The prevalence and role of human metapneumovirus in respiratory tract infections: a systematic review and meta-analysis of global data

Summary

Background

Human metapneumovirus (hMPV) is a significant respiratory pathogen, yet its global prevalence and epidemiological patterns remain poorly characterized. This study presents the first comprehensive systematic review and meta-analysis of hMPV prevalence among patients with respiratory tract infections (RTIs), addressing a critical gap in the literature, and clarifying the association between hMPV infection and RTIs.

Methods

We conducted a systematic search across PubMed, Scopus, and Web of Science, from 2002 to 2024. Studies reporting hMPV prevalence detected by Polymerase chain reaction-based methods in respiratory samples were included. Data was analyzed using random-effects models, with subgroup analyses by age, region, diagnostic methods, and patient’s demographic. The main outcome was hMPV prevalence among patients with RTIs.

Findings

The global pooled prevalence of hMPV among 2,236,820 patients with RTIs was 5.3% (95% CI: 5.0%–5.6%). Prevalence was highest in children under 5 years (6.7%) and lowest in adults (2.5%). Lower respiratory infections showed the highest prevalence (8.2%), with bronchiolitis (7.0%) and pneumonia (5.5%) being prominent. Inpatients exhibited a significantly higher prevalence of hMPV (6.1%) compared to outpatients (3.3%). Among hMPV-positive samples, type A was more prevalent than type B. A meta-analysis of 41 case-control datasets revealed a significant association between hMPV infection and respiratory infections (pooled odds ratio [OR] = 4.6; 95% CI: 3.7–5.6; I² = 14.4%).

Interpretation

This first-of-its-kind meta-analysis highlights hMPV as a major contributor to RTIs, particularly in young children and severe respiratory conditions. The findings show the need for improved diagnostic strategies, targeted surveillance, and vaccine development to mitigate hMPV’s global burden.

Funding

This study was financially supported by the Iran University of Medical Sciences (Grant Number: 1402-2-75-25881).

Research in context.

Evidence before this study

Before conducting this study, we conducted a systematic literature search using PubMed, Scopus, and Web of Science to identify studies on human metapneumovirus (hMPV) prevalence in patients with respiratory tract infections (RTIs). Our systematic search included studies published from 2002 to May 19, 2024, using Polymerase chain reaction-based detection methods. We identified 611 studies (698 datasets) that met the inclusion criteria covering diverse populations and geographic regions. Prior research reported varying hMPV prevalence rates influenced by patient demographics, geographic factors, and study methodologies. The overall quality of the evidence was assessed using a modified STROBE checklist, ensuring inclusion of studies with a validity score of at least 8 out of 12. A meta-analysis of case-control studies previously suggested a possible link between hMPV and RTIs, but global pooled estimates remained inconsistent.

Added value of this study

This study is the first comprehensive systematic review and meta-analysis to provide a global pooled prevalence estimate of hMPV among patients with RTIs. By analyzing data from over 2.2 million patients, we offer robust evidence supporting the role of hMPV as a significant respiratory pathogen. Our findings reveal that hMPV is particularly prevalent in young children and inpatients, with higher detection rates in lower respiratory infections, including bronchiolitis and pneumonia. The study also identifies a shift in hMPV prevalence during the COVID-19 pandemic, probably due to public health interventions. Furthermore, we highlight the predominance of hMPV type A over type B in infected patients, a key epidemiological finding with implications for diagnostics and vaccine development.

Implications of all the available evidence

The combined evidence shows the need for enhanced surveillance and diagnostic strategies to improve hMPV detection, particularly in vulnerable populations. Our findings suggest that hMPV should be considered a priority pathogen in respiratory disease research and vaccine development. Public health policies should integrate hMPV surveillance into routine RTI monitoring to better inform outbreak responses. Future research should explore the long-term burden of hMPV, assess vaccine efficacy, and investigate potential interactions with other respiratory viruses to develop comprehensive prevention strategies.

Introduction

Respiratory tract infections (RTIs) remain a significant global health burden, contributing to substantial morbidity and mortality across all age groups.¹ Among the myriad of pathogens responsible for RTIs, human metapneumovirus (hMPV) has emerged as a prominent viral agent since its discovery in 2001.² hMPV, a member of the Pneumoviridae family, is known to cause a spectrum of respiratory illnesses, ranging from mild upper respiratory tract infections to severe lower respiratory tract diseases, particularly in young children, the elderly, and immunocompromised individuals.³ Despite its clinical significance, hMPV often remains underdiagnosed due to its symptomatic overlap with other respiratory viruses, such as human respiratory syncytial virus (hRSV), influenza, and severe acute respiratory syndrome coronavirus 2.⁴

Research highlights hMPV as a leading cause of RTIs, particularly in vulnerable populations. The study conducted in 2024 across high-burden settings in Africa and Asia identified hMPV as the second leading cause of severe pneumonia in children under five, following hRSV. Additionally, a comprehensive analysis of over 155,000 pediatric hospitalizations for acute RTIs over 12 years revealed that hMPV was most prevalent in infants under one year and in children with comorbidities, with 2.34% of hMPV-positive cases requiring intensive care unit admission. hMPV is associated with a spectrum of clinical manifestations, ranging from mild upper respiratory tract infections (URTIs), such as the common cold, to severe lower respiratory tract conditions, including bronchitis, pneumonia, and asthma exacerbations. Severe disease is more frequently observed in young children, the elderly, and immunocompromised individuals, but emerging evidence underscores its impact in immunocompetent adults.⁴

Understanding the global prevalence of hMPV among patients with RTIs is crucial for informing public health strategies, improving diagnostic protocols, and guiding the development of targeted therapies and vaccines. However, the reported prevalence of hMPV varies widely across different regions and populations, influenced by factors such as geographic location, climate, demographic characteristics, and diagnostic methodologies.⁵ To address this variability and provide a comprehensive estimate of hMPV's global impact, systematic reviews and meta-analyses are essential tools for synthesizing data from diverse studies.

This article aims to systematically review and meta-analyze the global prevalence of hMPV among patients with RTIs, synthesizing evidence from studies conducted across different continents and populations. By doing so, it seeks to provide a clearer understanding of the epidemiology of hMPV and highlight gaps in current knowledge that warrant further investigation. The findings of this study will contribute to a more nuanced understanding of hMPV's role in RTIs and inform global efforts to mitigate its impact on public health.

Methods

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline served as the foundation for this systematic review and meta-analysis approach.⁶

Search strategy

To identify relevant studies, a systematic literature search was conducted using three electronic databases: PubMed, Scopus, and Web of Science. The search was limited to articles published from the inception of these databases up to May 19, 2024. The specific search terms used for each database are detailed in Supplementary Table S1. Additionally, the reference lists of relevant articles were manually reviewed to identify further studies that met the inclusion criteria. For efficient data management, the results of the systematic literature search were imported into EndNote software version X8 (Thomson Reuters, California, USA).

