Single or dual antiplatelet therapy after transcatheter aortic valve implantation. A meta-analysis of randomized controlled trials

INTRODUCTION

Transcatheter aortic valve implantation (TAVI) revolutionized the treatment of severe aortic valve stenosis over the past decade.1,2 Furthermore, in recent years, there has been a clear preference for TAVI over surgical techniques regarding the treatment of this valvular heart disease, with an increasing number of patients, including low surgical risk ones.3 Although post-TAVI infective endocarditis (IE) is a rare complication reported in 1% up to 6% of patients, it is often associated with grim clinical outcomes and high mortality rates despite diagnostic1,2 and therapeutic4 advances. Complications are expected to rise exponentially as the number of TAVIs continues to rise at a constant rate in our setting as well.5

Few studies have compared the incidence of IE on TAVI and surgical aortic valve replacement (SAVR). However, randomized clinical trials have shown similar annual incidences of IE after SAVR and TAVI.6,7 Due to the strict patient selection of such studies, the results obtained in each center during the routine clinical practice may vary, and few studies like this using dedicated databases have been conducted. A multicenter study8 revealed the characteristics of IE on TAVI vs SAVR and the outcomes of a large patient cohort, yet it did not analyze the incidence of IE. Assessing the outcomes of this severe complication at each center should be a priority due to its potential impact on decision-making for the heart team.3

The aim of the present study was to analyze both the incidence and mortality rates of IE on TAVI vs SAVR patients from a tertiary teaching hospital in a long series of patients with severe aortic stenosis.

METHODS

Study design and population

We conducted an observational, single-center, retrospective study of a prospective cohort including all cases of IE in symptomatic patients with severe aortic valve stenosis treated with TAVI or SAVR using a biological prosthetic valve, diagnosed, and followed at a tertiary teaching referral hospital by an endocarditis team from January 2008 through December 2022.

Endpoints and definitions

The primary endpoint of the study was to analyze the overall incidence and mortality rates of IE on TAVI at our center. Secondary endpoint was to compare both rates in patients with IE on TAVI and SAVR during the same period. Another secondary endpoint was to study the number of patients who underwent surgery for IE in both groups. IE was diagnosed using the Duke criteria9 or the 2015 European Society of Cardiology modified criteria10 depending on the time of diagnosis in the study period. Cases of IE in TAVI patients during this period were identified, and their characteristics were compared with those of IE in SAVR patients.

Follow-up

Follow-up events were defined based on the criteria established by the Valve Academic Research Consortium-2.11 All complications, need for surgery, and mortality at the follow-up were recorded. IE-related cardiac complications included decompensated heart failure, fistula, prosthetic dehiscence, abscess, and complete atrioventricular block. Systemic complications included acute kidney failure, sepsis, embolism, and disorders of the central nervous system. Early IE was defined as that occurring within the first year after TAVI or surgery based on European recommendations.10 Since 1987, a prospective protocol for inclusion and follow-up of all IE cases has been in place at our center, with a systematic registry including in-person visits, at least, annually for all patients, as well as phone consultations when needed.

Statistical analysis

Qualitative data are expressed as percentages, and continuous data as mean and standard deviation or median [interquartile range], depending on whether they follow a normal distribution. Inter-group comparisons were drawn using the chi-square test or Fisher’s exact test for qualitative variables, and the Student’s t-test or Mann-Whitney U test for continuous variables, as appropriate. Time-to-event analyses for all-cause mortality were conducted using Kaplan-Meier curves. All tests were two-sided, and P values < .05 were considered statistically significant. Statistical analyses were performed with SPSS software (version 24; IBM Corp, Armonk, NY, United States).

RESULTS

Study population and incidence rate

During the study period, a total of 778 patients successfully underwent TAVI, 70% of them with self-expanding valves. After a median follow-up of 49 months (p25-p75, 29–108 months), 10 cases of IE were eventually diagnosed, which amounts to an overall incidence rate of 1.29% (an incidence rate of 0.09/100 patient-years). Twenty-four of the 1457 patients treated with surgical bioprostheses were diagnosed with IE for an overall incidence rate of 1.64% (an incidence rate of 0.12/100 patient-years). The hazard ratio for the IE incidence rate between the 2 groups was 0.75 (95% confidence interval, 0.36–1.57). The incidence rate of IE on TAVI remained stable throughout the study period, as did the incidence of IE on surgical bioprostheses. Four of the 10 IE cases in TAVI occurred between 2008 and 2015—incidence of 1.21%—and 6 between 2016 and 2022 (incidence of 1.33%). The incidence rates of IE on surgical bioprostheses for the same periods were 1.62% and 1.67%, respectively. Clinical characteristics, treatment, and mortality rates were also similar across periods in both IE groups.

Characteristics of IE in TAVI and SAVR

Approximately half of the cases reported in both groups turned out to be early IE: 5 of the 10 TAVI IE cases and 11 of the 24 cases described on surgical bioprostheses occurred within the first year after implantation (50% TAVI and 46% SAVR; P = .56). The remaining cases were late prosthetic IE (table 1). Among TAVI IE cases, the 5 early ones were diagnosed 2, 4, 6, 8, and 11 months after implantation, while the 5 late ones were diagnosed within years 2 (3 cases) and 3 (2 cases). Among the IE reported on surgical bioprostheses, early cases were diagnosed within the first 2 months (2 cases), between months 3 and 6 (3 cases), and between months 6 and 12 (5 cases) after surgery. Regarding late cases, 5 occurred in year 2, 4 in year 3, and 4 > 3 years after surgery.

Table 1. Baseline characteristics of patients with infectious endocarditis after transcatheter aortic valve implantation or surgical aortic valve replacement

A possible source of infection different from implantation or surgery was identified in 2 of the 10 TAVI IE cases (20%) and in 2 of the 24 surgical bioprostheses. IE cases (8.3%). Only 1 of the early IE cases was associated with a possible infection source: a colonoscopy in a TAVI IE patient due to Enterococcus. None of the SAVR IE cases had an identified source, with implantation or the perioperative period being regarded as the probable sources of infection in the remaining patients. Among late IE cases, 1 TAVI IE case was associated with a dental procedure—despite proper antibiotic prophylaxis—and 2 SAVR IE cases with an upper gastrointestinal endoscopy and a dental visit, respectively. These patients’ clinical characteristics are shown in table 1. TAVI patients were older, with a median [interquartile range] age of 76 years [67–85] vs 63 years [49–77] of those treated with surgical bioprostheses (P < .001) and had higher rates of diabetes (70% vs 29%; P = .034). However, there were no significant differences in sex, other comorbidities, or symptoms. As expected, the TAVI group had a significantly higher EuroSCORE II vs the SAVR group (table 1). The profile of causative pathogens was very similar between the 2 groups (table 2). Enterococci were the pathogens most widely identified, followed by coagulase-negative staphylococci and Staphylococcus aureus, also with no significant differences being reported between the 2 groups. In 17% of SAVR patients and 20% of TAVI patients, the causative agent could not be identified (figure 1). Diagnostic echocardiographic findings of IE were also very similar between the 2 groups: transthoracic echocardiography identified IE only in half of the TAVI cases vs 37% of the cases treated with surgical bioprostheses (table 2). Transesophageal echocardiography diagnosed all IE cases in the TAVI group and slightly fewer in the SAVR group (table 2 and figure 1). In TAVI IE patients, the echocardiographic lesions identified included 9 endocardial vegetations—all affecting the leaflets—with 3 stent expansions (figure 2), 1 periaortic abscess, and 1 right ventricular perforation.

Table 2. Microbiological profile of the most common microorganisms and diagnostic lesions in the echocardiogram of patients with infectious endocarditis after transcatheter aortic valve implantation or surgical aortic valve replacement

Disease progression

The course of the disease was similar in the 2 groups (table 3). Half of the TAVI group experienced cardiac complications, as did one-third of the SAVR group, without any significant differences being reported. Half of the patients had a surgical indication due to IE (40% in the TAVI group vs 50% in the SAVR group; P = .49), but among those, only 20% of the TAVI group and 25% of the SAVR group eventually underwent surgery (P = .93) (table 3). The in- hospital mortality rate was similar (20% in the TAVI group vs 25% in the SAVR group; P = .51), without any significant differences being reported in the 1-year mortality rate, which remained high—at approximately 30%—in both groups (table 3, figure 1, and figure 3).

