SEARCH

SEARCH BY CITATION

Keywords:

  • Epidemic modelling;
  • human rhinovirus;
  • influenza A(H1N1)2009;
  • pandemics;
  • viral interference

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

Clin Microbiol Infect 2010; 16: 326–329

Abstract

In contrast to the experience in other European countries, the onset of the A(H1N1)2009 influenza virus epidemic was unexpectedly slow in France during the first part of autumn 2009. Our objective was to test the hypothesis that intense circulation of rhinoviruses might have reduced the probability of infection by A(H1N1)2009 virus at the beginning of autumn 2009. Systematic analysis for the detection of A(H1N1)2009 (H1N1) and human rhinovirus (HRV) was performed by RT-PCR from week 36 to week 48 on respiratory samples sent to the diagnostic laboratory by the paediatric hospital (= 2121). Retrospective analysis of the obtained data, using 2 × 2 contingency tables with Fisher’s exact test, revealed evidence of an inverse relationship between HRV and H1N1 detection. Between weeks 36 and 48 of 2009, both HRV and H1N1 were detected but in different time frames. HRV dispersed widely during early September, peaking at the end of the month, whereas the H1N1 epidemic began during mid-October and was still active at the end of this survey. During the co-circulation period of these two respiratory viruses (weeks 43–46), HRV detection appeared to reduce the likelihood of H1N1 detection in the same sample (OR = 0.08–0.24 p <0.0001). These results support the hypothesis that HRV infections can reduce the probability of A(H1N1) infection. This viral interference between respiratory viruses could have affected the spread of the H1N1 viruses and delayed the influenza pandemic at the beginning of autumn in France.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

Six months after its emergence in North America, the novel swine-origin influenza virus A(H1N1)2009 had spread worldwide, causing many countries to initiate their pandemic control plans [1]. It is believed that large-scale air transportation of humans with H1N1 infections were responsible for the early reports of outbreaks in Europe (Portugal, Great Britain, Spain) [2]. A combination of atmospheric conditions [3] and the introduced containment procedures may have been responsible for sporadic circulation of the virus in Europe from April to the end of August [4]. However, by September the dry and cold weather, together with the resulting altered social behaviour (increased indoor activity), usually leads to the appearance of human rhinovirus (HRV), seasonal influenza, and respiratory syncytial virus (RSV) infections. Therefore, it was predicted by most experts that the H1N1 virus would quickly reach epidemic levels, spreading throughout Europe, by early October, and that children would play a major role in the dissemination of the virus as observed in the southern hemisphere.

As anticipated the epidemic status of H1N1, based on reports of Influenza Like Illness (ILI), was declared in France during the first week of September (week 36) [5]. However, these reports contrasted with the low incidence of H1N1 infection reported in the community by the Groupe Regional d’Observation de la Grippe (GROG) – an influenza network based on laboratory confirmation of samples provided by volunteer practitioners [6]. From week 36 to week 43 this network was reporting sporadic H1N1 activity but did not report epidemic status of H1N1 until week 44 (mid-October). A similar pattern was observed in Sweden [7]. It has been suggested by Linde et al. [7] that HRV could have been responsible for the apparent increase of ILI reported in early September, and might have delayed onset of the H1N1 pandemic.

The aim of our study was to test the hypothesis that HRV reduced the probability of H1N1 infection in France at the beginning of Autumn 2009. From the beginning of September to the end of October, samples collected from ILI cases visiting the emergency ward of the paediatric hospital were tested by PCR-based methods for the two viruses. Whether or not there was a correlation between HRV and H1N1 incidence was analysed using a statistical approach.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

Clinical specimens and population

In total, 2121 paediatric samples were sent to the laboratory for H1N1 detection between weeks 36 and 48 by the practitioners at the Femme Mère Enfant Hospital.

Among these, 1456 samples (68.6%) were also tested for HRV positivity. The specimens consisted of nasopharyngeal swabs (84.9%), nasopharyngeal aspirates (9.5%), nasal swabs (4.6%), and tracheobronchial aspirates (1%).

