INTRODUCTION
Respiratory viral infections are community-wide concerns that extend beyond individual patient suffering, imposing considerable social costs (1, 2). Interest in respiratory viral infections has increased owing to the emergence of novel subtypes of influenza or new viruses such as severe acute respiratory syndrome Coronavirus-2 (SARS-CoV-2) (3, 4). Accordingly, the incidence of respiratory viral infections has significantly increased, leading to the need for monitoring systems for respiratory viral pathogens in the community because of the nature of pathogens associated with high morbidity. Respiratory infections that can cause fatal complications in infants and elderly populations often require laboratory testing and analysis because the causative pathogen is clinically difficult to identify.
In particular, eight types of respiratory viruses periodically cause infections in the community: respiratory syncytial virus (RSV), human coronavirus (hCoV), influenza virus (IFV), human metapneumovirus (hMPV), parainfluenza virus (PIV), human bocavirus (hBoV), adenovirus (ADV), and human rhinovirus (hRV). As these viruses can directly or indirectly cause large-scale infections within a community, a surveillance system is used in several regions, including Gwangju, which periodically checks the detection rate of the virus (5, 6, 7, 8). Several studies have investigated the characteristics of individual viruses. The prevalence of viruses can vary due to environmental factors such as temperature and humidity (9, 10, 11, 12).
In this study, we aimed to analyze the patterns of respiratory viruses detected in Gwangju over the past 10 years. The specimens used in the analysis were collected via a monitoring system operated by the Korea Centers for Disease Control and Prevention from 2014 to 2023, covering periods of the global prevalence of respiratory viruses, MERS-CoV, and SARS-CoV2, during which the impact of these new pathogens on existing viruses in the community was investigated (13, 14). We also aimed to analyze the genotypes of seasonal viruses, such as influenza and RSV, to determine whether each virus had its own epidemic pattern.
MATERIALS AND METHODS
Samples were collected from pediatric and otolaryngology patients with clinical symptoms in Gwangju from 2014 to 2023. Samples from throat examinations or nasal discharges of the patients were collected from medical institutions participating in the Korea Influenza and Respiratory Viruses Surveillance System (KINRESS) organized by the Korea Centers for Disease Control and Prevention (KDCA) and were transported in a virus transport medium. Nucleic acids were extracted from 140 µl of collected samples using a QIAamp RNA kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Commercial respiratory virus detection real-time reverse transcription polymerase chain reaction (PCR) kits (RSV A&B/hMPV, hCoV-229E, -NL63, -OC43, IFV A&B, PIV-1/2/3, hRV, Pandemic H1N1/H3N2, and IFV-B/Victoria & Yamagata) and real-time PCR kits (hBoV and ADV) (Kogenebiotech, Seoul, Korea) were used to detect target genes. The PCR amplification conditions were as follows: 50 °C for 30 min for reverse transcription, followed by 95 °C for 10 min for inactivation of reverse transcriptase and 40 cycles of 95 °C for 15 s and 60 °C for 1 min.
RESULTS
Total detection of respiratory viruses over 10 years
In total, 16,573 specimens were collected over the 10 years, of which 10,894 (65.73%) tested positive. The detection rates are listed as follows in a decreasing order: hRV 17.82% (2,954/16,573), ADV 11.17% (1,852/16,573), IFV 9.68% (1,605/16,573), PIV 7.11% (1,178/16,573), RSV 6.66% (1,104/16,573), hMPV 5.59% (926/16,573), hCoV 4.36% (722/16,573), and hBoV 3.34% (553/16,573). In 2020 and 2021, the positivity rate fell below 60%. Especially in 2021, hMPV and IFV were not detected (Table 1).
Table 1.
Respiratory virus detection rates in Gwangju from 2014 to 2023
Respiratory syncytial virus
RSV demonstrated typical seasonal peaks in winter, with high detection rates consistently observed from November through February prior to 2020. Notably, between 2014 and 2019, RSV activity was concentrated in the winter months, with little to no detection during the rest of the year. However, after the onset of the COVID-19 pandemic, the seasonal pattern shifted. RSV activity was suppressed during 2020, followed by a marked surge in December 2021, peaking in January 2022. Unlike previous years, detection rates increased again in September and October 2022 and, after a period of low detection during winter, rose once more in the spring of 2023 (Fig. 1A). In our surveillance, RSV subtypes A and B were differentiated using molecular assays targeting subtype-specific genetic markers. Before the COVID-19 pandemic, subgroup B generally exhibited higher and more distinct peaks than subgroup A particularly in 2014, 2016, 2018, and early 2021, suggesting that subgroup B was often predominant. Subgroup A peaks were smaller and increased every 2-3 years, with elevated detection in 2015, 2017, and 2019. After the onset of the pandemic, and the previous pattern of alternating predominance between A and B was no longer evident (Fig. 1B).
