Received: 21 September 2022 Revised: 22 November 2022 Accepted: 23 November 2022 DOI: 10.1111/irv.13085 ORIGINAL ARTICLE The burden of non-SARS-CoV2 viral lower respiratory tract infections in hospitalized children in Barcelona (Spain): A long-term, clinical, epidemiologic and economic study Jorgina Vila 1,2,3 | Esther Lera 2,4 | Cristina Andrés 5 | Maria Piñana 5 | Victoria Rello-Saltor 1 | Marc Tobeña-Rué 1,2 | Joan Balcells 2,6 | Zaira Benítez-Díaz 7,8 | Marta Beatriz Aller 7,8 | Rosario Muñoz 7,8 | Ana Vázquez 9 | Carlos Rodrigo 2,10,11 | Pere Soler-Palacín 2,3,12 | Andrés Anton 5 1 Paediatric Hospital Medicine, Department of Paediatrics, Hospital Universitari Vall d’Hebron, Barcelona, Spain 2 Department of Paediatrics, Obstetrics and Gynecology, Preventive Medicine and Public Health. Faculty of Medicine, Universitat Autònoma de Barcelona, Barcelona, Spain 3 Infection in Immunocompromised Paediatric Patients, Vall d’Hebron Research Institute, Barcelona, Spain 4 Paediatric Emergency Unit, Department of Paediatrics, Hospital Universitari Vall d’Hebron, Barcelona, Spain 5 Respiratory Viruses Unit, Virology Section, Microbiology Department, Hospital Universitari Vall d’Hebron, Barcelona, Spain 6 Paediatric Critical Care Unit, Department of Paediatrics, Hospital Universitari Vall d’Hebron, Barcelona, Spain 7 Department of Information Systems and Decision Support, Hospital Universitari Vall d’Hebron, Barcelona, Spain 8 Health Services Research Group, Vall d’Hebron Research Institute, Barcelona, Spain 9 Department of Applied Statistics, Universitat Autònoma de Barcelona, Barcelona, Spain 10 Department of Paediatrics, Hospital Universitari Germans Trias i Pujol, Barcelona, Spain 11 Germans Trias i Pujol Research Institute, Barcelona, Spain 12 Paediatric Infectious Diseases and Immunodeficiencies Unit, Department of Paediatrics, Hospital Universitari Vall d’Hebron, Barcelona, Spain Correspondence Pere Soler-Palacín, Department of Paediatrics, Abstract Hospital Universitari Vall d’Hebron, Vall Background: Viral lower respiratory tract infections (LRTI) are the leading cause of d’Hebron Barcelona Hospital Campus, Passeig Vall d’Hebron 119-129, Barcelona 08035, hospitalization in children. In Catalonia (Spain), information is scarce about the bur- Spain. den of viral LRTIs in paediatric hospitalizations. The aim of this study is to describe Email: pere.soler@vallhebron.cat epidemiological, clinical, virological and economic features of paediatric hospitaliza- Funding information tions due to viral LRTI. Health Research Fund, the Spanish Ministry of Economy and Competitiveness, Grant/Award Methods: From October 2012 to December 2020, children aged <16 years admitted Numbers: FIS PI18/00685, FIS PI14/01838; to a tertiary paediatric hospital in Catalonia (Spain) with confirmed viral LRTI were Spanish Network for Research in Infectious Diseases, Grant/Award Number: REIPI included in the study. Virus seasonality, prevalence, age and sex distribution, clinical RD16/0016/0003; Spanish National R + D + I characteristics, hospital costs and bed occupancy rates were determined. Plan 2008–2011; Carlos III Health Institute; European Regional Development Fund; Sanofi Results: A total of 3,325 children were included (57.17% male, 9.44% with comorbid- ities) accounting for 4056 hospitalizations (32.47% ≤ 12 months): 53.87% with wheezing/asthma, 37.85% with bronchiolitis and 8.28% with pneumonia. The most common virus was respiratory syncytial virus (RSV) (52.59%). Influenza A was This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. Influenza and Other Respiratory Viruses published by John Wiley & Sons Ltd. Influenza Other Respi Viruses. 2023;17:e13085. wileyonlinelibrary.com/journal/irv 1 of 9 https://doi.org/10.1111/irv.13085 2 of 9 VILA ET AL. associated with pneumonia (odds ratio [OR] 7.75) and caused longer hospitalizations (7 ± 31.58 days), while RSV was associated with bronchiolitis (OR 6.62) and was the most frequent reason for admission to the paediatric intensive care unit (PICU) (11.23%) and for respiratory support (78.76%). Male sex, age ≤12 months, chronic conditions and bronchiolitis significantly increased the odds of PICU admission. From October to May, viral LRTIs accounted for 12.36% of overall hospital bed days. The total hospitalization cost during the study period was €16,603,415. Conclusions: Viral LRTIs are an important cause of morbidity, hospitalization and PICU admission in children. The clinical burden is associated with significant bed occupancy and health-care costs, especially during seasonal periods. KEYWORDS cost of illness, global burden of disease, hospitalization, respiratory tract infections, viruses 1 | I N T RO DU CT I O N in order to improve understanding of the burden of viral LRTIs in hos- pitalized children. Viral lower respiratory tract infections (LRTI) are an important cause of morbimortality and hospital admissions in children, especially in the population aged under five years.1 The most common LRTIs in chil- 2 | METHODS dren are bronchiolitis, community acquired pneumonia (CAP), wheez- ing and asthma exacerbation.2–4 2.1 | Study design and participants Bronchiolitis is a viral infection with an annual incidence of 5.2/100 in children under six months of age in Spain, and an annual This is a descriptive, observational, retrospective, longitudinal study hospitalization