Letter to the Editor Regarding “Association Between Preoperative GLP-1 Receptor Analog Use and Postoperative Complications and Mortality Following Lumbar Fusion Surgery”

Abstract

We read with interest the study by Kesaria et al., which reported an association between preoperative glucagon-like peptide-1 receptor agonist (GLP-1 RA) use and fewer complications after lumbar fusion. 1 The lower 1-year pseudoarthrosis rate (8.9% vs 13.8%) is noteworthy, but it should not be interpreted as evidence of a direct drug effect on fusion. Randomized clinical data do not indicate a consistently osteoanabolic effect. After 52 weeks, semaglutide did not increase procollagen type I N-terminal propeptide; bone resorption increased, and bone mineral density at the hip and lumbar spine declined relative to placebo. 2 Likewise, liraglutide alone was associated with greater losses in hip and spine bone mineral density than exercise alone, whereas the combination of liraglutide and exercise preserved bone mineral density despite greater weight loss. 3 These findings suggest that changes in bone mineral density may not fully explain the association between GLP-1 RA exposure and arthrodesis; local fusion biology, metabolic improvement, and residual differences between treated and untreated patients remain competing explanations. An analysis designed around this causal question would be more informative than further adjustment for comorbidities alone. Longitudinal changes in weight and hemoglobin A1c could be specified as candidate mediators, while the specific agent, treatment indication, and treatment duration could be examined as effect modifiers. A mediation analysis could then estimate the extent to which metabolic changes mediate the observed association, provided that time-varying confounding is addressed. Outcome ascertainment is equally important because a diagnosis derived from electronic health records may not distinguish symptomatic nonunion from incidental radiographic findings and may also be influenced by differences in follow-up intensity. Computed tomography (CT) has the highest diagnostic accuracy among commonly used modalities for detecting pseudoarthrosis after thoracolumbar fusion. 4 A prospectively defined CT-based endpoint, supplemented by patient-reported outcomes and data on revision surgery, would therefore reduce misclassification and improve clinical interpretability. If the association persists after radiographic adjudication and mediation analysis, the evidence for a pathway-specific effect of GLP-1 RA therapy on fusion would be substantially stronger