The effect of the influenza vaccine in reducing the severity of clinical outcomes in patients with COVID-19: a systematic review and meta-analysis

Literature research and study selection

Our search collected 6058 articles from six online databases and 450 articles from other sources including manual search and Google Scholar. From the total of 6508 papers found, we excluded 1818 duplicates. Then 4667 papers were excluded from 4690 papers after title and abstract screening. The remaining 23 articles were checked against the eligibility criteria. 16 in total11,12,13,18,19,20,21,22,23,24,25,26,27,28,29,30 Papers were eventually included for final qualitative analysis, and only 1311,12,13,19,20,21,22,24,25,27,28,29,30 of these were subjected to a quantitative analysis (Fig. 1).

illustration 1

PRISMA flowchart summarizing the literature search and included studies.

study characteristics

Fourteen included studies were cohort studies12,13,18,19,20,21,22,23,24,25,26,28,29,30while the remaining two were case controls11:27. The included studies had a total sample size of 244,642 patients, including 191,496 COVID-19 positive patients. Most of the studies were conducted in Italy13,20,22,23,26,27 and the US12,24,25,28,29while the others were performed in Brazil18Poland19Spain21Iran11and UK30. All studies used the polymerase chain reaction (PCR) test to identify SARS-CoV-2 infection, but three studies did not report the SARS-CoV-2 infection identification test used12,22,29. The summary of all included studies is presented in (Table 1) and the patient data in (Table 2).

Table 1 Summary of included studies.
Table 2 Baseline characteristics of patients in the included studies.

The quality of the included studies

The quality assessment was based on the Newcastle-Ottawa scale. Based on the scale, nine11,12,19,20,21,22,25,26,28 of the sixteen included studies, 9/9 were evaluated, and the other seven studies13,18,23,24,27,29,30 were rated 8/9. All included studies achieved a low risk of bias. (Supplementary Tables 1 and 2).

Quantitative Synthesis

mortality rate

The pooled effect estimates from seven studies12,13,19,20,22,28,30 in a total sample size of 125,658 COVID-19 patients showed no significant difference between the influenza vaccinated group and the non-vaccinated group in terms of mortality rate (RR = 1.20, 95% CI [0.71, 2.04], P= 0.50; Fig. 2a). The pooled results were heterogeneous (P2= 98%). After introducing subgroups by country: USA12:28Italian13,20,22and other countries19.30 results remain non-significant in the US and other groups (RR = 0.82, 95% CI [0.60, 1.13], P= 0.22), (RR = 0.97, 95% CI [0.86, 1.09], P= 0.58) and also not significant in the Italian group (RR = 1.87, 95% CI [1.00, 3.49], P= 0.05), (Fig. 2b). The heterogeneity resolved in the US and other groups (P= 0.16, i2 = 49%) and (P= 0.88, i2 = 0%), while it remains heterogeneous in the Italian group (P= 0.0007, i2 = 86%).

figure 2
figure 2

Mortality rate. This figure shows the forest section of (a) Mortality rate between influenza vaccinated group and non-vaccinated group, (b) Mortality subgroups by country USA, Italy and other patients.

Admission to the hospital

The pooled effect estimates from eight studies12,13,19,22,25,27,29,30 (N=132,460 COVID-19 patients) showed no significant difference between the influenza vaccinated group and the non-vaccinated group in terms of length of hospitalization (RR=1.04, 95% CI [0.84, 1.29], P= 0.75); Fig. 3a). The pooled results were heterogeneous (P2 = 98%), the heterogeneity cannot be resolved. Wilcox et al.30 reported hospitalization and death, after removing it from the analysis, the results remained non-significant (RR = 1.01, 95% CI [0.71, 1.45], P= 0.94) and heterogeneous (P2 = 98%).

figure 3
figure 3

Hospital and ICU Outcomes. This figure shows the forest section of (a) hospital admission, (b) hospitalization time, (c) admission to the intensive care unit, (i.e) time in intensive care.

Hospitalization (days)

The pooled effect estimates from four studies12,19,20,28 in a total sample size of 1716 COVID-19 patients showed no significant difference between the influenza vaccinated group and the non-vaccinated group in terms of length of hospital stay (MD = −0.16, 95% CI [− 2.76, 2.45], P= 0.91; Fig. 3b). The pooled results were heterogeneous (P= 0.004, i2 = 78%), and the heterogeneity could be corrected by removing Conlon et al.12 but the results remained non-significant (RR = 0.97, 95% CI [− 1.23, 3.17], P= 0.39), P= 0.13, i2 = 50%).

Admission to the intensive care unit

The pooled effect estimates from six studies12,13,20,25,28,29(N = 116,399 COVID-19 patients) showed a non-significant difference between the influenza vaccinated group and the non-vaccinated group in terms of length of ICU admission (RR = 0.84, 95% CI [0.44, 1.62], P= 0.60); Fig. 3c. The pooled results were heterogeneous (P2= 95%), the heterogeneity cannot be resolved. Massari et al.13reported the rate of ICU admissions and death after this study was removed from the analysis, the results remained non-significant (RR = 0.71, 95% CI [0.44, 1.16], P= 0.17). The pooled results showed heterogeneity (P= 0.002, i2= 77%).

Intensive time (days)

The pooled effect estimates from two studies20.28(N = 202 COVID-19 patients admitted to the ICU) showed no significant difference between the influenza vaccinated group and the non-vaccinated group in terms of length of stay in the ICU (MD = 0.99, 95 % CI [− 2.15, 4.13], P= 0.54); Fig. 3d. The pooled results were homogeneous (P= 0.32, i2= 0%).

Mechanical ventilation

The pooled effect estimates of four12,19,20,28Studies in 1716 COVID-19 patients showed a significant preference for the influenza vaccinated group over the non-vaccinated group in terms of the duration of mechanical ventilation (RR = 0.72, 95% CI [0.54, 0.96], P= 0.03; Fig. 4a). The pooled results were homogeneous (P= 0.10, i2= 52%).

figure 4
figure 4

Mechanical ventilation and onset of symptoms. This figure shows the forest section of (a) mechanical ventilation, (b) symptomatic cases, (c) asymptomatic cases.

Symptoms experienced by the patient

The pooled effect estimates from three studies11,21,24of 1372 COVID-19 patients showed no significant difference between the influenza vaccinated group and the non-vaccinated group in terms of symptomatic and asymptomatic cases (RR = 0.55, 95% CI [0.23, 1.36], P= 0.20; Fig. 4b), (RR = 1.17, 95% CI [0.76, 1.80], P= 0.49; Fig. 4c). The pooled results were heterogeneous (P2= 95%), (P2= 94%), and the heterogeneity cannot be resolved.

Qualitative Synthesis

Finke et al.18a retrospective cohort study reports data from 53,752 Brazilian COVID-19 patients that patients who received an influenza vaccine had a 16% reduced likelihood of dying (95% CIs [0.78, 0.90]), 7% lower likelihood of needing ICU treatment (95% CIs [0.87, 0.98]) and 17% less likely to need respiratory support (95% CIs [0.77, 0.88]); while Pedote et al.26another retrospective cohort study reported no association between influenza vaccination and mortality (OR = 1.6, 95% CI [0.8–3.2], P= 0.165) or hospitalization (OR = 1.2, 95% CI [0.7–1.9], P= 0.51). Also Pastorino et al.23reported the same non-association results with influenza and pneumococcal vaccine.

Comments are closed.