Human coronavirus
The detection rate of hCoV at the baseline was higher than that of RSV. Similar to RSV, the annual hCoV detection rates before the 2020 COVID-19 pandemic showed a pattern that peaked between November and January. The detection rate of hCoV was not as high as that of RSV (32% RSV and 15% hCoV before the COVID-19 pandemic). In contrast to RSV, which showed a different detection rate pattern after COVID-19, no significant seasonal variation was noted in the detection rates of hCoV (Fig. 1C). Except in 2015, hCoV-229E and hCoV-NL63 were the most frequently detected viruses in January, the winter season, and the detection pattern of hCoV-OC43 was sporadic. However, during the winter of January-February 2023, at the end of the COVID-19 pandemic, we observed a sharp rise in detection rates, which were suppressed in 2021 and 2022. In particular, the detection rate of hCoV-OC43 increased significantly in 2022 and 2023 compared to that in the pre-COVID-19 pandemic period (Fig. 1D).
Influenza virus
In the case of IFV, the difference in detection rates between the epidemic period and base level was the most prominent among the eight types of viruses. During summer and fall, the detection rate of IFV was low (5%), whereas it was prevalently detected from early winter to early spring (November to May), with a detection rate of up to 66% (February 2014). Contrary to previous trends indicating that IFV is a winter virus, the detection rate of IFV remained low from March 2020 until the winter of 2023 and then began to rise in late autumn of that year, in October. However, the highest detection rate confirmed in the winter of 2023 was less than that observed in the previous winter (Fig. 2A). From 2015 to 2020, IFV-A became prevalent every winter, followed by the emergence and replacement by IFV-B. However, after the 2020 IFV-A epidemic, the disappearance of IFV during the COVID-19 pandemic continued. Since IFV-A was detected in the summer of 2022 (July) again, detection continued regardless of the season (Fig. 2B). The A/(H1N1)pdm09 and A/H3N2 subtypes showed repeated prevalence patterns from 2014 to 2019. This included a year-long prevalence of the A/(H1N1)pdm09 subtype after two consecutive years of prevalence of the A/H3N2 subtype. However, these patterns were not observed with IFVs detected after the COVID-19 pandemic (Fig. 2C). The detection rates of IFV subtypes are listed as follows in a decreasing order: A/(H1N1)pdm09 47.19% (1,043/2,210), A/H3N2 27.29% (603/2,210), B/Victoria 25.43% (562/2,210), and B/Yamagata 0.09% (2/2,210) (Fig. 2d).
Human metapneumovirus
Prevalent in spring and summer, hMPV tended to exhibit a relatively consistent detection rate from 2014 to 2019. Except for 2017 (when detection rates started to increase in February), the detection rate of hMPV peaked between March and May and was lower than 5% during the cold winter months. However, no spring-summer epidemic was observed during the COVID-19 pandemic, and the seasonality trend did not appear to have recovered, with a sudden increase in detection rates in September 2022 at a time when social distancing was eased in Gwangju and a resurgence in the summer of 2023. The delayed epidemic of hMPV was observed in the early summer of 2023, when the COVID-19 pandemic ended (Fig. 3A).
Parainfluenza virus
PIV detection rates peaked after the hMPV epidemic from 2014 to 2019. Except for the epidemic period, the detection rate of basal levels was higher for PIV than for hMPV, and the detection period was longer for PIV than that for hMPV, from April to September. The detection rate of PIV showed a sudden and sharp increase from September to November 2021 during the COVID-19 pandemic, followed by a decline. Unlike previous years when PIV typically circulated in the spring and summer, no such seasonal pattern was observed in 2022. Instead, PIV maintained a steady detection rate of approximately 10% from November 2022 through 2023 (Fig. 3B). From 2014 to 2019, when PIV epidemics followed previously known patterns, PIV 3, PIV 1, and PIV 2 accounted for 52.41% (326/622), 33.76% (210/622), and 13.83% (86/622) of detections, respectively (Fig. 3C). PIV 3 was predominantly detected in early summer, during a period when hMPV—typically circulating in spring (March to May)—was not observed, highlighting the distinct early-summer seasonality of PIV 3. In contrast, PIV 1 and PIV 2 showed less defined seasonality. However, after the COVID-19 pandemic, no differences in the serotypes and sporadic prevalence of PIV were observed (Fig. 3D).