incidence of 2.1 cases/100 children per year.2 It is performed from October 2012 (week 42) to December 2020 (week responsible for up to 13.5% of paediatric intensive care unit (PICU) 50) in the Vall d’Hebron University Hospital in Barcelona (Spain), the admissions during the winter months.5 The main etiological agent of largest hospital complex in the Catalan Health System, which is also a bronchiolitis is the respiratory syncytial virus (RSV), followed by the rhi- tertiary paediatric hospital catering for around 9,200 paediatric hospi- novirus (RV).5 CAP is also a major cause of illness in children in devel- talizations per year. The hospital has 16 PICU beds that can be oped countries. A virus is identified in 43–67% of cases of CAP, with increased to 20 during seasonal periods and 55 in general paediatric RV, RSV, and influenza viruses being the main culprits worldwide.3 ward. This is the reference hospital for the population living in the Finally, 30–50% of children suffer one episode of wheezing before northern part of Barcelona’s metropolitan area, including 30, 508 school age, and 30–40% present recurrent episodes. Asthma is the inhabitants under 15 years old (2.68% of Catalan population under leading cause of chronic disease worldwide, with a prevalence of 10% 15 years). Moreover, it is also the main paediatric hospital in Catalonia in Western Europe.4 Both diseases can be triggered by viral infection, for patients requiring PICU care or solid organ transplant as well as with RV being the most common etiological agent, followed by RSV.4 for patients suffering from oncohematologic diseases or some forms The introduction of molecular testing as standard laboratory prac- of rare diseases being the population covered actually higher than the tice to determine the presence RSV, influenza, and other respiratory above-mentioned. viruses has led to a better understanding of the burden of viral LRTIs All consecutive paediatric patients (<16 years old) with in hospitalized children. Interest in respiratory viral infections has laboratory-confirmed viral LRTI other than SARS-CoV-2 who required been heightened during the COVID-19 pandemic; however, there is hospitalization during the study period were included. Respiratory scant information about the surveillance of respiratory viruses other samples for laboratory confirmation were collected within 3 days than RSV, influenza, and severe acute respiratory syndrome before and 2 days after admission. The reason for hospitalization was coronavirus-2 (SARS-CoV-2) in hospitalized paediatric patients. Viral considered to be an LRTI if, according to the International Classifica- LRTIs are not only interesting from a microbiological and clinical per- tion of Diseases ninth (ICD-9) or tenth (ICD-10) revision (Table S1), spective, they also impact hospital bed occupancy rates and health- the main diagnosis was bronchiolitis (466, J21), CAP (480, 483–488, care costs, and data on the latter can help hospital managers plan the J12, J16-J18, J09-J11), wheezing, or asthma exacerbation (466, 493, distribution of their sources during annual epidemics. In this study, we J20, J45), or the main diagnosis was influenza (487, 488, J09-J11), analyse the demographic, clinical, and viral characteristics of LRTIs respiratory failure (518, J96) or neonatal infection (771, P28, P39, together with the cost of paediatric hospitalizations due to viral LRTI P96) with a secondary diagnosis of bronchiolitis, CAP, wheezing or other than SARS-CoV-2 in a referral hospital in Southwestern Europe asthma exacerbation. VILA ET AL. 3 of 9 Patients that met the inclusion criteria were classified in three 2.4 | Statistical analysis groups according to the main clinical diagnosis: bronchiolitis—defined as the first episode of viral LRTI in an infant aged under 2 years as per Continuous variables are described as mean and standard deviation 6 our local protocol, CAP and wheezing/asthma. (SD) and categorical variables as frequencies and proportions. Contin- gency tables were used to analyse the relationship between two quali- tative variables, and chi-square tests were used to assess associations. 2.2 | Variables analysed The Kruskal Wallis test was used to analyse the relationship between quantitative and qualitative variables. A logistic regression model was Demographical and clinical data were retrospectively gathered from used to characterize the response variables, and the results were pre- structured data in each patient’s electronic clinical record by the sented as an odds ratio (OR) with the corresponding 95% confidence Hospital Information Systems. These data included: main and sec- interval (95% CI). A multivariate logistic regression model was used to ondary diagnosis, respiratory viruses detected, age, sex, date of hos- determine the risk factors associated with PICU admission. In all pitalization and discharge, admission to the PICU, and if so, date of models, pair-wise comparisons were performed using Tukey correc- admission and discharge, need for respiratory support and maximum tion for multiple comparisons. The significance level was set at 0.05 support administered (conventional oxygen, high flow nasal cannula, for all tests. The results were analysed using SAS v9.4 (SAS Institute or non-invasive mechanical ventilation (HFNC/NIMV), invasive Inc., Cary, NC, USA). mechanical ventilation (IMV) and extracorporeal membrane oxygena- tion (ECMO) (Table S2). All the information was uploaded to a codi- fied database. 