Samples were collected in Virocult® transport medium, and immediately sent to the laboratory for analysis. The mean population age was 3.8 years (4.4 SD) with 55.3% being males. Many of the patients had been admitted to the paediatric emergency unit (73.2%) or the intensive care unit (4.6%).

Rhinoviruses and influenza RT-PCR assays

Nasopharyngeal samples were extracted by automatic standard procedures using NucliSens easyMAG instrument (Biomerieux, Marcy L’étoile). The detection of HRV was performed using an in-house real-time RT-PCR method. Primers and probes were derived from sequences located in the 5′ non-coding region of the HRV genome [8].

The detection of H1N1 viruses was performed using two RT-PCR assays. The first RT-PCR used universal primers based on M gene detection for type A influenza virus [9]. The primers used for the M RT-PCR were previously described for a classical RT-PCR and the probe was designed on Primer Express software. Specificity of oligonucleotides was assessed by a local alignment search (blastn; http://blast.ncbi.nlm.nih.gov./Blast.cgi). The second RT-PCR assay was designed for the specific detection of the H1N1 haemagglutinin, and was kindly provided by V. Enouf and S. van der Werf (NIC North of France, Institut Pasteur, Paris). The primers and probe designed for the specific H1 RT-PCR are available upon request (grippe@pasteur.fr).

The RT-PCRs were all performed on the ABI 7500 (Applied Biosystems, Foster City, CA, USA) platform, using SuperScript III Platinum One-Step Quantitative RT-PCR System (Invitrogen, Carlsbad, CA, USA) and optimized with 0.8 μm primer, 0.2μm probe and 0.5μL enzyme mix (primers and probe sequences are available on request). After reverse transcription at 50°C for 15 min and denaturation at 95°C for 2 min, a two-step amplification in 50 cycles was performed at 95°C for 15 s and 60°C for 40 s. Negative controls were included in each experiment.

Statistical analysis

Analysis for a correlation between HRV and H1N1 was performed using 2 × 2 contingency tables with Fisher’s exact test. Odds ratio (OR) and 95% confidence interval (CI) for the likelihood of co-detection were calculated (Table 2). Differences in age according to number of co-detections and detection of HRV were compared using the Kruskal–Wallis test. Significance level was set at p 0.05.

Table 2.   Odds ratio (OR) and 95% confidence interval (CI) for the likelihood of H1N1 detection in HRV positive samples
Outcome of interestFactors p <0.05Odds RatioConfidence interval 95%p-value
H1N1 positive weeks 36–48Rhinovirus positive0.150.09–0.24p <0.0001
H1N1 positive weeks 43–47Rhinovirus positive0.170.10–0.30p <0.0001
H1N1 positive age <1 yearRhinovirus positive0.160.05–0.5p 0.0002
H1N1 positive age 1–5 yearsRhinovirus positive0.150.06–0.33p <0.0001
H1N1 positive age 6–10 yearsRhinovirus positive0.140.05–0.39p <0.0001
H1N1 positive age >10 yearsRhinovirus positive0.120.02–0.97p 0.014

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

Virus detection between weeks 36 and 48

At least one virus each was detected in 925 specimens (43.6%). Relative frequencies of H1N1 and HRV were, respectively, 28.9% (525/1815) and 24% (415/1731); 15 co-infections were observed (0.7%). However, during the study period, HRV and H1N1 had distinctly different distributions (Fig. 1). An epidemic due to HRV began at week 37 (relative frequency 20.6%). It peaked at week 40 (relative frequency 36.8%) and then gradually diminished to a relative frequency of 4.6% by week 45. In contrast, the epidemic due to H1N1 began later, at week 43, with a relative frequency of 18.8%, and it was still active at week 48, when the HRV epidemic had subsided.

image

Figure 1.  Absolute numbers of H1N1 and rhinoviruses isolated in samples provided by Hôpital Femme Mère Enfant units, each week.

