reconstruction offerings.

Robot-assisted laparoscopic nephrectomy: early outcome measures with the implementation of multimodal analgesia and intrathecal morphine via the acute pain service.

PURPOSE: The objective of this study was to perform a retrospective cohort analysis, in which we measured the association of an acute pain service (APS)-driven multimodal analgesia protocol that included preoperative intrathecal morphine (ITM) compared to historic controls (i.e., surgeon-driven analgesia protocol without ITM) with postoperative opioid use. METHODS: This was a retrospective cohort study in which the primary objective was to determine whether there was a decrease in median 24-h opioid consumption (intravenous morphine equivalents [MEQ]) among robotic nephrectomy patients whose pain was managed by the surgical team prior to the APS, versus pain managed by APS. Secondary outcomes included opioid consumption during the 24-48 h and 48-72 h period and hospital length of stay. To create matched cohorts, we performed 1:1 (APS:non-APS) propensity score matching. Due to the cohorts occurring at the different time periods, we performed a segmented regression analysis of an interrupted time series. RESULTS: There were 76 patients in the propensity-matched cohorts, in which 38 (50.0%) were in the APS cohort. The median difference in 24-h opioid consumption in the pre-APS versus APS cohort was 23.0 mg [95% CI 15.0, 31.0] (p < 0.0001), in favor of APS. There were no differences in the secondary outcomes. On segmented regression, there was a statistically significant drop in 24-h opioid consumption in the APS cohort versus pre-APS cohort (p = 0.005). CONCLUSIONS: The implementation of an APS-driven multimodal analgesia protocol with ITM demonstrated a beneficial association with postoperative 24-h opioid consumption following robot-assisted nephrectomy.

Smallest Clinically Meaningful Improvement in Amputation-Related Pain and Brief Pain Inventory Scores as Defined by Patient Reports of Global Improvement After Cryoneurolysis: a Retrospective Analysis of a Randomized, Controlled Clinical Trial.

BACKGROUND: The smallest meaningful improvement in pain scores (minimal clinically important difference [MCID]) after an analgesic intervention is essential information when both interpreting published data and designing a clinical trial. However, limited information is available for patients with chronic pain conditions, and what is published is derived from studies involving pharmacologic and psychological interventions. We here calculate these values based on data collected from 144 participants of a previously published multicenter clinical trial investigating the effects of a single treatment with percutaneous cryoneurolysis. METHODS: In the original trial, we enrolled patients with a lower-limb amputation and established phantom pain. Each received a single-injection femoral and sciatic nerve block with lidocaine and was subsequently randomized to receive either ultrasound-guided percutaneous cryoneurolysis or sham treatment at these same locations. Investigators, participants, and clinical staff were masked to treatment group assignment with the exception of the treating physician performing the cryoneurolysis, who had no subsequent participant interaction. At both baseline and 4 months (primary end point), participants rated their phantom limb pain based on a numeric rating scale (NRS) and their interference of pain on physical and emotional functioning as measured with the Brief Pain Inventory's interference subscale. They subsequently qualitatively defined the change using the 7-point ordinal Patient Global Impression of Change (PGIC). The smallest clinically meaningful improvements in phantom limb pain and Brief Pain Inventory scores were calculated using an anchor-based method based on the PGIC. RESULTS: The median (interquartile range [IQR]) phantom pain NRS (0-10) improvements at 4 months considered small, medium, and large were 1 [1-1], 3 [3-4], and 4 [3-6], respectively. The median improvements in the Brief Pain Inventory interference subscale (0-70) associated with a small, medium, and large analgesic changes were 16 [6-18], 24 [22-31], and 34 [22-46]. The proportions of patients that experienced PGIC ≥5 were 33% and 36% in the active and placebo groups, respectively. The relative risk of a patient experiencing PGIC ≥5 in the active group compared to the sham group with 95% confidence interval was 0.9 (0.6-1.4), P = .667. CONCLUSIONS: Amputees with phantom limb pain treated with percutaneous cryoneurolysis rate analgesic improvements as clinically meaningful similar to pharmacologic treatments, although their MCID for the Brief Pain Inventory was somewhat larger than previously published values. This information on patient-defined clinically meaningful improvements will facilitate interpretation of available studies and guide future trial design.

