Impact of Nirsevimab Immunization on Pediatric Hospitalization Rates: A Systematic Review and Meta-Analysis (2024)

Abstract

A systematic review with a meta-analysis was performed to gather available evidence on the effectiveness of monoclonal antibody nirsevimab in the prevention of lower respiratory tract diseases (LRTDs) due to respiratory syncytial virus (RSV) in children and newborns (CRD42024540669). Studies reporting on real-world experience and randomized controlled trials (RCTs) were searched for in three databases (PubMed, Embase, and Scopus) until 1 May 2024. Our analysis included five RCTs, seven real-world reports, and one official report from the health authorities. Due to the cross-reporting of RCTs and the inclusion of multiple series in a single study, the meta-analysis was performed on 45,238 infants from 19 series. The meta-analysis documented a pooled immunization efficacy of 88.40% (95% confidence interval (95% CI) from 84.70 to 91.21) on the occurrence of hospital admission due to RSV, with moderate heterogeneity (I² 24.3%, 95% CI 0.0 to 56.6). Immunization efficacy decreased with the overall length of the observation time (Spearman's r = -0.546, p = 0.016), and the risk of breakthrough infections was substantially greater in studies with observation times ≥150 days compared to studies lasting <150 days (risk ratio 2.170, 95% CI 1.860 to 2.532). However, the effect of observation time in meta-regression analysis was conflicting (β = 0.001, 95% CI -0.001 to 0.002; p = 0.092). In conclusion, the delivery of nirsevimab was quite effective in preventing hospital admissions due to LRTDs. However, further analyses of the whole RSV season are required before tailoring specific public health interventions.

Keywords: RSV; immunization; monoclonal antibodies; viral pneumonia.

Figure A1

Forest plot for prevalence estimates of hospitalizations due to RSV-associated LRTIs among a pooled sample of 9365 children not immunized with nirsevimab. A pooled prevalence of 4.25% (95% CI 2.42 to 7.38) was calculated, affected by substantial heterogeneity (I² 98.8%, 95% CI 98.5 to 99.0; tau² = 1.481, Q = 2.98, p < 0.001), with no substantial differences between groups (chi-squared test 2.98, p value = 0.084) [59,61,78,79,82,83,84,85,88].

Figure A2

Forest plot for sensitivity analysis on hospitalization rates for respiratory syncytial virus (RSV)-associated lower respiratory tract infections in infants previously treated with nirsevimab. Sensitivity analysis was performed by excluding one study at a time and b reporting corresponding pooled estimates with heterogeneity estimates (I²) [59,61,78,79,82,83,84,85,88].

Figure A3

Forest plot for sensitivity analysis on hospitalization rates for respiratory syncytial virus (RSV)-associated lower respiratory tract infections in infants not treated with nirsevimab. Sensitivity analysis was performed by excluding one study at a time and reporting corresponding pooled estimates with heterogeneity estimates (I²) [59,61,78,79,82,83,84,85,88].

Figure A4

Forest plot for sensitivity analysis of the immunization efficacy of nirsevimab regarding the prevention of hospital admissions due to respiratory syncytial virus (RSV)-associated lower respiratory tract infections. Sensitivity analysis was performed by excluding one study at a time and reporting corresponding pooled estimates with heterogeneity estimates (I²) [59,61,78,79,82,83,84,85,88].

Figure A5

Funnel plots for prevalence data on the occurrence of hospital admissions due to respiratory syncytial virus (RSV)-associated lower respiratory tract infections. Subfigure (a) individuals treated with nirsevimab; subfigure (b) individuals not treated with nirsevimab [59,61,78,79,82,83,84,85,88].

Figure A6

Radial plots for prevalence data on the occurrence of hospital admissions due to respiratory syncytial virus (RSV)-associated lower respiratory tract infections with their corresponding results for the regression test of funnel plot asymmetry (Egger’s test). Subfigure (a) individuals treated with nirsevimab; subfigure (b) individuals not treated with nirsevimab [59,61,78,79,82,83,84,85,88].

Figure A7

Forest plot for the proportion of RSV-associated LRTIs among the total number of hospitalizations due to LRTIs. A pooled prevalence of 63.60% (95% CI 48.87 to 77.58) was calculated, affected by substantial heterogeneity (I² 60.5%, 95% CI 27.6 to 78.5; tau² = 1.103, Q = 30.40, p = 0.002), with substantial differences between groups (chi-squared test 4.02, p value = 0.044) [78,82,84,85].

Figure 1

Flowchart of included studies.

Figure 2

Summary of the risk of bias (ROB) estimates for observational studies [72,85]. Analyses were performed according to the National Toxicology Program (NTP)’s Office of Health Assessment and Translation (OHAT) handbook and respective ROB tools including all retrieved studies (n = 13).

Figure 3

Correlations between the following: (a) observation time and hospital admission rate as events per 100 people; (b) hospital admission rates among immunized and not immunized infants; (c) observation time and point estimates for immunization efficacy per single study. No substantial correlation was found between observation time and hospitalization rates for infants treated and not treated with nirsevimab (Spearman’s r = 0.340, 95% CI −0.149 to 0.696, and p = 0.154 and r = −0.076, 95% CI −0.523 to 0.404, and p = 0.758, respectively), while a positive correlation was identified between hospital admission rates (Spearman’s r = 0.473, 95% CI 0.009 to 0.769, and p = 0.041), and a negative correlation was found for immunization efficacy and observation time (Spearman’s r = −0.546, 95% CI −0.807 to −0.108, and p = 0.016) [59,61,78,79,82,83,84,85,88].

Figure 4

Forest plot for hospitalization rates due to RSV-associated LRTIs among a pooled sample of 33,884 children immunized with nirsevimab. A rate of 0.42% (95% CI 0.26 to 0.68) was calculated for 143 total hospital admissions (I² 86.1%, 95% CI 79.7 to 90.5; tau² = 0.912, Q 129.70, p < 0.001), with no substantial differences between randomized controlled trials (RCTs) and real-world experience (chi-squared test 0.05, p value = 0.830) [59,61,78,79,82,83,84,85,88].

Figure 5

Forest plot for immunization efficacy (IE) of nirsevimab against RSV-associated LRTIs in real-world experience and randomized controlled trials (RCTs). A pooled IE of 88.4% (95% CI 84.7 to 91.2) was calculated (I² 24.3%, 95% CI 0.0 to 56.6; tau² = 0.105, Q = 23.79, p = 0.162), with substantial differences between groups (chi-squared test 7.20, p value = 0.007) [59,61,78,79,82,83,84,85,88].

Figure 6

Bubble plot for the metaregression of the included studies by observation days. In the regression model, the observation time was characterized as a non-significant explanatory variable for IE (β = 0.001, 95% CI −0.001 to 0.002; p = 0.092). Pooled analyses were affected by low heterogeneity (I² = 19.0%, tau² = 0.055) [59,61,78,79,82,83,84,85,88].

Figure 7

The funnel plot (a) and radial plot (b) for studies reporting on the immunization efficacy of nirsevimab on the occurrence of hospital admissions due to respiratory syncytial virus (RSV)-associated lower respiratory tract infections. A regression test of funnel plot asymmetry (Egger’s test) ruled out substantial publication bias due to the small-study effect (t = −0.54, df = 17, p-value = 0.597, bias estimate: −0.324, SE = 0.601, tau² = 1.376) [59,61,78,79,82,83,84,85,88].