Download figure to PowerPoint

From week 43 to week 47 the HRV and H1N1 viruses co-circulated in this French population, with, respectively, 17.7% and 42.9% relative frequency. Interestingly, over all the study period there was a significant difference (p <0.0001) in age amongst the patients positive for HRV (mean of 2.4 years, SD 3.4) and H1N1 (mean of 5.6 years, SD 4.3) (Table 1).

Table 1.   Absolute numbers of cases of H1N1 and HRV detected in various age groups
 Age <1 yearAge 1–5 yearsAge 6–10 yearsAge >10 years
Cases of H1N1detected7320816678
Cases of HRV detected1601973919
Cases of co-infections detected3651

Also, the HRV detected during the studied period were from different sub-types (data not shown; I. Schuffenecker, National Enteroviruses Centre, Lyon, France).

Virus co-detection

Overall, 15 samples were positive for both H1N1 and HRV. This represented 4.6% and 25.9% H1N1 positive relative frequency, considering the HRV-positive and -negative specimens, respectively. The odds ratio was estimated to be 0.14 (0.08; 0.24 IC95) p <0.0001.

During the co-circulation period from weeks 43–47 and among the different age groups a higher proportion of H1N1 detections was observed in the HRV negative samples. As reflected by an odds ratio <1, HRV detection resulted in a reduced likelihood of co-detecting H1N1 (Table 2).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

Our results suggest that the ILI activity reported in France during early autumn 2009 was initially caused by HRV and then from mid-October by H1N1. Moreover, in any single respiratory sample analysed, detection of HRV resulted in reduced likelihood of detection of the H1N1 virus. Thus, the results showed a significant inverse relationship between HRV and H1N1 virus irrespective of the time period and age group analysed. Therefore, our data support a previous suggestion that the presence of HRV reduces the risk of infection by the H1N1 virus and thus, indirectly, the spread of the virus. The fact that several early studies had reported the absence of co-epidemic respiratory viruses [6,10,11] may explain why interference with influenza epidemicity has only recently been observed.

In contrast to previous studies, our samples were derived primarily from the paediatric emergency department and thus are not representative of the situation in adult populations. Nevertheless, our analysis of a paediatric cohort of patients infected with HRV and/or H1N1 supports and extends the findings of Linde et al. [7] who postulated that rhinovirus epidemics that occur after commencement of the school year may interfere with the spread of influenza during a period with a warm and humid climate that decreases spread of influenza by aerosol. Our observations are also consistent with the findings of Greer et al. [10] who identified similar negative associations between detection of HRV and co-detection of different respiratory viruses, including influenza A viruses.

Some known biological features of these viruses could provide insights into the possible mechanisms involved in these virus interactions. It is known that these viruses do not bind the same receptors in the cells of the respiratory tract, which correspond to human intercellular adhesion molecule-1 (ICAM-1) for HRV and sialic acid alpha 2–6 for human influenza viruses [12,13]. Interference could possibly result from the non-specific innate immune response. Indeed, it has been reported that the interferon response induces a refractory state to virus infection of neighbouring cells. Therefore, during HRV infection and shedding, a refractory period may develop when superinfection of respiratory cells by other respiratory viruses is inhibited [14].

The duration of this refractory period remains to be determined. Based on our data it appears that it could persist throughout the period when it remains possible to detect the interfering virus, i.e. several weeks [15]. Additional epidemiological investigations with multiple serial sampling will be required to extend our understanding of this phenomenon.

These data are important firstly because they may reflect the impact of such viral interference on prediction models of influenza virus dispersal. Such prediction models are a key feature for rational decision making in disease control. Secondly, it has been suggested that co-infection could trigger multiple pandemic waves [16]. Thirdly, current preliminary evidence [10] supports the notion that HRV may provide a temporary window of protection against influenza virus infection.

Further investigations are now in hand to elucidate other aspects of such multiple virus interactions such as the impact of H1N1 infection on children under 2 years of age when there is a high risk of infection by RSV. Finally, one can ask the question, does pandemic influenza virus limit the spread of seasonal influenza viruses, as has been reported in the southern hemisphere [17]?

