Prevalence and seasonality of viral respiratory infections in a temperate climate region: A 24‐year study (1997–2020)

Abstract Background Few long‐term reports have been published on the epidemiology of respiratory viruses despite their frequent involvement in extremely common infections. The aim here was to determine the frequency and distribution of respiratory viruses in a temperate climate area (Barcelona, Spain) throughout a 24‐year period. Methods We collected data on all respiratory viruses detected from 1997 to 2020 in our institution. Clinical specimens were analyzed mainly by conventional techniques, and molecular techniques were also used. Results Of the 59,579 specimens analyzed, 21,382 (35.9%) were positive for at least one virus. The number of positive samples during cold months was significantly higher than in warm months. Respiratory virus infections were detected in patients of all ages, above all in children under 3 years of age, who were most frequently infected with the respiratory syncytial virus, whereas Influenza A virus predominated in the other groups, especially in adults. A clear demographic and seasonal pattern was established for some viruses. Circulation of other respiratory viruses during the FLUAV H1N1pdm09 and SARS‐CoV‐2 pandemics was observed. Conclusions This long‐term study provides new knowledge about the prevalence of respiratory viruses in a Mediterranean region. Throughout the study period, the frequency of some viruses remained constant, whereas others varied with the year. A clear demographic and seasonal pattern was established for some viruses. Patients suffering from severe respiratory infections should be examined for a range of respiratory viruses regardless of gender, age, or season.

is essential to determine the etiological agent causing the infection. 1 The most common respiratory viruses are respiratory syncytial virus (RSV), adenoviruses (AdV), influenza virus types A and B (FLUAV and FLUBV), human metapneumovirus (hMPV), parainfluenza virus types 1-3 (PIV-1, PIV-2, and PIV-3), rhinoviruses (RV), enteroviruses (EV), and coronaviruses (CoV). Classical virological diagnosis of respiratory infections is based on cell culture and/or antigen detection techniques. Molecular techniques have improved the capacity to detect these and other viruses in respiratory tract specimens, especially those that cannot be identified by conventional methods. 2,3 The few long-term studies on respiratory virus epidemiology published to date [4][5][6][7][8][9] show seasonal patterns of circulation in defined geographical areas. However, most reports cover shorter periods. [10][11][12][13][14] The epidemiological studies carried out in Spain are restricted to a few seasons, 15 certain viruses, 16 or particular groups of patients, 17 and to our knowledge, none have focused on the prevalence of several respiratory viruses within a broad population over a long period of time.
The aim of the present study was to determine the frequency and distribution of respiratory viruses in a Mediterranean area throughout a 24-year period, including the first months of the SARS-CoV-2 pandemic.

| Study design and data collection
We collected all the available data on respiratory viruses detected from January 1997 to March 2020 in our institution, a tertiary referral teaching hospital covering an area of 407,550 inhabitants in Barcelona (Spain). We analyzed the circulation of respiratory viruses, including frequency and seasonal distribution, the relationship between the different viruses detected, and two baseline parameters of patients (age and sex). The Ethics Committee of Hospital de la Santa Creu i Sant Pau The clinical specimens analyzed were nasopharyngeal aspirates, nasal and pharyngeal exudates, bronchoalveolar lavages, and lung biopsies. The conventional techniques routinely used to identify respiratory viruses were immunochromatography, immunofluorescence (IF), and cell culture (CC), as described in the literature. 18 The respiratory viruses detected using these techniques were RSV, AdV, FLUAV, FLUBV, PIV1, PIV2, PIV3, RV, and EV; screening for hMPV was included from December 2008. At the start of 2009, an in-house PCR was introduced to detect FLUAV, FLUAV H1N1pdm09, and FLUBV.
Since February 2010, the molecular detection of these viruses has been performed by GeneXpert ® Dx (Cepheid), and since January 2015, this kit has also targeted RSV. A panel for the molecular detection of respiratory viruses (FilmArray™) was introduced in the routine workflow of the laboratory in May 2014, allowing the detection of PIV4 and different coronaviruses (HuCoV-229E, HuCoV-HKU1, HuCoV-NL63, and HuCoV-OC43). At the beginning of 2020, several molecular methods were set up for the detection of SARS-CoV-2 (GeneXpert ® , Simplexa ® , VIASURE ® Real Time PCR Detection Kit, SARS-CoV-2 ELITe MGB ® Kit, cobas ® SARS-CoV-2 Test, and Alinity m SARS-CoV-2 assay).
Specimens received from March 2020 were only studied for the presence of SARS-CoV-2 due to the exceptionality of that year, and they were analyzed separately.

