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.

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.

The association of depression and anxiety with postoperative opioid use following total joint arthroplasty.

Background: The devastating opioid epidemic in the United States has been exacerbated by health care practices as well as underlying individual factors. Total joint arthroplasty (TJA) is one of the most common surgical procedures performed annually and patients frequently require opioids for pain control. Patient anxiety and depression has been shown to be associated with increased pain and poorer outcomes after TJA. Our study sought to determine if there was an association between depression/anxiety and postoperative opioid use following TJA. Methods: In this retrospective cohort study, postoperative outcomes after TJA were compared among three cohorts of patients: 1) no depression; 2) mild depression; or 3) moderate or severe depression at our institution from 2017 to 2019. Our primary outcome was persistent opioid use ≥3 months after surgery. Secondary outcomes included postoperative day 1 opioid consumption and hospital length of stay (LOS). Multivariable regression modeling was performed to control for various potential confounders. Results: Of the 542 total patients that met inclusion criteria for this study, 53 (9.8%) had mild depression and 67 (12.4%) had moderate or severe depression. Persistent opioid use ≥3 months after surgery was found in 132 (24.3%) patients. Mild depression was associated with increased odds of persistent opioid use (odds ratio 4.11, 95% confidence interval 1.65-10.18, P = 0.002). Depression was not associated with immediate postoperative opioid use or hospital LOS. Conclusion: Mild depression was associated with persistent opioid use after surgery. Future studies should investigate if better management of this comorbidity could improve outcomes in patients undergoing joint arthroplasty.

A Point-Based Risk Calculator for Mortality After Hepatectomy.

BACKGROUND: Preoperative risk stratification for hepatectomy patients can aid clinical decision making. The objective of this retrospective cohort study was to determine postoperative mortality risk factors and develop a score-based risk calculator using a limited number of preoperative predictors to estimate mortality risk in patients undergoing hepatectomy. METHODS: Data were collected from patients that underwent hepatectomy from the National Surgical Quality Improvement Program dataset from 2014 to 2020. Baseline characteristics were compared between survival and 30-day mortality cohorts using the χ 2 test. Next, the data were split into a training set to build the model and a test set to validate the model. A multivariable logistic regression model modeling 30-day postoperative mortality was trained on the training set using all available features. Next, a risk calculator using preoperative features was developed for 30-day mortality. The results of this model were converted into a score-based risk calculator. A point-based risk calculator was developed that predicted 30-day postoperative mortality in patients who underwent hepatectomy surgery. RESULTS: The final dataset included 38,561 patients who underwent hepatectomy. The data were then split into a training set from 2014 to 2018 (n = 26,397) and test set from 2019 to 2020 (n = 12,164). Nine independent variables associated with postoperative mortality were identified and included age, diabetes, sex, sodium, albumin, bilirubin, serum glutamic-oxaloacetic transaminase (SGOT), international normalized ratio, and American Society of Anesthesiologists classification score. Each of these features were then assigned points for a risk calculator based on their odds ratio. A univariate logistic regression model using total points as independent variables were trained on the training set and then validated on the test set. The area under the receiver operating characteristics curve on the test set was 0.719 (95% confidence interval, 0.681-0.757). CONCLUSIONS: Development of risk calculators may potentially allow surgical and anesthesia providers to provide a more transparent plan to support patients planned for hepatectomy.

Racial and Ethnic Differences in the Use of Regional Anesthesia for Patients Undergoing Total Knee Arthroplasty.

BACKGROUND: There is an increasing body of evidence that suggests racial and ethnic disparities exist in medical care. In the field of anesthesiology, few studies have investigated the association of race and ethnicity with the provision of regional anesthesia for patients undergoing total knee arthroplasty. This analysis queried a large national surgical database to determine whether there were racial or ethnic differences in the administration of peripheral nerve blocks for patients undergoing total knee arthroplasty. METHODS: In this retrospective cohort study, data were collected from a large national database during the years 2017-2019. Multivariable logistic regressions were used to measure the association of race and ethnicity with utilization of regional anesthesia. The participants for the study were patients 18 years or older undergoing total knee arthroplasty. RESULTS: Our primary finding was that among patients undergoing total knee arthroplasty, Black patients had lower odds (adjusted odds ratio [aOR]: 0.93, 99% confidence interval [CI]: 0.89-0.98) of receiving regional anesthesia than White patients. Also, Hispanic patients had lower odds (aOR: 0.88, 99% CI: 0.83-0.94) of receiving regional anesthesia than non-Hispanic patients. Native Hawaiian/Pacific Islander patients had increased odds (aOR: 2.04, 99% CI: 1.66-2.51) of receiving regional anesthesia. CONCLUSION: This study demonstrated that there might be racial and ethnic differences in the provision of regional anesthesia for patients undergoing total knee arthroplasty. These differences underscore the need for more studies aimed at equitable access to high quality and culturally competent health care.