Human bocavirus
Before the COVID-19 pandemic (2014-2019), hBoV showed the highest detection rate in early summer. However, among the concurrently circulating viruses, it exhibited the slowest increase in detection rate, consistently remaining below 20% during the same period. The detection rate suddenly increased in the winter of 2020, the year the COVID-19 pandemic began, and increased until July 2021, followed by a decrease. After a high detection rate again in June 2022, the rate decreased from September 2022. The annual peak detection rate, which remained in the 10% range before the COVID-19 pandemic, increased in winter after the onset of the COVID-19 pandemic in 2020 to approximately 25% and lasted until the summer of 2021. The detection rate decreased, and the pattern re-emerged in the summer of 2022, with no summer prevalence pattern recorded in 2023 (Fig. 4A).
Adenovirus
With no known seasonal pattern, the detection rate of ADV was maintained every year from an average of 10% to 20% before the COVID-19 pandemic. Despite the quarantine system for the COVID-19 pandemic, including social distancing, ADV had a similar annual detection rate, which remained relatively high in 2023, when the COVID-19 pandemic ended. Considering that the detection rate, which had remained constant, increased so sharply that it approached 50% in September 2023, ADV was widespread throughout the community (Fig. 4B).
Human rhinovirus
Among the eight viruses, the basal-level detection rate of hRV was the highest for ten years. Before the COVID-19 pandemic, a constant detection rate of approximately 20% was maintained throughout the year, except for a slight decline in the winter months of January and February. However, the detection rate of hRV has increased sharply since the early summer of 2020 during the COVID-19 pandemic. hRV, a four-season epidemic virus, was confirmed to be widespread regardless of the season, even during the COVID-19 pandemic (Fig. 4C). In comparison to the IFV prevalence in winter, the lowest detection rate of hRV and the highest detection rate of IFV were simultaneously observed between January and February from 2014 to 2020. However, the COVID-19 pandemic began in 2020, and this trend disappeared; it was not observed during the 2023 influenza epidemic (Fig. 4D).
DISCUSSION
The COVID-19 pandemic brought greater focus to respiratory viral infections because of their wide-ranging social and economic effects, such as increased strain on healthcare, financial challenges, and disruption to society. Additionally, the pandemic altered how respiratory viruses behave in communities, including changes in seasonal trends, lower activity of some viruses, and unpredictable outbreak timings. These findings highlight the importance of improving surveillance, tailoring public health measures, and carefully distributing resources, especially in areas like Gwangju (1, 2).
For example, during the COVID-19 pandemic, some viruses, whose detection rates increased significantly compared to those before the pandemic, were not structurally enveloped, such as hRV and hBoV (15, 16, 17). Our results agreed with this, confirming that the detection rates of hRV and hBoV averaged 29.59% and 9.13% in 2021, whereas, in 2023, these values recovered to the same levels as those before the COVID-19 pandemic, at 16.97% and 1.99%, respectively. Studies that have analyzed the patterns of respiratory viruses prevalent in the community over the long term have provided various perspectives (5, 6, 7, 8).
In this study, we confirmed that 10 years of epidemiology of eight viruses, including the influenza B epidemic that occurred after the influenza A epidemic, which began in early winter in Gwangju. These findings can be observed in countries located geographically in the Western Pacific Region (18, 19, 20, 21). In addition, we confirmed that influenza types A and B were more predominantly detected in 2-year and 1-year cycles, respectively. While influenza viruses were identified every year, influenza A viruses exhibited increased prevalence approximately biennially, whereas influenza B viruses demonstrated a pattern of annual prevalence. Specifically, among the influenza A types, A/H3N2 and A/(H1N1)pdm09 were primarily detected in the 2-year and 1-year cycles, respectively.
