Abstract
Introduction
Influenza viruses are RNA viruses stratified into influenza A, influenza B, influenza C and influenza D viruses. Seasonal epidemics of influenza A viruses account for the major burden of influenza disease in humans. 1 Influenza A viruses is classified into 15 H and 9 N subtypes based on the antigenic properties of their surface hemagglutinin (HI) and neuraminidase (NA) glycoproteins. 2 In 2009, the emergence of a new subtype of influenza A (H1N1) virus caused the first influenza pandemic of this century. 2 Despite the lower rates of hospitalization and case fatality, 3 the disease burden of influenza A H1N1 in terms of life lost is more severe than that of other typical influenza. 4 Persons living with HIV(PLWH),5–15 pregnant women 16 and young people with chronic diseases 17 were associated with a greater risk of complications.
Acquired immune deficiency syndrome (AIDS) caused by HIV has challenged human health and public health security. 18 According to the Joint United Nations Programme on HIV/AIDS (UNAIDS) data report, by the end of 2022, there were about 39.0 million PLWH in the world, and 630,000 people died of AIDS-related illnesses in 2022 alone 19 Living with HIV may be positively associated with increased risk of severe influenza symptoms and excess in-hospital mortality. 20 Moreover, there is a risk of co-epidemic of influenza and COVID-19 this autumn and winter. Collectively, compelling evidence exists to sustain influenza vaccination in PLWH.
A review revealed a lower rate of seroconversion of influenza vaccine in PLWH compared with the general population. 21 Considering the enhanced morbidity and mortality of influenza in PLWH, two doses of influenza vaccine were recommended for PLWH in some countries and regions. 22 Nevertheless, the vaccination rate of influenza booster in PLWH is low 22 because of concerns about vaccine efficacy and adverse effects and misconceptions about influenza disease. 23 One published meta-analysis has evaluated the immunogenicity and safety of adjuvant influenza vaccination in PLWH, 24 and a relevant analysis of two-dose influenza vaccine is still lacking. We conducted a meta-analysis to compare the immunogenicity (seroprotection rate, seroconversion rate, GMT) and safety of booster doses in PLWH.
Methods
Search strategy
According to the meta-analysis (PRISMA) reporting guideline formulated search strategy, studies published from 1 January, 2009 until 13 April, 2023 were searched in PubMed, Web of Science, and Embase. The following search terms were used: “Influenza, Human” (such as “influenza,” “flu,” “grippe,” and Vaccines (such as “vaccination,” “immunization,” “Vaccines,” “adjuvant,” “unadjuvant,” “non-adjuvant” and HIV (such as “HIV,” “AIDS”).
Inclusion criteria
(1) Types of population: Studies on the immunogenicity and safety of two doses of influenza vaccine in PLWH were published in three databases (PubMed, Web of Science, Embase) since 2009. (2) Types of study design: RCTs, self-control and cohort studies. (3) Types of intervention: the experimental group received two doses of influenza vaccine, and the control group received single dose. Blood samples were collected before and within 21 to 28 days after vaccination. (4) Types of outcome: primary immunogenicity outcomes is assessed by Geometric mean titer (GMT), seroconversion (SC) and seroprotection (SP), criteria for evaluating safety of influenza vaccines in PLWH is the risk of systemic or local adverse effects after vaccination. The seroprotection rate (SPR) and seroconversion rate (SCR) of the vaccine were detected by hemagglutination-inhibition (HI) assay. SPR was defined as the proportion of patients with HI antibody titer ≥1∶40 after vaccination. SCR was defined as the proportion of patients with HI antibody titer <1:10 before vaccination and ≥1:40 after vaccination, or the proportion of patients with HI antibody titer >1:10 before vaccination and a 4-fold increase in HI antibody after vaccination.
Data extraction
Abstracts meeting the inclusion criteria were identified through three databases (PubMed, Web of Science, and Embase) by a reviewer for preliminary screening, and all articles included in this phase required a full-text review.
One researcher independently read the title and abstract of the literature for preliminary screening. The articles obtained from the initial screening were read in full according to the inclusion and exclusion criteria to determine the selection. Different opinions in the screening process were resolved through discussion. If there were still different opinions in the discussion, the third researcher would judge whether to include them. The extracted data included: first author, year of publication, study year, location, study design, participants, follow up duration, types of population, age, subtype of influenza virus, vaccine type, sample size.
Quality assessment
Cochrane’s tools were used to assess the quality of RCTs, mainly including random allocation methods, allocation scheme concealment, blinding, completeness of outcome data, selection bias, and other biases. The quality of cohort studies was assessed according to the Newcastle-Ottawa Scale (NOS), which mainly included the selection of study subjects (0–4), comparability between groups (0–2), and outcome measurement bias (0–3). NOS scores with comparability removed were used to assess the quality of self-control studies.