Transparency Declaration

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References

No extra funding was used. There is no commercial relationship or any potential conflict of interest of any nature.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Transparency Declaration
  9. References
  • 1
    Dawood FS, Jain S, Finelli L et al. Emergence of a novel swine-origin influenza A (H1N1) virus in humans. N Engl J Med 2009; 360: 26052615.
  • 2
    Khan K, Arino J, Hu W et al. Spread of a novel influenza A (H1N1) virus via global airline transportation. N Engl J Med 2009; 361: 212214.
  • 3
    Lowen AC, Mubareka S, Steel J, Palese P. Influenza virus transmission is dependent on relative humidity and temperature. PLoS Pathog 2007; 3: 14701476.
  • 4
    European Centre for Disease Prevention and Control (ECDC). [Weekly Influenza Surveillance Overview 48/2009]. [Accessed 8 October 2009]. English. http://ecdc.europa.eu/en/activities/surveillance/EISN/Newsletter/091204_EISN_Weekly_Influenza_Surveillance_Overview.pdf
  • 5
    Réseau Sentinelles, INSERM, UPMC [Sentiweb-hebdo 49/2009]. French. http://websenti.u707.jussieu.fr/sentiweb/b3e/index.php?zub=36&option=colendrier&annee=2009
  • 6
    Lina B, Valette M, Foray S et al. Surveillance of community-acquired viral infections due to respiratory viruses in Rhone-Alpes (France) during winter 1994 to 1995. J Clin Microbiol 1996; 34: 30073011.
  • 7
    Linde A, Rotzén-Östlund M, Zweygberg-Wirgart B, Rubinova S, Brytting M. Does viral interference affect spread of influenza? Euro Surveill 2009; 14(40): pii: 19354.
  • 8
    Andeweg AC, Bestebroer TM, Huybreghs M, Kimman TG, De Jong JC. Improved detection of rhinoviruses in clinical samples by using a newly developed nested reverse transcription-PCR assay. J Clin Microbiol 1999; 37: 524530.
  • 9
    Bouscambert Ducamp M, Casalegno JS, Gillet Y et al. Pandemic A(H1N1)2009 influenza virus detection by real time RT-PCR: is viral quantification useful? Clin Microbiol Infect 2010; 16: 317321.
  • 10
    Greer RM, McErlean P, Arden KE et al. Do rhinoviruses reduce the probability of viral co-detection during acute respiratory tract infections? J Clin Virol 2009; 45: 1015.
  • 11
    Casalegno JS, Bouscambert-Duchamp M, Morfin F, Lina B, Escuret V. Rhinoviruses, A(H1N1)v, RVS: the race for hivernal pandemics, France 2009–2010. Euro Surveill 2009; 14(44): pii: 19390.
  • 12
    Smith EC, Popa A, Chang A, Masante C, Dutch RE. Viral entry mechanisms: the increasing diversity of paramyxovirus entry. FEBS J 2009; 276: 72177227.
  • 13
    Nicholls JM, Chan RW, Russell RJ, Air GM, Peiris JS. Evolving complexities of influenza virus and its receptors. Trends Microbiol 2008; 16: 149157.
  • 14
    Haller O, Kochs G, Weber F. The interferon response circuit: induction and suppression by pathogenic viruses. Virology 2006; 344: 119130.
  • 15
    Jartti T, Lehtinen P, Vuorinen T, Koskenvuo M, Ruuskanen O. Persistence of rhinovirus and enterovirus RNA after acute respiratory illness in children. J Med Virol 2004; 72: 695699.
  • 16
    Merler S, Poletti P, Ajelli M, Caprile B, Manfredi P. Coinfection can trigger multiple pandemic waves. J Theor Biol 2008; 254: 499507.
  • 17
    Depoortere E, Mantero J, Lenglet A, Kreidl P, Coulombier D. Influenza A(H1N1)v in the southern hemisphere--lessons to learn for Europe? Euro Surveill 2009; 14(24): pii: 19246.