Cardiology · Medical Literature
Effect of empagliflozin on copeptin levels in patients with recent acute coronary syndrome and newly detected dysglycaemia: a post-hoc analysis of the SOCOGAMI randomized controlled trial.
Abstract
Copeptin, a surrogate marker for vasopressin secretion, is associated with cardiovascular disease, insulin resistance and dysglycaemia. The cardioprotective effects of sodium-glucose cotransporter 2 inhibitors (SGLT2i) may involve vasopressin modulation through fluid redistribution, but whether this effect persists long-term in high-cardiovascular-risk patients with newly detected dysglycaemia remains unknown. In this post-hoc analysis of the SOCOGAMI double-blind, placebo-controlled trial, 42 patients (mean age 67.5 years, 19% females) with impaired glucose tolerance or newly detected type 2 diabetes following an ACS and no heart failure were randomized to empagliflozin 25 mg/day (n = 20) or placebo (n = 22) for 7 months. Copeptin was measured during oral glucose tolerance tests (OGTT) at baseline, after 7 months on-treatment, and 3 months after treatment withdrawal. Treatment effects were assessed by repeated-measures ANOVA with treatment × time interaction and linear mixed-effects models. Haematocrit, but not copeptin, showed a significant between-group difference at 7 months (p = 0.03 and p = 0.63, respectively). Both markers returned toward baseline after treatment withdrawal, but the overall treatment × time interaction was not significant for either (p = 0.72 and p = 0.64 respectively). Results were unchanged after accounting for a baseline imbalance in diuretic use (35% vs. 14%). Copeptin was not associated with the glucose-lowering effect of empagliflozin and no differential copeptin response during the OGTT across groups or visits was observed. In exploratory analyses, copeptin correlated with arterial pulse wave velocity at baseline (rs = 0.40, unadjusted p = 0.03). In this post-hoc analysis, empagliflozin treatment was not associated with statistically significant sustained vasopressin secretion in post-ACS patients with newly detected dysglycaemia and preserved cardiac function. Due to the limited power and the absence of early on-treatment sampling these findings cannot exclude AVP modulation and warrant confirmation in adequately powered studies. EudraCT number 2015-004571-73.
Clinical relevance
This post-hoc analysis addresses whether the cardioprotective effects of SGLT2 inhibitors—specifically empagliflozin—are mediated through vasopressin (AVP) modulation, a plausible mechanistic hypothesis given SGLT2i-induced fluid redistribution and the known association of copeptin with cardiovascular risk and insulin resistance. The target population is a high-risk, underexplored cohort: post-ACS patients with newly detected dysglycaemia (impaired glucose tolerance or new-onset T2DM) and preserved cardiac function. Clarifying the vasopressin pathway is clinically relevant because it could inform mechanistic understanding of SGLT2i cardioprotection beyond glycaemic and haemodynamic effects.
Key findings
1. No significant between-group difference in copeptin levels was observed at 7 months of empagliflozin 25 mg/day versus placebo (p = 0.63), arguing against sustained AVP modulation as a primary cardioprotective mechanism in this setting. 2. Haematocrit showed a statistically significant between-group difference at 7 months favouring empagliflozin (p = 0.03), confirming the expected haemoconcentration effect of SGLT2 inhibition even in this small cohort. 3. The overall treatment × time interaction was non-significant for both copeptin (p = 0.64) and haematocrit (p = 0.72) across the three time points (baseline, 7 months on-treatment, 3 months post-withdrawal), indicating no sustained differential trajectory between arms. 4. Both copeptin and haematocrit trended back toward baseline values after treatment withdrawal in both groups, consistent with reversibility of SGLT2i haemodynamic effects. 5. Copeptin levels were not associated with the glucose-lowering effect of empagliflozin, and no differential copeptin response during the OGTT was observed across groups or visits. 6. In exploratory analyses, baseline copeptin correlated with arterial pulse wave velocity (Spearman rs = 0.40, unadjusted p = 0.03), supporting copeptin as a marker of vascular stiffness in this post-ACS dysglycaemic population, though this finding is hypothesis-generating only.
Practice implications
Clinicians should not yet incorporate AVP modulation into their mechanistic rationale for prescribing empagliflozin in post-ACS patients with newly detected dysglycaemia, as this analysis does not support copeptin suppression as a significant mediator of benefit in those with preserved cardiac function; established indications (glycaemic control, cardiorenal protection) remain the primary drivers. The confirmed haematocrit rise with empagliflozin reinforces the need for vigilance regarding haemoconcentration, particularly in post-ACS patients who may be volume-sensitive or on concomitant diuretics (35% vs. 14% at baseline in this trial). Given the study's limited power (n = 42), absence of early on-treatment sampling, and post-hoc design, these null findings should be interpreted cautiously and prospective, adequately powered trials with serial early copeptin sampling are warranted before definitively excluding AVP pathway involvement.