Statistical analysis
State 11.0 software was used to conduct a meta-analysis of the studies that met the inclusion criteria. According to the heterogeneity test, a random (fixed) effect model was used (if the heterogeneity was small and p > .1, the fixed effect model was selected). The RR values of seroconversion rate and seroprotection rate were combined and the 95%CI was calculated. The test level was α = 0.05, and the SMD and 95%CI were calculated from the mean and SD of the GMT. Sensitivity analysis was performed using the method of article by article exclusion, and the results were considered robust if the combined effect was less than 10%. Heterogeneity was classified as insignificant (0–25%), mild (26–50%), moderate (51–75%) and significant (>75%) based on I2. 25 Subgroup and sensitivity analyses were used to identify and reduce sources of variation between studies. Publication bias was assessed by funnel plots and Egger linear regression analysis.
Results
Search results
The initial literature search yielded 2124 results (Figure 1); 34 duplicate articles were removed by using EndNote. A total of 22 studies were read in full, and of these, eight studies were excluded due to inappropriate study design for not two doses versus one (n = 2), no full text (n = 2), duplicated date (n = 1) and no control group (n = 3). A total of 2436 participants from 14 studies were included. Table 1 summarizes the literature regarding the immunogenicity and safety of two doses of HIN1 influenza vaccine in PLWH, including location, study design, population type, sample size, vaccine type, vaccine dose, interventions, antibody measurement and NOS score. Of the 14 included studies, only two were from cohort studies, in Denmark
7
and Japan.
8
A cohort study examined the administration of both adjuvanted and non-adjuvanted vaccines.
7
Both RCTS5,6 used non-adjuvanted vaccines, and one
5
involved pregnant women. Ten self control studies8–14,26–28 were conducted in Australia, America, Germany, the United Kingdom, Brazil, France, and Switzerland; four studies involved children living with HIV and one study involved pregnant women living with HIV. Thirteen studies5–11,13,14,26–28 detected seroconversion rates, 135–15 assessed seroprotection rates, 135–11,13–15 assessed GMTS, and 55,8,9,12,27 assessed adverse effects (systemic or local adverse effects). Santini-Oliveira et al.
13
reported the administration of two doses of adjuvant vaccine and two doses of non-adjuvant vaccine, and another study
14
reported the administration of two doses of standard-dose vaccine and two doses of high-dose vaccine. If possible, we included them as separate experiments in the analysis. Flow chart of study selection. Summary of the characteristics of included studies. T: the test group; C: the control group; RCT: randomized controlled trial; NR: not reported; CI: confidence interval; HAI: hemagglutination inhibition assay; MN: microneutralization assay; NOS: the Newcastle–Ottawa scale for studies; S: the number of stars for selection section; C: the number of stars for comparability section; O: the number of stars for outcome section; T: the total number of stars.
Quality assessment
Table 1 illustrates the low overall risk of bias for the two cohort studies, with NOS scores of six and eight, respectively. Self controlled studies lack comparability, with modified NOS scores of four to six and general quality. The overall quality of the two RCTS was moderate; neither involved other sources of bias nor blinding, meanwhile, the study by Cooper et al. did not mention concealment of the allocation scheme.
Meta-analysis
Multiple studies have manifested that influenza vaccination for PLWH is considered safe, and two doses of vaccine can further enhance immunogenicity.6–13,26 However, some studies have also claimed that two doses of vaccine have limited immune enhancement effects on PLWH.5,27,28 To better understand the immunogenicity and safety of two doses of influenza vaccine, a meta-analysis was conducted on the relevant literature of all PLWH who received two doses of influenza vaccines.
Immunogenicity
GMT
Results of vaccine immunity analysis.

Subgroup analysis forest plot of vaccine types for GMT in immunogenicity.
Seroprotection
In 13 studies with a total of 2397 PLWH, the seroprotection rate increased in two-doses vaccinated population compared with single-dose vaccinated population (RR: 1.14, 95%CI: 1.08–1.21, p < .01, Table 2, Figure 3b). Subgroup analysis revealed no significant differences in seroprotection rates by dose, study population, age, vaccine type and study design. Likewise, the following Egger’s test corroborated no publication bias (Supplementary Figure 1b). The pooled RR manifested no remarkable differences after sensitivity analysis that each study was excluded. In the aggregate, the heterogeneity of the meta outcome cannot be explained by any single study. Summary of immunogenicity forest plots included in the analysis. (a) GMT forest plots of the 13 studies included in the analysis. (b) Forest plot of seroprotection rate from the 13 studies included in the analysis. (c) Forest plot of seroconversion rate from the 13 studies included in the analysis.
Seroconversion
Figure 3(c) manifests the aggregate seroconversion rate RR and corresponding 95% confidence interval (CI) of PLWH after receiving influenza vaccines (two doses vs single dose or no vaccine) in 13 studies. The pooled RR is 1.25 (95% CI: 1.16–1.34, p < .01, Table 2, Figure 3(c)). Moreover, there is substantial heterogeneity in this result, and subgroup analysis of factors (study design, age, study population, vaccine type, and vaccine dose) that may lead to heterogeneity did not manifeste significant differences. The funnel plot and Egger test results do not reveale the existence of publication bias (Supplementary Figure 1c). Similarly, the results of sensitivity analysis also reveals that a single study does not affect the stability of the results. Thus, none of the studies can explain the heterogeneity of the results.
