Which opioid to give a patient on a GLP-1 agonist, and CGM without insulin

The edition in brief

Today's diabetes edition opens with a large US claims cohort of 411,188 adults with type 2 diabetes on a GLP-1 receptor agonist who started an opioid. Oxycodone was associated with a higher 30-day risk of severe constipation, bowel obstruction or gastroparesis than hydrocodone or tramadol (0.51% vs 0.35% and 0.33%; RR about 1.5). Absolute risks were small and the design is observational. A randomised trial of 10-kHz spinal cord stimulation for painful diabetic neuropathy found 27 of 34 implanted participants reached at least 50% pain relief against 2 of 50 on medical management alone, with blinded sensory scores and nerve fibre density also improving at six months. On the regulatory side, the FDA lists Class II recalls of compounded tirzepatide and semaglutide combined with cyanocobalamin from one US compounder over lack of sterility assurance. The pearl is the blood ketone check for an unwell patient on an SGLT2 inhibitor whose glucose looks reassuring. The practice changer is the CONNECT trial in NEJM (14 October): in 283 adults with type 2 diabetes not using insulin, managed in primary care, CGM lowered HbA1c by 0.9 percentage points more than routine care at 26 weeks, with time in range 21 points higher and no rise in hypoglycaemia. The trial was industry-funded and short, and it did not measure complications, but it is the clearest randomised evidence yet that CGM helps beyond insulin users.