Diabetes Mellitus in Patients Undergoing Mitral Transcatheter Edge-to-Edge Repair—A Decade Experience in 1000+ Patients

Background: Diabetes mellitus worsens outcomes in patients suffering from heart disease undergoing cardiac procedures. Objectives: To investigate the impact of diabetes in patients undergoing mitral transcatheter edge-to-edge repair (M-TEER). Methods: 1118 patients treated with M-TEER for functional (FMR) and degenerative (DMR) mitral regurgitation (MR) between 2010 and 2021 were analyzed using the combined endpoint of death/rehospitalization for heart failure (HFH). Results: Among diabetics (N = 306; 27.4%), comorbidities such as coronary artery disease (75.2% vs. 62.7%; p < 0.001) and progressed (stage III/IV) chronic kidney disease (79.5% vs. 72.6%; p = 0.018) were more frequent. The rate of FMR was higher in diabetics (71.9% vs. 64.5%; p < 0.001). The combined endpoint occurred more frequently in diabetics (40.2% vs. 35.6%; log-rank = 0.035). While no difference was observed in FMR patients (36.8% vs. 37.6%; log-rank p = 0.710), rates of the combined endpoint differed significantly between diabetics and non-diabetics in DMR patients (48.8% vs. 31.9%; log-rank p = 0.001) only. However, diabetes did neither predict the combined endpoint in the overall (OR: 0.97; 95% CI 0.65–1.45; p = 0.890) nor in the DMR cohort (OR: 0.73; 95% CI 0.35–1.51; p = 0.389). Among diabetics treated with M-TEER, troponin (OR: 2.32; 95% CI 1.3–3.7; p = 0.002) and estimated glomerular filtration rate (OR: 0.52; 95% CI 0.3–0.88; p = 0.018) independently predicted the combined endpoint. Conclusions: Diabetes is associated with adverse outcomes after M-TEER, particularly in DMR patients. However, diabetes does not predict the combined endpoint. In diabetics undergoing M-TEER, biochemical markers associated with organ function and damage independently predict the combined endpoint of death and rehospitalization.


Introduction
Mitral transcatheter edge-to-edge repair (M-TEER) has already been used for over a decade for the treatment of symptomatic mitral regurgitation (MR). In patients with prohibitive surgical but acceptable interventional risk, it provides a treatment option in addition to drug therapy. Studies have shown M-TEER to be both safe and effective in reducing MR in patients suffering from functional (FMR) or degenerative (DMR) mitral regurgitation [1][2][3]. Notably, among patients with heart failure, M-TEER reduces rehospitalization for heart failure and mortality [4]. Diabetes is a frequent risk factor predisposing patients to cardiac disease and is often observed in patients undergoing M-TEER [1,2,[4][5][6][7]. Across the field of structural cardiac [8][9][10] and coronary interventions [11][12][13], ample evidence has been provided linking diabetes mellitus to adverse outcomes. In patients treated with M-TEER, a subgroup analysis from the COAPT (Cardiovascular Outcomes Assessment of the MitraClip Percutaneous Therapy for Heart Failure Patients with Functional Mitral Regurgitation) trial suggests worsened survival in diabetics with FMR [5], whereas others did not find a significant impact of diabetes on outcomes after M-TEER [14,15]. Given the conflicting evidence, we further investigated the impact of diabetes in >1000 patients undergoing M-TEER.

