High-dose quadrivalent influenza vaccine in older adults with or without cardiovascular risk: a meta-analysis of randomized trials on 511,890 participants

Abstract

Background

Older adults remain highly vulnerable to serious influenza-related complications despite routine vaccination. Whether high-dose quadrivalent influenza vaccine (HD-QIV) offers greater protection than standard-dose quadrivalent influenza vaccine (SD-QIV) against clinically important outcomes remains an important question.

Methods

We conducted a systematic review and pairwise meta-analyses of randomized controlled trials (RCTs) comparing HD-QIV with SD-QIV in adults aged 65 years or older. PubMed, Cochrane Library, Embase, Web of Science, and Scopus were searched up to February 2026. Odds ratios (ORs) with 95% confidence intervals (CIs) were pooled using random-effects models. Prespecified subgroup analyses were performed according to baseline cardiovascular disease (CVD) status.

Results

Four RCTs comprising 511,890 participants were included. HD-QIV was associated with significantly lower odds of cardio-respiratory hospitalization (OR 0.93, 95% CI [0.90–0.97]) and cardiovascular hospitalization (OR 0.93, 95% CI [0.89–0.98]), with no significant reduction in respiratory hospitalization (OR 0.90, 95% CI [0.80–1.01]). Significant reductions were also observed for hospitalization due to influenza (OR 0.61, 95% CI [0.50–0.74]), laboratory-confirmed influenza hospitalization (OR 0.69, 95% CI [0.56–0.86]), and heart failure hospitalization (OR 0.80, 95% CI [0.69–0.94]). No significant differences were found for all-cause hospitalization (P = 0.12), all-cause mortality (P = 0.413), myocardial infarction (P = 0.753), or serious adverse events (P = 0.419). No significant subgroup differences were observed according to baseline CVD status.

Conclusions

In older adults, HD-QIV was associated with modest but significant reductions in cardio-respiratory and cardiovascular hospitalization, as well as influenza-related and heart failure hospitalizations, without an apparent safety penalty. These findings support preferential use of high-dose influenza vaccination in this population.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12979-026-00571-x.

Introduction

Seasonal influenza remains a major cause of global morbidity and mortality, accounting for approximately 1 billion infections annually, including 3–5 million cases of severe illness and 290,000–650,000 respiratory deaths [1]. Older adults bear a disproportionate share of this burden; adults aged 65 years or older are at increased risk of severe influenza, hospitalization, and death, in part because age-related changes in immune function reduce the effectiveness of standard influenza vaccines [2, 3].

Beyond its respiratory manifestations, influenza is increasingly recognized as a cardiovascular trigger, with evidence linking acute infection to myocardial infarction, stroke, heart failure, and cardiovascular death [4, 5]. This interaction is especially important in older adults and in those with pre-existing cardiovascular disease, populations in whom influenza-related complications are both more frequent and more consequential [6, 7]. These considerations have heightened interest in enhanced influenza vaccines for older adults, particularly the high-dose quadrivalent inactivated influenza vaccine (HD-QIV), which is licensed for adults aged 65 years or older and contains 60 micrograms of hemagglutinin per strain—four times the antigen content of standard-dose quadrivalent inactivated influenza vaccines (SD-QIV) [8, 9].

More recently, pragmatic randomized controlled trials (RCTs) have expanded the evidence base for HD-QIV from immunogenicity and virologic efficacy to clinically important outcomes in older adults [10], including hospitalization for influenza or pneumonia, cardio-respiratory hospitalization, cardiovascular hospitalization, and death. Nevertheless, the cardiovascular relevance of these data remains uncertain when individual trials are considered in isolation.

Earlier meta-analyses had key limitations: they pooled different formulations, including trivalent-only or mixed trivalent and quadrivalent vaccines, and often lacked a focus on cardiovascular outcomes. Furthermore, a recent 2024 meta-analysis of five RCTs comprising 105,685 participants did investigate cardiovascular outcomes [11], but included only one study directly comparing quadrivalent high-dose and standard-dose vaccines [10]. Since then, three large pragmatic RCTs—DANFLU-2 (332,438 participants), GALFLU (133,882 across two seasons), and FinFluHD (33,093 participants )—have added much stronger evidence for HD-QIV [12–14]. These new data necessitate an updated synthesis. To our knowledge, this is the first RCT meta-analysis focused exclusively on inactivated HD-QIV and cardiovascular outcomes in older adults, featuring prespecified subgroup analyses based on baseline cardiovascular risk.

Methods

Protocol registration

This systematic review and meta-analysis was planned and conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [15] (Supplementary Table 1) and the Cochrane Handbook for Systematic Reviews [16]. The review protocol was registered prospectively in the International Prospective Register of Systematic Reviews (PROSPERO; CRD420261335386).

Eligibility criteria

Eligibility criteria were defined using the PICOS framework:

• Populations (P): Older adults (primarily aged ≥ 65).

• Intervention (I): High-dose inactivated influenza vaccine, typically quadrivalent (HD-QIV), containing 60 µg of hemagglutinin antigen per strain.

• Comparison (C): Standard-dose inactivated influenza vaccine, typically quadrivalent (SD-QIV), containing 15 µg of hemagglutinin antigen per strain.

• Outcomes: (O): Primary outcomes were hospitalizations for cardio-respiratory disease, respiratory disease, and cardiovascular disease. Secondary outcomes comprised hospitalizations for influenza or pneumonia, pneumonia alone, influenza alone, and laboratory-confirmed influenza, as well as all-cause hospitalization, major adverse cardiovascular events (MACE), all-cause mortality, hospitalizations for heart failure, myocardial infarction, stroke, and atrial fibrillation, in addition to serious adverse events and serious adverse reactions.