3 | RE SU LT S Hospitalization cost was based on the prices established by our hospital and calculated based on 2021 fares: €615 per day in the pae- During the study period, 29,511 children aged <16 years were admit- diatric ward and €2,350 per day in the PICU (internal data from the ted from the paediatric emergency department, 4,056 (13.74%) with a Catalan Health System data). Surveillance seasons lasted from week diagnosis of viral LRTI. The LRTI episodes corresponded to 3,325 chil- 40 in October to week 20 in May. dren, of whom 1,317 (32.47%) were under 12 months of age, 1,902 (57.20%) were male, 314 (9.44%) had a history of chronic disease, and 146 (3.60%) were immunocompromised. Four (0.10%) patients died. 2.3 | Procedures In total, 4,753 viruses were isolated from the 4,056 patients with viral LRTI (≥2 viruses were detected in 565 patients). Age distribution Respiratory tract specimens were collected within 3 days prior to of LRTI and the viruses isolated are shown in Table 1. The most com- admission for LRTI from all emergency department patients with mon virus causing LRTI was RSV, which was detected in 2,133 symptoms of respiratory infection. Thereafter, respiratory specimens (52.59%) children, followed by RV (1,210 [29.83%]), and HMPV were taken within two day of admission to the paediatric ward or (341 [8.41%]). Among the 84 hCoV diagnoses, 12 corresponded to the PICU. Samples were processed in the hospital microbiology lab- hCoV-229E, 28 to hCoV-NL63, and 44 to hCoV-OC43. Of the oratory within 24 hours for routine testing, and within six hours for 255 HPIV diagnoses, 39 were HPIV-1, 9 HPIV-2, 149 HPIV-3 and rapid detection tests for influenza viruses and RSV. From October 58 HPIV-4. The most frequent co-detections were RSV-RV and AdV- 2012 to May 2016, detection of respiratory viruses was routinely RV (83/565 and 64/565, respectively) (Table S3). Hospitalization performed by direct immunofluorescent antigen detection assay occurred during the surveillance seasons (from week 40 in October to (D3 Ultra 8™ DFA Respiratory Virus Screening & Identification Kit, week 20 in May) in 3,684 (90.82%) of study cases. According to clini- Diagnostic HYBRIDS, USA), which detects influenza A (FLUAV) and cal diagnosis, of the 4,056 patients hospitalized for viral LRTI, 2,185 B (FLUBV), RSV, adenovirus (AdV), human parainfluenza viruses 1–3 (53.87%) presented wheezing/asthma, 1,535 (37.84%) bronchiolitis, (HPIV 1–3) and human metapneumovirus (HMPV). From October and 336 (8.28%) CAP. Three-hundred and sixty patients with LRTIs 2014 to November 2016, real-time multiplex RT-PCR—first, the (8.87%) required PICU admission, representing 7.89% of all patients Anyplex™ II RV16 Detection kit (Seegene, South Korea), which was admitted to the PICU during the study period. Overall, 333/360 later replaced by the Allplex™ Respiratory Panel (Seegene, (92.50%) were admitted during the surveillance seasons. One hundred South Korea)—was included for routine testing for the foregoing and eighty children were admitted directly to the PICU from other respiratory viruses in addition to HPIV-4, bocavirus (BoV), enterovi- centres. Of the four non-surviving patients, one died due to FLUAV rus (EV), RV, and human coronaviruses (hCoV) NL63, OC43 and CAP, two due to CAP caused by RSV in children with a history of neu- 229E. Rapid RSV and influenza testing during epidemics was per- rologic comorbidities, and one due to RSV-related wheezing/asthma formed using an immunofluorescent assay (Sofia RSV or Influenza in a child with chronic lung disease. FIA, Quidel, CA, USA) or a rapid molecular test (GeneXpert Flu or Seasonal differences in the percentage of laboratory-confirmed Flu/RSV XC, Cepheid, CA, USA). If the rapid test was negative, the cases were observed in FLUAV, AdV, RSV, RV (p < 0.0001 for each), study was extended to include other respiratory viruses that were FLUBV (p = 0.0199), HMPV (p = 0.0006), HPIV-3 (p = 0.0037), and detected using routine testing techniques. BoV (p = 0.0112), especially during the 2020–2021 season when the 4 of 9 VILA ET AL. TABLE 1 Age distribution of PICU patients, LRTI, and isolated viruses Total (column) ≤12 months 12–24 months >24 month-4 years 5–15 years p value Total (row), n (%) 1,317 (32.47%) 1,120 (27.61%) 1,117 (27.54%) 502 (12.38%) Number of PICU patients, n (%) 360 227 (63.06%) 79 (21.94%) 35 (9.72%) 19 (5.28%) <0.0001 LRTI, n (%) Bronchiolitis 1,535 1,187 (77.33%) 348 (22.67%) 0 (0.0%) 0 (0.0%) <0.0001 Wheezing/asthma 2,185 116 (5.31%) 712 (32.58%) 938 (42.93%) 419 (19.18%) <0.0001 CAP 336 14 (4.17%) 91 (27.08%) 154 (45.83%) 77 (22.92%) <0.0001 Virus, n (%) FLUAV 136 22 (16.18%) 42 (30.88%) 50 (36.76%) 22 (16.18%) 0.0009 FLUBV 53 10 (18.87%) 11 (20.75%) 19 (35.85%) 13 (24.53%) 0.0130 RSV 2,133 1,002 (46.98%) 583 (27.33%) 469 (21.99%) 79 (3.70%) <0.0001 HMPV 341 104 (30.50%) 122 (35.78%) 92 (26.98%) 23 (6.74%) 0.0004 AdV 229 16 (6.99%) 91 (39.74%) 99 (43.23%) 23 (10.04%) <0.0001 hCoVa 84 17 (20.48%) 19 (22.89%) 36 (43.37%) 11 (13.25%) 0.0097 RV 1,210 183 (15.12%) 279 (23.06%) 415 (34.30%) 333 (27.52%) <0.0001 BoV 169 6 (3.55%) 93 (55.03%) 68 (40.24%) 2 (1.18%) <0.0001 a HPIV 255 55 (21.74%) 80 (31.62%) 83 (32.81%) 35 (13.83%) 0.0041 EV 143 5 (350%) 50 (34.97%) 