Study Population
For this single-center study, we retrospectively analyzed 1118 patients treated with M-TEER for FMR or DMR at our institution between 1 January 2010 and 31 December 2021. All patients undergoing M-TEER during the inclusion time frame were screened (N = 1171). Patients undergoing reintervention (N = 50) were excluded from this study. Reintervention rates were equally distributed among diabetics and non-diabetics (4.4% vs. 4.2%; p = 0.920). A patient was grouped into the diabetic group if a diagnosis of diabetes mellitus type II could be retrieved from current medical records at the time of the procedure. Patients with type I diabetes (N = 3) were excluded from this analysis. No further exclusion criteria existed.
Patients eligible for TEER suffered from chronic, symptomatic MR (grade III or IV) confirmed by transesophageal echocardiography (TEE) despite guideline-directed medical therapy (GDMT). Patients were evaluated by an interdisciplinary heart team and directed towards M-TEER by a joint decision based on guidelines [16][17][18] currently in place at the time of the procedure. All subjects were participants of the prospective MiTra ULM registry. All patients gave written informed consent for retrospective and prospective data collection. This research was approved by the local ethics committee.
The intervention was performed under general anesthesia, using echocardiographic guidance (TEE) and fluoroscopy. Precise details of M-TEER have been described elsewhere [2]. In brief, venous access is established via the groin and a guiding catheter is advanced towards the mitral valve after transseptal puncture from the right to the left atrium. One or multiple devices are positioned within the MV to achieve maximum MR reduction. The array of devices used in this study includes all commercially available edge-to-edge repair devices. For estimation of the glomerular filtration rate, the chronic kidney disease epidemiology collaboration (CKD-EPI) formula was used.

Patient Follow-Up
Standardized patient follow-up was completed by a routine clinical visit or telephone interview at 1, 3, 6 and 12 months and yearly thereafter. Patients not routinely seen in the outpatient department were followed up by a telephone interview carried out by trained study nurses.

Statistical Analysis
For statistical analysis, patients were grouped according to the presence of preexisting diabetes for the overall cohort and subgroups (FMR and DMR). The distribution of variables was analyzed using histograms and Q-Q plots. Continuous variables were expressed using mean and standard deviation. Categorical variables are shown as frequencies and percentages. Continuous variables were compared using Mann-Whitney Test or Student's T-test depending on the distribution of variables. Outcomes were analyzed using the combined endpoint of all-cause death or rehospitalization for heart failure. Time-to-event analysis for the endpoint was performed using Kaplan-Maier curves and the log-rank test. The time-to-event data are shown as the median and its respective 95% confidence interval (CI). To identify parameters impacting the time-to-event patients were also grouped according to the endpoint. Univariate and multivariate Cox proportional hazards regression was used to quantify the impact of these parameters. Pearson and Spearman's correlation coefficients as well as variance inflation factors (VIFs) were used to exclude correlation among variables before inclusion in multivariate models. Significant variables in univariate regression were included in the multivariate regression model using backwards conditional inclusion. For continuous variables, cut-off values were calculated using the Youden Index to maximize sensitivity and specificity. A p-value of <0.05 was considered significant for all statistical testing. Statistical analysis was carried out using SPSS, Version 29 (SPSS Statistics, IBM, Chicago, IL, USA).