• Study design (S): RCTs published in peer-reviewed journals only.

No restrictions were applied with respect to publication date, sex, ethnicity, language, or healthcare setting. We excluded animal studies, observational studies, non-randomized or quasi-experimental studies, conference abstracts, study protocols, letters, case reports, case series, books, theses, and studies with incomplete outcome data. Additionally, only RCTs directly comparing HD-QIV with SD-QIV were eligible. Trials evaluating trivalent formulations or mixed formulation comparisons (e.g., high-dose trivalent vs. standard-dose quadrivalent) were excluded.

Data sources and search strategy

A comprehensive search of PubMed, the Cochrane Library, Embase, Web of Science, and Scopus was conducted from database inception through 3 February 2026, without restrictions on language or publication date. The search strategy combined controlled vocabulary and free-text terms related to “Influenza”, “Influenza Vaccination”, “High-Dose Vaccine”, and “Randomized Controlled Trials”, the full detailed strategy outlined in Supplementary Table 2. Reference lists of relevant studies were also reviewed to identify any additional eligible reports not captured by electronic search.

Study Selection

All retrieved records were imported into Rayyan (Rayyan Systems Inc., Cambridge, MA, USA) [17], where duplicates were identified and removed. Screening was conducted by three reviewers (M.Q., M.A.A., and A.S.) in two stages. First, titles and abstracts were reviewed against the predefined eligibility criteria. Second, the full texts of potentially relevant studies were assessed for final inclusion. Studies that did not meet the eligibility criteria at either stage were excluded. Any discrepancies during the selection process were resolved through consultation with the senior author (M.S.E.).

Data extraction

Data were independently extracted by four reviewers (M.A.F., M.Q., M.A.A., and A.S.), with each data item reviewed by at least two investigators, using a standardized and pilot-tested extraction form. Extracted data included:

• i.Study characteristics: included study identification, study design, registration number, geographic setting, total sample size, intervention and comparator groups, influenza season, inclusion criteria, primary outcome, and maximum follow-up duration.
• ii.Baseline participant characteristics: included age, sex, diabetes, cardiovascular disease history, heart failure, atrial fibrillation, ischemic heart disease, chronic obstructive pulmonary disease, prior myocardial infarction, hypertension, cancer, immunodeficiency, and influenza and COVID-19 vaccination status during the same season, where available.
• iii.Outcome data: included the prespecified clinical and safety outcomes.

Any discrepancies were resolved by consultation with the senior author (M.S.E.).

Quality assessment

The risk of bias of included RCTs was assessed independently using the Cochrane Risk of Bias 2 (RoB 2) tool [18] by two reviewers (M.A.F. and M.Q.). Each study was evaluated across the relevant domains of bias, and an overall judgment was assigned in accordance with RoB 2 guidance. In addition, the certainty of evidence for each outcome was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach with GRADEpro software [19].

Statistical analysis

Pairwise meta-analysis

Pairwise meta-analyses were conducted in (version 4.4.2; R Foundation for Statistical Computing, Vienna, Austria) using the “meta” package with a random-effects model regardless of heterogeneity. Dichotomous outcomes were summarized as odds ratios (ORs) with 95% confidence intervals (CIs). For studies with zero-event data, a continuity correction of 0.5 was applied. Statistical heterogeneity was assessed using both the I² statistic and tau-squared (τ²). The magnitude of inconsistency according to I² was interpreted as not important (up to 40%), moderate (30% to 50%), substantial (50% to 75%), or considerable (75% to 90%).

Sensitivity & subgroup analyses

Leave-one-out sensitivity analyses were performed for outcomes with substantial statistical heterogeneity (I² > 50%) and for MACE, given the between-trial differences in its component definition. Prespecified subgroup analyses were performed according to baseline cardiovascular disease status, stratifying participants into those without prior cardiovascular disease and those with pre-existing cardiovascular disease, where such data were available from the included studies.

Trial sequential analysis

Trial sequential analysis (TSA) was performed using the “RTSA” package to determine whether the available evidence was sufficient to support reliable conclusions regarding the intervention effect [20]. Evidence was considered conclusive when the cumulative Z-curve crossed the trial sequential monitoring boundary (TSMB) for benefit (and/or the conventional significance boundary) and the required information size (RIS) reached; indicating that further studies were unlikely to materially change the overall findings. If these criteria were not met, additional research was considered necessary. The analysis was conducted using a two-sided alpha of 0.05, 80% power (beta = 0.20), and an anticipated relative risk reduction of 10%.

Number needed to treat

We calculated the number needed to treat (NNT) as the reciprocal of the absolute risk reduction (ARR), using pooled event rates for the primary outcomes in the intervention and control groups, in accordance with the method described by Cook and Sackett [21]. Confidence intervals for the NNT were derived by first estimating the 95% confidence interval of the ARR using the Wilson score method and then inverting its limits [22].

Meta-regression

To assess whether baseline CVD prevalence modified treatment effect, we performed an exploratory study-level meta-regression using baseline CVD prevalence (%) as the covariate and the treatment effect (log odds ratio [OR]) for each primary outcome as the dependent variable. Each trial contributed one independent effect estimate per outcome. Meta-regression was performed under a random-effects model with inverse-variance weighting and restricted maximum likelihood (REML) estimation of between-study variance (τ²). Bubble plots of OR versus baseline CVD prevalence were constructed, with bubble size proportional to study weight.