Selection criteria

Studies were eligible for inclusion if they met the following criteria: (1) cross-sectional or case-control studies reporting data on hMPV prevalence in both children and adults with respiratory symptoms, published in English in peer-reviewed journals. In this systematic review, the control groups were defined as those specified within the included case-control studies that investigated the association between hMPV positivity and symptomatic RTIs. Specifically, controls were individuals who did not exhibit respiratory symptoms or have a clinical diagnosis of RTIs, as determined by the standardized criteria established in each included case-control study. These criteria typically involved the absence of clinical signs such as fever, cough, dyspnea, or other respiratory symptoms. The control groups were selected by the original studies to be appropriately matched to cases (individuals with confirmed RTIs) based on factors such as age and sex, ensuring methodological rigor and comparability; (2) studies reporting the prevalence of hMPV using either upper respiratory tract specimens (e.g., nasopharyngeal, oropharyngeal, nasal, or pharyngeal swabs) or lower respiratory tract specimens (e.g., sputum, bronchoalveolar lavage, or tracheal aspirates); (3) studies using Polymerase chain reaction (PCR)-based methods for hMPV detection; (4) studies examining hMPV prevalence regardless of care setting (inpatient, outpatient, or community); and (5) original research articles or short communications containing sufficient data.

Studies were excluded if they met any of the following conditions: (1) focused on hMPV prevalence in patients with underlying conditions such as human immunodeficiency virus infection, asthma, chronic obstructive pulmonary disease, transplant recipients, hematological disorders, cancer, acute otitis media, sickle cell disease, or chronic heart failure. Studies focusing on at-risk populations with underlying conditions were excluded to ensure generalizability of our prevalence estimates to the broader population. These conditions are associated with increased susceptibility to respiratory infections, which could elevate hMPV prevalence and skew global estimates unrepresentative of typical RTI populations. By focusing on studies of general populations, we aimed to provide robust, comparable prevalence data for public health applications, such as vaccine development; (2) cohort or prospective studies investigating hMPV incidence in patients with respiratory symptoms. Prospective studies were included if they reported prevalence data, as these align with our objective of estimating the proportion of RTIs attributable to hMPV. However, prospective studies reporting only incidence rates were excluded, as our focus was on cross-sectional prevalence to ensure consistency and comparability across heterogeneous global settings, minimizing biases from longitudinal incidence reporting; (3) used non-respiratory samples, such as blood or cerebrospinal fluid; (4) employed detection methods other than PCR, such as ELISA, metagenomic next-generation sequencing, immunochromatography, immunofluorescence, enzyme immunoassays, or reverse transcription loop-mediated isothermal amplification (RT-LAMP). Studies employing non-PCR detection methods were excluded because PCR offers superior sensitivity and specificity, ensuring reliable and comparable hMPV detection across diverse studies, particularly for low-viral-load cases. This focus on PCR minimized false negatives and enhanced methodological rigor, though we acknowledge that non-PCR methods may provide complementary data in specific contexts; (5) examined hMPV antibody seroprevalence; (6) included asymptomatic individuals or patients without respiratory symptoms; (7) reported hMPV prevalence in patients undergoing antiviral treatment. The primary objective of our meta-analysis was to assess the prevalence and role of hMPV in symptomatic RTIs in untreated or naturally progressing cases, to better understand the virus’s natural epidemiology and clinical impact. Studies that reported hMPV prevalence in patients receiving antiviral treatment were excluded because antiviral therapies could potentially alter the detectable viral load, the duration of viral shedding, or the clinical presentation of RTIs. Such alterations could confound the assessment of hMPV’s true prevalence and its association with RTI outcomes, leading to potential bias in our meta-analysis. By excluding these studies, we aimed to ensure that our results reflected the natural course of hMPV infection in RTI cases, thereby maintaining consistency and comparability across the included studies. Additionally, the use of antiviral treatments in the context of hMPV is not standardized, and the specific agents, timing, and duration of treatment varied widely across studies. This heterogeneity posed challenges in interpreting hMPV prevalence data; (8) were letters, notes, case reports, case series, comments, reviews, posters, or conference abstracts; or (9) were published in languages other than English.

Data extraction and quality assessment

Three reviewers independently screened the titles and abstracts of all identified studies, removing those unrelated to the research topic. The reviewers then obtained and thoroughly evaluated the full texts of the selected studies, excluding any that did not meet the inclusion criteria. Any discrepancies among the reviewers were resolved through consultation with a fourth reviewer. A quality assessment of the included studies was conducted using a modified checklist based on the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.^7,8 The checklist included 12 questions designed to evaluate various methodological aspects of the studies. Only studies that achieved a validity score of 8 or higher out of a maximum of 12 were deemed eligible for inclusion in the main meta-analysis. Three reviewers were responsible for extracting the following data from each eligible article: the first author's last name, year of publication, year of sampling, study location, study design, sample size, type of sample, age ranges and groups of patients, gender distribution, number of hMPV-positive cases, hMPV detection methods, types of patient care, type of respiratory disease, and hMPV genotype. The extracted data were systematically recorded in a pre-designed Excel spreadsheet (Microsoft Corporation, Redmond, Washington, USA).

Statistical analysis

We calculated the pooled prevalence of hMPV among patients with respiratory tract infections using the metaprop package.⁹ We utilized a random-effects meta-analysis model and performed subgroup analyses stratified by gender, geographic location, age group, diagnostic methods, sampling period, type of respiratory disease, patient care settings, and hMPV genotypes. Regarding age groups, studies were categorized based on whether they reported broader age groups (e.g., 0–5 years, 6–18 years) or finer subgroups (e.g., 0–6 months, 7–12 months, etc.). Broader age group data were pooled into the 0–5 years and 6–18 years categories to include all relevant studies, with data from narrower age ranges (e.g., under 2 years) incorporated into the 0–5 years category. Finer subgroup analyses were conducted for the 0–5 years range (0–6 months, 7–12 months, 13–24 months, 25–36 months, 37–48 months, 49–60 months) when studies provided such breakdowns. Random-effects models and the I² statistic were used to account for heterogeneity, with sensitivity analyses performed to explore the impact of age group variability.

Additionally, we conducted meta-analyses to assess risk estimates for respiratory diseases associated with hMPV exposure in case-control studies, reporting pooled odds ratios (ORs) with 95% confidence intervals (CIs). Throughout the data extraction, analysis, and reporting processes, we adhered to the MOOSE guidelines for meta-analyses of observational studies. The pooled estimate of the OR along with its 95% CI was calculated using the DerSimonian and Laird method under a random-effects model.¹⁰ Statistical heterogeneity between studies was evaluated with Cochran’s Q test and quantified by I² statistic.¹¹ To assess potential publication bias, we employed a combination of methods, including visual examination of funnel plots, which plotted the logarithmically transformed ORs against the standard error of the corresponding log. Additionally, we conducted Begg’s and Egger’s tests to statistically evaluate bias. All statistical tests were two-tailed, with a significance threshold of less than 0.05, except for the heterogeneity test, which used a threshold of less than 0.1. All analyses were performed using Stata 14.1 (Stata Corp, College Station, TX, USA).