Discussion
We investigated the impact of diabetes in >1000 patients undergoing M-TEER for treatment of DMR and FMR treated at a high-volume tertiary center. To the best of our knowledge, we were able to present data and analyze outcomes of diabetics in the largest real-world M-TEER cohort so far. The main findings of this study are as follows: -Patients with diabetes can safely and effectively be treated with M-TEER. -Diabetes is associated with adverse outcomes after M-TEER but does not independently predict these outcomes. - In diabetics treated with M-TEER, well-established markers of advanced heart failure, troponin T and eGFR independently predict adverse outcomes.
Across studies reporting on M-TEER prevalence of diabetes ranges from 21.9 to 39.4% [1,2,[4][5][6][7] making it a relevant comorbidity. Recently, an outcome analysis of diabetics [5] treated within the randomized-controlled (RCT) COAPT trial (Transcatheter Mitral-Valve Repair in Patients with Heart Failure) [4] was published. COAPT compared M-TEER to GDMT in FMR patients [4]. The investigators around Shahim et al. found a higher 2-year mortality risk in diabetics compared to non-diabetics. Yet, M-TEER consistently reduced 2-year mortality compared to guideline-directed medical therapy alone in diabetics and non-diabetics [5]. Unlike COAPT investigators, we did not see a difference in outcome between diabetics and non-diabetics in FMR patients. Moreover, the adverse outcomes associated with diabetes in our study were driven by DMR patients (see Figure 1). However, our patient population represents real-world data and does not seem to be strictly comparable to that from the randomized-controlled COAPT trial: Perioperative risk (assessed by STS Score) was similar between diabetics and non-diabetics with FMR in our study (5.5 ± 4.6% vs. 5.3 ± 6.2%; p = 0.667), while diabetics in the COAPT trial had higher perioperative risk (STS Score 5.1 ± 4.6% vs. 6.9 ± 6.5%; p < 0.001) compared to non-diabetics [5]. Renal disease was more frequent in their study in diabetics (65.1% vs. 51.9%; p = 0.002), while being balanced equally in our study (75.5% vs. 77.0%; p = 0.678) among FMR patients. Nevertheless, an overall greater comorbidity burden was seen among diabetics in our study, an observation also made by COAPT investigators [5] and others [14]. Apart from that, evidence regarding the outcome of diabetics after M-TEER is scarce: In a study by Hellhammer et al., diabetes independently predicted NT-proBNP non-response (≤30% decrease) at 6-month follow-up in a mixed cohort of 58 patients [19]. In another investigation of 79 patients with FMR by Paulus et al., diabetes predicted a lack of improvement in the six-minute walking test distance [20]. Nevertheless, these studies [15,20] neither unveil increased rehospitalization nor mortality rates in diabetics. Additionally, in a very recent study, Kirschfink et al. reported results from their real-world cohort of 340 patients with DMR and FMR: At 1-year follow-up, neither mortality nor rehospitalization differed between diabetics and non-diabetics [14].
With regard to the prevalence of comorbidities in patients with FMR and DMR, some interesting observations became evident: Hyperlipidemia (HLP) was significantly more prevalent among diabetics with FMR (p < 0.01), whereas the rate of current smokers was significantly higher in DMR patients (p = 0.019). These effects were not observed viceversa with similar rates of HLP (p = 0.734) and smoking status (0.182) in DMR and FMR patients, respectively. These findings are interesting; however, unlikely to have any effect on the outcome as neither variable was found to predict outcome in the overall cohort nor in diabetics.
In our study, rates of periprocedural adverse events were low in the overall cohort with no difference in diabetics and non-diabetics. Furthermore, the results of M-TEER did not vary either. Consequently, our study is in line with others [5,14], confirming the safety and effectiveness of M-TEER in diabetics.

Limitations
We presented results from an observational retrospective single-center investigation with all limitations inherent in such a study. Our study's population reflects treatment strategy and patient selection in a German high-volume tertiary center over the course of more than 10 years in >1000 patients. We demonstrated that these patients differ from those treated in a large RCT (COAPT), which operates on tighter inclusion and exclusion criteria. Moreover, guidelines as well as patient selection for M-TEER have changed in recent years and any effects thereof, which might act as confounders, cannot be ruled out. Data on diabetes onset and glycemic control over time were not available retrospectively. Our results could therefore not be adjusted for these factors and a confounding effect cannot be ruled out. Moreover, poor glycemic control was associated with worse outcomes in previous studies in patients suffering from diabetes [32] as well as in diabetics with heart failure [33].

Conclusions
Diabetes is associated with adverse outcomes after M-TEER, particularly in DMR patients. However, biochemical markers associated with organ function (eGFR) and damage (Troponin) predict the combined endpoint of death and rehospitalization, whereas diabetes mellitus itself does not.
Supplementary Materials: The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/jcm12103502/s1, a supplementary file containing additional tables is attached to this manuscript. Table S1. Baseline patient characteristics, echocardiography and procedural outcomes separated by etiology. Table S2. Patients (overall cohort) grouped according to the endpoint of death/rehospitalization for heart failure. Table S3. Patients (DMR only) grouped according to the endpoint of death/rehospitalization for heart failure. Table S4. Diabetics grouped according to the endpoint of death/rehospitalization for heart failure. Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the local Review Board. All patients gave written informed consent for retrospective and prospective data collection. Informed Consent Statement: All patients gave written informed consent for retrospective and prospective data collection. Data Availability Statement: All relevant data are included within the manuscript or its Supplemental Information.

Conflicts of Interest:
The authors declare no conflict of interest.