Results

Search results

The systematic search yielded 1,316 records from electronic databases. After removal of 447 duplicates, 869 records remained for title and abstract screening, of which 856 were excluded at this stage. Thirteen reports were retrieved for full-text assessment, and all were evaluated for eligibility. Following full-text review, 8 reports were excluded. In the end, 4 trials reported across 5 publications [10, 12–14, 23], met the inclusion criteria, and were incorporated into the final review. The study selection process is illustrated in the PRISMA flow diagram (Fig. 1).

Fig. 1.

PRISMA flow chart for the systematic search and selection process

Characteristics of the included trials

A total of 4 RCTs [10, 12–14], enrolling 511,890 participants, were included in the review. The principal characteristics of the included trials are summarized in Table 1. All studies were conducted in Denmark, Spain, and Finland, and compared HD-QIV (60 µg hemagglutinin per strain) with SD-QIV (15 µg hemagglutinin per strain). The included trials spanned influenza seasons from 2019/2020 to 2024/2025, with a follow-up period ranging from 3 to 8 months.

Table 1.

Summary characteristics of the included trials

Study ID	Study Design	Register No	Location	Total	Study arms		Influenza Seasons	Inclusion criteria			Primary Outcome	MFUP
					Intervention (s)	Comparator (s)		Age (Y)	Population	CV Risk
Johansen 2025 (DANFLU-2)	Open-label, RCT	NCT-05517174	Denmark	332,438	QIV-HD (Efluelda Tetra / Fluzone)	QIV-SD (Vaxigrip Tetra)	2022/23, 2023/24, & 2024/25	≥ 65	General older adults	Mixed (27.4%)	Cardio-respiratory hospitalization	8 mon
Pardo-Seco 2025 (GALFLU)	Open-label, RCT	NCT-06141655	Spain	133,882	QIV-HD (Efluelda)	QIV-SD (Influvac Tetra)	2023/24 & 2024/25	65 to 79	Community-dwelling	Mixed (12.5%)	Hospitalization for influenza or pneumonia	6 mon
Palmu 2024 (FinFluHD)	Double-blind, RCT	NCT-04137887	Finland	33,093	QIV-HD (Efluelda / Fluzone)	QIV-SD (Vaxigrip Tetra)	2019/20	≥ 65	Older adults in health stations	Mixed (16.4%)	Cardio-respiratory hospitalization	7 mon
Johansen 2023 (DANFLU-1)	Open-label, RCT	NCT-05048589	Denmark	12,477	QIV-HD (Efluelda / Fluzone)	QIV-SD (Influvac Tetra)	2021/22	65 to 79	Community-dwelling	Mixed (20.4%)	Hospitalization for influenza or pneumonia	3 mon

RCT Randomized controlled trial, QIV-HD high-dose quadrivalent influenza vaccine, QIV-SD standard-dose quadrivalent influenza vaccine, CV Cardiovascular, MFUP Maximum follow-up period, Y years

Overall, allocation was nearly equal between study arms, with 50.03% of participants assigned to the high-dose vaccine group and 49.97% to the standard-dose group. The mean age of the study population was 73.2 ± 5.4 years, and 48.1% of participants were women. Cardiovascular disease was frequently represented at baseline, with an overall prevalence of 22.6%. The study populations predominantly comprised older adults, including community-dwelling individuals and older adults receiving care in health stations, with mixed cardiovascular risk profiles. Detailed baseline characteristics are provided in Table 2.

Table 2.

Baseline characteristics of the included patients

Study ID	Groups	Total	Age, mean (SD), y	Sex (Female), No. (%)	Diabetes, No. (%)	CVD, No. (%)	Heart Failure, No. (%)	Atrial Fibrillation, No. (%)	Ischemic Heart Disease, No. (%)	COPD, No. (%)
Johansen 2025 (DANFLU-2)	High-dose	166,218	73.7 ± 5.8	80,779 (48.6)	21,929 (13.2)	45,688 (27.49)	5,201 (3.13)	17,076 (10.27)	15,666 (9.43)	6,799 (4.1)
	Standard-dose	166,220	73.7 ± 5.8	80,753 (48.58)	21,952 (13.2)	45,338 (27.28)	5,209 (3.13)	17,009 (10.23)	15,446 (9.29)	6,689 (4)
Pardo-Seco 2025 (GALFLU)	High-dose	67,093	72.3 ± 4.2	31,028 (46.24)	3,233 (4.8)	8,373 (12.5)	1,296 (1.9)	2,103 (3.1)	3,367 (5.0)	1,453 (2.2)
	Standard-dose	66,789	72.3 ± 4.3	31,115 (46.59)	3,230 (4.8)	8,301 (12.4)	1,327 (2.0)	2,034 (3.0)	3,363 (5.0)	1,411 (2.1)
Palmu 2024 (FinFluHD)	High-dose	16,549	72.6 ± 5.7	8,273 (49.99)	2,572 (15.54)	2,774 (16.8)	169 (1.02)	1420 (8.58)	N/A	N/A
	Standard-dose	16,544	72.5 ± 5.6	8241 (49.81)	2,506 (15.15)	2,648 (16)	160 (0.97)	1328 (8.03)	N/A	N/A
Johansen 2023 (DANFLU-1)	High-dose	6,245	71.8 ± 3.9	2956 (47.3)	574 (9.2)	1,227 (19.7)	137 (2.2)	458 (7.3)	450 (7.2)	227 (3.6)
	Standard-dose	6,232	71.7 ± 3.9	2921 (46.9)	588 (9.4)	1,313 (21.1)	138 (2.2)	420 (6.7)	512 (8.2)	190 (3.0)

Data are presented as mean ± SD or as proportions, n (%)

No Number, SD Standard deviation, CVD Cardiovascular disease, COPD Chronic obstructive pulmonary disease, N/A Not available.