Table 3. Complications, rate of surgical procedures, and 1-year mortality rate in patients with infectious endocarditis after transcatheter aortic valve implantation or surgical aortic valve replacement

Figure 1. Infective endocarditis (IE) after transcatheter aortic valve implantation in a cohort of 778 patients vs a cohort of patients with surgical bioprostheses-related IE. SAVR, surgical aortic valve replacement; TAVI, transcatheter aortic valve implantation; TEE, transesophageal echocardiography; TTE, transthoracic echocardiography.

Figure 2. Transesophageal echocardiogram showing a large vegetation on the leaflets of a transcatheter aortic valve (arrows).

Figure 3. Kaplan-Meier curves showing the 1-year mortality rate in patients with infective endocarditis after transcatheter aortic valve implantation (TAVI) or surgical aortic valve replacement (SAVR).

DISCUSSION

The main finding of our study was that the incidence rate of IE on TAVI is low in our setting and similar to that of surgical aortic bioprostheses, despite being patients with higher surgical risk, older age, and more comorbidities. These results are similar to those reported in the literature (1% up to 6%);4 however, more recent large TAVI trials suggest lower rates. The PARTNER 3 trial reported annual rates of 0.2%,1 and similarly, the low-risk Evolut study2 found incidence rates of 0.1% and 0.2% at 30 days and 1 year, respectivelu, which is more consistent with our rates. Studies directly comparing the incidence of IE after SAVR and TAVI are scarce, and some have produced contradictory results.7,12 However, most observational studies based on large national databases and randomized clinical trials have found no statistically significant differences on this regard,6,13,14 even though TAVI patients are older and generally have more comorbidities, which is consistent with our results. There were also no differences in our study regarding early IE incidence (50%), which was similar in both groups, although slightly lower for IE on TAVI vs what has been reported by the literature (rates up to 64%.)15 However, a multicenter study conducted in Spain found a higher rate of early IE in TAVI vs SAVR patients (78.1% vs 39.3%; P = .001), which is significantly higher than that described in our series for the TAVI group.16 This could be explained by the different antibiotic prophylaxis regimens and procedures used at each center.

Secondly, another relevant finding from our study is that enterococci were the main cause of IE in both patient groups, which is consistent with the literature on TAVI-related IE17,18 but not on surgical bioprostheses-related IE, of which S. aureus is usually the main microorganism involved. These differences are difficult to explain, as the increase in enterococcal incidence as a causal agent in TAVI is primarily associated with older patient age and transfemoral access but still would not explain why it is also the most common pathogen in SAVR-related IE.4 Compared with a multicenter Spanish series of similar characteristics to ours,16 there are some differences: the most common causative microorganism of IE—in both TAVI and SAVR— was Staphylococcus epidermidis, followed by enterococci— less frequent than in our series—and thirdly, S. aureus. These differences could be explained by varying antibiotic prophylaxis regimens used in different settings, underscoring the critical importance of understanding the most common microorganisms involved in prosthetic IE, in general, to apply the most appropriate and effective prophylaxis regimen.

Another noteworthy aspect of this study is that echocardiographic lesions diagnostic of IE could be identified in 100% of TAVI- related IE cases vs 67% of SAVR-related IE cases in surgical pros­theses. This contrasts sharply with most reports, which indicate that the combined sensitivity rate of transthoracic and transesophageal echocardiography was 67.8% in TAVI patients, 73% in SAVR patients, and nearly 90% in native valves.19 However, a very recent study comparing patients with IE after TAVI or SAVR confirmed that vegetations were identified via echocardiogram in up to 82% of the TAVI group, more in line with our findings and significantly higher than the diagnostic rate of IE reported in surgical prostheses (62.5%; P < .001).8 Our results also differ from those reported in the literature in that most lesions found via echocardiography were vegetations, whereas other authors, such as Salaun et al.,20 reported vegetations in only 5 out of 11 cases diagnosed with TAVI-related IE, with the remaining cases being atypical lesions.

Lastly, this study also highlights the course of the disease, with a 1-year mortality rate of 30%, which is lower than that reported in other published studies (between 33% and 66%).4,12,21-23 The in-hospital mortality rate was also lower than that reported in other studies, such as the international multicenter registry of post-TAVI endocarditis21 (36% vs 20% in our series) and much lower than the Spanish multicenter trial (35%).16 These differences may be due to the wide variability in patient characteristics across studies. Of note, there were no significant differences in mortality when comparing surgical bioprostheses-related IE, unlike other series reporting lower mortality for the latter vs TAVI-related IE.23 However, a study published by Panagides et al.,8 comparing early and 1-year mortality in IE after TAVI and SAVR using propensity score matching found no significant differences between the 2. In our series, only 20% of TAVI patients with a surgical indication underwent surgery, which is consistent with other studies where surgical rates were similar to ours, with optimal medical therapy being the most common strategy.4,12,16,22,23 Some studies found no improvement in prognosis for these patients undergoing surgery, with similar mortality rates associated with optimal medical therapy,24 while a meta-analysis published by Tinica et al.25 showed that the surgical strategy was significantly superior to conservative therapy. In any case, given the lack of strong evidence, the treatment strategy for TAVI-related IE remains unclear, even if the presence of complications warrants surgical indication, and decisions should be based on local expertise.

Study limitations

Our study has the inherent limitations of its observational design, with data collected over many years, during which diagnostic criteria for IE, valve types—towards better technical models—and procedural approaches—towards less invasive methods—have evolved. The small number of IE cases in the 2 groups means results should be interpreted with caution. As this is a crude analysis, the presence of confounding bias cannot be ruled out; nevertheless, data show real-life outcomes. Additionally, patients with mechanical valves were not included, so results cannot be extrapolated to this group.

CONCLUSIONS

In our setting, TAVI-related IE has a low overall incidence rate, with no significant differences vs SAVR-related IE, despite involving older and more comorbid patients. Regarding the causative microorganism of IE—both in surgical and percutaneous surgical bioprostheses—enterococci were the most common pathogen. There were no differences in mortality between the 2 types of aortic valves, and treatment for IE was predominantly conservative.

FUNDING

None declared.

ETHICAL CONSIDERATIONS

The study was conducted in full compliance with the Declaration of Helsinki and approved by the Clinical Research Ethics Committee at the beginning of the registry in 1987. However, for patients with TAVI or SAVR, all participants signed an informed consent form authorizing the collection and analysis of their data for research purposes. Since this was an observational study, patient treatment was not impacted. For this new registry analysis, approval was obtained from our center ethics committee. Data were handled completely anonymously in full compliance with the provisions of Organic Law 3/2018 of the Spanish Data Protection Authority.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence was used in the development of this work.

AUTHORS’ CONTRIBUTIONS

A. Roldán participated in data collection and analysis. C. Urbano participated in data collection and analysis. N. Aguayo participated in patient identification, data collection, and analysis. M. Crespín, J. López, and J.C. Castillo participated in the conception of the article and its interpretation. R. González contributed to statistical analysis and result interpretation. D. Mesa and M. Ruiz contributed to data processing, analysis and result interpretation, and collaborated in the revision and preparation of the manuscript for publication. J. Perea, I. Gallo, J. Suárez de Lezo, and S. Ojeda participated in the conception of the work, helped gather patient information, and provided guidance on literature review and manuscript drafting. M. Pan and M. Anguita supervised all stages of manuscript drafting, from conception, data collection, and result interpretation, to revision, correction, and preparation of the article for submission.

CONFLICTS OF INTEREST

S. Ojeda is an associate editor of REC: Interventional Cardiology. The journal’s editorial procedure to ensure impartial handling of the manuscript has been followed. The remaining authors declared no conflicts of interest whatsoever.

REFERENCES

INTRODUCTION

Transcatheter aortic valve implantation (TAVI) has become an established option for symptomatic severe aortic stenosis across risk groups.1,2 The implantation technique is moving towards a less invasive procedure with shorter hospitalizations.3,4 Because the length of hospital-stay has been related to major adverse cardiovascular events during follow-up,5 optimizing the procedure remains a concern. Different studies have demonstrated that patients can be safely discharged 72 and even 48 hours after TAVI, and various protocols with a minimalist approach have been described so far, mainly with balloon-expandable valves.6-9 Nevertheless, the potential need for a permanent pacemaker implantation (PPI) frequently prolongs hospital admissions for the purpose of rhythm monitorization after TAVI.10,11 This issue is particularly relevant for self-expanding valves,12-14 since the rate of PPI is highest amongst these models and most early-discharge strategies have been reported for balloon-expandable prostheses.9,15

The rate of PPI can be reduced by guiding TAVI by the cusp-overlap projection (COP) to isolate the non-coronary cusp and better control prothesis implantation depth.16,17 In addition, sequential rapid atrial pacing (RAP) after deployment has been proposed to screen for atrioventricular (AV) conduction disorders.18 However, the effect of combining both preventive and diagnostic techniques on clinical outcomes after self-expanding TAVIs remains unknown. Therefore, the present study was designed to determine whether compared to conventional care, the combined use of COP and RAP can improve the management of conduction disturbances and the early outcomes after self-expandable TAVI procedures.