Ethics

Due to the nature of the research, the present study did not require informed consent or approval from any Ethics Committee.

Role of the funding source

The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

Results

Literature search

In the initial search, 9769 papers were identified, with an additional 20 papers found through manual review of reference lists from relevant studies. After removing 4534 duplicate records, a further 4416 papers were excluded based on a manual screening of titles and abstracts. The full texts of the remaining 839 papers were then thoroughly assessed for eligibility, leading to the exclusion of 219 studies. Using a modified STROBE checklist, 611 articles were determined to be of high quality (scoring 8 or higher), while 9 papers did not meet the quality threshold. Ultimately, this systematic review and meta-analysis included 611 articles, comprising 698 datasets. The selection process for relevant studies is summarized in Fig. 1.

Fig. 1.

Flowchart presenting the steps of literature search and selection.

Study characteristics

Out of the 698 research, 657 were cross-sectional and 41 were case-control in design. The articles' publication dates varied from 2002 to 2024. The largest research involved 686,199 patients with respiratory infections,¹² while the smallest contained 24 cases.¹³ Out of the 698 datasets included in this meta-analysis, 95 research examined the gender distribution of hMPV infection, and 89 research performed hMPV typing. Among the age group of children under 5 years old, the largest population included in the studies were children under 1 year of age (n = 60,508). Overall, children under 5 years of age (n = 249,695) were the predominant population participating in the studies compared to children aged 6–18 years (n = 33,291). In this meta-analysis examining the prevalence of hMPV, the primary population analyzed consisted of children under the age of 18. The study included a total of 840,766 participants in the pediatric age group (≤18 years), while the adult population over 18 years old accounted for 199,414 individuals. The majority of studies (n = 87) were conducted in China, followed by United States (n = 47), Brazil (n = 36), and Italy (n = 31). In terms of the number of participants, Australia ranked first with 703,294 cases, followed by China, South Korea, and the United States with 362,001, 286,559, and 230,596 cases, respectively. Many studies spanned multiple seasons or entire years, with several covering multi-year periods and sampling across various seasons, capturing hMPV circulation beyond the influenza season. The characteristics of included studies in this systematic review and meta-analysis are summarized in Supplementary Table S2.

Among the 41 case control studies, subgroup analyses showed that the control group in 5 (12%), 10 (24.39%), and 26 (63.41%) of studies were health care unit admitted asymptomatic, healthy community asymptomatic, and hospital admitted asymptomatic individuals, respectively.

Prevalence of hMPV infection among patients with respiratory infections

The overall pooled prevalence of hMPV infection among 2,236,820 patients with different respiratory infections was 5.3% (95% CI: 5.0%–5.6%; I² = 98.7%, P < 0.0001). Real-time PCR was the most frequently used method (316 datasets) and reported a prevalence of 5.0% (95% CI: 4.6%–5.4%). RT-PCR, with 162 datasets, had the highest prevalence at 6.8% (95% CI: 6.1%–7.5%), while multiplex PCR (96 datasets) showed the lowest prevalences 4.3% (95% CI: 3.5%–5.2%)

Regarding the type of disease, lower respiratory infection (LRI) had the highest prevalence of hMPV infection at 8.2% (95% CI: 6.8%–9.7%), while influenza-like illness (ILI) had the lowest prevalence at 3.0% (95% CI: 2.6%–3.4%). Bronchiolitis and pneumonia also showed relatively high prevalences of 7.0% (95% CI: 5.5%–8.7%) and 5.4% (95% CI: 4.3%–6.8%), respectively. Overall, lower respiratory tract infection (LRTI), with 254 datasets, had a higher prevalence of 6.4% (95% CI: 5.7–7.1) compared to URTI, which had a prevalence of 3.0% (95% CI: 2.6%–3.4%) based on 98 datasets.

The period 2001–2010 had the highest number of datasets and reported a prevalence of 6.6% (95% CI: 6.0–7.3). The most recent period, 2020–2023, had the lowest prevalence at 3.1% (95% CI: 2.3–3.9), suggesting a possible decline in hMPV infections in recent years.

Regarding the sample type, nasopharyngeal samples were the most commonly used for the detection of hMPV infection (332 datasets) and had a prevalence of 5.7% (95% CI: 5.3%–6.2%). Nasal samples, with 61 datasets, showed the highest prevalence at 6.7% (95% CI: 5.4%-8.1). Overall, upper respiratory samples (586 datasets) had a higher prevalence of 5.4% (95% CI: 5.1%–5.7%) compared to lower respiratory samples (13 datasets), which had a prevalence of 2.1% (95% CI: 1.4%–3.0%).

The prevalence of hMPV was comparable between male patients (6.0%, 95% CI: 5.2%–6.9%) and female patients (5.9%, 95% CI: 5.0%–6.9%), with no statistically significant difference (P = 0.93).

Among children under the age of 5 years, the age group 25–36 months had the highest prevalence of hMPV infection at 10.1% (95% CI: 6.1–14.7), while the age group 49–60 months had the lowest prevalence at 2.0% (95% CI: 0.0–11.8). Among pediatric patients, children ≤5 years had the highest number of datasets and the highest prevalence of 6.7% (95% CI: 6.2%–7.2%) compared with children between 6 and 18 years (2.6%, 95% CI: 2.0%–3.4%). In addition to the analyses conducted in various age groups of children, a separate analysis was also performed on the overall group of children and adults. The results of this analysis revealed that the prevalence of hMPV was significantly higher in the children's group (6.3%, 95% CI: 5.9%–6.7%) compared to the adult group (2.5%, 95% CI: 2.1%–2.9%).

Regarding the patient type, inpatients had a remarkable higher prevalence of hMPV (6.1%; 95% CI: 5.6%–6.6%) compared to outpatients (3.3%; 95% CI: 2.8%–3.9%). Regarding the type of hMPV distribution, a total of 89 datasets has addressed the typing of hMPV in patients with respiratory symptoms. Analyses in this context indicate that among 6413 hMPV-positive samples, the frequency of type A (64.6%, 95% CI: 59.4%–69.7%) is significantly higher than that of type B (44.0%, 95% CI: 38.8%–49.1%).

In the case of regional prevalence, the highest prevalence was observed in the Eastern Mediterranean Region (6.6%; 95% CI: 5.7%–7.5%) and Region of the Americas (6.3%; 95% CI: 5.3%–7.1%), respectively. On the other hand, African Regions (3.7%; 95% CI: 3.0%–4.4%) and Western Pacific Region (4.5%; 95% CI: 4.0%–5.0%) had the lowest prevalence. Fig. 2 shows the worldwide prevalence of hMPV based on the WHO world regions.