Risk of bias

Overall, all included studies were judged to be at low risk of bias using the RoB 2 tool (Fig. 2). FinFluHD was least susceptible to bias from deviations from intended interventions through its modified double-blind design [13], whereas GALFLU, DANFLU-1, and DANFLU-2 were conducted as open-label trials [10, 12, 14]. Although this may raise some concern in principle, the potential impact was likely minimal given the use of objective outcomes.

Fig. 2.

Overview of the risk of bias of the included randomized controlled trials

Primary outcome. Cardio-respiratory hospitalization

The high-dose group, compared to the standard-dose group, showed a statistically significant reduction in cardio-respiratory hospitalization (OR 0.93, 95% CI: [0.90; 0.97], P = 0.0004) (Fig. 3a) and cardiovascular hospitalization (OR 0.93, 95% CI: [0.89; 0.98], P = 0.009) (Fig. 3b). However, there was no statistically significant difference in respiratory hospitalization (OR 0.90, 95% CI: [0.80–1.01], P = 0.072) (Fig. 3c).

Fig. 3.

Forest plots comparing high-dose vs. standard-dose quadrivalent influenza vaccine for primary hospitalization outcomes in older adults. a Cardio-respiratory hospitalization. b Cardiovascular hospitalization. c Respiratory hospitalization. OR odds ratio, CI confidence interval

No significant heterogeneity was detected for cardio-respiratory hospitalization or cardiovascular hospitalization (I² = 0% for both outcomes). In contrast, respiratory hospitalization showed moderate heterogeneity (I² = 50.5%) (Fig. 3c). Leave-one-out sensitivity analysis indicated that heterogeneity was most reduced after exclusion of DANFLU-2 (I² = 15.6%) [12]. Notably, omitting DANFLU-2 also rendered the pooled effect statistically significant (OR 0.84, 95% CI [0.73–0.98]). However, excluding any single study did not meaningfully alter the overall effect (Supplementary Fig. 1a).

Subgroup analysis according to baseline CVD status showed that, for cardiovascular hospitalization, the high-dose group had a statistically significant reduction among participants without CVD (OR 0.91, 95% CI: [0.84; 0.98]), whereas no significant difference was observed among those with CVD (OR 0.94, 95% CI: [0.88; 1.01]); however, the test for subgroup differences was not significant (P = 0.49) (Supplementary Fig. 3). No significant subgroup differences were observed according to baseline CVD status for either cardio-respiratory hospitalization (P = 0.83) (Supplementary Fig. 2) or respiratory hospitalization (P = 0.47) (Supplementary Fig. 4).

Based on pooled crude event rates, ARR and corresponding NNT were 0.00141 and 708 (95% CI 456–1,580) for cardio-respiratory hospitalization, 0.00080 and 1,256 (95% CI 719–4,944) for cardiovascular hospitalization, and 0.00066 and 1,516 (95% CI 847–7,190) for respiratory hospitalization.

Secondary outcomes

Respiratory hospitalization

The high-dose group, compared with the standard-dose group, showed a statistically significant reduction in influenza hospitalization (OR 0.61, 95% CI: [0.50; 0.74], P < 0.0001) (Fig. 4b) and lab-confirmed influenza hospitalization (OR 0.69, 95% CI: [0.56; 0.86], P = 0.0008) (Fig. 4c). However, there was no statistically significant difference in influenza or pneumonia hospitalization (OR 0.82, 95% CI: [0.66; 1.04], P = 0.097) (Fig. 4a) or pneumonia hospitalization (OR 0.98, 95% CI: [0.90; 1.06], P = 0.59) (Fig. 4d).

Fig. 4.

Forest plots of influenza- and pneumonia-related hospitalization outcomes with high-dose vs. standard-dose quadrivalent influenza vaccine in older adults. a Influenza or pneumonia hospitalization. b Influenza hospitalization. c Lab-confirmed influenza hospitalization. d Pneumonia hospitalization. OR odds ratio, CI confidence interval

No significant heterogeneity was observed for influenza hospitalization or pneumonia hospitalization (I² = 0% for both outcomes) (Fig. 4). Lab-confirmed influenza hospitalization showed low-to-moderate heterogeneity (I² = 33.8%) (Fig. 4c). By contrast, influenza or pneumonia hospitalization was associated with substantial heterogeneity (I² = 71.5%) (Fig. 4a). Leave-one-out sensitivity analysis showed that heterogeneity was most reduced after exclusion of DANFLU-1 (I² = 49.7%) [10]. However, omission of any single study did not materially alter the pooled effect estimate (Supplementary Fig. 1b).

Cardiovascular hospitalization

MACE definitions were broadly comparable but not identical across trials: GALFLU [14] and FinFluHD [13] used the same ischemic/cerebrovascular definition, whereas DANFLU-2 [12] used a slightly broader composite that additionally included heart failure hospitalization and cardiovascular death (Supplementary Table 4).

Compared with the standard-dose group, the high-dose group showed a statistically significant reduction in hospitalization for heart failure (OR 0.80, 95% CI: [0.69; 0.94], P = 0.0050) (Fig. 5b). In contrast, no statistically significant differences were observed for MACE (OR 0.95, 95% CI: [0.85; 1.07], P = 0.394) (Fig. 5a), hospitalization for myocardial infarction (OR 1.04, 95% CI: [0.83; 1.29], P = 0.753) (Fig. 5c), hospitalization for atrial fibrillation (OR 0.96, 95% CI: [0.87; 1.05], P = 0.383) (Supplementary Fig. 5a), or stroke (OR 0.95, 95% CI: [0.85; 1.06], P = 0.374) (Supplementary Fig. 5b).