METHODS

Study population

This is a one-center retrospective analysis of consecutive patients who underwent transfemoral self-expandable TAVI between January 2018 to January 2022. All patients had symptomatic severe aortic stenosis and were deemed eligible for TAVI by a multidisciplinary heart team. We included all transfemoral cases irrespectively of surgical risk. From this cohort, we selected patients in sinus rhythm undergoing a self-expandable prothesis implantation procedure. We excluded patients with permanent atrial fibrillation or permanent pacemakers, and patients who did not undergo follow-up in our center. Procedures performed before 2018 were excluded from our analysis to test our hypothesis on a homogeneously contemporary series.

The institutional Ethics Committee approved the study, and all subjects provided written informed consent.

TAVI procedure

Pre-TAVI workup included baseline electrocardiogram (ECG), transthoracic echocardiography, coronary and peripheral angiography, as well as computed tomography. Implanting technique was standard among operators, following the conventional “minimal approach” previously described3,4,19: all invasive lines were minimized, and procedures were performed under minimal sedation and local anesthesia. Femoral access was obtained using ultrasound guidance and double percutaneous vascular closure (Perclose, Abbott, MN, United States) was placed before non-fractionated heparin administration. During the procedure, all patients received a temporary pacemaker via femoral venous access using a balloon-tipped bipolar catheter (5 French, Arrow, Teleflex, USA) positioned in the right ventricular apex. Pre and post valvular dilation during the intervention were performed according to the operator’s judgement. After implantation, a supravalvular angiography was routinely performed to check for paravalvular leak. At the end of the procedure, an angiography of the main vascular access was performed on all patients to verify arteriotomy closure success as well as ilio-femoral vessel integrity. Until May 2020 (standard-of-care cohort), the temporary pacemaker was kept for 24 hours with monitorization in the cardiac intensive care unit, with later transfer to the cardiology ward. Permanent pacemaker was implanted in the presence of persistent or new-onset complete or high degree AV block.

Since June 2020, the procedure was modified to routinely include: a) COP during valve deployment to assess implantation depth, and b) RAP at the end of the procedure in the absence of complete AV block, at rates of 70 to 120 beats/min (or until pacing-induced Wenckebach heart block was observed) in 10 beats/min increments for a total of 20 beats at each increment.18 In patients with persistent complete AV block, permanent pacemaker was placed in the same day. In the absence of immediate complications, patients were transferred to a standard monitored ward bed minimizing ICCU stay unless clinically needed, and temporary pacemaker surveillance was only maintained for 24 hours in the presence of Wenckebach phenomenon or in patients with AV block until permanent pacemaker was implanted. Final determination of patient discharge was based on patient’s characteristics and clinical status by the referring physician.

Follow-up

After implantation, assessment included ECG, echocardiogram, in-hospital, and 90-day clinical follow-up. All patients’ clinical, procedural, in-hospital information, and follow-up data were collected from the electronic medical records.

Study endpoints

The primary safety and efficacy outcomes were complete AV block after implantation with COP and the diagnostic performance of RAP added to COP in self-expandable valves.

The secondary endpoints included the length of hospitalization following TAVI and the composite major adverse cardiovascular events, defined as cardiovascular mortality, PPI, rehospitalization for heart failure decompensation, stroke, and major bleeding and vascular complications occurring 90 days after hospital discharge. In addition, rehospitalizations for healthcare-associated infections within the same period were analyzed separately.

Statistical analysis

Categorical variables are presented as counts (percentages) and were compared using the chi-square test. Continuous variables were tested for normality of distribution by using the Shapiro-Wilk test. According to their distribution, continuous variables are expressed as mean ± standard deviation or median values with interquartile range [IQR], and were compared using Student t test, ANOVA or Wilcoxon test, as appropriate. For the multivariate analysis, we first selected baseline variables that showed a significant association with the outcome (P < .2) and performed single logistic regression. Variables with a P-value < .05 were entered into a multiple logistic regression analysis. Odds ratios (OR) were calculated using uni and multivariate logistic regression, along with 95% confidence intervals. Statistical analyses were conducted using RStudio 4.1.20 P-values < .05 were considered significant.

RESULTS

Baseline characteristics

A total of 273 patients met the inclusion criteria for the analysis. Among them, 134 underwent standard implantations, and 139 underwent the combined use of COP and RAP (COP+RAP) (figure 1). Mean patient age was 81 ± 7 years old and 51.3% were female, with similar prevalence of cardiovascular risk factors and comorbidities between groups. Patients in the standard-of-care group showed a worse NYHA functional class, and there were no differences in the surgical risk (table 1). Both cohorts presented similar rates of conduction disturbances at baseline. The mean transvalvular pressure gradient by ultrasound was 47 [41-58] mm Hg with no significant differences on baseline findings between groups. The COP+RAP group showed a higher valvular calcium score (2429 [1577-3557] Agatston units) and larger left ventricular outflow tract and annular perimeters (P = .006 and P = .04 respectively) (table 1).

Figure 1. Patient flow chart. AF, atrial fibrillation; COP, cusp-overlap projection; RAP, rapid atrial pacing TAVI, transcatheter aortic valve implantation.

Table 1. Demographics, baseline electrocardiographic and imaging characteristics

Procedure and in-hospital outcomes

Procedural details are listed in table 2. All patients underwent TAVI under conscious anesthesia, with more frequent use of midazolam and fentanyl in the COP+RAP cohort in place of dexmedetomidine (P < .001). Also, in line with current recommendations for self- expandable valves, the rate of predilation was significantly higher in the COP+RAP group (P < .001).

The occurrence of intraprocedural conductance disorders was lower with COP in the COP+RAP group (25.8% vs 14.4%; P = .007; OR, 0.49; 95%CI, 0.27-0.91; P = .026) mainly due to a reduction in temporary complete AV block (table 2), whereas rates of new-onset first degree AV block and atrioventricular block (LBBB) remained similar between groups. Same day permanent pacemaker was implanted in 16 patients of the COP+RAP group with persistent AV block. RAP revealed Wenckebach phenomenon at a median stimulation rate of 115 bpm in 24 patients and allowed to reduce the need for temporary pacemaker surveillance from 91.8 % in the standard-of-care group to 28.1 % of the cases in the COP+RAP group (P < .001). Among the patients in the COP+RAP cohort who did not develop Wenckebach phenomenon (n = 98/122, 80.3%), permanent pacemaker was required in only 3% of patients (OR, 0.16; 95%CI, 0.03-0.76; P = .02; negative predictive value = 97%; 95%CI, 91-99%. Patients with pacing-induced Wenckebach had a higher risk of PPI during admission (OR, 6.33; 95%CI, 1.30-34.29; P = .021; positive predictive value 79% 95%CI, 58-93%).

Table 2. Procedural characteristics, electrocardiographic and imaging outcomes

Incidence of AV block and timing of PPI after TAVI is shown in figure 2. Delayed AV block (developed > 48 hours after TAVI) during admission occurred in 8.2% patients in the standard group and 3.6% of the COP+RAP group (P = .14) mostly in the presence of new-onset LBBB (68% of the cases). Adding both maneuvers did not impact the procedure length or led to adverse events, with a low and similar rate of in-hospital complications among groups (table 3). Additionally, reduction of the implantation depth did not lead to a higher rate of residual significant aortic regurgitation (P = .32).

Figure 2. Immediate complete AVB and on-admission permanent pacemaker implantation trends. Upper: intraprocedure cusp-overlap projection and Wenckebach phenomenon during atrial pacing. Midd: comparison of immediate AVB in standard and protocol groups. Bottom: trend in pacemaker implantation during admission. AVB, atrioventricular block; COP, cusp-overlap projection; RAP, rapid atrial pacing TAVI, transcatheter aortic valve implantation.

Table 3. Immediate, and short-term outcomes

The reduction of stay in the intensive care unit along with the lower rate of 24-hour temporary pacemaker surveillance, significantly decreased the length of hospitalization in the COP+ RAP cohort (−3.0 [−2.5 to −3.5] days, P < .0001), enabling to discharge patients within 48 hours in 56.8% of the cases compared to 7.5% in the standard-of-care group (P < .001).

