key: cord-0918641-igeyiw0t
authors: Gerna, G.; Piralla, A.; Rovida, F.; Rognoni, V.; Marchi, A.; Locatelli, F.; Meloni, F.
title: Correlation of rhinovirus load in the respiratory tract and clinical symptoms in hospitalized immunocompetent and immunocompromised patients
date: 2009-06-23
journal: J Med Virol
DOI: 10.1002/jmv.21548
sha: 9e67dd8c89f5ac641dc2dacf45d7d80da9e36ae8
doc_id: 918641
cord_uid: igeyiw0t

While human rhinoviruses (HRVs) are well accepted as a major cause of common cold syndromes (rhinitis), their role in the etiology of lower respiratory tract infections is still controversial, and their detection in asymptomatic patients is relatively common. The HRV pathogenic role in four groups of hospitalized patients (pediatric immunocompetent and immunocompromised patients, and adult immunocompetent and immunocompromised patients) was investigated by quantifying HRV load in nasopharyngeal aspirates or bronchoalveolar lavage samples by real‐time reverse transcription PCR (RT‐PCR). Real‐time RT‐PCR was performed in duplicate on all respiratory samples resulting positive by qualitative RT‐PCR. In addition, molecular typing allowed detection of all known HRV species (A, B, and C). In immunocompetent pediatric patients HRVs were mostly associated with lower respiratory tract infections (in the absence of other viral agents) and wheezing, when viral load was ≥10(6) RNA copies/ml. In young immunocompromised patients (stem cell transplantation recipients), an inverse correlation between HRV persistence over time and time at which the infection occurred after transplantation was observed, whereas in adult immunocompromised patients (lung transplant recipients) HRVs could be detected at a medium–low level (<10(5) RNA copies/ml) in bronchoalveolar lavage samples taken routinely from asymptomatic patients. In conclusion, when detected at high viral load, HRVs may cause severe upper and lower respiratory tract infections, whereas when detected at a medium–low viral load, an event more frequent in immunocompromised subjects, they may represent only bystander viruses. J. Med. Virol. 81:1498–1507, 2009. © 2009 Wiley‐Liss, Inc.

Following their identification in 1956, human rhinoviruses (HRVs) were mostly considered as the responsible agent of "common cold" syndromes and, thus, having a minor impact on human pathology . However, more recently, their pathogenic role has been repeatedly claimed in acute sinusitis of adults , acute otitis media of children [Pitkaranta et al., 1998] , and asthma exacerbations of children, adolescents, and adults [Nicholson et al., 1993; Johnston et al., 1995] . Furthermore, the pathogenic role of HRVs in lower respiratory tract infections of infants, elderly and immunocompromised patients has been reported. In more detail, 12.5% of viral pneumonia cases in neonates were attributed to HRVs [Abzug et al., 1990] , as well as 24% of community-acquired pneumonia cases in hospitalized children [Juven et al., 2000] , often in association with concurrent bacterial infections. In addition, HRVs are considered the second most common cause of bronchiolitis after RSV, often in association with RSV [Ong et al., 2001; Papadopoulos et al., 2002] . Furthermore, HRVs have been reported to be associated with hospitalization of children with a history of wheezing/ asthma [Miller et al., 2007] . In a prospective study covering a 3-year follow-up, first year wheezing caused by HRV was the strongest predictor of subsequent 3rd year wheezing [Lemanske et al., 2005] .

In adults, the impact of HRV infections on the lower respiratory tract in asthma, chronic bronchitis, and cystic fibrosis exacerbations [Nicholson et al., 1993; Collinson et al., 1996; Seemungal et al., 2001] has been reported along with the frequent association of HRV with lower respiratory tract infections in immunocompromised hosts and lung transplant recipients Kaiser et al., 2006 ]. In addition, HRVs have been reported to be present in respiratory secretions of asymptomatic individuals or patients [Van Kraaij et al., 2005] .

Recently, HRVs, which were classified previously into two genetic groups (HRV-A including 76 serotypes and HRV-B including 25 serotypes), have been extended to include a third group (HRV-C) consisting of many previously unrecognized non-cultivable HRVs [Lau et al., 2007; Lee et al., 2007] .

In the present study, the relationship was investigated between HRV load in respiratory secretions and occurrence of clinical symptoms in the respiratory tract in young immunocompetent and immunocompromised subjects as well as in adult immunocompetent and immunocompromised patients admitted to hospital.