68 (47.55%) 20 (13.99%) <0.0001 Abbreviations: AdV, adenovirus; BoV, bocavirus; CAP, community acquired pneumonia; EV, enterovirus; FLUAV, influenza A virus; FLUBV, influenza B virus; hCoV, human coronaviruses; HMPV, human metapneumovirus; HPIV 1–4, human parainfluenza viruses 1–4; LRTI, lower respiratory tract infection; PICU, paediatric intensive care unit; RSV, respiratory syncytial virus; RV, rhinovirus. a One patient with hCoV-229E and hCoVNL63 coinfection and 2 patients with hPIV-3 and hPIV-4 coinfection were only taken into account once. annual FLUAV, FLUBV, RSV and HMPV epidemics were not observed, positive for each virus with those that tested negative for that virus and RV and AdV were the main viruses circulating that season (Figure 1). FLUAV was associated with pneumonia (OR 7.75, 95% CI (Figure S1 and Table S4). Additionally, statistically significant differ- 5.24–11.49; p < 0.0001), RSV was associated with bronchiolitis ences were observed in the monthly trends of FLUAV, FLUBV, RSV, (OR 3.57, 95% CI 3.13–4.17; p < 0.0001), and RV caused mainly HMPV, AdV, RV, HPIV 3–4, and EV (p < 0.0001 for each) (Figure S2). wheezing/asthma (OR 3.58, 95% CI 3.07–4.18; p < 0.0001). During seasonal periods, RSV was the first virus to peak (in December), We also performed multivariate analyses to identify the host fac- followed by FLUAV peaking in February, FLUBV in March, and HMPV tors that increased the risk of PICU admission (Table S5): male sex between March and April. (adjusted p = 0.0048), age ≤ 12 months (adjusted p = 0.0002 com- Analysing the main clinical findings related to laboratory- pared with 12–24 months), history of chronic conditions (adjusted confirmed LRTI cases for respiratory viruses (Table 2), we found that p < 0.0001), and bronchiolitis (adjusted p = 0.0004). We were unable 3,068 (75.64%) of patients with LRTIs required respiratory support, of to confirm any correlation with viruses. which 2,750 (89.63%) needed conventional oxygen supply, Our analysis of hospitalization costs showed that during surveil- 234 (7.63%) HFNC/NIMV, 82 (2.67%) IMV, and 2 (0.07%) ECMO. The lance seasons, viral LRTIs accounted for 12.36% of overall bed days two cases requiring ECMO were a previously healthy four year-old girl (17,921/144,900), 14.39% of paediatric ward bed days with no disease history who presented severe pneumonia due to (15,294/106,260) and 6.79% of PICU bed days (2,627/38,640). Dur- FLUAV-Streptococcus pneumoniae co-infection, and a 4 year old girl ing the entire study period, total hospitalization costs were with severe neurologic sequelae due to a brain tumour who presented €16,603,415 (being only €92,930 attributable to the 2020–2021 sea- RSV CAP, which was ultimately fatal. FLUAV caused longer hospitali- son), with a cost per patient of €4,093.54 (Table S6). zations than the other viruses (7.51 ± 31.58 days, p = 0.0077). The highest proportion of patients needing admission to the PICU (241/2,133; 11.23%, p < 0.0001) and respiratory support 4 | DI SCU SSION (1,680/2,133; 78.76%, p < 0.0001) presented RSV infection. In con- trast, RV-related disease was milder (length of stay [LOS]: 4.04 In this study, we reviewed the burden of viral LRTIs other than SARS- ± 9.45 days, p = 0.0065; PICU admission: 6.94% of patients, CoV-2 in children aged <16 years in one of the largest tertiary paedi- p = 0.0057). atric hospitals in Catalonia (Spain) from October 2012 to December To determine whether certain respiratory viruses were associated 2020. Our results show that viral LRTIs are an important cause of with specific LRTIs, we compared the number of patients who tested morbidity and high healthcare costs in hospitalized children during VILA ET AL. TABLE 2 Association between causative viruses and mean hospital LOS, PICU admission and respiratory support hCoVs (OC-43, FLUAV FLUBV RSV HMPV AdV 229-E, and NL-63)a RV BoV hPIVs (1-4)a EV p valueb Total 136 53 2,133 341 229 83 1,210 169 253 143 LOSMean 7.51 (31.58) 5.35 (3.57) 5.16 (12.51) 4.49 (3.32) 3.78 (3.41) 4.28 (4.75) 4.04 (9.45) 3.86 (5.07) 4.85 (6.07) 3.54 (2.7) =0.0126 (SD) p valuec 0.0077 0.7361 0.0184 0.6164 0.2088 0.6521 0.0065 0.3489 0.9864 0.2298 PICU admission, 5 (3.68%) 4 (7.55%) 241 (11.23%) 21 (6.16%) 9 (3.93%) 8 (9.64%) 84 (6.94%) 16 (9.47%) 12 (4.74%) 11 (7.69%) <0.0001 N (%) p valuec 0.0378 0.7272 <0.0001 0.0709 0.0102 0.7984 0.0057 0.7803 0.0193 0.6315 Respiratory support, 97 (71.32%) 39 (73.58%) 1,680 259 (75.95%) 155 (67.69%) 58 (69.88%) 905 (74.79%) 131 (77.51%) 161 (63.64%) 108 <0.0001 N (%) p valuec 0.2615 0.7084 (78.76%) 0.9077 0.0072 0.2596 0.4485 0.5876 <0.0001 (75.52%) <0.0001 0.9933 Maximum respiratory 0.0779 0.4517 0.0027 0.1533 0.2033 0.4129 0.2917 0.0139 0.5864 0.5180 support, p valuec ECMO, N (%)d 1 (0.74%) 0 1 (0.05%) 0 0 0 0 0 0 0 IMV, N (%)d 4 (2.94%) 0 51 (2.39%) 5 (1.47%) 3 (1.31%) 1 (1.20%) 25 (2.07%) 2 (1.18%) 3 (1.19%) 2 (1.4%) HFNC/NIMV, N (%)d 4 (2.94%) 2 (3.77%) 153 (7.17%) 12 (3.52%) 6 (2.62%) 8 (9.64%) 59 (4.89%) 11 (6.51%) 5 (1.98%) 7 (4.9%) d O2, N (%) 88 (64.71%) 37 (69.81%) 1,475 242 (71.00%) 146 (63.76%) 49 (59.04%) 821 (67.85%) 118 (69.82%) 153 (60.47%) 99 (69.23%) (69.15%) Abbreviations: AdV, adenovirus; BoV, bocavirus; ECMO, extracorporeal membrane oxygenation; EV, enterovirus; FLUAV, influenza A virus; FLUBV, influenza B virus; hCoV, human coronaviruses; HFNC/NIMV, high flow nasal oxygen/non-invasive mechanical ventilation; HMPV, human metapneumovirus; HPIV 1–4, human parainfluenza viruses 1–4; IMV, invasive mechanical ventilation; LOS, length of hospital stay; LRTI, lower respiratory tract infection; O2, conventional oxygen; PICU, paediatric intensive care unit; RSV, respiratory syncytial virus; RV, rhinovirus; SD, standard deviation. a One patient with hCoV-229E and hCoVNL63 coinfection and 2 patients with hPIV-3 and hPIV-4 coinfection were only taken into account once for the study of virus-related disease. b p values are calculated comparing positive cases of each virus. c p values are calculated comparing cases positive for each virus with those that tested negative for the virus studied. d From the total of each virus. 