Fig. 5.

Forest plots of cardiovascular outcomes with high-dose vs. standard-dose quadrivalent influenza vaccine in older adults. a Major adverse cardiovascular events (MACE). b Heart failure. c Myocardial infarction. OR odds ratio, CI confidence interval

No heterogeneity was detected for heart failure, atrial fibrillation, or stroke (I² = 0% for all outcomes) (Supplementary Fig. 5). Low-to-moderate heterogeneity was observed for MACE (I² = 33.7%), whereas myocardial infarction showed moderate heterogeneity (I² = 45.3%). (Fig. 5)

In a sensitivity analysis, exclusion of the trial with a different outcome definition (DANFLU-2 2025) did not materially change the overall association, and the pooled estimate remained statistically nonsignificant (OR 0.98, 95% CI 0.84–1.14). Between-study heterogeneity became substantial after exclusion of DANFLU-2 2025 (I² = 65.4%) (Supplementary Fig. 6).

General clinical outcomes

High-dose QIV showed no statistically significant difference in total hospitalization (OR 0.97, 95% CI: [0.93; 1.01], P = 0.12) (Supplementary Fig. 7a), with no significant heterogeneity (I² = 0%). Likewise, no statistically significant difference was observed for all-cause mortality (OR 0.93, 95% CI: [0.77; 1.11], P = 0.413) (Supplementary Fig. 7b), although substantial heterogeneity was detected (I² = 70.7%). Leave-one-out sensitivity analysis for all-cause mortality showed that heterogeneity was best reduced after excluding DANFLU-1 2023 (I² = 51.9%) [10]. However, omission of any single study did not materially alter the pooled effect estimate, indicating that the overall result was robust (Supplementary Fig. 1c).

Safety outcomes

High-dose QIV, compared to standard-dose group, showed no statistically significant difference in serious adverse events (OR 0.98, 95% CI: [0.93; 1.03], P = 0.419) (Supplementary Fig. 8a) or serious adverse reactions (OR 0.49, 95% CI: [0.10; 2.35], P = 0.371) (Supplementary Fig. 8b). No significant heterogeneity was detected for either serious adverse events or serious adverse reactions (I² = 0% for both outcomes).

Trial sequential analysis

TSA for cardio-respiratory hospitalization and cardiovascular hospitalization showed that the cumulative Z-curve crossed TSMB for benefit, indicating that the pooled effects remained statistically significant after adjustment for repeated significance testing. In both outcomes, the cumulative information size exceeded the RIS, suggesting that the available evidence is sufficient to support firm conclusions (Supplementary Figs. 9 & 10a).

In contrast, TSA for respiratory hospitalization showed that the cumulative Z-curve did not cross TSMB for benefit and remained within the sequential boundaries. Moreover, the cumulative information size did not reach the RIS, indicating that the current evidence is not yet conclusive and that additional adequately powered trials are required to provide firm evidence and to minimize the risk of random error (Supplementary Fig. 10b).

Meta-regression

In study-level meta-regression analyses, baseline CVD prevalence did not significantly modify the relative effect of HD-QIV versus SD-QIV for any primary outcome as shown in Supplementary Tables 5 and Supplementary Fig. 11. The regression coefficient was 0.0024 (95% CI: [-0.0038 to 0.0086], P = 0.4546) for cardiorespiratory hospitalization, -0.0021 (95% CI: [-0.0105 to 0.0063], P = 0.6288) for cardiovascular hospitalization, and 0.0084 (95% CI: [-0.0008 to 0.0176], P = 0.0733) for respiratory hospitalization.

Certainty of evidence

According to the GRADE assessment, the certainty of evidence was high for hospitalization for cardio-respiratory disease and cardiovascular disease, supporting strong confidence in the observed benefit of QIV-HD over QIV-SD for these outcomes. The certainty was moderate for hospitalization for influenza and heart failure, owing to the low number of events. By contrast, the certainty of evidence for hospitalization for influenza or pneumonia was low, primarily because of significant heterogeneity and imprecision. (Table 3)

Table 3.

Grading of recommendations assessment, development, and evaluation (GRADE) evidence profile

Certainty assessment						№ of patients		Relative Effect (95% CI)	Certainty
№ of studies	Risk of bias	Inconsistency	Indirectness	Imprecision	Others	QIV-HD	QIV-SD
Hospitalization for Cardio-Respiratory Disease
4 RCTs	not serious	not serious	not serious	not serious	none	5,093/256,105 (2.0%)	5,450/255,785 (2.1%)	OR 0.93 (0.90 to 0.97)	⨁⨁⨁⨁ High
Hospitalization for Cardiovascular Disease
4 RCTs	not serious	not serious	not serious	not serious	none	2,945/256,105 (1.1%)	3,145/255,785 (1.2%)	OR 0.93 (0.89 to 0.98)	⨁⨁⨁⨁ High
Hospitalization for Influenza or Pneumonia
4 RCTs	not serious	seriousa	not serious	seriousb	none	1,380/256,105 (0.5%)	1,521/255,785 (0.6%)	OR 0.82 (0.66 to 1.04)	⨁⨁◯◯ Lowa, b
Hospitalization for Influenza
3 RCTs	not serious	not serious	not serious	seriousc	none	170/249,860 (0.1%)	279/249,553 (0.1%)	OR 0.61 (0.50 to 0.74)	⨁⨁⨁◯ Moderatec
Hospitalization for Heart Failure
3 RCTs	not serious	not serious	not serious	seriousc	none	294/249,860 (0.1%)	366/249,553 (0.1%)	OR 0.80 (0.69 to 0.94)	⨁⨁⨁◯ Moderatec

