Maternal immunisation with RSVpreF and infant immunisation with the long-acting monoclonal antibodies nirsevimab and clesrovimab are reshaping prevention of respiratory syncytial virus (RSV) disease. Trials showed that RSVpreF reduced severe RSV-associated lower respiratory tract illness and that both antibodies reduced RSV-associated hospitalisations in infants.1, 2, 3 As RSV disease declines, however, a larger share of remaining respiratory admissions may test positive for other viruses. Distinguishing that proportional change from a true increase in their incidence is an urgent surveillance question for Latin America.
If RSV hospitalisations decline, rhinovirus, human metapneumovirus, adenovirus, or parainfluenza virus could account for a larger proportion of the remaining hospitalisations even if their absolute incidence is unchanged. This mathematical shift could be mistaken for viral replacement. A change in test positivity alone is equally difficult to interpret if the number or selection of children tested has changed.
Demonstrating an epidemiological change requires population-based rates, testing denominators, stable diagnostic criteria, and comparable time series before and after implementation.
Rhinovirus warrants particular attention. An analysis of 432,399 detections reported to WHO FluNet by 50 countries between 2016 and 2025 found prolonged circulation, with a median epidemic duration of 31 weeks, frequent secondary peaks, and greater temporal variability in tropical regions. Peak timing was broadly stable between the pre-pandemic and post-pandemic periods, whereas changes in amplitude differed among countries.4 These data do not establish a uniform worldwide increase, but they show why a single seasonal baseline is unlikely to describe every setting.
Latin American observations show why comparable baselines matter. In Santander, Colombia, the proportion of samples positive for rhinovirus was 4.8% during 2020–21 and 28.0% during 2023–24, with greater genotype diversity in the latter period.5 These periods span the COVID-19 pandemic and do not assess the effect of RSV immunisation. In Costa Rica, during epidemiological weeks 1–30 of 2026, 585 of 1185 positive viral results from respiratory samples of hospitalised patients at the National Children’s Hospital were rhinovirus detections (49.4%).
In epidemiological week 31, rhinovirus ranked first among viruses reported by the national laboratory network of the Costa Rican Social Security Fund.6
These observations do not demonstrate replacement of RSV. The data combine different populations, samples, and surveillance settings. Changes in molecular panels, testing indications, and diagnostic capacity further limit historical comparisons.
Some multiplex assays also report rhinovirus and enterovirus as a combined target. Moreover, detection in an upper respiratory sample does not always establish that the detected virus caused the hospitalisation, especially when pathogens are co-detected.
Evidence after nirsevimab implementation does not establish replacement either. In the randomised MELODY trial, medically attended lower respiratory tract infections caused by non-RSV pathogens occurred at similar rates in the nirsevimab and placebo groups, without evidence of replacement.7 A Spanish multicentre study found fewer bronchiolitis hospitalisations after introduction but described a second seasonal peak dominated by rhinovirus and human metapneumovirus.8 Another prospective Spanish study observed fewer hospitalisations associated with RSV, human metapneumovirus, and adenovirus after implementation; its observational design cannot establish that nirsevimab caused the non-RSV changes.9 Clesrovimab also reduced RSV-associated hospitalisation in a trial.3 Post-introduction surveillance should assess the absolute incidence and relative distribution of other respiratory viruses. Seasonal variation, testing practices, and the proportional effect of reducing RSV remain alternative explanations.
Latin America’s tropical and temperate settings, differing seasonality, and uneven diagnostic capacity make local baselines essential. Surveillance should record weekly numbers of patients assessed, tests performed, and results by pathogen; distinguish rhinovirus from enterovirus where possible; maintain stable panels and sampling criteria; and stratify findings by age, care setting, severity, co-detection, maternal vaccination, and receipt of monoclonal antibodies. Findings should include test positivity and population-based rates of consultation, hospitalisation, and severe disease.
Where population denominators are unavailable, consistently sampled sentinel cohorts can still provide comparable trends, provided changes in eligibility and testing are documented.4, 10
Existing sentinel networks offer a starting point. Updated WHO guidance supports integrated surveillance of influenza, SARS-CoV-2, and RSV.10 Selected sites could additionally test consistently for other respiratory viruses, including rhinovirus, and use comparable time series to investigate changes in disease burden. This proposed extension should be evaluated for feasibility and laboratory capacity in each setting.
The objective is to establish a baseline before an unexpected signal emerges. Success should be measured through RSV disease averted and the total respiratory burden, alongside the aetiology of admissions that remain. Before concluding that the viral landscape has changed, we need the data to observe it.
Contributors
MLA conceptualized the paper, formulated the central question, and wrote the paper. LHFA, JB, XSL, and AG critically reviewed the paper, provided comments, and validated the references. All authors contributed to the final document.
Declaration of interests
The authors declare no competing interests.