Ultrasound-Guided Cryoneurolysis for the Treatment of Painful Diabetic Neuropathy of the Foot: A Case Series.

Diabetic peripheral neuropathy is one of the most common causes of chronic neuropathic pain. Treatment of peripheral neuropathy has been limited to either treating the underlying cause or using medications, such as tricyclic antidepressants and anticonvulsants, to manage the symptoms. In this case series, we report the use of ultrasound-guided percutaneous cryoneurolysis of the superficial peroneal nerves to treat diabetic neuropathy of the feet. This demonstrates the potential effectiveness and safety of using cryoneurolysis for painful peripheral diabetic neuropathy.

Development of a predictive model for risk stratification of acute kidney injury in patients undergoing cytoreductive surgery with hyperthermic intraperitoneal chemotherapy.

Acute kidney injury (AKI) following hyperthermic intraperitoneal chemotherapy (HIPEC) is common. Identifying patients at risk could have implications for surgical and anesthetic management. We aimed to develop a predictive model that could predict AKI based on patients' preoperative characteristics and intraperitoneal chemotherapy regimen. We retrospectively gathered data of adult patients undergoing HIPEC at our health system between November 2013 and April 2022. Next, we developed a model predicting postoperative AKI using multivariable logistic regression and calculated the performance of the model (area under the receiver operating characteristics curve [AUC]) via tenfold cross-validation. A total of 412 patients were included, of which 36 (8.7%) developed postoperative AKI. Based on our multivariable logistic regression model, multiple preoperative and intraoperative characteristics were associated with AKI. We included the total intraoperative cisplatin dose, body mass index, male sex, and preoperative hemoglobin level in the final model. The mean area under the receiver operating characteristics curve value was 0.82 (95% confidence interval 0.71-0.93). Our risk model predicted AKI with high accuracy in patients undergoing HIPEC in our institution. The external validity of our model should now be tested in independent and prospective patient cohorts.

Differences in the Receipt of Regional Anesthesia Based on Race and Ethnicity in Colorectal Surgery.

BACKGROUND: Health equity in pain management during the perioperative period continues to be a topic of interest. The authors evaluated the association of race and ethnicity with regional anesthesia in patients who underwent colorectal surgery and characterized trends in regional anesthesia. METHODS: Using the American College of Surgeons National Surgical Quality Improvement Program database from 2015 to 2020, the research team identified patients who underwent open or laparoscopic colorectal surgery. Associations between race and ethnicity and use of regional anesthesia were estimated using logistic regression models. RESULTS: The final sample size was 292,797, of which 15.6% (n = 45,784) received regional anesthesia. The unadjusted rates of regional anesthesia for race and ethnicity were 15.7% white, 15.1% Black, 12.8% Asian, 29.6% American Indian or Alaska Native, 16.3% Native Hawaiian or Pacific Islander, and 12.4% Hispanic. Black (odds ratio [OR] 0.93, 95% confidence interval [CI] 0.90-0.96, p < 0.001) and Asian (OR 0.76, 95% CI 0.71-0.80, p < 0.001) patients had lower odds of regional anesthesia compared to white patients. Hispanic patients had lower odds of regional anesthesia compared to non-Hispanic patients (OR 0.72, 95% CI 0.68-0.75, p < 0.001). There was a significant annual increase in regional anesthesia from 2015 to 2020 for all racial and ethnic cohorts (p < 0.05). CONCLUSION: There was an annual increase in the use of regional anesthesia, yet Black and Asian patients (compared to whites) and Hispanics (compared to non-Hispanics) were less likely to receive regional anesthesia for colorectal surgery. These differences suggest that there are racial and ethnic differences in regional anesthesia use for colorectal surgery.

Repeated Ultrasound-Guided Percutaneous Intercostal Cryoneurolysis for the Management of Chronic Postmastectomy Pain Syndrome: A Case Report.

