Feasibility and Findings of Including Self-Identified Adult Congenital Heart Disease Patients in the INVESTED Trial

Background Adult congenital heart disease (ACHD) patients have significant morbidity and rise in cardiac admissions. Their outcome with high-dose influenza vaccination is unknown in comparison to those without ACHD. Objectives The purpose of this study was to compare all-cause mortality or cardiopulmonary hospitalizations in self-identified ACHD versus non-ACHD patients receiving high- or low-dose influenza vaccination within the INfluenza Vaccine to Effectively Stop cardioThoracic Events and Decompensated heart failure trial. Methods We prospectively included ACHD patients in the INVESTED (INfluenza Vaccine to Effectively Stop cardioThoracic Events and Decompensated heart failure) trial. The primary endpoint was all-cause death or hospitalization for cardiovascular or pulmonary causes. Results Of the 272 ACHD patients, 132 were randomly assigned to receive high-dose trivalent and 140 to standard-dose quadrivalent influenza vaccine. Compared to the non-ACHD cohort (n = 4,988), ACHD patients were more likely to be younger, women, smokers, have atrial fibrillation, and have a qualifying event of heart failure. The primary outcome was 49.8 events versus 42.8 events per 100 person-years (adjusted HR: 1.17; 95% CI: 0.95-1.45; P = 0.144) in the ACHD group and non-ACHD group, respectively. The interaction between ACHD status and randomized treatment effect was not significant for the primary outcome (P = 0.858). Vaccine-related adverse events were similar in both groups. Conclusions Patients who self-identify as being ACHD had similar primary outcome of all-cause death or hospitalization for cardiovascular or pulmonary causes compared to non-ACHD cohort. High-dose influenza vaccination was similar to standard-dose influenza vaccination on the primary outcome in patients who self-identify as ACHD.

2][3] Immunization is an important public health measure through which the cardiovascular sequelae of influenza can be abated. 4,5ult congenital heart disease (ACHD) patients have significant morbidity leading to an increase in the number of hospital admissions and health care costs. 6,7With advancement of medical and surgical care for this cohort, survival has increased such that today, >90% of infants with CHD survive to adulthood, 8,9 and the ACHD population now outnumbers the pediatric CHD population. 10In addition, adults admitted with heart failure (HF) have a higher mortality and rehospitalization rate if they have a history of CHD compared to those without a history of CHD. 11Previous data have demonstrated that immunization rates in ACHD are lower than the non-ACHD population. 12,13Therefore, any effort to curb influenza-related morbidity and mortality in this growing population may have significant implications.
In addition to immunizing, offering higher dose vaccine to those with reduced antibody-mediated responses is another public health strategy to potentially reduce cardiovascular sequelae of influenza.
Current eligible patients for high-dose vaccination include adults 65 years and older.High-dose trivalent versus standard-dose quadrivalent vaccination was further explored in the INVESTED (INfluenza Vaccine to Effectively Stop cardioThoracic Events and Decompensated heart failure trial. 14This was a multicenter, double-blind, active comparatorcontrolled randomized clinical trial conducted in 5,260 patients with high-risk cardiovascular disease in which high-dose trivalent inactivated influenza vaccine (HD-IIV3) did not significantly reduce all-cause mortality or cardiopulmonary hospitalizations compared with standard-dose quadrivalent inactivated influenza vaccine (SD-IIV4). 14 hypothesized that self-identified ACHD patients represent a high-risk group for the primary endpoint in the INVESTED trial, namely allcause death, or cardiopulmonary hospitalization.We therefore performed a prespecified subgroup analysis of the INVESTED trial to describe the patient characteristics, clinical events, adverse events, and potential interaction with randomized treatment arms comparing patients with or without ACHD separated into those receiving HD-IIV3 or SD-IIV4.In addition, we compared ACHD status to other traditional risk factors in predicting all-cause mortality or cardiopulmonary hospitalizations.

METHODS
STUDY DESIGN.The methodology, design, and protocol of the trial have been summarized previously. 14,15The INVESTED trial was a randomized,  Dehghani et al  16 ACHD status by treatment interaction was assessed in a similar model as above with the ACHD status by treatment interaction term as an additional covariate in the model.The number of events of all-cause mortality (without hospitalization) and recurrent cardiopulmonary hospitalizations (before all-cause death) was analyzed using a proportional means model. 17atistical analyses was completed using both SAS (SAS Corporation) and R (The R Foundation).OUTCOMES ACCORDING TO ACHD STATUS.The median duration of follow-up was 9 months for the non-ACHD group and 9.07 months in the
OUTCOMES ACCORDING TO TREATMENT GROUP.
Among patients with ACHD, there were 52 primary events (47 hospitalizations for cardiovascular or pulmonary causes and 5 deaths from any cause) in 132 unique participants (event rate 50.5 per 100 patient-years) among 170 participant-seasons in the high-dose group compared with 52 primary events (47 hospitalizations for cardiovascular or pulmonary causes and 5 deaths from any cause) in 139 unique participants (event rate 49.1 per 100      Dehghani et al