Collectively, the meta-analysis results of three immunogenicity related indicators (GMT, seroprotection rate and seroconversion rate) corroborate the better humoral response of PLWH receiving two doses of influenza vaccine compared to single dose of vaccine.
Safety
Adverse reaction analysis of vaccination.

Forest plots of statistically different adverse effects. (a) Forest plot of fever. (b) Forest plot of myalgia.
Discussion
Influenza vaccinations are strongly advocated to PLWH to reduce the risk of H1N1 infection and associated severe disease. However, the immune response to receiving single dose of influenza vaccine are marginal, and a cohort study involving 410 children living with HIV revealed suboptimal effectiveness with a single dose of influenza vaccine. 29 The immunogenicity data of two-dose influenza vaccine are controversial,5–15 resulting in a paucity of comparisons of the immunogenicity and safety of influenza vaccines administered with two doses versus a single dose. The United States advocates two doses of non-adjuvant vaccine for PLWH, and there are also studies that support the better immunogenicity of adjuvant vaccination. Moreover, concerns about vaccine safety can lead to a decrease in vaccination rates. 30 Based on this occurrence, we conducted this meta-analysis to further explore the immunogenicity and safety of two-dose vaccination regimens in PLWH. The results of this study could provide guidance for the optimization of influenza vaccination strategies in PLWH.
By consulting the literature, conclusive evidence was found to sustain the vaccination of PLWH against influenza. A study involving 151 adults living with HIV in the United States from 2013 to 2016 corroborated that participants who received single dose of influenza vaccine had an increase in antibody titers; 31 yet data regarding immune response 31 and durability 32 were unsatisfactory. Concurrently lower CD4+ T cell counts were associated with poorer humoral immunity.21,33 In contrast, PLWH who receive two doses of influenza vaccine tend to have better immunogenicity6–13,26 and persistence data. 34 Despite better and more durable utility to two doses of vaccination, uptake rates remain low, mainly due to vaccine hesitancy related to vaccine safety. 30 Multiple studies on vaccine safety have shown that administering two doses of vaccine is not associated with serious or fatal adverse reactions.5,27,28 To improve this situation, it is essential to help PLWH correctly understand the safety of influenza vaccination.
The results of the meta-analysis are consistent with previous studies (retrieved literature). GMT (SMD: 0.42, 95%CI: 0.35–0.49) and seroprotection rate (RR: 1.14, 95%CI: 1.08–1.21) and seroconversion rate (RR: 1.25, 95%CI: 1.16–1.34) are superior than single dose, manifesting that two doses of influenza vaccine might be a more efficaceous vaccination regimen. Meanwhile, two doses of adjuvant vaccines produce higher GMT compared to two doses of non-adjuvant vaccines. Therefore, it would be more beneficial to promote two-dose adjuvant vaccination in PLWH to control the risk of complications. Besides, the safety analysis revealed that the fever risk of the two doses of influenza vaccine was 3.42-fold higher against that single dose, with a decreased risk of myalgia (RR = 0.75, 95%CI: 0.58–0.98). But there were no statistical difference in other local or systemic serious adverse events. In general, the results of this meta-analysis corroborate superior immunogenicity and an acceptable safety profile of two dose vaccination.
Several limitations should be considered in the interpretation of the results of our review. First, we included the English literature, so the findings cannot be generalized to studies published in non-English languages. Second, self-controlled studies were included due to limited data. The research quality of self-controlled studies is worse than that of randomized controlled studies and cohort studies, which may attenuate the level of evidence. Third, the results of the meta-analysis were moderately heterogeneous. Except for subgroup analysis of GMT, which found that vaccine type was an important source of heterogeneity, the sources of heterogeneity in other results were still ambiguous and need further evaluation and study.
Conclusion
This meta-analysis manifests that two-dose influenza vaccine regimens provide better humoral responses than single-dose regimens, and two-dose adjuvant influenza vaccination regimens were superior to the two-dose non-adjuvant influenza vaccination schedule. In terms of safety, the two-dose vaccine regimen was associated with a higher risk of fever and lower risk of myodynia, but there were no statistically significant differences in other local or systemic adverse reactions. Reasonable evidence supports the use of two doses of influenza vaccine in PLWH.
Supplemental Material
Supplemental Material - A meta-analysis of immunogenicity and safety of two versus single-doses of influenza A (H1N1) vaccine in person living with HIV
Supplemental Material for A meta-analysis of immunogenicity and safety of two versus single-doses of influenza A (H1N1) vaccine in person living with HIV by Teng Zhang, Juan Geng, Yazhe Du, Haiyan Yang, Yuefei Jin, Shuaiyin Chen and Guangcai Duan in International Journal of STD & AIDS.
Footnotes
Author contributions
SC and TZ designed and implemented this research. TZ and JG did literature search, quality assessment and data extraction. All authors wrote and revised the manuscript together. All authors read and approved the final manuscript.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by National Science and Technology Specific Projects (2018ZX10301407), National Natural Science Foundation of China (82273695, 82073618).
Data availability statement
All data generated or analysed during this study are included in this published article.
Supplemental Material
Supplemental material for this article is available online.
Appendix
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