Although influenza A viruses were detected every season, their prevalence tended to spike more markedly approximately every two years, indicating a biennial pattern of increased activity. Importantly, this biennial fluctuation does not imply a continuous increase in case numbers, but rather reflects periodic peaks in prevalence amid ongoing seasonal circulation. This pattern was observed only during the pre-COVID-19 pandemic period, and whether it will reemerge in the post-pandemic era remains to be determined.
Subtype analysis showed that was most frequently detected, followed by A/H3N2 and the B/Victoria lineage, while B/Yamagata was rarely observed. This changing dominance among A subtypes and the low circulation of B/Yamagata influence vaccine strain selection and surveillance efforts. The relatively lower detection rate of IFV-B/Victoria lineage and minimal presence of B/Yamagata lineage reflect shifting IFV-B circulation patterns, which are important for optimizing vaccine strain selection and public health surveillance priorities (22). From a clinical and public health perspective, biennial peaks of A/H3N2 have been linked to more severe influenza seasons, possibly due to its antigenic variability and increased likelihood of vaccine mismatch (23). Tracking these cyclic patterns and subtype shifts is therefore critical for optimizing vaccination programs and anticipating epidemic intensity. Moreover, mode, modeling studies have emphasized that vaccine effectiveness depends significantly on antigenic distance between circulating strains and vaccine strains, particularly in older adults who are at higher risk (24).
Understanding these dynamics is essential for continuous surveillance and tailoring vaccination strategies in the evolving post-pandemic respiratory virus landscape. For RSV, three regular 2-year cycle patterns were observed in Gwangju from 2014 to 2018, which differed from observations in Europe in the late 1990s (25). This pattern was not observed after COVID-19; however, the two subgroups were detected together, as the detection rate, which was suppressed during the COVID-19 pandemic (26, 27), increased sharply by the end of 2021. The detection of hCoV, which had been relatively sporadic, has tended to increase rapidly since the onset of the COVID-19 pandemic, especially for hCoV-OC43 and hCoV-NL63. Among them, hCoV-OC43 is hypothesized to be more closely related to SARS-CoV-2, both in terms of genetic similarity and its higher resistance to certain chemicals (28, 29).
Previous studies on respiratory viruses in South Korea over a long period have focused on the comparison before and during the COVID-19 quarantine system (15, 16, 17). However, our study is consequential because it provides findings regarding respiratory viruses before and after the implementation of control measures for COVID-19. However, further studies are needed to examine respiratory virus characteristics that have changed during the COVID-19 pandemic. For example, in the case of hBoV and hRV, whose detection rates have increased, unlike other viruses, during the COVID-19 pandemic, further research is needed to determine whether they show different genetic characteristics from previous epidemics.
However, the limitations of this study include the regional limitations of Gwangju and the fact that samples were collected from small hospitals and clinics. Our study offers valuable insights into the long-term trends of respiratory virus circulation in our region. However, a significant limitation lies in interpreting detection rates, particularly during the COVID-19 pandemic. During this period, diagnostic resources were predominantly allocated towards SARS-CoV-2 testing. Patients presenting with respiratory symptoms were often screened first for COVID-19, and if positive, further testing for other respiratory viruses may have been given lower priority. This likely led to an underestimate of the true prevalence of non-SARS-CoV-2 respiratory viruses. This phenomenon isn't unique to our study; similar trends have been observed worldwide, with a dramatic decrease in the detection of common respiratory viruses like influenza and RSV during intense COVID-19 activity (30, 31).
While non-pharmaceutical interventions (NPIs) certainly played a role, the impact of these altered testing strategies on surveillance data cannot be overlooked. Therefore, when interpreting shifts in respiratory virus patterns observed in our data during the COVID-19 pandemic, the potential influence of diagnostic prioritization is a crucial consideration. The lower detection rates for non-SARS-CoV-2 viruses during this time may not solely reflect a reduction in viral circulation, but also arise from altered testing algorithms and reduced diagnostic throughput for other pathogens. Nevertheless, it is crucial to analyze the patterns of viruses that have been prevalent in a region for a decade to obtain fundamental data on the epidemiology of the virus.
In conclusion, the prevalence of respiratory viruses in the community showed a constant pattern, whereas the existing pattern changed owing to environmental factors, such as the emergence of new viruses. Therefore, we need to continue to examine future respiratory virus patterns, including post-COVID-19, to develop a better system of measures to prevent and control respiratory viral infections.