5 of 9 6 of 9 VILA ET AL. F I G U R E 1 Association between causative viruses and related LRTIs (wheezing/asthma, bronchiolitis and community acquired pneumonia) expressed by OR and 95% CI. The OR is calculated by comparing the number of positive cases with those that tested negative for each virus. p value is only specified in statistically significant cases. AdV, adenovirus; BoV, bocavirus; 95CI, 95% confidence interval; EV, enterovirus; FLUAV, influenza A virus; FLUBV, influenza B virus; hCoVs, human coronaviruses; HMPV, human metapneumovirus; HPIVs, human parainfluenza viruses; LRTI, lower respiratory tract infection; OR, odds ratio; RSV, respiratory syncytial virus; RV, rhinovirus seasonal periods, particularly in children aged <1 year, males, and in trends described in our study can be explained by the interaction of children with chronic conditions. Interestingly, 90% of study patients environmental, host, and viral factors, in addition to the use of differ- had no history of chronic conditions. An analysis by virus type showed ent laboratory diagnostic techniques.11,12 The interaction of the fore- that RSV was the main etiological agent. going factors was particularly important during the first half of the RSV, RV, wheezing/asthma, and bronchiolitis were the most prev- 2020–2021 season, in which the introduction of non-pharmaceutical alent viruses and diseases during the study period; bronchiolitis interventions and restrictions on international travel during the increased the risk of requiring intensive care. As previously reported COVID-19 pandemic prevented the spread not only of SARS-CoV-2, in the literature, and confirmed by our results, host factors such as but also of other respiratory viruses, particularly RSV and influenza male sex, younger age, and history of chronic conditions increase the viruses.8,10 Nevertheless, RV and AdV were still detected, probably severity of viral LRTIs, and therefore the likelihood of PICU because non-pharmaceutical interventions were not as effective in admission.5,7 these non-enveloped viruses as in other respiratory viruses, including Before the COVID-19 pandemic, the burden of respiratory SARS-CoV-2.13,14 viruses was particularly high during the winter months in temperate In addition to the prevalence and seasonality of LRTIs, one of the countries such as Spain due to the annual RSV, influenza, and HMPV most interesting factors to be considered is their burden in hospital- epidemics.8–10 Differences in seasonal and monthly respiratory virus ized children. According to the Global Burden of Diseases 2017, the VILA ET AL. 7 of 9 burden of LRTI in children aged <5 years was lower in pre-pandemic years old.26 Other strategies currently under development are anti- seasons due to improvements in hygienic and sanitary conditions and viral treatments based on humanized monoclonal antibodies against increased vaccine coverage.15 Nevertheless, our results show that RSV, which can prevent 77.3% of hospital admissions for RSV-related viral LRTIs are still an important cause of morbidity in children. The LRTI in children entering their first RSV season.27 few studies that have analysed the burden of LRTIs in paediatric hos- This study has several limitations, apart from its retrospective pitals have mainly focused on the most virulent viruses, such as RSV, design. Factors such as viral-bacterial and viral-viral co-detections influenza viruses, and HMPV.16–18 We, however, found that respira- were not taken into consideration when evaluating clinical evolution, tory viruses other than RSV, influenza, and HMPV were responsible and may modify clinical severity. However, although viral-bacterial co- for a third of all cases of LRTIs, showing that the burden of these detections are known to increase disease severity, there is no conclu- viruses is significant in terms of bed occupancy rates and healthcare sive evidence that viral-viral co-detections increase the disease bur- costs. Overall, RSV has been described as the main cause of hospital den compared with single infections.20,28 From a microbiological point admission and LRTI-related deaths worldwide, followed by influenza of view, we used various diagnostic techniques with varying sensitivi- viruses, and HMPV, particularly in children aged <1 year.1,16–18 RSV ties to detect viral species. This makes it difficult to compare data and influenza viruses were also the main causes of morbidity and mor- between seasons, and may have led us to underestimate the true bur- tality in our study, as both were associated with prolonged LOS. In den of RV, BoV, hCoV, HPIV-4 and EV, especially