90-day outcomes

At 90-days, the cumulative rate of PPI in the COP+RAP cohort persisted lower than in the standard-of-care group (29.5% vs 18.7%; P = .049). Of all 65 pacemaker implantations, 58 (89%) occurred during the index admission and 7 (11%) following hospital discharge. Among the 3 patients from the COP+RAP requiring late PPI, 2 developed with new-onset LBBB (QRS width 127 and 136 ms respectively), and both had displayed Wenckebach phenomenon at 120 bpm. They were monitored with temporary pacemaker for 24 hours without events but developed high degree AV block within 48 hours after discharge (day 4 after TAVI). The third patient had new-onset first degree AV block (PR interval 240 ms), Wenckebach phenomenon at 120 bpm and narrow QRS (118 ms) but developed high-degree AV block at day 55. No patients with a negative RAP test required late PPI. The univariate logistic analysis of potential baseline and procedural factors of PPI after TAVI is reported in table 4. Interestingly, differences in sedation, predilation, valve model (P = .98) and bicuspid valves (P = .71) did not influence the risk of PPI after TAVI. In the multivariate analysis, prior 1st degree AV block, baseline right bundle branch block, prosthesis posdilation and new-onset LBBB were associated with PPI. The combination of COP with RAP decreased the need of PPI on a 90-day follow-up (OR, 0.48; 95%CI, 0.24-0.92; P = .031) (figure 3).

Figure 3. Odds plot of multivariate analysis of 90-day permanent pacemaker implantation predictors. 95%CI, 95% confidence interval; AVB, atrioventricular block; COP, cusp-overlap projection; LBBB, left bundle branch block; PPI, permanent pacemaker implantation; RAP, rapid atrial pacing; RBBB, right bundle branch block.

Table 4. Main factors associated with PPI after TAVI: unadjusted logistic regression analysis of PPI predictors at 90 days

The COP+RAP group presented a significantly lower percentage of the secondary composite endpoint than the standard-of-care cohort (20.1% vs 10.1%; P = .031). The rates of HF readmission, and mortality were low and similar across groups during the follow-up period (table 3). None of the 3 deaths during follow-up were due to rhythm disturbances. At 90 days, there was a reduction in the number of readmissions due to admission-related infections in the COP+RAP cohort (OR, 0.35; 95%CI, 0.12-0.89 P = .036) (figure 4), with an 8.6 % reduction in the odds of infection for each day of reduced hospitalization (OR, 0.91; 95%CI, 0.85-0.99; P = .023). The most common were urinary (47.4%) and respiratory (42.1%) infections. None of the patients presented infections related to PPI.

Figure 4. Central illustration. Main immediate and short-term findings of combining cusp-overlap projection and rapid atrial pacing when implanting self- expandable valves. OR, odds ratio; TAVI, transcatheter aortic valve implantation.

DISCUSSION

In the present study we demonstrate that combining the COP with RAP, COP reduces the risk of complete AV block after self- expanding TAVI and negative RAP after implantation adequately dismiss underlying AV conduction disturbances requiring PPI.

This strategy reduces the need of temporary PPI and the appearance of late onset AV block, and safely allows an early discharge in sinus rhythm patients receiving a self-expandable valve. Reducing hospital stays lead to a significant reduction in admission-related infections at short-term follow-up (figure 4).

A minimalistic approach in TAVI has become progressively relevant as a way of simplifying the intervention and promoting early recovery.4,6-8 Multiple studies have shown that early discharge after TAVI is possible, preserving the high standards of effectiveness, patient safety and outcomes.3,9,21 However, most of these studies included highly selected patients, with a predominance of balloon-expandable valves, and might not be representative in the current TAVI scenario. Evidence at the beginning of minimalistic TAVI inception found self-expanding prothesis to be a predictor for rhythm disturbances and delayed discharge.5,6 Since their risk of AV conduction disturbances remain in the range of 5% to 30% even with contemporary transcatheter heart valves10,12-14 and there is no appropriate way to assess the risk of AV block development, self-expandable valves have been underrepresented in early discharge protocols so far.3,7,22,23

Postimplantation conduction disturbances remains one of the current weaknesses in TAVI, and great heterogeneity of recommendations persist regarding their management.24,25 In this regard, the aim of our study was to minimize the appearance and improve the assessment of conduction disturbances without increasing invasiveness in patients receiving self-expandable valves, by combining 2 well-described maneuvers.

Guiding the procedure using the COP has already been established as an effective method to guide implantation depth.16,17 Isolating the non-coronary cusp allows a better identification of the most inferior point of the aortic annulus, leading to a higher (more aortic) valve positioning, which in turn reduces the risk of post-TAVI conduction disturbances. Various observational studies have described the COP technique in distinct types of self-expandable valves,15-17,26 and highlighted its potential benefits in decreasing the PPI rate in the Evolut (Medtronic, United States) and the Portico (Abbott Structural Heart, United States) valves.27 In line with these studies, in our cohort, COP significantly decreased the emergence of intra-procedural transient and late onset ( > 48 hours) high-grade AV block irrespectively of the self-expandable valve type, with no noticeable impact on the rate of new-onset LBBB, that remained around 29% in both cohorts.

After implantation, current consensus documents recommend temporary pacemaker surveillance for at least 24 hours. Further management depends on baseline intra-procedure, and post-procedure conduction disturbances,25,28 that can result in longer duration of temporary pacemaker placement or prolonged inpatient rhythm monitoring. In this context RAP, emerges as an easy and interesting strategy to discriminate which patients may require monitoring. Since its publication in 2020,18 different works have recognized the underuse of this technique, although is an important component of routine electrophysiological studies to “stress” the AV conduction.10 It has the limitation of assessing specifically at the level of the atrio-His interval but seems to be a good indicator of patients with underlying conduction disorders with a described strong negative predictive value for PPI in the absence of Wenckebach phenomenon. In our study, the rate of Wenckebach phenomenon during dynamic atrial stimulation was 19.8% and showed a moderate positive predictive value. However, most importantly, the absence of RAP-induced AV block displayed a very high negative predictive value of PPI during follow-up. These results are in agreement with the previous report of RAP on balloon-expandable valves,18 showing adequate reliability of AV Wenckebach testing over delayed conduction disturbances immediately at the end of the procedure in self-expandable valves. The improvement in valve deployment and risk stratification of AV conduction disturbances allowed us to reduce temporary pacemaker surveillance in 63.1% of cases, with less late-onset high-degree AV block and no impact in procedure length or complications, leading to early recovery and minimizing functional decline. Moreover, although early discharge protocols described so far have focused on demonstrating safety, it should be noted that in our simplified protocol cohort there was a significant reduction in the healthcare-related infection readmission rate at 90 days. The reduction in the incidence of admission-related respiratory and urinary infections was likely attributable to the shortened length of hospitalization, that has been associated with adverse events.5 Observed differences in major adverse cardiovascular events, although significant, should be interpreted cautiously, as part of the findings might be a result of technological advances and cumulative expertise.

The adequacy of these combined maneuvers was patent in the multivariate analysis and in the maintained differences in PPI rate at a 90-day follow-up. Nevertheless, special caution should be recommended in patients developing persistent LBBB after TAVI, as it was a predictor of delayed AV block and PPI. These patients would probably benefit from longer monitoring, and in the presence of new-onset LBBB or Wenckebach-induced phenomenon, a formal electrophysiological study or continuous ECG monitoring might have the potential to risk-stratify patients with unclear indications of PPI following TAVI.29-31

Limitations

This is a single-center study with a modest sample size and an observational design; therefore, there is an inherent potential bias due to patient selection. First, both cohorts are not contemporary, which may introduce bias related to technological advancements or increased operator experience over time. Second, RAP is restricted to patients in sinus rhythm, and the cutoff of 120 beats/min for maximum RAP rate was selected according to previous literature but may be insufficient as low risk patients tend to be young and might request testing at higher heart rates. Also, RAP stresses the AV conduction at the atrio-His level, and therefore our results might not properly assess patients with broadened QRS, as high-grade conduction block occurs at the His-Ventricular segment.18,30,31 Third, our protocol has the inconvenience of requiring temporary pacemaker so it can be moved to the atrium at the end of the procedure, which is not feasible in procedures performing pacing through the left ventricular wire. However, we believe it might be a safe and efficient way to risk-stratify patients immediately after the procedure. Finally, our study was focused on patients in sinus rhythm and self-expanding prothesis and therefore the results and conclusions should be interpreted in this context.