Insights From the INVESTED Trial
A P R I L 2 0 2 4 : 1 0 0 8 9 7 differences between self-identified ACHD and non-ACHD groups in the high-dose group (see Table 4).
In  [20][21][22] Stroke rate was also significantly higher: 19.5% of ACHD patients had a previous history of stroke compared to 7% in the non-ACHD group in INVESTED, and 7% to 9% in other trials. 19,20Bouchardy et al 23 found that the presence of an atrial arrhythmia in ACHD increased the risk of stroke and death by close to 50%.Lanz et al 24 found the prevalence of stroke in ACHD to be 1 in 11 for men and 1 in 15 for women, and the 30-day mortality after ischemic stroke in ACHD was 5.1%.Despite the self-identified ACHD patients being younger by an average of 6 years, prevalence of renal impairment and hypertension was similar to the non-ACHD cohort.While this may reflect INVESTED trial's inclusion criteria of a higher risk patient population, it could also speak to altered vascular hemodynamics in CHD such as those with coarctation or Fontan physiology.The higherthan-expected smoking rate as seen in our ACHD ACHD ¼ adult congenital heart disease; HD-IIV3 ¼ high-dose trivalent inactivated influenza vaccine; SD-IIV4 ¼ standard-dose quadrivalent inactivated influenza vaccine.
Dehghani et al Insights From the INVESTED Trial cohort has been previously described in previous ACHD reports 25 and may also explain the higher rates of chronic obstructive pulmonary disease (COPD) in our cohort.
Why is it that despite having higher rates of atrial fibrillation, stroke, and smoking status, our selfidentified ACHD cohort had similar risk for death and cardiopulmonary hospitalization, a finding that is contrary to the literature 11 ?This may be because our ACHD cohorts were younger, and more likely to be women, essentially counterbalancing overall risk.Our ACHD cohort may also have had a disproportionally higher rate of mild anatomical complexity and/or less advanced physiologic staging, a hypothesis that cannot be confirmed due to limitation of our current data set.Similar outcomes between the ACHD and non-ACHD cohorts can also be explained by the overall cardioprotective properties of the influenza vaccine offered to everyone in this trial.Although low uptake of influenza vaccination in ACHD patients is well-documented, 12,13 this was not observed in our study-likely a reflection of selection bias, as patients interested and willing to enroll in a vaccination trial may have greater knowledge of influenza vaccination as compared with the general public.Finally, our median duration follow-up of 9 months may have been inadequate.Contemporary data suggest that CHD-related mortality no longer occurs at the time of surgical intervention, 26,27 but rather, later in life.A median duration follow-up of 9 years was required in a 7,000 ACHD patients British registry to demonstrate a mortality rate of 7.7%. 26spite having a similar LVEF as those without ACHD patients, self-identified ACHD patients were more likely to have a history of HF hospitalization (25% vs 17%), and their index event was more likely to be congestive HF (82% vs 62%).This presumably relates to a lower risk for the other possible reason for inclusion, being MI, in the younger self-identified ACHD subgroup.Given a higher anticipated prevalence of right ventricular causes of "HF" in the ACHD cohort, it may be surprising that the LVEF was similar in the ACHD and non-ACHD cohorts.This may be explained by inadequate data collection on systemic ventricular dysfunction in patients with morphologic right ventricles.Even though ACHD patients are living longer, and their comorbidities and HF admissions are increasing, 28 relying solely on LVEF may be Dehghani et al