A Neural Network Model Using Pain Score Patterns to Predict the Need for Outpatient Opioid Refills Following Ambulatory Surgery: Algorithm Development and Validation.

BACKGROUND: Expansion of clinical guidance tools is crucial to identify patients at risk of requiring an opioid refill after outpatient surgery. OBJECTIVE: The objective of this study was to develop machine learning algorithms incorporating pain and opioid features to predict the need for outpatient opioid refills following ambulatory surgery. METHODS: Neural networks, regression, random forest, and a support vector machine were used to evaluate the data set. For each model, oversampling and undersampling techniques were implemented to balance the data set. Hyperparameter tuning based on k-fold cross-validation was performed, and feature importance was ranked based on a Shapley Additive Explanations (SHAP) explainer model. To assess performance, we calculated the average area under the receiver operating characteristics curve (AUC), F1-score, sensitivity, and specificity for each model. RESULTS: There were 1333 patients, of whom 144 (10.8%) refilled their opioid prescription within 2 weeks after outpatient surgery. The average AUC calculated from k-fold cross-validation was 0.71 for the neural network model. When the model was validated on the test set, the AUC was 0.75. The features with the highest impact on model output were performance of a regional nerve block, postanesthesia care unit maximum pain score, postanesthesia care unit median pain score, active smoking history, and total perioperative opioid consumption. CONCLUSIONS: Applying machine learning algorithms allows providers to better predict outcomes that require specialized health care resources such as transitional pain clinics. This model can aid as a clinical decision support for early identification of at-risk patients who may benefit from transitional pain clinic care perioperatively in ambulatory surgery.

The association of Serratus Anterior Plane blocks with postoperative opioid use and analgesia following simple lumpectomy: a retrospective cohort analysis.

BACKGROUND: The potential benefit of regional interventions for simple lumpectomy breast cancer surgeries has not been well investigated. Understanding which patients to not offer a regional intervention to can be just as important as knowing which would benefit. It is unclear whether fascial plane blocks, such as serratus anterior plane (SAP) block, should be routinely performed for less extensive breast surgeries. Therefore, our goal in this retrospective cohort study was to evaluate the association of integrating SAP blocks into a standard perioperative multimodal analgesia plan in patients undergoing simple lumpectomies (without node biopsies) with perioperative opioid consumption. As secondary outcomes, we also analyzed postoperative pain scores and post-anesthesia care unit (PACU) length of stay. METHODS: This was a single institution retrospective cohort study (surgical site infiltration only versus SAP block cohorts) assessing the association of SAP blocks to our outcomes of interest. In the adjusted analysis, we created matched cohorts using 1:1 (surgical site infiltration only: SAP block) propensity-score matching using nearest neighbor-matching without replacement. To compare the primary and secondary outcomes in the matched cohorts, we used the Wilcoxon signed rank test. A P-value of < 0.05 was considered statistically significant. RESULTS: There were 419 patients included in the analysis, in which 116 (27.7%) received a SAP block preoperatively in addition to our standard perioperative analgesia plan. In an unadjusted analysis, no differences were seen in perioperative opioid consumption, PACU pain scores, and PACU length of stay. Among the matched cohorts, the median [quartile] perioperative opioid consumption in the surgical site infiltration only versus SAP block cohorts were 10 mg [10, 13.25 mg] and 10 mg [7, 15 mg], respectively (P = 0.16). No differences were seen in the other outcomes. CONCLUSIONS: In this study, we evaluated the impact of SAP blocks on patients undergoing simple lumpectomies, which are relatively less involved breast surgeries. We concluded that routine use of preoperative regional anesthesia is not beneficial for these specific patients. Future studies should focus on identifying patients that would directly benefit from regional interventions.