CONCLUSIONS

Combining COP during deployment with RAP immediately after TAVI diminishes the incidence of postprocedural AV block and improves the screening of underlying atrio-His conduction disturbances in patients receiving a self-expandable valve. In turn, this allows to reduce the length of temporary pacemaker surveillance and admission days, with differences in PPI rate maintained at a 90-day follow-up. Optimizing strategies of care after TAVI has a significant impact at short-term, driven by a significant reduction in readmissions due to infections.

FUNDING

M. Tamargo was partially supported by grants from the Fundación para la Investigación Biomédica Gregorio Marañón, Spain, and Rio Hortega CM20/00054 from the Instituto de Salud Carlos III, Spain. J. Bermejo was partially supported by INT22/00025 from the Instituto de Salud Carlos III, Spain.

ETHICAL CONSIDERATIONS

The institutional Ethics Committee approved the study, and all subjects provided written informed consent. This manuscript complies with the guidelines of the SAGER guidelines, and possible gender and/or age differences have been taken into account, without significant differences being observed.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence has been used in the development of this paper.

AUTHORS’ CONTRIBUTIONS

All authors comply with the international characteristics of authorship of scientific articles. Conception and design: M. Tamargo, J. García Carreño, E. Gutiérrez, J. Bermejo, F. Fernández-Avilés. Data collection: M. Tamargo, M. Huanca, E. Ludeña, J. García Carreño. Analysis: M. Tamargo E. Gutiérrez, J. Bermejo. The manuscript was reviewed and approved by all authors.

CONFLICTS OF INTEREST

Nothing to disclose.

REFERENCES

INTRODUCTION

In clinical research, statistical methodologies used for survival analyses range from the easiest non-parametric models—such as Kaplan-Meier estimates—to semi-parametric models such as the Cox proportional hazards model.1 When multiple adverse events are of interest, it is common practice to create a composite variable, such as major adverse cardiovascular events (MACE), to indicate whether a patient experienced any event at the follow-up, facilitating the application of these models,2 which provide numerous advantages with some associated limitations; for instance, they usually only consider the time to the index event for each patient, regardless of which component of the composite variable triggered such an event, which complicates the interpretation of the intervention effect since reliable conclusions cannot be drawn about the effects on individual components due to potential bias from competing risks.3,4

For this reason, multi-state models have gained traction in recent years, providing a framework to analyze disease progression.4 In medical applications, states in a multi-state model can represent various adverse events that patients may experience over time.5 A multi-state model is defined by its state structure, consisting of states and transitions across them. This structure allows for defining certain states as absorbing (from which the patient cannot exit, such as death) or transient (intermediate states between the initial and the absorbing states). These models extend competitive risk models—a multi-state model with 1 initial and multiple mutually exclusive absorbing states—by enabling any state structure, for example, extending analysis to what happens after a transient event.6 Additionally, they allow assessment of variables that impact the patients’ probability of transitioning from one state to the other by modeling these transitions, which is particularly useful in long-term clinical trials.7

Multi-state models can incorporate several covariates, such as demographic characteristics or biomarkers, to assess their effects on event rates and time-to-event. This aids in identifying risk factors and understanding their impact on patient prognosis, thus facilitating the efficacy evaluation of various treatments or interventions, and the selection of the most suitable strategy for each patient.8-10

The aim of this study is to model disease progression in the patients of a cardiology study by using a multi-state model and evaluating its applicability and limitations.

METHODS

Data

The database used is the updated version of the multicenter and retrospective PACO-PCI registry (Antithrombotic strategies in elderly patients with atrial fibrillation revascularized with drug-eluting stents),11 which included a total of 1057 patients older than 75 years with atrial fibrillation on oral anticoagulant therapy after revascularization with drug-eluting stents from 2015 through 2019. The endpoints of this registry included MACE (death, acute myocardial infarction [AMI], revascularization) and bleeding 12 months after treatment. Updated data extend patient follow-up to 5 years. Previous results from the PACO-PCI study11 demonstrated the efficacy of various antithrombotic therapies regarding the onset of MACE and major bleeding events. This study uses such data to illustrate the application of multi-state models, focusing on factors influencing the occurrence of the events of interest to achieve a more individualized model of disease progression.

Data analysis

The multi-state model used includes 4 states (1 initial state, 2 transient ones, and 1 absorbing state) and the possible transitions across them (figure 1). Specifically, a patient enters initial state 1 (treatment) at the time of the intervention. From this state, they can transition to transient state 2 (bleeding) if a major bleeding event occurs, transient state 3 (AMI/revascularization) if they experience an AMI or require re-intervention, or absorbing state 4 (death) if they die. From state 2 (bleeding), patients can transition to state 3 (AMI /revascularization) if they experience an AMI or require re-intervention or vice versa if a new bleeding event occurs. Patients can transition to the absorbing state from any state if they die. Compared with traditional methods—event composition and competing risks—this model distinguishes the severity of adverse events while maintaining a certain simplicity.

Figure 1. Proposed multi-state model. Upon intervention, patients enter the initial state 1 (treatment), from which they can move to transient state 2 (bleeding) if they experience a major bleed, or transient state 3 (acute myocardial infarction [AMI] or revascularization) if they experience an AMI or require re-intervention, or state 4 (absorbing, death) if they die. Patients in state 2 (bleeding) can transition to state 3 (AMI or revascularization) if they experience an AMI or require re-intervention. Patients in state 3 (AMI or revascularization) can move to state 2 (bleeding) if they experience a new bleed and can also transition to state 4 (death) if they die. The number of patients experiencing each adverse event is indicated alongside each transition, based on an initial cohort of 1057 patients. CKD, chronic kidney disease; LMCA, left main coronary artery; LVEF, left ventricular ejection fraction.

The multi-state model was adjusted using the msm12 package for R,13 which employs an exponential model for the time spent in each state. This package allows fitting a general multi-state model to survival data, requiring a complete data matrix; missing data for quantitative variables were completed with the corresponding mean value.

The proposed model for the time spent in each state allows including factors affecting each transition. For complete model determination purposes, variables associated with each transition were selected, ie, factors influencing the probability of transitioning from one state to the other. We chose a starting set of variables that could impact the occurrence of adverse event based on clinical criteria, as shown in table 1 of the supplementary data. This set includes the most important baseline characteristics, the number of vessels treated, and the scores obtained on the PreciseDAPT,14 HAS-BLED,15 and CHA²DS²-VASc16 scales. Afterwards, a multi-state model was adjusted including these variables each one at a time to identify their individual influence on each transition. Results of this analysis are shown in table 1 of the supplementary data. Subsequently, we tested different combinations of influential variables to achieve models with the best fit based on the Akaike information criterion, which favors model fit with the fewest covariates.12 Finally, we selected the model with the lowest value for this criterion that provided the most clinically relevant information.

Table 1. Covariates selected for the survival model in each state transition

Afterwards, we conducted a goodness-of-fit study of the multi-state model to determine whether the exponential model adequately fit the observed time in each state. This analysis revealed that the model overestimates event-free survival after 1000 days (slightly more than 2.5 years), mainly because it underestimates the prevalence of death beyond this period. Therefore, a maximum follow-up of 1000 days was considered for the final analysis.

Furthermore, we conducted a traditional survival analysis to compare it with our model. Specifically, a Cox regression model was fitted for the MACE variable, defined as AMI, revascularization, bleeding, or death. We conducted Univariate Cox regression analyses to determine factors affecting the occurrence of MACE (table 1 of the supplementary data) using the same initial set considered for the multi-state model. Based on these results, we selected a subset of variables to adjust a multiple Cox regression model. Again, the Akaike information criterion was used to select the best model among all possible models. We performed all calculations with the statistical program R, version 4.1.1; in particular, the Survival17 package was used for the above-mentioned traditional survival analysis.

RESULTS

The database includes information on 20 Spanish centers for a total of 1057 patients older than 75 years who underwent percutaneous coronary intervention with drug-eluting stents from 2015 through 2019. The patients’ mean age is 81 ± 4.2 years, and most (almost 70%) are men. Diabetes mellitus—a known risk factor for various cardiovascular diseases—was present in 42.4% of the population, and most patients (about 80%) had experienced a prior cardiovascular event. The patients’ baseline characteristics are shown in table 2 of the supplementary data. Only 5 variables had missing data: anemia (4.4%), chronic kidney disease (CKD) (0.9%), left ventricular ejection fraction (LVEF) (3.2%), PreciseDAPT score (0.6%), and treated left main coronary artery disease (LMCAD) (< 0.1%).