RCT Randomized controlled trial, CI Confidence interval, OR Odds ratio, QIV-HD High-dose quadrivalent influenza vaccine, QIV-SD Standard-dose quadrivalent influenza vaccine

Explanations: a. I^2 ≥ 50%; shows significant heterogeneity, b. Cross no effect line (1.0), does not exclude the risk of appreciable benefit/harm, c. Low number of events

Discussion

Summary of findings

In this meta-analysis, we tested the hypothesis that the HD-QIV provides better clinical protection than SD-QIV in older adults. Across four RCTs including 511,890 participants, HD-QIV significantly reduced cardio-respiratory and cardiovascular hospitalizations, although no significant reduction was observed for respiratory hospitalization overall. Among secondary outcomes, HD-QIV was associated with lower rates of influenza hospitalization, laboratory-confirmed influenza hospitalization, and heart failure hospitalization. However, it did not significantly reduce influenza or pneumonia hospitalization, pneumonia hospitalization alone, MACE, myocardial infarction, atrial fibrillation, stroke, total hospitalization, or all-cause mortality. Importantly, both vaccines showed a similar safety profile, with no significant differences in serious adverse events or serious adverse reactions. Trial sequential analysis indicated that the evidence for the primary significant outcomes was sufficient, supporting the reliability of the observed benefits of HD-QIV.

Interpretation and clinical relevance

The synergy between seasonal influenza and acute cardiovascular decompensation has emerged as a focal point in contemporary preventive cardiology. Rather than acting as a localized respiratory insult, influenza serves as a profound systemic trigger for atherothrombotic events, particularly in high-risk populations [24, 25]. Robust epidemiological data have highlighted a dramatic six-fold elevation in the risk of acute myocardial infarction within the initial seven days of laboratory-confirmed infection [26]. This risk extends beyond ischemic events; community-wide surges in influenza-like illness demonstrate a striking temporal correlation with hospitalizations for heart failure exacerbations [27]. This “Cardiovascular-Influenza Axis” underscores a pathophysiological state where systemic inflammation, metabolic stress, and plaque destabilization converge to drive significant cardiovascular morbidity [26]. Given the high stakes of these secondary complications, maximizing vaccine-mediated protection is essential, as standard formulations may fail to provide an adequate immunological shield for the aging cardiovascular system.

In our findings, HD-QIV significantly reduced cardio-respiratory and cardiovascular hospitalizations compared with SD-QIV. These findings are consistent with the meta-analysis by Skaarup et al., who analyzed randomized trials of high-dose versus standard-dose influenza vaccines in adults aged 65 years or older and reported that high-dose vaccination was associated with lower rates of pneumonia or influenza hospitalization and all-cause hospitalization, although that study was not limited to quadrivalent vaccines and did not specifically focus on cardiovascular outcomes [11]. Similarly, the DANFLU-2 trial conducted by Johansen et al. evaluated 332,438 older adults randomized to HD-QIV or SD-QIV and observed lower rates of cardio-respiratory and cardiovascular hospitalization with the high-dose strategy [12]. This may be explained by the known effects of influenza on the cardiovascular system because influenza infection can trigger systemic inflammation, endothelial dysfunction, sympathetic activation, plaque instability, and a prothrombotic state, processes that may precipitate acute cardiovascular events and hospitalization in vulnerable older adults [27].

An important observation in our analysis was the reduction in influenza hospitalization and laboratory-confirmed influenza hospitalization with HD-QIV, while broader respiratory outcomes such as pneumonia or the composite of influenza or pneumonia hospitalization showed less clear differences. This pattern is consistent with the prespecified pooled FLUNITY-HD analysis, in which Johansen et al. reported that high-dose influenza vaccination in older adults reduced several severe cardio-respiratory outcomes, including influenza-related hospital events, compared with standard-dose vaccination [23]. A similar trend was observed in subgroup analyses of the DANFLU-2 trial, where the relative effectiveness of high-dose vaccination against influenza hospitalization remained consistent across different cardiovascular risk groups [28]. One possible explanation is that outcomes directly linked to influenza infection are more sensitive to differences in vaccine immunogenicity. In contrast, pneumonia-based endpoints are influenced by multiple competing causes such as bacterial infections, aspiration, chronic lung disease, and differences in diagnostic coding, which may dilute the measurable effect of vaccination on influenza-specific respiratory outcomes [11, 23].

In practical terms, NNT of 708 for cardio-respiratory hospitalization and 1,256 for cardiovascular hospitalization means that, over one influenza season, one additional cardio-respiratory hospitalization would be prevented for every 708 older adults vaccinated with HD-QIV instead of SD-QIV, and one additional cardiovascular hospitalization would be prevented for every 1,256 vaccinated older adults. Although these absolute benefits are modest, they remain clinically relevant because influenza vaccination is delivered at scale to very large numbers of older adults, so even small effects may prevent a substantial number of hospitalizations and reduce seasonal healthcare utilization.

