Background and Aims Non‐alcoholic steatohepatitis (NASH) is a common cause of hepatocellular carcinoma (HCC) worldwide. Emerging data suggests NASH‐induced HCC could be associated with less response to immune checkpoint inhibitor (ICI)‐based therapy. Metformin has been associated with improved outcomes in cancers like melanoma treated with ICIs, but its impact on HCC is not well defined. The purpose of this study was to examine the effect of metformin on clinical outcomes in patients with advanced HCC treated with ICIs. Methods We retrospectively analysed patients with advanced HCC treated with ICIs in first and later‐line settings between 2015 and 2021. The primary endpoints were overall survival (OS), progression‐free survival (PFS), and objective response rate (ORR) as assessed according to Response Evaluation Criteria in Solid Tumours (RECIST) version 1.1. Patients were stratified based on their usage of metformin. Results Our study included 18 patients in the metformin group and 93 patients in the non‐metformin group. The most common causes of HCC were viral hepatitis (52%), NASH (29%), and alcohol (8%). ORR was 5.6% in the metformin group vs 22.6% in the non‐metformin group (P = .0987). Median OS was 10.8 months versus 45.9 months (HR = 1.99, 95% CI = 0.95–4.21, P = .065) and median PFS was 2.5 months versus 6.6 months (P = .077) in the metformin and non‐metformin groups, respectively. Regardless of metformin usage, OS was significantly worse in patients with poor ECOG performance status, HCC aetiology of NASH, MELD score 10–23, AFP >= 400, and use of ICIs in later lines of therapy. Conclusions Metformin use was associated with a trend, although not statistically significant, toward a worse ORR, OS and PFS in advanced HCC patients treated with ICIs.

Objective Postoperative opioid prescriptions tend to exceed children's analgesic needs, but awareness of the opioid epidemic may have driven changes in prescribing behaviors. This study evaluated opioid prescribing patterns after major pediatric ear surgery. Methods This study reviewed all cases of tympanoplasty, tympanomastoidectomy, mastoidectomy, cochlear implantation, otoplasty, and aural atresia repair at a pediatric hospital during 2010–2021. Regressions were conducted to identify opioid prescribing trends over time. Potential covariates were assessed. Returns to the system were reviewed as a balancing measure. Results Even without a targeted protocol, opioid prescribing declined significantly. After prescribing peaked in 2012–2013, significant negative trends yielded lower rates of opioid prescriptions, fewer doses per prescription, smaller patient‐weight‐standardized dose sizes, and less variability (all p < 0.001). In 2012, 96.1% of patients received opioid prescriptions; the rate fell to 13.5% by 2021. For patients ages, 0–6, the annual rate of opioid prescriptions dropped from a maximum of 96.3% in 2012 to 0.0% in 2021. The annual average supply of doses per prescription decreased by 68% between 2013 and 2021, reducing the total days' supply to an evidence‐based 3.1 ± 1.6 days. Regressions did not detect changes in returns to the system. Pain‐related returns were rare (0.9%) and did not vary by opioid prescriptions (p = 0.37). Prescribing trends were closely correlated with a tonsillectomy‐focused protocol that our institution implemented in 2019. Conclusion Surgeon‐driven opioid stewardship has improved with no resultant change in revisit rates. Procedure‐specific quality improvement interventions may have broader off‐target effects on prescribing behaviors. Level of Evidence IV Laryngoscope, 133:1987–1992, 2023