Table 2. Results of the Cox multiple regression model for major cardiovascular adverse events (1000-day follow-up)

The mean follow-up was 854.8 days (2 years and 4 months), with the shortest follow-up being 2 days and the longest one, 2018 days. Figure 1 and table 3 of the supplementary data illustrate that death is the most common event among patients (14.1%), followed by major bleeding (10.6%). After the intervention and stent treatment, a significant number of patients experience a new AMI or require re-intervention (7.9%) as their first adverse event.

Table 3. Characteristics (risk factor values) of hypothetical patients used to demonstrate the predictive capabilities of the multistate model

Multi-state model

We obtained an estimate and 95% confidence interval (95%CI) of the hazard ratio (HR) for each variable and transition following the variable selection process. Table 1 shows the estimated risk associated with each variable in each transition. For the PACO-PCI study data, the resulting multi-state model revealed, for example, that a higher score on the PreciseDAPT scale increases the risk of bleeding after treatment (HR, 1.05; 95%CI, 1.03–1.06), and that LVEF is a protective factor vs after bleeding (HR, 0.95; 95%CI, 0.92–0.97). The transition from treatment to death is influenced by the number of vessels treated and LVEF, and by the PreciseDAPT and HAS-BLED scores. The transition from treatment to bleeding is related to anemia and the PreciseDAPT score. After a bleeding event, the likelihood of experiencing a new AMI or revascularization is associated with treated LMCAD. The transition from bleeding to death depends on the LVEF and previous coronary artery bypass graft. The transition from treatment to AMI or revascularization is related to diabetes and the PreciseDAPT score. After a new AMI has occurred or revascularization has been performed, the likelihood of bleeding is influenced by the HAS-BLED score. Lastly, the transition from AMI or revascularization to death is determined by CKD and LVEF.

Comparison between the multi-state model and Cox regression analysis

The results of the MACE variable study with a Cox regression analysis—which provides the HR for each MACE predictor—are shown in table 2. The factors included in the best multiple Cox regression model were diabetes, CKD, anemia, the PreciseDAPT and HAS-BLED scales, LVEF, and the number of vessels treated (Table 2).

Although the variables treated LMCAD and previous coronary artery bypass graft were not significant predictors of MACE in the multiple Cox regression analysis, they were significant for some transitions in the multi-state model. Diabetes, CKD, anemia, the PreciseDAPT and HAS-BLED scales, LVEF, and the number of vessels treated were significant predictors in the univariate Cox regression analysis (table 1 of the supplementary data) and for some transitions in the multi-state model.

Utility of the multi-state model

In contrast to the Cox regression model, a fitted multi-state model, like the one proposed, can predict the probability of a patient transitioning across states within a specified period of time, that is, the probability of experiencing each type of event after treatment or after experiencing another transient event within a specified timeframe. For example, it is possible to calculate the probability that a patient with certain baseline characteristics who has experienced major bleeding will die within 1 year.

To illustrate the predictive capability of the model, we defined 2 types of patients—low- and high-risk—whose characteristics are shown in table 3. We used the multi-state model to predict the probability of each of these hypothetical patients in each of the possible transitions within the first year after treatment or after an exit event. These predictions for the 2 types of patients are shown in figure 2. For example, the PACO-PCI data reveal that that the probability rates of death 1 year after major bleeding are 75% and 10% for high- and low-risk patients, respectively.

Figure 2. Event-free survival graphs (A, E) and survival graphs (B, C, D, F, G) show the probability of low- (green) and high-risk (red) patients experiencing an adverse event within 1000 days (a little more than 2.5 years). A: probability of bleeding after treatment. B: probability of experiencing a new acute myocardial infarction or revascularization after treatment. C: probability of death after treatment. D: probability of a new acute myocardial infarction or revascularization after bleeding. E: probability of bleeding after a new acute myocardial infarction or revascularization. F: probability of death after bleeding. G: probability of death after a new acute myocardial infarction or revascularization.

DISCUSSION

Results demonstrate the added value of multi-state models in survival analyses within biomedical research. Multi-state models introduce additional predictive variables beyond those identified by traditional survival analyses, and provide information on the expected time and probability of transitioning from one state to the other based on risk factors, treatment characteristics, and previous disease progression. Traditional analyses only provide information on general significant variables, without clarifying which specific adverse event they predict.

In a prior study, a multi-state model with a 3-state structure was applied to data from the Synergy ACS study,9 selecting the most determinant variables for each type of adverse event. Specifically, diabetes mellitus, the number of diseased vessels, and CKD were analyzed in relation to the time elapsed from treatment administration to the occurrence of a new AMI or revascularization; age, LVEF, and previous percutaneous coronary intervention for the time elapsed from treatment administration to death; and diabetes mellitus, the number of diseased vessels, and stent thrombosis for survival from post-treatment AMI or revascularization. In the PACO-PCI study data10 given the patients’ advanced age and baseline characteristics, we observed a high probability of bleeding after treatment, so this variable was included as a transient state in the model. There are common predictors in the 2 studies, such as the number of diseased or treated vessels, LVEF, and CKD, though not all factors corresponded to the same transition in the multi-state model. Moreover, it is notable that each database includes unique variables not found in the other.

In the current dataset, factors such as age and stent thrombosis are not statistically significant due to the patient profile and the fact that most experienced a prior adverse event. Consequently, predictive scales such as the Precise-DAPT and the HAS-BLED—which include multiple events—are crucial regarding adjusting the model.

Multi-state models have been used in other cardiology studies2,6,18-20 with different state structures. In heart failure, using the model applied by Upshaw et al.18, both LVEF and diabetes mellitus were found to be predictors of death. CKD is related directly to death and to death following hospitalization for heart failure. Postmus et al.1 used a multi-state model that was similar to the disability model to predict hospitalization for heart failure and death, identifying AMI, diabetes mellitus, LVEF, and CKD as predictors.

The proposed multi-state model has certain limitations. Regarding data, it is a retrospective observational registry affected by the limitations of all observational studies. Specifically, the most significant limitations of this study are: 1) the heterogeneity of follow-up, which can introduce significant biases; 2) its limited statistical power for a model with 7 transitions; and 3) its retrospective design without event adjudication, implying that many deaths may have been due to unreported ischemic or hemorrhagic events. It is also worth noting that the variables included in the model were selected not only based on statistical criteria but also subjectively by the researcher, meaning that results should be interpreted with caution. Although the management of missing data through multiple imputation would have accounted for variability due to data loss, model selection with missing data in multi-state models has not yet been resolved in the literature.

Finally, multi-state models are currently widely used in fields outside the cardiovascular clinical trial,2 hematology,21 and oncology settings.22,23 Despite their proven utility, there are 3 main limitations in performing multi-state model analysis. First, the msm package in R assumes the Markov property, meaning that in our model, survival after a transient event does not depend on the time from the initial intervention to the corresponding event. Second, multi-state models require sufficient observed events to have statistical power and make reliable predictions. Third, most software for multi-state model analysis is integrated into statistical packages and is not easy to use; for example, each requires a different data structure. Interested readers can consult a systematic review of existing programs.24

CONCLUSIONS

Multi-state models are essential for describing disease progression due to their capacity to adapt to various events or factors through their state structure. Another advantage is that they consider all available follow-up data, including patients who may have or experienced an event. Additionally, they provide information on the estimated time to an event along with the probability of transitioning across states, making them an essential tool in cardiovascular event analysis by providing more accurate estimates of future event risk.

FUNDING

None declared.

ETHICAL CONSIDERATIONS

The original study (PACO-PCI) was approved by the reference CEIm of the Health Areas of León and El Bierzo (Spain) on 11-26-2019, reference no. 19167. Since this study involved new statistical analysis of observed results without new tests or data collection, ethical committee review was deemed unnecessary.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence was used in the development of this article.

AUTHORS’ CONTRIBUTIONS

All authors contributed equally to the design of the multi-state model. J.M. de la Torre-Hernández and J.L. Ferreiro provided the data. N. Montoya and A. Quirós conducted the data analysis and model implementation. N. Montoya, A. Quirós, and A. Pérez de Prado drafted the manuscript, and all authors substantially contributed to the review process.

CONFLICTS OF INTEREST

J.M. de la Torre-Hernández is the editor-in-chief of REC: Interventional Cardiology; A. Pérez de Prado is an associate editor of REC: Interventional Cardiology; in both cases, the journal’s editorial procedure to ensure impartial handling of the manuscript has been followed. The remaining authors declared no conflicts of interest whatsoever.