during the first sea- the case of RSV, this was probably due to the need for respiratory son studied, when these viruses were not routinely detected. In support and PICU admission for bronchiolitis,19 while in the case of addition, the use of rapid tests targeting only RSV or influenza viruses influenza, it was likely due to the use of parenteral antibiotics and during annual epidemics may have led us to underestimate the role of respiratory support for secondary bacterial infection in patients with viral co-detection during the winter months. Our use of structured 20 CAP. FLUAV and RSV were associated with the four fatalities in our data allowed us to export a considerable volume of information; how- study, all of which occurred in children with a history of comorbidities, ever, although a group of experts reviewed the disease codes, we can- as previously reported.19,20 Despite the lower number of HMPV not rule out the existence of disease and respiratory support coding cases, the risk of severe infection has been described as similar to that errors. Nevertheless, large amounts of structured patient data from 17,21 of RSV or influenza viruses, especially in children aged <1 year. hospital information system are relatively easy to obtain, and will play However, we observed milder outcomes in HMPV infection, probably a key role in the clinical and virological surveillance of viral LRTIs in due to the lower proportion of children aged <1 year and to differ- children. The total costs may be biassed because of COVID-19 pan- ences in HMPV genotypes circulating in each region.9 RV, despite demic, especially in the 2020–2021 season. Additionally, the calcu- being the second most prevalent LRTI-related virus, caused milder dis- lated cost for this season is underrated as patients were only included ease compared with RSV or influenza viruses, probably due to less until December 2020. Also, costs may have changed along the study respiratory cytokine production leading to a milder proinflammatory period but we were unable to retrieve appropriate yearly data due to response.22 Our data showed that a high proportion of hCoV- and changes in the measurement system and economic databases so they BoV-related LRTIs required PICU admission, and a high number of were all calculated based on 2021 fares. Finally, despite the relevance BoV-related infections required respiratory support, although this of our findings, the single-centre design of our study prevents us from occurred mostly in cases of co-infection with more virulent viruses, extrapolating our results to other geographical areas. Similar studies in such as RSV.23,24 other regions are needed to compare findings on viral LRTIs. Aside Viral LRTIs are generally associated with high bed occupancy from these limitations, our study stands out from previous research rates during surveillance seasons. In our study, the total direct health conducted in our geographical area, insofar as we describe viral LRTIs care cost due to LRTI was €16,603,415, with a per-patient cost of from the perspective of microbiological data, clinical outcomes, and €4,093.54—a considerable economic burden on the health care sys- consumption of healthcare resources.6,9,29 Finally, one of the tem, considering that the per capita budget of the Catalan Health strengths of our study is the inclusion of a large patient sample, which Institute is approximately €1,293.1/year, a figure that falls far short of allowed to study the burden of viral LRTIs in hospitalized children the real cost of these hospitalizations.25 In addition to direct hospital over a long period of time. costs, the cost of treating children in primary care and secondary costs In conclusion, viral LRTIs place a heavy burden on hospital arising from parental work absenteeism, together with travel and meal resources, and on the healthcare system in general due to their associ- expenses during the child’s hospital stay, should be taken into ated morbidity. LRTI-related mortality in our region, however, is account. This underscores the importance of research and the imple- scarce, and associated with previous chronic conditions. Continuous mentation of preventive measures to reduce morbidity, clinical costs, surveillance of LRTI will improve healthcare planning, show changes in and bed occupancy rates. Our results suggest that these measures prevalence and severity trends based on previous seasons, which may should focus on the prevention of RSV and influenza viruses during indicate the circulation of a new viral variant, and allow hospital epidemics, especially in younger children, and should be offered to authorities to monitor the efficacy of preventive or therapeutic mea- the entire paediatric population, regardless of their comorbid status. sures in the paediatric population. Our experience of the recent Preventive measures include influenza vaccination, which can prevent COVID-19 pandemic has shown that active LRTI microbiological and up to 50% of influenza-related hospitalizations in children under eight clinical surveillance should be a public health priority. Finally, it is 8 of 9 VILA ET AL. essential to increase the limited number of antivirals available for Rodrigo: Conceptualization; investigation; methodology; project influenza and RSV by developing new prophylaxis measures and drugs administration; resources; visualization; writing-original