Fig. 2.

Global prevalence of human metapneumovirus infection among patients with respiratory infections according to the WHO regions.

The subgroup analysis of the prevalence of hMPV infection among patients with respiratory infections is presented in Table 1.

Table 1.

Subgroup analysis of the prevalence of human metapneumovirus infection among patients with respiratory infections.

	Group	Number of datasets	Total sample size	Pooled prevalence (%) (95%CI)	Heterogeneity test I2%, P-value	Differences between subgroups; χ2 test (P-value)
Overall	–	698	2,236,820	5.3 (5.0–5.6)	98.7%, P < 0.0001	–
Study period	<1 Year	232	137,807	5.8 (5.2–6.4)	94.9%, P < 0.0001	<0.0001
	≥1 Year	457	2,042,580	5.0 (4.6–5.3)	99.0%, P < 0.0001
WHO global regions	African Regions	45	51,741	3.7 (3.0–4.4)	93.0%, P < 0.0001	<0.0001
	Eastern Mediterranean Region	80	114,976	6.6 (5.7–7.5)	96.4%, P < 0.0001
	European Region	202	281,303	5.3 (4.8–5.9)	97.1%, P < 0.0001
	Region of the Americas	125	315,828	6.3 (5.6–7.1)	98.1%, P < 0.0001
	South-East Asia Region	51	68,310	4.9 (4.3–5.6)	92.6%, P < 0.0001
	Western Pacific Region	194	1,403,328	4.5 (4.0–5.0)	99.2%, P < 0.0001
Detection method	Nested PCR	46	186,075	4.9 (3.7–6.3)	98.6%, P < 0.0001	<0.0001
	Multiplex PCR	96	292,164	4.3 (3.5–5.2)	98.5%, P < 0.0001
	Real-time PCR	316	695,041	5.0 (4.6–5.4)	98.1%, P < 0.0001
	RT-PCR	162	240,118	6.8 (6.1–7.5)	97.1%, P < 0.0001
	PCR- Hybridization	50	36,710	5.9 (4.7–7.1)	93.8%, P < 0.0001
Type of disease	ILI	84	181,858	3.0 (2.6–3.4)	93.4%, P < 0.0001	<0.0001
	URI	14	13,681	3.3 (2.1–4.7)	90.6%, P < 0.0001
	LRI	89	91,429	8.2 (6.8–9.7)	98.1%, P < 0.0001
	Pneumonia	91	242,001	5.4 (4.3–6.8)	98.5%, P < 0.0001
	SARI	38	54,210	4.4 (3.6–5.2)	94.2%, P < 0.0001
	Bronchiolitis	23	6683	7.0 (5.5–8.7)	80.1%, P < 0.0001
	Wheezing	12	2935	6.2 (4.0–8.9)	82.8%, P < 0.0001
	ARDS	2	2445	3.7 (3.0–4.5)	NA
Type of disease [Overall]	URTI	98	195,539	3.0 (2.6–3.4)	93.4%, P < 0.0001	<0.0001
	LRTI	254	399,455	6.4 (5.7–7.1)	98.4%, P < 0.0001
Sampling time	1991–2000	2	919	3.8 (2.6–5.1)	NA	<0.0001
	2001–2010	275	208,061	6.6 (6.0–7.3)	96.9%, P < 0.0001
	2011–2019	300	1,278,305	4.6 (4.2–4.9)	98.3%, P < 0.0001
	2020–2023	71	429,224	3.1 (2.3–3.9)	99.0%, P < 0.0001
Season	Spring	27	5186	8.0 (5.3–11.1)	91.9%, P < 0.0001	<0.0001
	Summer	18	4786	1.4 (0.4–2.9)	81.3%, P < 0.0001
	Autumn	21	4238	1.8 (0.6–3.4)	88.4%, P < 0.0001
	Winter	50	30,395	5.1 (4.1–6.2)	93.2%, P < 0.0001
Covid-19 status	Before COVID-19	577	1,487,285	5.5 (5.2–5.8)	98.0%, P < 0.0001	<0.0001
	After COVID-19	71	429,224	3.1 (2.3–3.9)	99.0%, P < 0.0001
Sample type	NP	332	438,974	5.7 (5.3–6.2)	96.8%, P < 0.0001	<0.0001
	OP	35	50,748	4.8 (3.8–5.8)	95.3%, P < 0.0001
	Nasal	61	42,461	6.7 (5.4–8.1)	96.3%, P < 0.0001
	Pharyngeal	9	24,827	4.6 (2.7–6.9)	97.3%, P < 0.0001
	Mixed Upper	150	376,346	4.6 (4.0–5.1)	97.9%, P < 0.0001
	Sputum	4	1112	3.4 (0.8–7.6)	86.0%, P =0.0001
	BAL	4	798	3.2 (0.7–7.2)	79.2%, P =0.002
	Tracheal	1	55	1.8 (0.05–9.7)	NA
	Mixed Lower	4	6032	1.9 (1.5–2.2)	0%, P =0.6
	Mixed Upper Lower	58	329,220	4.0 (3.1–4.9)	99.1%, P < 0.0001
Sample type (Overall)	URS	586	933,108	5.4 (5.1–5.7)	97.1%, P < 0.0001	<0.0001
	LRS	13	7997	2.1 (1.4–3.0)	72.4%, P < 0.0001
	MRS	58	329,220	4.0 (3.1–4.9)	99.1%, P < 0.0001
Gender	Male	95	124,642	6.0 (5.2–6.9)	96.8%, P < 0.0001	0.9317
	Female		91,879	5.9 (5.0–6.9)	96.3%, P < 0.0001
Age (month) [Under 5 years]	0–6	32	23,655	4.7 (3.5–6.0)	93.0%, P < 0.0001	0.0032
	7–12	27	9821	5.9 (4.4–7.6)	86.1%, P < 0.0001
	0–12	101	60,508	6.2 (5.7–6.8)	93.3%, P < 0.0001
	13–24	34	12,688	7.1 (5.5–8.9)	85.8%, P < 0.0001
	25–36	13	1070	10.1 (6.1–14.7)	69.5%, P = 0.0002
	37–48	10	564	7.9 (4.5–11.9)	27.8%, P = 0.1
	49–60	9	250	2.0 (0.0–11.8)	71.5%, P = 0.0005
Age (year) [Children]	≤5	307	249,695	6.7 (6.2–7.2)	95.6%, P < 0.0001	<0.0001
	6–18	73	33,291	2.6 (2.0–3.4)	87.9%, P < 0.0001
Age (year) [All ages]	≤18	492	840,766	6.3 (5.9–6.7)	98.2%, P < 0.0001	<0.0001
	>18	145	199,414	2.5 (2.1–2.9)	95.7%, P < 0.0001
Patient type	Outpatients	83	114,698	3.3 (2.8–3.9)	94.7%, P < 0.0001	<0.0001
	Inpatients	393	684,234	6.1 (5.6–6.6)	98.3%, P < 0.0001

WHO: World Health Organization; ILI: Influenza-like illness; URI: Upper respiratory infection; LRI: Lower respiratory infection; SARI: Severe acute respiratory infection; ARDS: Acute respiratory distress syndrome; URTI: Upper respiratory tract infection; LRTI: Lower respiratory tract infection; NP; Nasopharyngeal, OP; Oropharyngeal; BAL: Bronchoalveolar lavage; URS: Upper respiratory sample; LRS: Lower respiratory sample; MRS: Mixed respiratory sample.