Ultrasound-guided percutaneous cryoneurolysis of the intercostal nerves is an effective analgesic modality for the management of postmastectomy pain syndrome. The duration of analgesia provided by intercostal cryoneurolysis for chronic postmastectomy pain syndrome may vary. Furthermore, limited studies have described the repeated use of cryoneurolysis to manage chronic pain in these patients. In this case report, we describe the repeated use of ultrasound-guided cryoneurolysis of the intercostal nerves to manage chronic postmastectomy pain for 2.5 years. No adverse events or long-term neurologic sequelae were reported.

Factors Associated With Same Day Discharge Post-Spinal Cord Stimulator Placement.

BACKGROUND: Spinal cord stimulator (SCS) surgeries, whether performed using the open or percutaneous approach, are becoming increasingly common for a range of neuropathic pain conditions, including post-laminectomy syndrome and complex regional pain syndrome. However, there is limited knowledge regarding the factors linked to same-day discharge patterns following SCS. OBJECTIVE: The purpose of this study was to identify factors associated with same-day discharge after SCS placement. The primary outcome was same-day discharge. STUDY DESIGN: Retrospective, cohort study using a nationwide database. METHODS: Inclusion criteria included patients who underwent percutaneous or open SCS from January 1, 2014 through December 31, 2021. Exclusion criteria included patients with missing data (n = 178) and those with SCS implants for unlisted indications (n = 1,817). A multivariable analysis was conducted on the outcome data and co-variates associated with same-day discharge after SCS. RESULTS: After applying inclusion and exclusion criteria, a total of 18,058 patients remained in the final data set, including 7,339 patients who underwent percutaneous SCS and 10,719 patients who underwent open SCS procedures. After analysis, factors associated with increased rates of same-day discharge after SCS included men (odds ratio [OR] 1.16; 95% CI, 1.09 -1.24;  P < 0.001), patients on Medicaid (OR 1.64; 95% CI, 0.1.34 - 2.01; P < 0.001), and hospitals in the US Midwest (OR 1.66; 95% CI, 1.45 - 1.90; P < 0.001) and hospitals in the US West (OR 1.32; 95% CI, 1.20 - 1.46; P < 0.001). Factors associated with decreased rates of same-day discharge after SCS included the open approach (OR 0.21; 95% CI, 0.19 - 0.23; P < 0.001), Hispanic ethnicity (OR 0.61; 95% CI, 0.54 - 0.69; P < 0.001) and increased age (OR 0.99; 95% CI, 0.98 - 0.99; P < 0.001). LIMITATIONS: Since our study is retrospective, the data are subject to various biases, including variable confounding, human error in data entry, and generalizability of the results. CONCLUSION: These results can be used to help determine hospital bed needs post-SCS surgery. Future research should focus on identifying the specific reasons certain demographic and geographic factors might influence same-day discharge rates. Our study provides important insights into the factors associated with same-day discharge rates post open and percutaneous SCS implant and highlights the need for patient-centered, evidence-based approaches to health care delivery.

Public Health Informatics and the Perioperative Physician: Looking to the Future.

The role of informatics in public health has increased over the past few decades, and the coronavirus disease 2019 (COVID-19) pandemic has underscored the critical importance of aggregated, multicenter, high-quality, near-real-time data to inform decision-making by physicians, hospital systems, and governments. Given the impact of the pandemic on perioperative and critical care services (eg, elective procedure delays; information sharing related to interventions in critically ill patients; regional bed-management under crisis conditions), anesthesiologists must recognize and advocate for improved informatic frameworks in their local environments. Most anesthesiologists receive little formal training in public health informatics (PHI) during clinical residency or through continuing medical education. The COVID-19 pandemic demonstrated that this knowledge gap represents a missed opportunity for our specialty to participate in informatics-related, public health-oriented clinical care and policy decision-making. This article briefly outlines the background of PHI, its relevance to perioperative care, and conceives intersections with PHI that could evolve over the next quarter century.

Long-Term Outcomes After Spinal Cord Stimulator Placement in Patients with Pre-procedural Active Opioid Use Versus Patients Who Were Opioid-Naïve.