Among the cardiovascular outcomes, the reduction in heart failure hospitalization is particularly notable. In our analysis, HD-QIV was associated with a significant decrease in heart failure hospitalization, whereas other ischemic endpoints such as myocardial infarction and stroke, were not significantly different. A similar pattern was observed in the DANFLU-2 trial, where Johansen et al. reported that the reduction in cardiovascular hospitalization with high-dose quadrivalent influenza vaccination was largely driven by fewer admissions for heart failure in older adults randomized to the high-dose strategy [28]. Consistent findings were also suggested in the pooled analysis of the FLUNITY-HD trials, which indicated that HD-QIV may provide broader protection against severe cardiovascular and respiratory events among older adults [23]. The pathophysiologic basis for this finding is strong as the influenza infection can worsen heart failure through fever-related tachycardia, increased metabolic demand, hypoxemia, inflammatory myocardial depression, fluid imbalance, and increased pulmonary vascular stress. Supporting this link, Kytömaa et al. in the ARIC study found that periods of increased influenza-like illness activity were associated with increased heart failure hospitalizations at the population level [27].

Our subgroup analyses stratified by present of baseline cardiovascular risk offer important clinical nuance. Although the reduction in cardiovascular hospitalizations reached statistical significance exclusively among participants without pre-existing cardiovascular disease—falling short of significance in the established disease cohort—the formal interaction testing was unequivocally negative. Furthermore, we observed no significant subgroup heterogeneity for the broader cardio-respiratory or respiratory endpoints. Moreover, exploratory meta-regression analyses did not identify a significant association between baseline CVD prevalence and treatment effect for any of the primary outcomes.

These findings are robustly corroborated by recent landmark analyses. A prespecified pooled analysis of the FLUNITY-HD trials demonstrated that the relative clinical advantage of high-dose vaccination is not substantially modified by a patient’s cardiovascular history [23]. In a remarkably similar vein, a dedicated sub-analysis of the DANFLU-2 trial revealed that while individuals with baseline atherosclerotic cardiovascular disease endure a predictably higher absolute burden of cardiopulmonary events, the relative vaccine effectiveness of the high-dose formulation compared to the standard dose remains stable regardless of underlying disease status [28].

The conclusion that this cardioprotective envelope extends broadly across varying risk profiles is further supported by large-scale observational data. An analysis of nearly 200,000 adults utilizing English electronic health records found that influenza vaccination significantly reduces the incidence of first acute cardiovascular events, maintaining this protective effect consistently across all age groups and across both high and low baseline cardiovascular risk strata [29]. Ultimately, this convergence of trial and real-world data indicates that the preventative value of HD-QIV is not confined merely to secondary prevention in high-risk cardiac patients, but serves a critical, highly effective role in primary prevention for the broader older adult population. Conversely, because patients with established cardiovascular disease operate at a drastically higher baseline risk of hospitalization, their absolute clinical benefit from the enhanced vaccine remains uniquely substantial.

Our findings strongly align with the shifting paradigm in global cardiovascular guidelines, which increasingly recognize immunization not merely as an infectious disease intervention, but as a core pillar of secondary cardiovascular prevention. Recent consensus statements, including those from the European Society of Cardiology (ESC) and the American College of Cardiology/American Heart Association (ACC/AHA), explicitly endorse annual influenza vaccination for patients with established cardiovascular disease—including stable atherosclerotic disease, heart failure, and acute coronary syndromes—to mitigate the risk of MACE and mortality [30, 31]. These findings suggest that, for adults aged 65 years and older, particularly those with underlying cardiovascular disease, the high-dose formulation may transition from a preferred alternative to the standard of care.

Regarding safety, our findings further support the tolerability of HD-QIV in older adults. In this meta-analysis, HD-QIV was not associated with a significant increase in serious adverse events or serious adverse reactions compared with SD-QIV, indicating that the reduction in hospitalization outcomes was not accompanied by major safety concerns. This is consistent with Skaarup et al., who also reported improved clinical outcomes with high-dose influenza vaccination without an excess in all-cause mortality. Although high-dose vaccines may cause slightly stronger short-term reactions, these do not appear to translate into serious complications, while the enhanced immune response may help prevent influenza-related cardiopulmonary events that lead to hospitalization [11].

The heterogeneity observed in respiratory outcomes likely reflects differences across trials in endpoint definitions, healthcare systems, influenza seasons, and diagnostic coding practices. The studies were conducted in different countries and across distinct influenza seasons, including the COVID-19–affected 2019–2020 period. Respiratory endpoints were also not fully uniform: FinFluHD used primary ICD-10 discharge ‘J’ and ‘I’ codes to define acute respiratory or cardiovascular hospitalization, whereas the other trials generally used principal discharge ICD-10 codes for influenza, pneumonia, or broader cardio-respiratory outcomes. These factors contributed to the moderate heterogeneity for respiratory hospitalization and the substantial heterogeneity for influenza or pneumonia hospitalization.

Cost effectiveness

As healthcare systems move toward the broader implementation of these enhanced vaccines, the health-economic implications of this transition require meticulous fiscal and clinical scrutiny. While our findings clearly establish the clinical superiority of HD-QIV over its standard-dose counterpart, the current literature still lacks robust, direct comparisons between high-dose trivalent (HD-TIV) and HD-QIV formulations. This evidentiary gap is particularly notable when contextualizing landmark studies like the INVESTED trial, which compared a high-dose trivalent formulation against a standard-dose quadrivalent vaccine in a high-risk cardiopulmonary population and found no significant difference in hard clinical outcomes [32].