Patients were admitted to the Fondazione IRCCS Policlinico San Matteo, University Hospital, and respiratory specimens were sent to the Viral Diagnostic Service for routine viral diagnosis. Patients examined (n ¼ 602) were classified into four groups: (i) pediatric immunocompetent patients admitted to hospital with a diagnosis of acute respiratory tract infection, mostly interesting the lower respiratory tract, and associated with wheezing (n ¼ 290); (ii) children and young adults with oncohematologic disorders and an acute respiratory tract infection (n ¼ 68), some (n ¼ 35) who had received a hematopoietic stem cell transplantation; (iii) adult immunocompetent patients (n ¼ 95) admitted to hospital with different pathologies and an acute respiratory tract infection; and (iv) adult immunocompromised patients (n ¼ 149) including adult stem cell transplantation recipients (n ¼ 25) with an acute respiratory tract infection, and lung transplant recipients examined either because of an acute respiratory tract infection or following a routine bronchoscopy performed to monitor the occurrence of chronic rejection (n ¼ 67).

Respiratory specimens, either nasopharyngeal aspirates or bronchoalveolar lavage samples, were collected from November 1, 2006 to May 31, 2007. Samples were collected according to a standardized procedure both at admission and discharge for most patients, and in some cases also at one or more follow-up medical visits. Respiratory samples positive for respiratory viruses at admission were included even if negative at subsequent visits during follow-up. The period June-October 2007 was not included in this study.

Nasopharyngeal aspirates and bronchoalveolar lavage samples were aliquoted and handled as previously reported [Rovida et al., 2005; Sarasini et al., 2006 ]. Respiratory specimens were tested for influenzaviruses A and B (Flu A and B), human parainfluenzaviruses (hPIV) 1-4, human respiratory syncytial virus (RSV), human adenoviruses (hAdV), human coronaviruses (hCoV) 229E and HKU1, and human metapneumoviruses (hMPV) by monoclonal antibodies (MAbs) and direct fluorescent antibody (DFA) staining. MAbs to Flu A and B, hPIV 1-4, RSV and hAdV were obtained from the same source (Chemicon International, Inc., Temecula, CA), while MAbs to hMPV Gerna et al., 2006b] , hCoV 229E [Gerna et al., 2006a] and HKU1 were developed in the laboratory.

Specimens positive by DFA were quantified by realtime reverse-transcription PCR (RT-PCR) for hMPV A and B [Kuypers et al., 2005] , and RSV A and B [Perkins et al., 2005] . In addition, specimens were tested by realtime PCR for human bocavirus [hBoV, Lu et al., 2006] . Respiratory samples were also tested on a qualitative basis for HRV by the method of Steininger et al. [2001] , and for hCoV by the method of Mo€ es et al. [2005] . Specimens positive for HRV were quantified by real-time RT-PCR [Deffernez et al., 2004] . Finally, the presence of HRV-C strains was verified in the group of young oncohematologic patients by sequencing and phylogenetic analysis using the method of Lee et al. [2007] . This approach was limited to this group of patients, in order to verify whether different types of HRV infected these patients during follow-up.