| Statistical analyses
Statistical analysis was performed by means of SPSS v26 (IBM Corp, Armonk, New York, USA). Continuous data and categorical data were analyzed using the t test and Chi-square test, respectively.

| General results
In the 1997-2019 period, 59,579 respiratory specimens were received by the laboratory, with a median of 2,590 per year (range 952-4,883). Of these, 49,712 samples were processed by conventional methods and the other 9,867 only by molecular methods. A total of 21,382 samples (35.9%) were positive for at least one virus.
Overall, 21,939 virus detections were made, with more than one identified in 551 specimens (2.6%). The rate of positivity was significantly higher in samples acquired during the cold versus warm months, matching the seasonal distribution of samples received (p < 0.001).
In 2020, a total of 66,616 respiratory specimens were collected, 2,903 during the three first months of the year, and 832 (28.6%) respiratory virus detections were made. All specimens received after the WHO declaration of the SARS-CoV-2 pandemic were analyzed only for SARS-CoV-2. A total of 6,566 samples were positive for SARS-CoV-2 in 2020.

| Seasonal distribution of respiratory viruses
The different respiratory viruses identified throughout the study period are shown in Table 1. The most frequently detected were FLUAV (31.4%) and RSV (28.7%), which were co-detected in 25.5% PIV3, n (%)  Table 1).
The other respiratory viruses were not detected every year.  Table 1).
The monthly distribution of each respiratory virus and their seasonal patterns during the 23 years prior to the SARS-CoV-2 pandemic are shown in Figures 1 and 2, respectively. The relation between season and prevalence was statistically significant for all viruses (p < 0.001). Among the viruses detected in the cold months, RSV detection was highest from November to January, reaching a maximum in December. The prevalence of FLUAV was highest from December to March, normally reaching a peak in January, whereas for FLUBV, it was from January to March, with a peak usually 1 month after FLUAV. hMPV was most prevalent from February to May, and PIV-2, from September to December, with a peak usually in November.
The other respiratory viruses in this study were detected throughout the year: AdV was most prevalent from November to May; EV and RV usually peaked in autumn and spring; PIV-1, from September to April with a peak in October; and PIV-3, from April to June and October to December.

| Respiratory viruses during the FLUAV H1N1 pdm09 pandemic
The circulation of other respiratory viruses during the FLUAV H1N1

| Correlation between respiratory viruses and the age/sex of patients
The highest number of samples was collected from patients over 60 years (group I) and the lowest from those between 18-29 years (group F). The detection frequency for each respiratory virus varied between the different age groups (Table 3) Regarding the gender of patients, 54.6% (n = 32,587) and 44.8% (n = 26,688) of specimens were from males and females, respectively.
Respiratory viruses were detected in 11,394 samples from males and 9,873 from females. Although more specimens were received from male patients, the percentage of positivity was higher for females (36.9%) than for males (34.9%) (p < 0.001). Differences in the positivity rates between the genders stemmed from the groups of patients more than 18 years old.