BACKGROUND: Outcome optimization after the placement of a spinal cord stimulator (SCS) is critical. The objective of this study was to determine if an association existed between pre-procedural opioid use (compared to patients who were opioid-naïve) and postoperative long-term outcomes following SCS placement. OBJECTIVE: To examine the impact of preprocedural opioid use on long-term outcomes after SCS therapy. STUDY DESIGN: Cohort study utilizing a nationwide database. SETTING: Retrospective. METHODS: With the use of data from HCA Healthcare's national database, a retrospective cohort study was performed to analyze differences in outcomes between opioid-naïve patients and preoperative opioid users who underwent SCS placements. The primary outcome of interest was device explantation at 6 months and 12 months. Secondary outcome measurements included reoperations and readmissions at 6 months and 12 months, as well as operative complications. Multivariable logistic regression models were performed to analyze the association of preoperative opioid use with those outcomes. The odds ratio (OR), 95% confidence intervals (CI), and P values were reported for the independent variables. RESULTS: The final study population consisted of 13,893 patients who underwent SCS placements. In univariate analyses, patients who used opioids preoperatively had higher 6-month (3.6% vs. 2.6%) and one-year removal rates (3.6% vs. 2.8%) (all P < 0.009). On multivariable logistic regression, those using opioids preoperatively had higher odds of removal at 6 months (OR = 1.290, 95% CI 1.05-1.58, P = 0.01) and at one year (OR = 1.23, 95% CI 1.01-1.50, P = 0.04). There was no difference between patients requiring preoperative opioids and patients who were opioid-naive as far as the odds of 6- or 12-month readmissions were concerned. Compared to the opioid-naive group, patients requiring preoperative opioids had increased odds of reoperation at 6 months (OR = 1.2, 95% CI 1.02-1.40, P = 0.03). There were no differences in the odds of complications between both cohorts. LIMITATIONS: Opioid use in this study was defined as using opioids preoperatively in the 30 days leading up to surgery. CONCLUSION: Patients requiring preoperative opioids before SCS placements had increased odds of SCS explantation at 6 months and 12 months, as well as increased odds of reoperation at 6 months.

Association of race and ethnicity with pediatric burn outcomes: A population study of the Kids' Inpatient Database.

BACKGROUND: While most studies on burn outcomes have focused on adults, it is unclear if the same socioeconomic and environmental inequalities affect paediatric patients. This study aims to investigate the impact of race and ethnicity on outcomes in paediatric burn patients. METHODS: The Kids' Inpatient Database is released by Agency for Healthcare Research and Quality, and is the largest publicly available database for the United States inpatient paediatric population. All paediatric burned patients in 2016 and 2019 were identified. Race and/or ethnicity was the primary exposure variable, and the primary outcome was a composite of several in-hospital morbidities. Secondary outcomes included death, non-routine disposition, and length of stay. Fine-Gray competing risks regression and multivariable logistic regression were used to analyze length of stay and all other outcomes, respectively. Analysis also isolated subgroups related to socioeconomic status and case severity. RESULTS: We included12,582 pediatric burn patients in this study. No difference was found in composite morbidity between White patients and those of other race or ethnicity groups. Hispanic ethnicity was associated with longer lengths of stay and increased odds of routine (i.e. home) discharge. Black patients had increased length of stay compared to White patients only in severe burn cases. CONCLUSIONS: Our study implies that race- or ethnicity-associated mechanisms driving outcome disparities in adults does not necessarily apply in paediatric burn patients.

Improving case duration accuracy of orthopedic surgery using bidirectional encoder representations from Transformers (BERT) on Radiology Reports.

PURPOSE: A major source of inefficiency in the operating room is the mismatch between scheduled versus actual surgical time. The purpose of this study was to demonstrate a proof-of-concept study for predicting case duration by applying natural language processing (NLP) and machine learning that interpret radiology reports for patients undergoing radius fracture repair. METHODS: Logistic regression, random forest, and feedforward neural networks were tested without NLP and with bag-of-words. Another NLP method tested used feedforward neural networks and Bidirectional Encoder Representations from Transformers specifically pre-trained on clinical notes (ClinicalBERT). A total of 201 cases were included. The data were split into 70% training and 30% test sets. The average root mean squared error (RMSE) were calculated (and 95% confidence interval [CI]) from 10-fold cross-validation on the training set. The models were then tested on the test set to determine proportion of times surgical cases would have scheduled accurately if ClinicalBERT was implemented versus historic averages. RESULTS: The average RMSE was lowest using feedforward neural networks using outputs from ClinicalBERT (25.6 min, 95% CI: 21.5-29.7), which was significantly (P < 0.001) lower than the baseline model (39.3 min, 95% CI: 30.9-47.7). Using the feedforward neural network and ClinicalBERT on the test set, the percentage of accurately predicted cases, which was defined by the actual surgical duration within 15% of the predicted surgical duration, increased from 26.8 to 58.9% (P < 0.001). CONCLUSION: This proof-of-concept study demonstrated the successful application of NLP and machine leaning to extract features from unstructured clinical data resulting in improved prediction accuracy for surgical case duration.