ACKNOWLEDGMENTS

We wish to thank all researchers of the PACO-PCI registry.

SUPPLEMENTARY DATA

REFERENCES

INTRODUCTION

Percutaneous coronary intervention (PCI) is the first-line therapy in patients with ST-segment-elevation myocardial infarction (STEMI).1 The STEMI-related culprit vessel is usually considered as one of the most important prognostic factors in STEMI patients.2,3 This assumption comes from previous studies –conducted in the pre-reperfusion or thrombolysis era– which showed that left anterior descending artery (LAD)-related STEMIs were associated with worse clinical outcomes compared with right coronary (RCA) and left circumflex artery (LCX)-related lesions.4-9

However, in the contemporary era of primary PCI there are limited data about the prognostic impact of LAD as the STEMI-related culprit vessel especially in a very long follow-up.10,11

Therefore, the aim of this study was to investigate the impact of the LAD as the STEMI-related culprit vessel on very long-term clinical outcomes in STEMI patients undergoing primary PCI enrolled in the EXAMINATION-EXTEND study (10-year follow-up of the EXAMINATION trial).

METHODS

Study design and patients

The EXAMINATION trial (NCT00828087) was an all-comer, multicenter, prospective, 1:1 randomized, 2-arm, single-blind, controlled trial conducted at 12 centers across 3 countries to assess the superiority of EES (Xience V) vs BMS (Multilink Vision, Abbott Vascular) in STEMI patients regarding the primary endpoint of all-cause mortality, any myocardial infarction, and any revascularization at 1 year. The study had broad inclusion criteria and few exclusion criteria to ensure an all-comer STEMI population representative of the routine clinical practice. The study outcomes have been reported up to the year 5.12,13 After that, it was reinitiated as the EXAMINATION- EXTEND study to evaluate patient- and device-oriented composite endpoints at 10 years. The latter is registered at ClinicalTrials.gov (NCT04462315) as an investigator-driven extension of follow-up of the EXAMINATION trial. An independent study monitor (ADKNOMA, Barcelona, Spain) verified the adequacy of the extended follow-up and events reported. All events were adjudicated and classified by an independent event adjudication committee blinded to the therapy groups (Barcicore Lab, Barcelona, Spain). The 10-year primary endpoint results of the EXAMINATION-EXTEND study have been previously published.14 For the aim of this study, baseline, procedural characteristics and outcomes were stratified according to the STEMI-related culprit vessel (LAD vs others). All centers participating in the EXAMINATION trial received the approval of their Medical Ethics Committee, and all enrolled patients who had already signed their written informed consent forms. Medical ethics committee approval for EXAMINATION- EXTEND was granted at the institutions of the principal investigators (Hospital Clínic and Hospital Bellvitge, Barcelona, Spain), and the requirement to obtain informed consent to gather information on 10- year events was waived. The study complied with the Declaration of Helsinki.

Study endpoints

The primary endpoint of this study was the patient-oriented composite endpoint of all-cause mortality, any myocardial infarction, or any revascularization at 10 years. Secondary endpoints were each individual components of the primary endpoint, device-oriented composite endpoint (cardiac death, target-vessel myocardial infarction, target lesion revascularization), its individual components and stent thrombosis. Detailed descriptions of the study endpoints and definitions have been published previously.15

Statistical analysis

Continuous variables are expressed as median (interquartile range; IQR), and categorical variables as absolute and relative frequencies (percentages).

Baseline clinical, angiographic, and procedural characteristics were compared between the groups stratified by the STEMI-related artery (LAD vs other vessels) using the Wilcoxon rank sum test, the chi-square, or Fisher’s exact test, where appropriate.

Time-to-event curves for POCE and all-cause death were plotted using the one minus the Kaplan-Meier estimate and the cumulative incidence function for other outcomes. The incidence of events at the follow-up was compared between groups using log-rank or Grey’s test. Landmark analyses were also performed, setting landmark points at 1 and 5 years.

The association between LAD as a STEMI-related culprit vessel and events was analyzed in univariable and multivariable cause-specific Cox regression models. Covariates were added to the multivariable model in 2 blocks. The first model included all clinically relevant baseline characteristics variables with P < .1 in the between-groups comparison (LAD vs other vessels), i.e., sex, smoking status, peripheral vascular disease, previous PCI, previous CABG, previous MI, and Killip class. The second model (expanded adjustment) included both the baseline characteristics and the left ventricular ejection fraction (LVEF) at discharge.

Two-tailed P-value < .05 was considered statistically significant. All statistical analyses were performed using R (R Core Team (2022). R: a language for statistical computing. R Foundation for Statistical Computing, Austria) with the following packages: survival, tidycmprsk, jskm, and gtsummary.

RESULTS

Patient characteristics

In 631 (42%) out of the 1498 STEMI patients included in the EXAMINATION EXTEND trial, the LAD was the culprit vessel (LAD-STEMI group), whereas in 867 patients (58%) it was not (non- LAD-STEMI group). Patients’ inclusion flowchart is shown in figure 1.

Figure 1. Study flowchart. A total of 1498 patients were initially recruited. At 10 years, clinical follow-up was obtained in 95.2% of the patients. LAD, left anterior descending artery; STEMI, ST-elevation myocardial infarction.

LAD-STEMI group had a higher incidence of active smokers, advanced Killip class and more depressed LVEF vs the non-LAD-STEMI group, which, however, exhibited a higher incidence of peripheral vascular disease, previous MI and previous PCI (table 1). Also, although non-statistically significant, the frequency of late comers and bailout PCI was numerically higher in the LAD-STEMI group.

Table 1. Baseline clinical characteristics

Regarding procedural data, LAD-STEMI group received smaller stent diameter (3.12 mm vs 3.26 mm; P = .001) and had a lower incidence of ST-segment resolution than the non-LAD-STEMI group (73%, vs 50%; P = .001) (table 2). The use of GP IIb/IIIa inhibitors was numerically lower in the LAD-STEMI group, although the differences between groups were not statistically significant. Of note, almost half of the patients (46%) with LAD-STEMI had the lesion in the proximal LAD compared with 44% of them who had it in the mid/distal LAD.

Table 2. Angiographic and procedural characteristics

Ten-year outcomes

At the 10-year follow-up, POCE did not differ between LAD-STEMI and non-LAD-STEMI group (adjusted HR, 0.95; 95%CI, 0.79-1.13; P = .56) (figure 2). Moreover, no differences were found in terms of each individual component of POCE (all-cause mortality, MI, any revascularization) (figure 3) and other secondary endpoints (figure 1 of the supplementary data). Furthermore, when the expanded adjustment was performed and LVEF was included in the multivariable analysis, there were no inter-group differences between (table 3).

Figure 2. Central illustration. Outcomes of patients with ST-segment elevation myocardial infarction according to the culprit vessel at the 10-year follow-up. LAD, left anterior descending coronary artery; STEMI: ST-segment elevation myocardial infarction; POCE: patient-oriented composite endpoint.

Figure 3. Time-to-event curves for the patient-oriented composite endpoint (A), all-cause mortality (B), myocardial infarction (C), and any revascularization (D) in patients stratified according to the culprit vessel. LAD, left anterior descending coronary artery; MI, myocardial infarction; STEMI, ST-segment elevation myocardial infarction; POCE, patient-oriented composite endpoint.

Table 3. Ten-year outcomes

Landmark analyses

POCE landmark analysis showed no differences between the 2 groups across different time points. (figure 4A). Looking specifically at the various POCE individual components, the LAD-STEMI group exhibited a higher rate of all-cause mortality within the first year vs the non-LAD-STEMI group (p = 0.041), but this difference disappeared thereafter (figure 4B). Between years 0 and 1, there was also a trend toward a lower rate of myocardial infarction in the LAD-STEMI group vs the non-LAD-STEMI group (p = 0.081), which disappeared after year 1 (figure 4C). No differences were ever found regarding any revascularization (figure 4D) or other secondary endpoints between the 2 groups (figure 2 of the supplementary data).

Figure 4. Landmark analysis for the patient-oriented composite endpoint (A), all-cause mortality (B), myocardial infarction (C), and any revascularization (D) in patients stratified according to the culprit vessel. LAD, left anterior descending coronary artery; MI, myocardial infarction; STEMI, ST-segment elevation myocardial infarction; POCE, patient-oriented composite endpoint.