Insights From the INVESTED Trial
A P R I L 2 0 2 4 : 1 0 0 8 9 7 misleading. 260][31] For instance, in some patients with transposition of the great arteries (TGA), the systemic ventricle has right ventricular physiology, therefore less robust in its response to volume and pressure overload. 29Similarly, patients with Fontan circulation are at high risk of HF with moderate to severe systolic dysfunction described only in one third of such patients. 29,32Additionally, ACHD is associated with excessive neurohormonal activation, thought to be secondary to abnormal cardiac myocyte distension, with increased activation of this system likely contributing to cardiac deterioration. 29,33]34 Unfortunately, in addition to complex pathophysiology that precludes uniform approaches to HF management in ACHD, there is also a paucity of evidence to guide management in this population.ACHD are part of a larger underrepresented group in clinical trials of HF, including racial and ethnic minorities who share the common trait of higher rates of HF but have been least studied in clinical trials. 35For example, landmark clinical trials of HF which form the evidence for optimal goal-directed medical therapy, such as the RALES (Randomized Aldactone Evaluation Study) and SHIFT (Systolic Heart failure treatment with the I f inhibitor ivabradine Trial) specifically excluded ACHD. 19,36Our trial was able to successfully include a prespecified subgroup analysis of ACHD patients, demonstrating that this is a population that can be effectively and safely included in clinical trials.
Despite a different profile of risk, high-dose compared with low-dose influenza vaccination did not reduce the primary outcome in ACHD patients.
One explanation is that our trial did not examine rates of influenza infection, which high-dose vaccination may have reduced based on previous data. 37,38Additionally, this trial compared high-dose trivalent vaccine to standard-dose quadrivalent vaccine; the quadrivalent vaccine used for the standard-dose group contains an additional B/Yamagata strain which could have attenuated differences that would have been seen due to vaccine dosage. 16Although this trial did not demonstrate a difference between the 2 vaccine formulations in ACHD, we want to emphasize that this does not preclude the benefit of vaccination  and above, which also holds true for ACHD patients. 39UDY LIMITATIONS.This study and associated subgroup analysis has several limitations.Most importantly, our study did not include granular information on anatomical and/or physiologic classification of their underlying congenital lesions.Adults with CHD represent a heterogenous group of patients, with clear incremental prognostic implications depending on degree of anatomic complexity and/or physiologic staging. 31,40,41Additionally, ACHD status was self-determined and not externally validated.
Although incorrect classification is possible, it is plausible that adult self-reporting was accurate in our series as the ACHD cohort were younger, and had a higher prevalence of stroke and atrial fibrillation, characteristics that are well established in ACHD patient population. 42Furthermore, this subgroup analysis was performed on a trial with neutral results that was stopped early because the required event rate was surpassed.While our 5% prevalence of selfreported ACHD is higher than the accepted 0.8 to 1% prevalence of births with CHD, 43 the INVESTED population was already enriched with cardiac comorbidities with an admission related to HF or MI.A 7% prevalence of MI in ACHD has been documented, 44 which is considered higher than that of the general population.Lastly, influenza infection rates were not assessed, as the study was structured as a pragmatic trial examining clinical outcomes. 14Therefore, we cannot exclude the possibility that high-dose vaccine could reduce the rate of influenza vaccination in ACHD. 14,45 patients-years) among 178 participant-seasons in the standard-dose group (HR: 1.034; 95% CI: 0.687-1.555;P ¼ 0.87).Among patients without ACHD, there were 923 primary events (836 hospitalizations for cardiovascular or pulmonary causes and 87 deaths from any cause) in 2,474 unique participants (event rate 44.2 per 100 patient-years) among 3,407 participant-seasons in the high-dose group compared with 872 primary events (799 hospitalizations for cardiovascular or pulmonary causes and 73 deaths from any cause) in 2,465 unique participants (event rate 41.5 per 100 patients-years) CENTRAL ILLUSTRATION Insights on Adult Congenital Heart Disease Patients From the INVESTED Trial Dehghani P, et al.JACC Adv.2024;3(4):100897.Distinct differences exist in the clinical demographics between ACHD patients and non-ACHD patients, including the prevalence of all cause death or hospitalization for cardiovascular or pulmonary causes.Probabilities stratified by groups for the primary outcome.The interaction was at P ¼ 0.858.ACHD ¼ adult congenital heart disease; HF ¼ heart failure; INVESTED ¼ INfluenza Vaccine to Effectively Stop cardioThoracic Events and Decompensated heart failure; MI ¼ myocardial infarction.

J
A C C : A D V A N C E S , V O L . 3 , N O . 4 Insights From the INVESTED Trial among 3,399 participant-seasons in the standard dose group (HR: 1.06; 95% CI: (0.965-1.177);P ¼ 0.21).The P value for interaction between ACHD status and randomized treatment effect for the primary outcome was not significant with P ¼ 0.858 (

FIGURE 1
FIGURE 1 Primary Outcome According to Congenital Heart Disease Status

FIGURE 2
FIGURE 2 Selected Baseline Characteristics as Risk Factors for Primary Outcome in Univariate Analysis

JInsights
A C C : A D V A N C E S , V O L . 3 , N O . 4 From the INVESTED Trial compared to the unvaccinated.The Centers for Disease Control and Prevention (CDC) recommends annual influenza vaccination in those aged 6 months

FIGURE 3 4 Insights
FIGURE 3 Selected Baseline Characteristics as Risk Factors for Primary Outcome in Multivariate Analysis

TABLE 1
Baseline Characteristics of Patients With and Without Congenital Heart Disease Unique participant numbers of 2,474 in the high-dose group and 2,465 in the low-dose group add up to 4,939, rather than the 4,988.ACHD ¼ adult congenital heart disease; BMI ¼ body mass index; HD-IIV3 ¼ high-dose trivalent inactivated influenza vaccine; HF ¼ heart failure; LVEF ¼ left ventricular ejection fraction; MI ¼ myocardial infarction; SD-IIV4 ¼ standard-dose quadrivalent inactivated influenza vaccine; SD ¼ standard deviation.
Values are mean AE SD or n (%). a Represents all randomized participants.b
ADVERSE EVENTS.Overall, influenza vaccinations were well tolerated and there were no notable

TABLE 2
Outcomes According to Congenital Heart Disease Status a Based on Cox proportional hazards model with robust variance estimate.ACHD ¼ adult congenital heart disease; ICD ¼ implantable cardioverter-defibrillator; mITT ¼ modified intention-to-treat.

TABLE 3
Primary and Secondary Outcomes in a Study of the Effect of High-Dose vs Standard-Dose Influenza Vaccine According to ACHD Status a Based on Cox proportional hazards model with robust variance estimate.

TABLE 4
Postvaccination Adverse Events (Within 1 Week) in a Study of the Effect of High-Dose vs Standard-Dose Influenza Vaccine