draft; writing- to treat viral LRTI. This will improve the health of the paediatric popu- review and editing. Pere Soler Palacín: Conceptualization; funding lation, and reduce the consumption of hospital resources and associ- acquisition; investigation; supervision; writing-original draft; writing- ated costs. review and editing. Andrés Anton: Conceptualization; data curation; funding acquisition; investigation; methodology; project administra- ACKNOWLEDGEMEN TS tion; resources; supervision; validation; visualization; writing-original This work was partially supported by the Spanish National R + D + I draft; writing-review and editing. Plan 2008–2011 and by the Carlos III Health Institute, the Subdirectorate-General of Networks and Cooperative Research Cen- PE ER RE VIEW ters, the Spanish Ministry of Economy and Competitiveness, the The peer review history for this article is available at https://publons. Spanish Network for Research in Infectious Diseases (REIPI com/publon/10.1111/irv.13085. RD16/0016/0003), and by the Health Research Fund, the Spanish Ministry of Economy and Competitiveness (grants FIS PI14/01838 DATA AVAILABILITY STAT EMEN T and FIS PI18/00685), and the European Regional Development Fund Anonymized and de-identified participant data will be available upon (ERDF). Editorial assistance for revision of the English language and request until 2 years after publication. Data will be available for edition of the manuscript was provided by Vanessa Marfil, medical researchers who provide a methodologically sound proposal. Requests writer from MSC (Spain) and funded by Sanofi. may be sent to the corresponding author, and to gain access data requestors will need to sign a data access agreement. CONF LICT OF IN TE RE ST ORCID All authors certify that they have no affiliations with or involvement Pere Soler-Palacín https://orcid.org/0000-0002-0346-5570 in any organization or entity with any financial interest or non- financial interest in the subject matter or materials discussed in this RE FE RE NCE S manuscript. 1. GBD 2016 Lower Respiratory Infections Collaborators. Estimates of the global, regional, and national morbidity, mortality, and aetiologies E TH I CS S T A TE M E N T of lower respiratory infections in 195 countries, 1990–2016: a sys- All procedures were performed in accordance with the ethical stan- tematic analysis for the Global Burden of Disease Study 2016. Lancet dards of the Vall d’Hebron Barcelona Hospital Campus Research Infect Dis. 2018;18(11):1191-1210. 2. Muñoz-Quiles C, Lo  pez-Lacort M, Úbeda-Sansano I, et al. Popula- Ethics Committee and the 1964 Helsinki Declaration and its subse- tion-based analysis of bronchiolitis epidemiology in Valencia. Spain quent amendments, or comparable ethical standards. The study was Pediatr Infect Dis J. 2016;35(3):275-280. doi:10.1097/INF. approved by the Vall d’Hebron Hospital Clinical Research Ethics Com- 0000000000000993 mittee (PR[AMI]556/2020). 3. Ruuskanen O, Lahti E, Jennings LC, Murdoch DR. Viral pneumonia. Lancet. 2011;377(9773):1264-1275. doi:10.1016/S0140-6736(10) 61459-6 AUTHOR CONTRIBUTIONS 4. Jartti T, Gern JE. Role of viral infections in the development and Jorgina Vila: Conceptualization; data curation; investigation; method- exacerbation of asthma in children. J Allergy Clin Immunol. 2017; ology; project administration; resources; supervision; validation; visu- 140(4):895-906. doi:10.1016/j.jaci.2017.08.003 5. Ghazaly M, Nadel S. Characteristics of children admitted to intensive alization; writing-original draft; writing-review and editing. Esther care with acute bronchiolitis. Eur J Pediatr. 2018;177(6):913-920. Lera: Conceptualization; methodology; resources; supervision; valida- doi:10.1007/s00431-018-3138-6 tion; writing-original draft; writing-review and editing. Cristina 6. Vila J, Lera E, Peremiquel-Trillas P, et al. Management of hospitalized Andrés: Data curation; investigation; writing-original draft; writing- respiratory syncytial virus bronchiolitis in the pediatric ward in Spain: assessing the impact of a new clinical practice protocol. Pediatr review and editing. Maria Piñana: Data curation; investigation; Drugs. 2022;24(1):63-71. doi:10.1007/s40272-021-00488-6 writing-review and editing. Victoria Rello-Saltor: Investigation; 7. Heinonen S, Rodriguez-Fernandez R, Diaz A, Rodriguez-Pastor SO, writing-original draft; writing-review and editing. Marc Tobeña-Rué: Ramilo O, Mejias A. Infant immune response to respiratory viral Investigation; writing-original draft; writing-review and editing. Joan infections. Immunol Allergy Clin North Am. 2019;39(3):361-376. doi: Balcells: Investigation; writing-original draft; writing-review and edit- 10.1016/j.iac.2019.03.005 8. European Centre for Disease Prevention and Control. Seasonal ing. Zaira Benítez-Díaz: Data curation; investigation; methodology; influenza- Annual Epidemiological Report for 2020-2021. Available resources; software; validation; writing-review and editing. Marta from: ecdc.europa.eu/en/publications-data/seasonal-influenza- Beatriz Aller: Data curation; investigation; methodology; resources; annual-epidemiological-report-2020-2021 [Accessed 5 May 2022]. software; supervision; validation; visualization; writing-review and 9. Piñana M, Vila J, Maldonado C, et al. Insights into immune evasion of human metapneumovirus: novel 180- and 111-nucleotide duplica- editing. Rosario Muñoz: Data curation; resources; writing-review and tions within viral G gene throughout 2014-2017 seasons in editing. Ana Vazquez: Formal analysis; investigation; methodology; Barcelona, Spain. J Clin Virol. 