The association between hMPV infection and respiratory infections

In the next step, the second meta-analysis was performed on case-control studies to examine the relationship between hMPV infection and the risk of respiratory infections. From a total of 41 datasets with case control design, 25,802 were patients with respiratory infections and 9619 were controls. Cases were patients diagnosed with RTIs, while controls were individuals without respiratory symptoms or with confirmed non-hMPV etiologies (e.g., PCR-negative for hMPV or positive for other respiratory viruses), concurrently assessed for hMPV prevalence using PCR-based diagnostics. Meta-analysis with random effect model showed that the pooled OR for hMPV infection and risk of respiratory infections was 4.6 (95% CI: 3.7–5.6), based on 41 datasets, with low heterogeneity (I² = 14.4%, P = 0.2) (Fig. 3). The pooled OR was 4.6, indicating that patients with RTIs were 4.6 times more likely to have hMPV detected compared to controls, suggesting a significant causal association. The subgroup analysis of the association between hMPV infection and the risk of respiratory infections is presented in Table 2.

Fig. 3.

Forest plot of the association between human metapneumovirus infection and respiratory infection risk in patients according to the random effect model in case case-control studies.

Table 2.

Subgroup analysis of association between human metapneumovirus infection and respiratory infections risk.

Characteristics	Categories	No. of datasets	Pooled ORs (95% CI)	Heterogeneity: I2%, P value
Overall	–	41	4.6 (3.7–5.6)	14.4%, P = 0.2
Type of disease	LRI	9	3.2 (2.0–5.0)	51.0%, P = 0.03
	Pneumonia	19	6.3 (4.8–8.3)	0.0%, P = 0.7
	URI	1	3.1 (0.1–55.3)	NA
Type of disease	LRTI	28	4.6 (3.4–6.0)	36.2%, P = 0.03
	URTI	1	3.1 (0.1–55.3)	NA
Detection method	RT-PCR	3	4.6 (1.7–12.7)	0.0%, P = 0.8
	Real-time PCR	31	4.5 (3.5–5.7)	30.1%, P = 0.06
	Nested PCR	2	11.9 (1.5–89.6)	0.0%, P = 0.9
	Multiplex PCR	2	2.9 (0.3–22.2)	0.0%, P = 0.6
Sample type	Nasopharyngeal	24	5.0 (3.9–6.3)	0%, P = 0.8
	Nasal	3	5.3 (1.2–22.4)	0%, P = 0.8
	Mixed upper	13	4.1 (2.6–6.4)	56.2%, P = 0.007
	BAL	1	2.7 (0.2–28.3)	NA
Sample type	Upper respiratory samples	40	4.6 (3.7–5.7)	16.2%, P = 0.2
	Lower respiratory samples	1	2.7 (0.2–28.3)	NA
Sampling time	2001–2010	9	4.1 (2.7–6.0)	0%, P = 0.8
	2011–2019	29	4.6 (3.5–6.1)	27.3%, P = 0.08
	2020–2023	1	3.0 (1.0–8.8)	NA

LRI: Lower respiratory infection; URI: Upper respiratory infection; LRTI: Lower respiratory tract infection; URTI: Upper respiratory tract infection; RT-PCR: Reverse Transcription Polymerase Chain Reaction; BAL: Bronchoalveolar lavage; N.A: Not applicable.

We assessed publication bias with visual inspection of the funnel plot, and statistical tests. The results showed no evidence of publication bias for the association between hMPV infection and respiratory infections risk (P = 0.37, for Begg’s adjusted rank correlation test and P = 0.20 for Egger’s regression asymmetry test) (Fig. 4).

Fig. 4.

Funnel plot for assessment of publication bias.

When stratified by the type of disease, hMPV infection was significantly associated with pneumonia (OR = 6.3, 95% CI: 4.8–8.3; I² = 0.0%, P = 0.7) and LRI (OR = 3.2, 95% CI: 2.0–5.0; I² = 51.0%, P = 0.03). However, the association with upper respiratory infections (URI) was not statistically significant (OR = 3.1, 95% CI: 0.1–55.3). Overall, hMPV infection was associated with LRTI (OR = 4.6, 95% CI: 3.4–6.0; I² = 36.2%, P = 0.03), but not with URTI (OR = 3.1, 95% CI: 0.1–55.3).

Regarding detection methods, hMPV infection was significantly associated with respiratory infections across various PCR-based methods, including nested PCR (OR = 11.9, 95% CI: 1.5–89.6; I² = 0.0%, P = 0.9), RT-PCR (OR = 4.6, 95% CI: 1.7–12.7; I² = 0.0%, P = 0.8), and Real-time PCR (OR = 4.5, 95% CI: 3.5–5.7; I² = 30.1%, P = 0.06). However, multiplex PCR showed a non-significant association (OR = 2.9, 95% CI: 0.3–22.2; I² = 0.0%, P = 0.6).

In terms of sample types, hMPV infection was significantly associated with respiratory infections in nasal samples (OR = 5.3, 95% CI: 1.2–22.4; I² = 0%, P = 0.8), nasopharyngeal samples (OR = 5.0, 95% CI: 3.9–6.3; I² = 0%, P = 0.8), and mixed upper respiratory samples (OR = 4.1, 95% CI: 2.6–6.4; I² = 56.2%, P = 0.007). However, the association was not significant for bronchoalveolar lavage (BAL) samples (OR = 2.7, 95% CI: 0.2–28.3). Overall, the analysis was stratified by sample type, distinguishing between upper and lower respiratory samples. For upper respiratory samples, which included 40 datasets, the pooled OR was 4.6 (95% CI: 3.7–5.7), indicating a significant association between hMPV infection and an increased risk of respiratory infections using upper respiratory samples. The heterogeneity among these studies was low, with an I² value of 16.2% (P = 0.2), suggesting consistent findings across the included datasets. In contrast, only one dataset was available for lower respiratory samples, resulting in a pooled OR of 2.7 (95% CI: 0.2–28.3). Due to the limited number of studies, the heterogeneity could not be assessed, and the wide CI reflects the uncertainty of this estimate.

