Context Obesity is a pro-inflammatory metabolic state that may play a role in the development of immune-related adverse events (irAEs) associated with immune checkpoint inhibitor therapy. Objective To characterize the association between body mass index (BMI) and thyroid irAEs. Methods We performed a single center, retrospective analysis of 185 cancer patients treated with anti-PD-1/L1 from January 2014 to December 2018. Patients with normal thyroid function at baseline and available BMI were included. Main outcome measures The primary endpoint was difference in BMI in patients who developed overt thyroid dysfunction vs. remained euthyroid following anti-PD-1/L1 initiation. Additional endpoints included any (overt or subclinical) thyroid dysfunction, overt thyrotoxicosis or overt hypothyroidism, and time to development of dysfunction according to BMI. Results Any thyroid dysfunction developed in 72 (38.9%) patients and 41 (22.1%) developed overt thyroid dysfunction. Mean BMI was higher in those with overt thyroid dysfunction vs. euthyroid (27.3±6.0 vs. 24.9±4.5, p=0.03). Development of overt thyrotoxicosis vs. remaining euthyroid was associated with higher BMI (28.9±5.9 vs. 24.9±4.5; p<0.01), whereas overt hypothyroidism was not (26.7±5.5 vs. 24.9±4.5, p=0.10). Overt thyrotoxicosis developed within 57.5 (IQR 31.8-78.8) days of treatment in the low-normal BMI group, 38.0 (IQR 26.8-40.5) days in the overweight and 23.0 (IQR 21.0-28.0) days in the obese group (p=0.02). Conclusions Patients treated with PD-1/L1 inhibitors were more likely to develop thyroid irAEs, specifically overt thyrotoxicosis, with increasing BMI. Overt thyrotoxicosis occurred earlier in obese vs. leaner patients. These data highlight the complex interplay between obesity and immune response in ICI-treated patients.

Context: The association of inpatient glucose measurements with amputations in patients admitted with acute diabetic foot has not been described. Objective: To evaluate the relationship of hyperglycemia, hypoglycemia, and glucose variability during hospitalization with amputations in patients hospitalized with acute diabetic foot. Design: Retrospective cohort study. Setting: Academic tertiary hospital. Patients: We reviewed demographic, clinical, laboratory, and point-of-care glucose data in patients hospitalized with acute diabetic foot in the Diabetic Foot Unit during 2015 through 2017. Main Outcome Measures: The primary outcomes were any or major amputations during hospitalization. Secondary outcomes included length of hospitalization and in-hospital mortality. Results: During the study period, 418 patients were hospitalized in the Diabetic Foot Unit and 45,496 glucose measurements were taken. Patients experiencing any hyperglycemia and any or severe hypoglycemia were more likely to undergo any or major amputations during hospitalization. High glycemic variability was associated with major amputations. Peripheral vascular disease (PVD), high Wagner score, and hypoglycemia were independent predictors of amputations. Older age, PVD, previous amputation, elevated white blood cell level, high Wagner score, and hypoglycemia were independent predictors of major amputations. Conclusions: In-patient hypoglycemia emerged as an independent risk factor for any and major amputations. Although it is unclear whether hypoglycemia directly contributes to adverse outcomes or is simply a biomarker of disease severity, efforts to minimize in-hospital hypoglycemic events are warranted.

Romosozumab is a potent treatment for osteoporosis, with significant effects on bone density and fracture prevention. This study evaluated the cardiovascular safety of romosozumab in a real-world cohort of postmenopausal women at high fracture risk. We retrospectively evaluated postmenopausal women who initiated treatment with romosozumab between January 1, 2020, and June 30, 2023. We examined the occurrence of a major adverse cardiovascular event (MACE) across two distinct segments during the treatment period and after its conclusion. After applying inclusion and exclusion criteria, 847 women were followed for a median of 729 days (IQR: 445-1060). The incidence rate of MACE was 24.0 (95% CI 17.7-32.5) per 1000 person-years during the study period. The change in the rate of MACE from 0-90 days and 90-365 days post-treatment initiation was 0.04 and 0.06 events per 1000 days, respectively. The difference in the rate between these intervals was not statistically significant (p = .09). After 1 yr of treatment, the slope of MACE increased to 0.10, differing significantly from the preceding 12 mo on treatment (p<.001). The incidence of MACE was higher in those with a background of previous cardiovascular disease or diabetes at all timepoints, as expected. The consistency in event rates during treatment suggests that romosozumab is not associated with an increase in MACE in postmenopausal women. This finding challenges reports suggesting an increase in cardiovascular events within the first year of romosozumab treatment.