2020;132:104590. doi:10.1016/j.jcv. software; supervision; validation; writing-review and editing. Carlos 2020.104590 VILA ET AL. 9 of 9 10. Coma E, Vila J, Méndez-boo L, Anto  n A. Respiratory syncytial virus Pediatr Infect Dis J. 2022;41(4):284-289. doi:10.1097/INF. infections in young children presenting to primary care in Catalonia 0000000000003416 during the COVID-19 pandemic. J Pediatric Infect Dis Soc. 2022; 22. Adams O, Weis J, Jasinska K, Vogel M, Tenenbaum T. Comparison of 11(2):69-72. doi:10.1093/jpids/piab121 human metapneumovirus, respiratory syncytial virus and rhinovirus 11. Moriyama M, Hugentobler WJ, Iwasaki A. Seasonality of respiratory respiratory tract infections in young children admitted to hospital. viral infections. Annu Rev Virol. 2020;7(1):83-101. doi:10.1146/ J Med Virol. 2015;87(2):275-280. doi:10.1002/jmv.24025 annurev-virology-012420-022445 23. Heimdal I, Moe N, Krokstad S, et al. Human coronavirus in hospital- 12. Wu A, Mihaylova VT, Landry ML, Foxman EF. Interference between ized children with respiratory tract infections: a 9-year population- rhinovirus and influenza A virus: a clinical data analysis and based study from Norway. J Infect Dis. 2019;219(8):1198-1206. doi: experimental infection study. Lancet Microbe. 2020;1(6):e254-e262. 10.1093/infdis/jiy646 doi:10.1016/S2666-5247(20)30114-2 24. Petrarca L, Nenna R, Frassanito A, et al. Human bocavirus in children 13. Tuladhar E, Hazeleger WC, Koopmans M, Zwietering MH, Duizer E, hospitalized for acute respiratory tract infection in Rome. World J Beumer RR. Reducing viral contamination from finger pads: hand- Pediatr. 2020;16(3):293-298. doi:10.1007/s12519-019-00324-5 washing is more effective than alcohol-based hand disinfectants. 25. Government of Catalonia, Department of Health. La Salut del 2022. J Hosp Infect. 2015;90(3):226-234. doi:10.1016/j.jhin.2015.02.019 Pressupost 2022. Available from: govern.cat/govern/ 14. Leung NHL, Chu DKW, Shiu EYC, et al. Respiratory virus shedding in docs/2021/11/12/17/58/bd6c19a3-ca28-41db-910b- exhaled breath and efficacy of face masks. Nat Med. 2020;26(5): baf36a3b4044.pdf [Accessed 11 July 2022]. 676-680. doi:10.1038/s41591-020-0843-2 26. Feldstein LR, Ogokeh C, Rha B, et al. Vaccine effectiveness against 15. Troeger CE, Khalil IA, Blacker BF, et al. Quantifying risks and influenza hospitalization among children in the United States, interventions that have affected the burden of lower respiratory 2015-2016. J Pediatric Infect Dis Soc. 2021;10(2):75-82. doi:10. infections among children younger than 5 years: an analysis for the 1093/jpids/piaa017 Global Burden of Disease Study 2017. Lancet Infect Dis. 2020;20(1): 27. Hammitt LL, Dagan R, Yuan Y, et al. Nirsevimab for prevention of 60-79. doi:10.1016/S1473-3099(19)30410-4 RSV in healthy late-preterm and term infants. N Engl J Med. 2022; 386(9):837-846. doi:10.1056/NEJMoa2110275 16. Wang X, Li Y, O’Brien KL, et al. Global burden of respiratory infec- 28. Meskill SD, O’Bryant SC. Respiratory virus co-infection in acute tions associated with seasonal influenza in children under 5 years in respiratory infections in children. Curr Infect Dis Rep. 2020;22(1):3. 2018: a systematic review and modelling study. Lancet Glob Health. doi:10.1007/s11908-020-0711-8 2020;8(4):e497-e510. doi:10.1016/S2214-109X(19)30545-5 29. Gimferrer L, Vila J, Piñana M, et al. Virological surveillance of human 17. Wang X, Li Y, Deloria-Knoll M, et al. Global burden of acute lower respiratory syncytial virus A and B at a tertiary hospital in Catalonia respiratory infection associated with human metapneumovirus in (Spain) during five consecutive seasons (2013-2018). Future children under 5 years in 2018: a systematic review and modelling Microbiol. 2019;14(5):373-381. doi:10.2217/fmb-2018-0261 study. Lancet Glob Health. 2021;9(1):e33-e43. doi:10.1016/S2214- 109X(20)30393-4 18. Li Y, Wang X, Blau DM, et al. Global, regional, and national disease SUPPORTING INF ORMATION burden estimates of acute lower respiratory infections due to respi- Additional supporting information can be found online in the Support- ratory syncytial virus in children younger than 5 years in 2019: a sys- ing Information section at the end of this article. tematic analysis. Lancet. 2022;399(10340):2047-2064. doi:10.1016/ S0140-6736(22)00478-0 19. Pham H, Thompson J, Wurzel D, Duke T. Ten years of severe respi- How to cite this article: Vila J, Lera E, Andrés C, et al. The ratory syncytial virus infections in a tertiary paediatric intensive care unit. J Paediatr Child Health. 2020;56(1):61-67. doi:10.1111/jpc. burden of non-SARS-CoV2 viral lower respiratory tract 14491 infections in hospitalized children in Barcelona (Spain): A long- 20. Kalil AC, Thomas PG. Influenza virus-related critical illness: term, clinical, epidemiologic and economic study. Influenza pathophysiology and epidemiology. Crit Care. 2019;23(1):258. Other Respi Viruses. 2023;17(1):e13085. doi:10.1111/irv. doi:10.1186/s13054-019-2539-x 21. Hani N, Gornick W, Huff B, Atienza J, Singh J. Human metapneumo- 13085 virus infection in a children’s hospital—it should get more attention.