When stratified by sampling time, hMPV infection was significantly associated with respiratory infections in both the 2001–2010 period (OR = 4.1, 95% CI: 2.7–6.0; I² = 0%, P = 0.8) and the 2011–2019 period (OR = 4.6, 95% CI: 3.5–6.1; I² = 27.3%, P = 0.08). Data for the 2020–2023 period were limited, with a non-significant association (OR = 3.0, 95% CI: 1.0–8.8).

Meta-regression

Meta-regression was performed to explore sources of heterogeneity in hMPV prevalence, using three categorical variables: region (African, Eastern Mediterranean, European, Americas, Asia, Western Pacific), Sampling time ( ≤ 2010, 2011–2019, 2020–2023), Study Design and Sample size. In the meta-regression results, the heterogeneity was explained by the Sample size (P < 0.001), and Sampling time (P < 0.001), but the heterogeneity was not explained by the study type (P = 0.10). For different regions, Eastern Mediterranean and Americas had a significantly higher prevalence than African (P < 0.05), but no significant difference was found for European, Asia and Western Pacific (P > 0.05) (Table 3).

Table 3.

Meta-regression analysis results in the study of pooled prevalence of human Metapneumovirus prevalence among patients with respiratory tract infections.

Covariate	Coefficient	Standard error	t	Two-sided P	95% Conf. Interval
					lower	upper
Sample size	−5.08e-06	1.10e-06	−4.62	<0.001	−7.24e-06	−2.92e-06
Eastern Mediterranean	7,439,278	0.19	3.84	<0.001	0.36	1.12
European	0.31	0.17	1.82	0.07	−0.025	0.66
Americas	0.53	0.18	2.97	0.003	0.17	0.88
South-East Asia Region	0.23	0.20	1.14	0.25	−0.17	0.64
Western Pacific Region	0.15	0.17	0.90	0.36	−0.18	0.49
2011–2019	−0.48	0.08	−6.03	<0.001	−0.64	−0.32
2020–2023	−0.41	0.12	−3.29	0.001	−0.66	−0.16
Study design	−0.25	0.15	−1.63	0.10	−0.56	0.05
_cons	−2.33	0.21	−10.66	<0.001	−2.76	−1.90

Sensitivity analysis

In a sensitivity analysis by successively removing a particular study at a time to assess the influence of every single study on pooled results, a significant positive association [range of summary ORs 4.28–4.73] between hMPV infection and respiratory infections was observed consistently and did not alter the pooled results, which indicated that the meta-analysis model is robust.

Discussion

hMPV is a significant respiratory pathogen that affects individuals across all age groups, but recent studies highlight its disproportionate impact on young children.¹⁴ Our study showed that among children under the age of 5 years, the prevalence of hMPV infection varies significantly by age group. The highest prevalence was observed in children aged 2–3 years, with 10.1% testing positive for the virus. This suggests that toddlers in this age range are particularly vulnerable to hMPV, possibly due to their developing immune systems and increased exposure to respiratory pathogens in daycare or preschool settings. In contrast, the lowest prevalence was found in children aged 4–5 years, at just 2.0%, indicating a potential decline in susceptibility as children grow older and their immune systems mature.

When comparing pediatric patients more broadly, children aged 5 years and younger had the highest prevalence of hMPV infection at 6.7%. This was significantly higher than the prevalence observed in older children aged 6–18 years, which stood at 2.6%. These findings underscore the heightened risk of hMPV infection among younger children, particularly those in the first five years of life.

Similar to our findings, the prevalence of some other respiratory viruses is notably higher among children under 5 years of age compared to those in older age groups. For instance, hRSV and parainfluenza virus are more frequently detected in younger children, often leading to more severe clinical outcomes in this demographic. According to a study by Jain et al., hRSV was detected more commonly in children younger than 5 years of age than in older children (37% vs. 8%).¹⁵ Similarly, parainfluenza virus infections, particularly types 1 and 3, were more prevalent in younger children, as highlighted by Henrickson et al.¹⁶ The higher prevalence in this age group may be attributed to factors such as increased exposure to respiratory viruses in communal settings, as well as the lack of prior immunity to hMPV. The disparity in hMPV prevalence becomes even more pronounced when comparing children to adults. A separate analysis of the overall population revealed that children had a significantly higher prevalence of hMPV infection at 6.3% compared to adults, who had a prevalence of just 2.5%. This stark difference highlights the greater susceptibility of children to hMPV, likely due to their immature immune systems and limited prior exposure to the virus. In contrast, adults may have developed partial immunity through previous infections, reducing their likelihood of testing positive for hMPV.

Accurate detection of hMPV is critical for understanding its prevalence and guiding clinical management. Among the various diagnostic methods available, molecular techniques such as PCR-based assays are frequently used diagnostic methods due to their high sensitivity and specificity.⁵ According to the results of the present study, Real-time PCR, the most frequently used method with 316 datasets, reported a prevalence of 5.0%. This method’s widespread use and consistent results make it a reliable tool for hMPV detection in both clinical and research settings. Similarly, Nested PCR, with 46 datasets, also reported a prevalence of 4.9%, demonstrating its comparability to Real-time PCR in terms of accuracy and reliability. However, not all PCR-based methods yielded identical results. Reverse transcription PCR (RT-PCR), with 162 datasets, reported the highest prevalence of hMPV at 6.8%. This suggests that RT-PCR may be particularly effective at detecting hMPV, possibly due to its ability to amplify viral RNA with high efficiency. On the other hand, multiplex PCR, which allows for the simultaneous detection of multiple pathogens, showed the lowest prevalence at 4.3% across 96 datasets. While multiplex PCR is valuable for identifying co-infections, its slightly lower prevalence rate for hMPV may reflect differences in sensitivity or the prioritization of other pathogens in the testing process. Another notable method, PCR-Hybridization, demonstrated a high prevalence of hMPV at 5.9% across 50 datasets. This technique, which combines PCR amplification with hybridization to detect specific genetic sequences, appears to strike a balance between sensitivity and specificity, making it a promising option for hMPV detection. The variation in prevalence rates across these methods underscores the importance of selecting the appropriate diagnostic tool based on the clinical or research context. These findings have significant implications for both clinicians and researchers. While Real-time PCR remains a widely used and reliable method,¹⁷ the higher prevalence detected by RT-PCR suggests that it may be particularly useful in settings where hMPV is suspected but not easily confirmed. Conversely, the lower prevalence reported by multiplex PCR highlights the need for careful interpretation of results when using this method for hMPV detection. As diagnostic technologies continue to evolve, understanding the strengths and limitations of each method will be essential for accurately assessing the burden of hMPV and improving patient outcomes.

