Background The objective of this study is to develop predictive models for persistent opioid use following lower extremity joint arthroplasty and determine if ensemble learning and an oversampling technique may improve model performance. Methods We compared various predictive models to identify at-risk patients for persistent postoperative opioid use using various preoperative, intraoperative, and postoperative data, including surgical procedure, patient demographics/characteristics, past surgical history, opioid use history, comorbidities, lifestyle habits, anesthesia details, and postoperative hospital course. Six classification models were evaluated: logistic regression, random forest classifier, simple-feedforward neural network, balanced random forest classifier, balanced bagging classifier, and support vector classifier. Performance with Synthetic Minority Oversampling Technique (SMOTE) was also evaluated. Repeated stratified k-fold cross-validation was implemented to calculate F1-scores and area under the receiver operating characteristics curve (AUC). Results There were 1042 patients undergoing elective knee or hip arthroplasty in which 242 (23.2%) reported persistent opioid use. Without SMOTE, the logistic regression model has an F1 score of 0.47 and an AUC of 0.79. All ensemble methods performed better, with the balanced bagging classifier having an F1 score of 0.80 and an AUC of 0.94. SMOTE improved performance of all models based on F1 score. Specifically, performance of the balanced bagging classifier improved to an F1 score of 0.84 and an AUC of 0.96. The features with the highest importance in the balanced bagging model were postoperative day 1 opioid use, body mass index, age, preoperative opioid use, prescribed opioids at discharge, and hospital length of stay. Conclusions Ensemble learning can dramatically improve predictive models for persistent opioid use. Accurate and early identification of high-risk patients can play a role in clinical decision making and early optimization with personalized interventions.

BACKGROUND: The Acute Pain Service (APS) was initially introduced to optimize multimodal postoperative pain control. The aim of this study was to evaluate the association between the implementation of an APS and postoperative pain management and outcomes for patients undergoing cytoreductive surgery with hyperthermic intraperitoneal chemotherapy (CRS-HIPEC). METHODS: In this propensity-matched retrospective cohort study, we performed a before–after study without a concurrent control group. Outcomes were compared among patients undergoing CRS-HIPEC when APS was implemented versus historical controls (non-APS). The primary objective was to determine if there was a decrease in median total opioid consumption during postoperative days 0–3 among patients managed by the APS. Secondary outcomes included opioid consumption on each postoperative day (0–6), time to ambulation, time to solid intake, and hospital length of stay. RESULTS: After exclusion, there were a total of 122 patients, of which 51 and 71 were in the APS and non-APS cohort, respectively. Between propensity-matched groups, the median (quartiles) total opioid consumption during postoperative days 0–3 was 27.5 mg intravenous morphine equivalents (MEQs) (7.6–106.3 mg MEQs) versus 144.0 mg MEQs (68.9–238.3 mg MEQs), respectively. The median difference was 80.8 mg MEQs (95% confidence interval, 46.1–124.0; P < .0001). There were statistically significant decreases in time to ambulation and time to solid diet intake in the APS cohort. CONCLUSIONS: After implementing the APS, CRS-HIPEC patients had decreased opioid consumption by >50%, as well as shorter time to ambulation and time to solid intake. Implementation of an APS may improve outcomes in CRS-HIPEC patients.

Acute post-mastectomy pain is frequently challenging to adequately treat with local anesthetic-based regional anesthesia techniques due to its relatively long duration measured in multiple weeks. We report three cases in which preoperative ultrasound-guided percutaneous intercostal nerve cryoneurolysis was performed to treat pain following mastectomy. Across all postoperative days and all three patients, the mean pain score on the numeric rating scale was 0 for each day. Similarly, no patient required any supplemental opioid analgesics during the entire postoperative period; and, no patient reported insomnia or awakenings due to pain at any time point. This was a significant improvement over historic cohorts. Ultrasound-guided percutaneous cryoanalgesia is a potential novel analgesic modality for acute pain management which has a duration that better-matches mastectomy than other currently-described techniques. Appropriately powered randomized, controlled clinical trials are required to demonstrate and quantify both potential benefits and risks.

Background Paravertebral and serratus plane blocks are both used to treat pain following breast surgery. However, it remains unknown if the newer serratus block provides comparable analgesia to the decades-old paravertebral technique. Methods Subjects undergoing unilateral or bilateral non-mastectomy breast surgery were randomized to a single-injection serratus or paravertebral block in a subject-masked fashion (ropivacaine 0.5%; 20 mL unilateral; 16 mL/side bilateral). We hypothesized that (1) analgesia would be non-inferior in the recovery room with serratus blocks (measurement: Numeric Rating Scale), and (2) opioid consumption would be non-inferior with serratus blocks in the operating and recovery rooms. In order to claim that serratus blocks are non-inferior to paravertebral blocks, both hypotheses must be at least non-inferior. Results Within the recovery room, pain scores for participants with serratus blocks (n=49) had a median (IQR) of 4.0 (0–5.5) vs 0 (0–3.0) for those with paravertebral blocks (n=51): 0.95% CI −3.00 to −0.00; p=0.001. However, the difference in morphine equivalents did not reach statistical significance for superiority with the serratus group consuming 14 mg (10–19) vs 10 mg (10–16) for the paravertebral group: 95% CI −4.50 to 0.00, p=0.123. Since the 95% CI lower limit of −4.5 was less than our prespecified margin of −2.0, we failed to conclude non-inferiority of the serratus block with regard to opioid consumption. Conclusions Serratus blocks provided inferior analgesia compared with paravertebral blocks. Without a dramatic improvement in safety profile for serratus blocks, it appears that paravertebral blocks are superior to serratus blocks for postoperative analgesia after non-mastectomy breast surgery. Trial registration number NCT03860974.

Background: Same-day discharge, defined as discharge from the hospital within 24 h of surgery, has been shown to be safe for joint arthroplasty. We examined demographic and clinical factors associated with same-day discharge and unplanned readmission following shoulder arthroplasty in adult patients.Methods: Utilizing data from the American College of Surgeons National Surgical Quality Improvement Program database, we extracted information of all patients that underwent shoulder arthroplasty. The primary and secondary outcome of interest was same-day discharge and 30-day unplanned readmission, respectively. We utilized multivariable logistic regression to identify covariates associated with these outcomes. Results: There were 17,011 patients analyzed when identifying predictors for same-day discharge. There was an increase in same-day discharge from 2007 to 2016. The odds of same-day discharge were significantly better for males (P < 0.001). The odds of same-day discharge was significantly decreased for every 10-year increase in age and for patients with insulin dependent diabetes, poor functional status, chronic obstructive pulmonary disease, congestive heart failure, bleeding disorder, and comorbidity burden (all P < 0.001). There were 14,276 patients analyzed for hospital readmission. The odds of unplanned readmission were significantly higher for every 10-year increase in age and for patients with poor functional status, congestive heart failure, bleeding disorder, and higher comorbidity burden (all P < 0.001). Conclusions: The results of this study show that preoperative comorbidities and advanced age reduce the odds of same-day discharge. Risk stratification, preoperative optimization, and coordinated care after surgery may be helpful to optimize patients for same-day discharge.