The transition from trivalent (TIV) to quadrivalent (QIV) influenza vaccines directly addresses the clinical risk posed by B-lineage strain mismatch [33]. Economically, systematic reviews indicate that while QIV involves higher initial acquisition costs, it is more effective and frequently proves to be cost-effective or even cost-saving compared to TIV [33]. Its clinical and economic value is realized through the prevention of influenza-related hospitalizations and mortality (Warmath et al., 2023).Regional data further reinforce this value proposition. In the Netherlands, QIV was found to be cost-effective for adults aged $\ge$60 at a €20,000 per quality-adjusted life-year threshold [34]. Similarly, Portuguese evidence identifies QIV as a dominant strategy for the elderly, significantly reducing the healthcare burden of seasonal epidemics [35]. While results remain sensitive to annual strain circulation, the collective evidence supports the quadrivalent formulation as a superior value-based investment [33, 34].

Clinical implications

These findings highlight the value of stronger influenza vaccination strategies for improving outcomes in older adults. For patients, HD-QIV may lower the chance of serious flu-related complications that often lead to hospitalization, especially in those of advanced age or living with multiple comorbidities. For clinicians, the evidence supports including higher-antigen influenza vaccines in routine preventive care for older adults to reduce infection-related cardiopulmonary events during flu season. At the institutional level, wider use of high-dose vaccines could help decrease seasonal admissions and reduce healthcare burden in systems serving large elderly populations. From a public health standpoint, giving more immunogenic influenza vaccines to older adults may reduce influenza-related illness while preserving a strong safety profile. Clinicians should therefore consider preferential use of HD-QIV vaccines in adults aged 65 years and older, particularly those with cardiovascular or chronic respiratory disease. Health systems and national immunization programs should also prioritize vaccine access and availability. Future trials should consider prespecifying pericarditis and myopericarditis as exploratory cardiovascular outcomes. These complications are rare but potentially clinically significant manifestations of influenza-related cardiac involvement, yet they were not reported uniformly enough across the included trials to allow pooled analysis in the present review [36].

Strengths

Our meta-analysis has several strengths. First, it represents the first meta-analysis that exclusively evaluates cardiovascular outcomes of HD-QIV, whereas most previous meta-analyses have examined high-dose influenza vaccines more broadly or focused on trivalent formulations rather than quadrivalent vaccines alone. Second, the analysis included only RCTs, which showed an overall low risk of bias across the included studies. By synthesizing data from over 511,000 randomized participants, we achieved the substantial statistical power necessary to detect modest but clinically vital differences in specific cardiovascular outcomes—differences that smaller trials are inherently unequipped to measure accurately.

In addition, we performed prespecified subgroup analyses based on baseline cardiovascular disease status for the primary outcomes, allowing a more detailed interpretation of treatment effects in clinically relevant populations. Furthermore, TSA helped determine whether the accumulated evidence was sufficient to support firm conclusions and whether additional studies were likely to change the findings. Additionally, our strict inclusion criteria, focusing exclusively on comparing high-dose to standard-dose QIV, eliminate the confounding variables of strain coverage mismatches that plagued earlier meta-analyses. Finally, the GRADE approach was used to assess the certainty of evidence for each outcome, with high certainty for two primary outcomes, thereby improving the clinical interpretation of the results.

Limitations

This study also has several limitations. First, only four RCTs were included, despite the very large pooled population, which limits the available evidence for some secondary and safety outcomes. Second, generalizability to non-European settings should be interpreted cautiously. All included trials were conducted in European populations within high-income healthcare systems with established influenza vaccination programs and registry-based outcome ascertainment. Applicability to other regions, particularly low- and middle-income countries, may differ because of variations in influenza epidemiology, healthcare access, hospitalization thresholds, vaccine uptake, and diagnostic coding. Nevertheless, the biological rationale for improved protection with high-dose vaccination in older adults is likely relevant across settings, although confirmation in more diverse populations is needed. Third, three of the included trials were open-label, which may have introduced some performance or detection bias, particularly through potential differences in healthcare-seeking behavior or hospitalization thresholds. Nevertheless, this risk was likely reduced because the main outcomes were objective hospitalization-based events derived from routinely collected registry or administrative data, although it cannot be completely excluded. Fourth, follow-up duration was relatively short, ranging from 3 to 8 months, which restricts conclusions mainly to seasonal effects rather than long-term outcomes. In addition, some outcomes showed considerable heterogeneity, such as influenza or pneumonia hospitalization and all-cause mortality, and the certainty of evidence was lower for certain endpoints due to heterogeneity, imprecision, or low event numbers. Additionally, the definition of MACE was not identical across studies, which may have contributed to variability in cardiovascular outcomes. Finally, more large RCTs are needed to further clarify these findings.

Conclusion

In conclusion, among older adults (≥ 65 years), HD-QIV was associated with a significant reduction in cardio-respiratory and cardiovascular hospitalizations compared with SD-QIV, while maintaining a similar safety profile. Notably, our findings indicate that this protective benefit remains consistent regardless of baseline cardiovascular risk. These findings suggest that HD-QIV may provide additional protection against severe influenza-related and cardiovascular complications in this population. However, given the limited number of available RCTs and the relatively short follow-up duration, further large trials are warranted to confirm these findings.

Authors’ contributions

M.S.E.: Validation. M.S.E. and A.S.: Conceptualization, material preparation, investigation, methodology, visualization, and writing — original draft. A.A.: Investigation, data curation, and writing — original draft. M.R.: Data curation and formal analysis. M.Q., M.I., M.A.F., and M.A.A.: Investigation and writing — original draft. M.I., M.A.: Supervision and writing — review and editing. No paper mills or generative AI were used. All authors read and approved the final manuscript.

Funding

No funding was received for conducting this study.

Data availability

All required data are included in the manuscript or the supplementary material. Any additional data can be obtained from the corresponding author upon a reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.