DISCUSSION

The main findings of this study can be summarized as follows: a) STEMI patients with LAD as the culprit vessel have a different baseline clinical profile vs STEMI patients with other culprit vessels; b) in the contemporary era of primary PCI, LAD as the STEMI-related culprit vessel did not bring worse very long-term outcomes compared with other coronary vessels; c) nevertheless, between years 0 and 1 the LAD-STEMI group exhibited a higher all-cause mortality rate, which disappeared thereafter compared with non-LAD-STEMI group.

Cardiology community knows (as reflected by the ESC guidelines on the management of acute coronary syndromes) that STEMI with LAD involvement as culprit vessel is a clinical marker of high risk of further events.1 LAD-related STEMI represents, approximately, 40% up to 50% of all STEMIs,12,16 and its worse prognosis has been related to the large myocardium covered by the LAD flow compared with the myocardium supplied by other coronary vessels. Of note, those studies were performed in the pre-reperfusion4-7 and early thrombolysis/PCI era,8,9 when PCIs were still not widely available. In the PCI era, there are very few studies (with short or mid-term follow-ups ranging from 1 to 3 years) reporting that LAD-STEMI is associated with an increased risk of stroke, heart failure, all-cause mortality10,17 and cardiovascular death11 after the PCI.

In our analysis, conducted in a cohort where the PCI was extensively performed, LAD as the STEMI culprit vessel did not appear to confer a worse prognosis to patients at the 1- or even 10-year follow-up. Of interest, LAD-STEMI patients exhibited the classical clinical features related to LAD, such as advanced Killip class at the time of presentation, lower ST-segment resolution and lower LVEF, which is similar to previous studies.8-11,17 All these unfavorable clinical characteristics are indeed related to the large amount of myocardium damaged in a LAD-STEMI with subsequent heart failure and ventricular arrhythmias.17-19 Nevertheless, this did not translate into a worse, very long-term clinical outcome. Significantly, even after accounting for variations in LVEF (which we addressed separately in our model due to its perceived role in the outcome cascade) the results showed no differences. This observation stands in contrast to earlier evidence, where the higher mortality rate in this cohort had been partially attributed to the subsequent decline in LVEF after STEMI.9,10

Several explanations may be claimed to understand our main finding. It may be hypothesized that worse outcome related to anterior STEMI may have been overcome by the introduction of the PCI with quick myocardial reperfusion. Pharmacological treatment has been also improved from thrombolysis to the PCI era, not only in terms of antiplatelet agents, but also in terms of secondary prevention (high intensity statins and angiotensin converting enzyme inhibitors/angiotensin receptor blockers or angiotensin receptor/neprilysin inhibitors for left ventricular dysfunction).20-23 Furthermore, in our study, the LAD-STEMI group had a higher proportion of active smokers. Smoking cessation remains the most critical preventive measure for coronary artery disease. The relationship between smoking and cardiovascular outcomes has been a matter of discussion, as some studies have suggested improved cardiovascular outcomes, even in the long term, among smokers who experienced STEMI.24 However, many of these studies were observational registries conducted in the pre-PCI era. Recent evidence indicates that smoking is associated with more post-PCI long-term adverse outcomes.25 Therefore, the so-called “smoker’s paradox” might be better explained by factors such as younger age and a lower prevalence of other risk factors among smokers. Indeed, in our study, while the LAD-STEMI group had a higher proportion of smokers, they had a lower prevalence of other risk factors, such as peripheral vascular disease and a history of prior PCI or MI.

Last, but not least, in landmark analysis we found that between years 0 and 1, all-cause mortality was more common in the LAD-STEMI group. Notably, in this period, there was a numerically higher number of cardiac deaths (although not statistically significant, P = .12), a similar finding to other existing evidence that found a higher relatively short-term mortality in the LAD-STEMI group within the first 30 days. In these studies, the elevated short-term mortality was associated with acute sequelae, such as heart failure and was also speculated to be connected to other lethal complications, such as ventricular arrhythmias, cardiogenic shock or mechanical complications.10,11 In our cohort, we found a trend towards a higher rate of reinfarction in the non-LAD-STEMI group (P = .081) that was largely unrelated to TLR, TVMI, or stent thrombosis. This observation contrasts with previous literature that reported a more common occurrence of reinfarction at the follow-up in patients with the SVG as the culprit vessel26 as well as the LAD,8 but not in LCx or the RCA.9-11

Our 10-year follow-up revealed similar clinical event rates between LAD-STEMI and non-LAD-STEMI group, indicating absence of long- term divergence. Previous studies showed a favorable post-acute phase prognosis for LAD-STEMI patients,10,11 which is consistent with our findings. In fact, non-cardiac factors seem to impact long-term mortality more than infarct location does.19 Thus, patients with STEMI should receive uniform management focused on secondary prevention strategies, regardless of the culprit vessel. Unfortunately, insufficient long-term data collection limits deeper insights into these outcomes (such as the presence of heart failure, optimal medical therapy, or other comorbidities).

Limitations

This study presents several limitations. First, this is a non-prespecified post-hoc analysis of the EXAMINATION-EXTEND study and therefore its conclusions must be considered only hypothesis generating. The association between infarction and outcomes may be driven by confounders which have not been recorded in the study. Then, several clinical and procedural characteristics were not available for the analysis, such as specified in-hospital or follow-up clinical data, like optimal medical treatment or compliance to medication at the follow-up.

CONCLUSIONS

In a contemporary cohort of STEMI patients, there were no differences in POCE between LAD as the STEMI-related culprit vessel and other vessels at the 10-year follow-up. However, within the first year after STEMI, all-cause death was more common in the LAD-STEMI group. Our results should be considered as hypothesis-generating. Further studies are needed to specifically assess the relationship between infarction location and outcomes in a contemporary setting where interventional and medical treatments are optimized.

FUNDING

The EXAMINATION-EXTEND study was funded by an unrestricted grant of Abbott Vascular to the Spanish Society of Cardiology (promoter). P. Vidal Calés has been supported by a research grant provided by Hospital Clínic at Barcelona, Spain.

ETHICAL CONSIDERATIONS

The study fully complied with the Declaration of Helsinki and was approved by our Institutional Review Committee. All patients signed a written informed consent form before being included in this study. The clinical ethics committee gave its approval for the analysis of the data collected. In this work, SAGER guidelines regarding sex and gender bias have been followed.

STATEMENT ON THE USE OF ARTIFICIAL INTELLIGENCE

No artificial intelligence tools were used during the preparation of this work.

AUTHORS’ CONTRIBUTIONS

The authors declare they meet the full criteria and requirements for authorship and have reviewed and agree with the content of the article. P. Vidal Calés, K. Bujak, R. Rinaldi, A. Salazar Rodríguez, S. Brugaletta and M. Sabaté contributed to conceptualization, design, data analysis and drafting of the manuscript. L. Ortega-Paz, J. Gómez-Lara, V. Jiménez-Diaz, M. Jiménez, P. Jiménez-Quevedo, R. Diletti, P. Bordes, G. Campo, A. Silvestro, J. Maristany, X. Flores, A. De Miguel-Castro, A. Íñiguez, A. Ielasi, M. Tespili, M. Lenzen, N. Gonzalo, M. Tebaldi, S. Biscaglia, R. Romaguera, J.A. Gómez-Hospital and P. W. Serruys reviewed and edited the manuscript.

CONFLICTS OF INTEREST

M. Sabaté declares he has received consulting fees from Abbott Vascular and iVascular outside the submitted work. R. Romaguera is associate editor of REC: Interventional Cardiology. The journal’s editorial procedure to ensure impartial handling of the manuscript has been followed. The rest of the authors declared no conflicts of interest whatsoever.

WHAT IS KNOWN ABOUT THIS TOPIC?

• – In STEMI patients, the culprit vessel is often regarded as a crucial prognostic factor.
• – This assumption is based on earlier studies conducted during the pre-reperfusion or thrombolysis era, which demonstrated that STEMIs involving the left anterior descending coronary artery (LAD) were linked to poorer clinical outcomes vs those involving other vessels.
• – In the current PCI era, there is limited data on the long-term prognostic impact of the LAD as the culprit vessel in STEMI patients.

WHAT DOES THIS STUDY ADD?

• – Patients with LAD as the STEMI-related culprit vessel have a higher all-cause mortality within the first year after STEMI.
• – However, our study found that this difference did not persist beyond the initial year suggesting that the prognostic impact of the culprit vessel might pertain to the immediate post- STEMI period.
• – Moreover, our results support that (irrespective of the location of the infarction) all STEMI patients should receive uniform medical care in the long-term focused on implementing secondary prevention strategies.

SUPPLEMENTARY DATA

REFERENCES