| Circulation of other respiratory viruses during the SARS-CoV-2 pandemic
The circulation of other respiratory viruses besides SARS-CoV-2 was

| DISCUSSION
The aim of the present study was to describe the frequency and distribution of respiratory viruses in a specific geographical area with a temperate climate over a 24-year period up to 2020. In the 23 years prior to the SARS-CoV-2 pandemic, the overall positivity rate was 36.8% of all the samples analyzed, which is in accordance with other studies also using conventional methods. 4,5,[10][11][12]19 The main differences with other reports are due to the different viruses evaluated, as not all include RV, EV, or herpes viruses. 4,5,11 As expected, our positivity rates are lower than in studies using molecular techniques, which are more sensitive and can identify a wider range of viruses than conventional methods. 6,13,14,[20][21][22] The phenomenon of respiratory viral interference, in which infection by a particular virus may interfere with the timing or rate of other viral outbreaks, is well documented. 22 The long period of our study The detection frequency of the viruses in our study is similar to that of previous reports. 5,9,24 The differences observed may be explained by the type of population under study, the season in which the study was performed, its length, and/or the methodologies used. 4,10,12,19 Thus, we observed a lower RSV frequency compared to studies focusing only on a pediatric population. 4,9,10,12 The detection frequency of FLUAV and FLUBV also differed from the literature, as although infections caused by these viruses follow a yearly epidemical pattern, their intensity fluctuates. 4,6,7,9,12 Furthermore, some viruses were found in lower frequencies (especially hMPV, RV, EV, and some PIV) than in other studies using molecular methods. 13,[20][21][22] In temperate climates, respiratory viruses are usually expected during the cold months, whereas in tropical climates, they may be found throughout the year or are associated with the rainy season. 4,7,9,12,19,25,26 In the present study, RSV, influenza viruses, and hMPV were mainly detected in winter and early spring and hardly detected in summer, in agreement with other studies performed in temperate climates. 6,12,22,[27][28][29] RSV was the first virus detected at the start of the cold season, displaying a marked seasonal distribution between November and January. FLUAV was found mainly between December and March. Although not detected every year, FLUBV circulated mainly between January and March, usually after the FLUAV epidemics, 30   or every year and vary between different geographic areas. 10,28 In our study, PIV-1 was detected essentially in autumn and nearly every year, whereas PIV-2 was mainly found between September and December.
In to light its role in respiratory infections in adults. 5,8 We observed an increasing tendency in FLUAV detection with patient age, as described in the literature. 5,22,29  Infections caused by PIV-1 and PIV-3 were mainly observed in infants under 6 months, whereas those caused by PIV-2 were predominant not only in these patients but also in children aged 1 to 3 years.
These results are comparable to those of Henrickson, 33 who reported that children suffered more severe PIV-1 and PIV-2 infections in their first year, and PIV-3 caused symptoms similar to those of RSV.
The overall rate of virus detection in the present study was higher in females than in males, in contrast with previously reported data. 4,11,19,34 This difference was observed only in adult female patients, which might be partly due to their having closer contact with young children compared to men. 27 In our study, using antigen detection and viral isolation methods, co-detections were obtained in 2.6% of positive samples. These data are similar to those obtained by other studies based on conventional methods 4,10 and lower than those only using molecular methods, where the frequency of co-detections ranges from 10% to 50%. 14,21,22 The clinical impact of co-detection remains unknown, although a more severe clinical course has been associated with the detection of more than one respiratory virus. 35 In this work, the most frequent co-detection was of RSV and FLUAV, as reported in the literature, 5 whereas other authors have found RV and AdV to be the most common, 21,22 attributing it to their year-round circulation. In our study, cases of co-detection were observed mainly during the winter months, the season associated with were also detected during the first year of the SARS-CoV-2 pandemic. 36 We conclude that patients suffering from severe respiratory infection should be screened for a wide range of respiratory viruses regardless of gender, age, and season.

ACKNOWLEDGMENTS
The authors would like to thank Dr. Adela Retana for her collaboration from the pediatric department. Special thanks to all the staff of the Microbiology Department of the Hospital de la Santa Creu i Sant Pau, who during all these years has collaborated with the processing of samples and data logging.

FUNDING INFORMATION
This work was supported by own funding.

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/irv.12972.

DATA AVAILABILITY STATEMENT
Data sharing not applicable to this article as no datasets were generated or analysed during the current study ORCID Núria Rabella https://orcid.org/0000-0002-3409-0468