Tracking persistent postoperative opioid use: a proof-of-concept study demonstrating a use case for natural language processing.

BACKGROUND: Large language models have been gaining tremendous popularity since the introduction of ChatGPT in late 2022. Perioperative pain providers should leverage natural language processing (NLP) technology and explore pertinent use cases to improve patient care. One example is tracking persistent postoperative opioid use after surgery. Since much of the relevant data may be 'hidden' within unstructured clinical text, NLP models may prove to be advantageous. The primary objective of this proof-of-concept study was to demonstrate the ability of an NLP engine to review clinical notes and accurately identify patients who had persistent postoperative opioid use after major spine surgery. METHODS: Clinical documents from all patients that underwent major spine surgery during July 2015-August 2021 were extracted from the electronic health record. The primary outcome was persistent postoperative opioid use, defined as continued use of opioids greater than or equal to 3 months after surgery. This outcome was ascertained via manual clinician review from outpatient spine surgery follow-up notes. An NLP engine was applied to these notes to ascertain the presence of persistent opioid use-this was then compared with results from clinician manual review. RESULTS: The final study sample consisted of 965 patients, in which 705 (73.1%) were determined to have persistent opioid use following surgery. The NLP engine correctly determined the patients' opioid use status in 92.9% of cases, in which it correctly identified persistent opioid use in 95.6% of cases and no persistent opioid use in 86.1% of cases. DISCUSSION: Access to unstructured data within the perioperative history can contextualize patients' opioid use and provide further insight into the opioid crisis, while at the same time improve care directly at the patient level. While these goals are in reach, future work is needed to evaluate how to best implement NLP within different healthcare systems for use in clinical decision support.

Association of thoracic epidural analgesia and hospital length of stay for patients undergoing thoracotomy: a retrospective cohort analysis.

INTRODUCTION: Regional anesthesia has been shown to be efficacious for analgesia in patients who underwent thoracotomies. The objective of this study was to analyze the association of epidurals and peripheral regional anesthesia with time to hospital discharge for these patients. METHODS: This was a retrospective cohort study using National Surgical Quality Improvement Program dataset from 2014 to 2020. Propensity-matched cohorts were assembled based on use of regional anesthesia, peripheral regional anesthesia, or epidural. Fine-Gray competing risk regressions were used to explore the association between regional anesthesia use and rate of discharge. The subdistribution hazard ratio (HR) represented relative discharge rates, and in-hospital death was a competing event. A sensitivity analysis was subsequently performed in which patients with American Society of Anesthesiologists score ≥4 were removed. RESULTS: There were 4350 patients included in this analysis, in which 472 (10.8%) received a peripheral regional anesthesia nerve block and 565 (13.0%) received thoracic epidural analgesia. The subdistribution HR for rate of discharge in the epidural versus non-epidural cohort was 1.09 (95% CI 1.01 to 1.18), thus epidurals were associated with an increased rate of discharge over time. However, this benefit was no longer apparent with the sensitivity analysis. The subdistribution HR for rate of discharge in the peripheral regional anesthesia versus no regional anesthesia cohort was 1.26 (95% CI 1.15 to 1.39), thus peripheral regional anesthesia was associated with an increased rate of discharge over time. This benefit remained even with the sensitivity analysis. CONCLUSIONS: Thoracic epidural use when compared with no regional anesthesia was associated with decreased length of stay following thoracotomy in our primary analysis. The difference was no longer apparent with the sensitivity analysis. Peripheral regional anesthesia was associated with decreased length of stay even after sensitivity analysis.