hMPV is a leading cause of respiratory infections, but its prevalence varies significantly depending on the type of disease. According to our results, LRIs have the highest prevalence of hMPV infection at 8.2%. This underscores the virus's propensity to cause more severe respiratory conditions, particularly those affecting the lungs and airways. In contrast, ILI had the lowest prevalence of hMPV at 2.9%, suggesting that hMPV is less frequently associated with milder, flu-like symptoms. These findings highlight the importance of considering hMPV as a potential pathogen in cases of severe respiratory illness, especially when symptoms align with LRIs. Among specific respiratory conditions, bronchiolitis and pneumonia also showed relatively high prevalences of hMPV infection. Bronchiolitis, a common condition in young children characterized by inflammation of the small airways,¹⁸ had a prevalence of 7.0%. Similarly, pneumonia, a more serious infection affecting the lungs, had a prevalence of 5.4%. These results emphasize the significant role hMPV plays in driving hospitalizations and severe respiratory outcomes, particularly in vulnerable populations such as infants and the elderly. Clinicians should remain vigilant for hMPV in patients presenting with these conditions, as early detection can guide appropriate management and treatment strategies. When comparing broader categories of respiratory infections, LRTIs had a higher prevalence of hMPV at 6.4% based on 254 datasets. This was notably higher than the prevalence observed in URTIs, which stood at 3.0% based on 98 datasets. Given the high prevalence of hMPV in severe respiratory conditions like bronchiolitis, pneumonia, and other LRIs, increased awareness and diagnostic testing for hMPV are essential. Additionally, efforts to develop targeted therapies or vaccines for hMPV could significantly reduce the burden of severe respiratory infections, particularly in high-risk populations.

The analysis of sampling time reveals interesting trends in the prevalence of hMPV infection over different periods. The highest prevalence was observed in the 2001–2010 period, with a pooled prevalence of 6.6%, indicating a significant burden of hMPV during this decade. This was followed by a slight decline in the 2011–2019 period, with a prevalence of 4.6%. Notably, the most recent period, 2020–2023, showed the lowest prevalence at 3.1%, which may reflect the impact of public health interventions, changes in viral circulation patterns, or increased awareness and diagnostic practices during the COVID-19 pandemic.

During the COVID-19 pandemic, the prevalence of many other respiratory viruses, such as influenza, hRSV, and human rhinovirus (HRV), significantly decreased.¹⁹ This decline was primarily due to the implementation of non-pharmaceutical interventions, including mask-wearing, social distancing, improved hand hygiene, and travel restrictions.²⁰ For example, reports from various countries indicated that influenza activity during the 2020–2021 season reached unprecedented lows.¹⁹ Similarly, the incidence of hRSV dropped significantly in many regions, although a resurgence was observed after restrictions were lifted. These shifts in prevalence patterns highlight the importance of preventive measures in controlling respiratory viruses.²⁰

The analysis of hMPV infection prevalence based on sample type reveals significant variations across different respiratory samples. Overall, upper respiratory tract samples showed a pooled prevalence of 5.4%, which was higher than that of lower respiratory tract samples at 2.1%. Among specific sample types, nasal swabs had the highest prevalence at 6.7%, followed by nasopharyngeal samples at 5.7%. In contrast, lower respiratory samples such as sputum, bronchoalveolar lavage (BAL), and tracheal aspirates exhibited lower prevalence rates, ranging from 1.8% to 3.4%. These findings suggest that hMPV detection is more effective in upper respiratory samples, particularly nasal swabs, which may be attributed to the virus's tropism for the upper respiratory tract.

The analysis of patient type reveals significant differences in the prevalence of hMPV infection between outpatient and inpatient populations. Among outpatients, the pooled prevalence of hMPV infection was 3.3%, while inpatients exhibited a higher prevalence of 6.1%. This disparity suggests that hMPV infection may be more severe or more frequently detected in individuals requiring hospitalization, potentially reflecting a higher disease burden or more severe respiratory symptoms in this group.

The distribution of hMPV types among patients with respiratory symptoms reveals a notable predominance of type A over type B. This disparity in type distribution suggests that hMPV type A may be more prevalent or more easily transmissible than type B in populations with respiratory symptoms.

The subgroup analysis presented in Table 1 provides valuable insights into the association between hMPV infection and the risk of respiratory infections, stratified by various characteristics such as type of disease, detection method, sample type, and sampling time. The overall pooled OR of 4.6 (95% CI: 3.7–5.6) indicates a significant association between hMPV infection and respiratory infections, with low heterogeneity (I² = 14.4%, P = 0.2), suggesting consistent findings across the included datasets. When examining the type of disease, the results reveal that hMPV infection is strongly associated with pneumonia, with pooled ORs of 6.3 (95% CI: 4.8–8.3). The association with pneumonia is particularly noteworthy, as it demonstrates the highest OR among the subgroups, with no observed heterogeneity (I² = 0.0%, P = 0.7), indicating robust and consistent results across studies. In contrast, the association with URIs is less clear. Overall, the analysis of LRTIs and URTIs aligns with these findings, further emphasizing the significant role of hMPV in severe respiratory conditions.

The detection method used for hMPV also appears to influence the observed associations. Real-time PCR, the most commonly used method across the studies (n = 31), yielded a pooled OR of 4.5 (95% CI: 3.5–5.7), which is consistent with the overall estimate. Interestingly, nested PCR showed the highest OR of 11.9 (95% CI: 1.5–89.6), although this result is based on only two datasets and should be interpreted with caution due to the wide CI. The use of RT-PCR and multiplex PCR also demonstrated significant associations, albeit with fewer datasets, suggesting that the choice of detection method may impact the strength of the observed association.

The observed regional variation in hMPV prevalence shows important geographic differences that may reflect underlying epidemiologic, climatic, and health-system factors. The highest prevalence was noted in the Eastern Mediterranean Region (6.6%) and the Region of the Americas (6.3%), which suggests either more intensive surveillance and diagnostic capacity in these areas or genuinely higher transmission rates. In contrast, the African Region exhibited the lowest pooled prevalence (3.7%), which may in part be attributable to under-ascertainment, limited laboratory infrastructure, or competing respiratory pathogens that overshadow hMPV in clinical testing panels.

Global laboratory surveillance consistently shows that hMPV activity is highest in the “spring” season and lowest in the summer/autumn months. For example, aggregated data indicate a spring prevalence of ∼8% (vs. 5.1% in winter and only ∼1.4–1.8% in summer/autumn). Environmental factors are primary drivers of this seasonality. In many temperate-region studies, hMPV incidence is negatively correlated with ambient temperature.²¹ Cold, dry air prolongs the stability and transmissibility of enveloped respiratory viruses, and experimental models (e.g., for hRSV) show enhanced viral shedding at low temperature due to greater stability of virus in secretions.^21,22 By contrast, warm or humid conditions reduce aerosolized virus survival and dilution by ventilation. Therefore, low winter temperatures create conditions that favor hMPV transmission and survival on surfaces and in droplets.²¹

The subgroup analysis based on sample type revealed significant associations between hMPV infection and respiratory infections across different sample types. Nasal samples demonstrated the highest pooled OR of 5.3 (95% CI: 1.2–22.4), indicating a strong association between hMPV detection in nasal specimens and respiratory infection risk, although with wider CIs likely due to the smaller number of datasets (n = 3). Similarly, nasopharyngeal samples showed a comparable pooled OR of 5.0 (95% CI: 3.9–6.3). Overall, upper respiratory samples collectively showed a robust association (pooled OR: 4.6, 95% CI: 3.7–5.7), while lower respiratory samples, represented by only one dataset, had a weaker association (pooled OR: 2.7, 95% CI: 0.2–28.3). These findings suggest that hMPV detection in upper respiratory samples, particularly nasal specimens, is strongly linked to respiratory infection risk, highlighting the importance of sample type in diagnostic and epidemiological studies.

The pooled OR of 4.6, derived from case-control studies, underscores hMPV’s significant association with respiratory infections, indicating that hMPV is a notable causative agent across age groups. By comparing hMPV prevalence in RTI cases to controls without respiratory symptoms or with non-hMPV etiologies, assessed concurrently via PCR, our analysis isolates hMPV’s specific contribution. However, we acknowledge that other respiratory viruses or co-infections may contribute to RTI symptoms, and the OR does not imply sole causality. Limitations, such as potential co-infections or variability in control group definitions, may affect precision. Future studies could further investigate hMPV’s role in co-infections to strengthen causal inferences.

This study has several limitations that should be considered when interpreting the findings. First, the high heterogeneity observed across studies (I² = 98.7%) suggests significant variability in prevalence estimates, which may be attributed to differences in study design, diagnostic methods, sample types, and geographic regions. While subgroup analyses were conducted to address some of these factors, residual heterogeneity may still influence the results.

Second, the reliance on PCR-based diagnostic methods, although highly sensitive, may introduce variability due to differences in assay protocols, primer designs, and laboratory practices. Our exclusive use of PCR-based diagnostics, selected for their high sensitivity, specificity, and standardized application across studies, may underestimate hMPV prevalence, as studies combining PCR with paired serology could potentially enhance detection. Excluding antibody tests may underestimate hMPV prevalence, as serology could detect additional cases, particularly in retrospective analyses. However, serology-based studies were excluded due to variability in diagnostic performance, challenges in accurately timing seroconversion, potential cross-reactivity, and inconsistent protocols, which could introduce heterogeneity and compromise comparability. While the standardized use of PCR ensures reliable prevalence estimates, this constraint may limit the detection of some hMPV cases. The absence of active surveillance for hMPV in community and inpatient settings, coupled with a lack of robust testing guidelines, likely contributes to underestimating prevalence. Many studies included in our meta-analysis relied on passive or inconsistent testing practices, potentially missing undetected hMPV cases and limiting the representativeness of our estimates. This gap hinders accurate assessment of the true disease burden across diverse settings. While our focus on PCR-based diagnostics ensured reliable detection, the lack of standardized surveillance protocols remains a constraint.

Third, the study predominantly included data from hospitalized or clinically severe cases, potentially overestimating the prevalence of hMPV in the general population. Additionally, the exclusion of studies in non-English languages and those focusing on specific patient populations (e.g., immunocompromised individuals) may limit the generalizability of the findings.

Case definitions used in some studies, such as ILI requiring fever, may underestimate hMPV prevalence, particularly in adults, where hMPV infections are often present without fever, potentially excluding atypical cases.

Biases from enrollment, testing, and healthcare-seeking behavior may influence our findings, as hMPV testing is often not prioritized in routine diagnostics due to costs and lack of targeted interventions. Testing is frequently limited to high-risk patients, potentially overestimating hMPV prevalence in inpatient settings while underestimating it in community or outpatient settings, complicating comparisons of disease burden across these contexts. While our meta-analysis included PCR-based studies from diverse settings to capture general population prevalence, these biases likely affect the representativeness of our estimates. Additionally, the exclusion of unpublished studies introduces potential publication bias, which may overestimate prevalence by favoring studies with positive or significant findings.

The observed decline in hMPV prevalence during the COVID-19 pandemic may reflect temporary changes in viral transmission patterns due to public health interventions, and long-term trends remain uncertain. These limitations highlight the need for standardized diagnostic protocols and more inclusive studies to better understand the global epidemiology of hMPV.

Finally, regarding the importance of the issue, many research papers are being published on this topic, making it impossible to include all the latest published papers after data extraction and analysis. Papers published until May 2024 were included in the study, while an updated systematic search showed 28 papers published from then until August 2025 (Supplementary Table S3). It is important to mention that analysis of those papers could impact on the regional prevalence of hMPV as 18 of the papers were from Western Pacific Region. Also, some other factors such as gender and age groups as well as prevalence after COVID-19 pandemics could slightly change. However, due to high number of the included datasets, addition of the recent not-included papers could not make a considerable impact on the overall outcomes of the current study.

In conclusion, hMPV is a significant respiratory pathogen that disproportionately affects young children, particularly those under the age of 5, with the highest prevalence observed in toddlers aged 2–3 years. This heightened susceptibility is likely due to their developing immune systems and increased exposure in communal settings such as daycare or preschool. The prevalence of hMPV declines with age, underscoring the role of immune maturation in reducing infection risk. Diagnostic methods, particularly RT-PCR, have proven effective in detecting hMPV, with variations in prevalence rates across techniques highlighting the importance of selecting appropriate diagnostic tools based on clinical context. hMPV is strongly associated with severe respiratory conditions, such as LRIs, bronchiolitis, and pneumonia, emphasizing the need for increased awareness and targeted diagnostic testing in high-risk populations. The observed decline in hMPV prevalence during the COVID-19 pandemic further underscores the impact of public health interventions on respiratory virus transmission. Overall, these findings highlight the critical need for continued research, improved diagnostic strategies, and potential vaccine development to mitigate the burden of hMPV, particularly in vulnerable populations such as young children and hospitalized patients.

Contributors

A.T designed and administrated the study. H.S and S.G performed all statistical analyses. P.K, M.H.R, and S.G performed search strategy and data extraction. P.K, M.H.R, S.G, and A.T accessed and verified the data. P.K and A.T wrote the initial draft. M.H.R and H.S constructed all maps and graphs. A.M and A.T performed intellectual interpretation. All authors were involved in revision. All authors read and approved the final draft.

Data sharing statement

All data included in this study are available upon request from the corresponding author.

Editor note

The Lancet Group takes a neutral position with respect to territorial claims in published maps and institutional affiliations.

Declaration of interests

The authors have no conflict of interest.