RNA was extracted from 200 ml nasopharyngeal aspirates or bronchoalveolar lavage by the automated extraction kit Nuclisens Ò easy MAGÔ (BioM erieux, Lyon, France) and eluted in 50 ml elution buffer. Reverse transcription (RT) of viral RNA was performed with SuperScriptÔ II RNase H-Reverse Transcriptase (Invitrogen, Carlsbad, CA) in a reaction mixture containing 5 ml extracted RNA (corresponding to 20 ml nasopharyngeal aspirates or bronchoalveolar lavage), 10 ml 5Â buffer (250 mM Tris-HCl [pH 8,3], 375 mM KCl, 15 mM MgCl 2 ), 5 ml dithiothreitol (0.1 M), 1 ml Super-Script II RNase H-Reverse Transcriptase, 5 ml deoxynucleoside triphosphate (10 mM), 2.5 ml random primers p(dN) 6 , and 21.5 ml nuclease-free water in a final volume of 50 ml. The RT thermal profile was: 10 min at 25 C, 50 min at 42 C, and 15 min at 70 C. Real-time PCR was performed on 10 ml cDNA (corresponding to 4 ml nasopharyngeal aspirates or bronchoalveolar lavage), according to Deffernez et al. [2004] , in a final volume of 50 ml reaction mixture containing 25 ml TaqMan Ò Universal PCR Master Mix, no AmpErase Ò UNG (Applied Biosystems International, Foster City, CA), 900 mM forward, and 600 mM reverse primer, and 200 mM of each of the two probes. Primer and probe sequences were as follows: forward primer, 5 0 -GCACTTCTGTTTCCCC-3 0 ; reverse primer, 5 0 -GGCAGCCACGCAGGCT-3 0 ; probe 1, FAM-5 0 -AGCCTCATCTGCCAGGTCTA-3 0 TAMRA; probe 2, VIC-5 0 -AGCCTCATCGACCAAACTA-3 0 TAMRA. Thus, probe 1 (for HRV-A) was labeled at the 5 0 end with the reporter dye 6-carboxyfluorescein (FAM) and at the 3 0 end with the quencher dye 6-carboxytetramethylrodamine (TAMRA), and probe 2 (for HRV-B) was labeled at the 5 0 end with the reporter dye VIC (Applied Biosystems, PE Europe, Basel, Switzerland), and at the 3 0 end with the same quencher dye TAMRA. DNA amplification was performed in 96-well microplates using the ABI PRISM Ò 7000 Sequence Detection System (Applied Biosystems) according to the following thermal profile: 2 min at 50 C, 10 min at 95 C, followed by 45 cycles of 15 sec at 95 C and 60 sec at 55 C. Each real-time RT-PCR test run included the amplification of two DNA scales (one per each probe) consisting of two linearized plasmids obtained by cloning two reference HRV serotypes (serotype 13 for probe 1, and serotype 3 for probe 2) with the TA cloning kit (Invitrogen). External standards were amplified from 10 6 to 10 1 DNA copies, allowing the construction of two standard curves, from which amounts of viral load present in clinical samples were extrapolated. Results were then multiplied by a factor of 250 (referring to 1.0 ml nasopharyngeal aspirates or bronchoalveolar lavage). Thus, negative results were indicated as <250 RNA copies/ml. The two curves were adopted in order to detect both HRV-A and HRV-B groups [Deffernez et al., 2004] . RT-PCR methods for HRV detection by Steininger et al. [2001] , as well as the RT-PCR for HRV molecular typing (based on phylogenetic analysis of a 260-bp variable sequence in the 5 0 non-coding region of HRV and performed on the group of young oncohematologic patients only) according to Lee et al. [2007] were performed as reported. In experiment replicas, mean intraassay and interassay coefficients of variability were both less than 10% for the PCR assay (plasmid scales).

Comparison between medians was performed with the Wilcoxon test for paired data, while the Mann-Whitney U-test was used for unpaired data. The correlation between onset of HRV infection after transplantation and duration of HRV infection was investigated by means of a nonlinear curve fitting analysis.

During the 2006-2007 winter-spring season, 602 patients admitted to hospital were examined for respiratory viruses in nasopharyngeal aspirates or bronchoalveolar lavage samples. On the whole, 348 (57.8%) were found to carry one or more respiratory viruses, whereas 254 (42.2%) patients had virus-free respiratory samples collected during the hospital stay or during medical control visits in the out-patient unit. The total number of respiratory samples examined was 967. Of these, 519 (53.7%) were positive for one or more respiratory viruses, whereas 448 (46.3%) were negative. Since a fair number of patients were followed in the outpatient unit for a sustained period of time after hospital discharge, episodes of respiratory virus infection were considered different from one another for the same patient, when the virus responsible for one episode disappeared from respiratory secretions for at least 2 weeks (i.e., upon two subsequent examinations), while a subsequent episode was considered to be caused by a different virus. On this basis, 330 were single infection episodes, and 62 were co-infections by two or three respiratory viruses, resulting in a total number of 392 respiratory infection episodes.

The two viruses causing by far the largest number of episodes of single infection and co-infection were HRVs and RSV (Fig. 1B ,C). In detail, HRVs were associated with 107 (27.3%) episodes of single infection and 41 (10.5%) episodes of co-infection (Table I ). In parallel, RSV was associated with 120 (30.6%) episodes of single infection and 34 (8.7%) episodes of co-infection (Table I) .

On the whole, HRVs and RSV were associated with 148/ 392 (37.8%) and 154/392 (39.3%) episodes of respiratory infection, respectively (Table I) . When considering the 18 episodes of HRV þ RSV co-infection, the total number of respiratory episodes involving HRV and/or RSV was 284/392 (72.4%).

The absolute numbers of patients and respiratory samples examined and the numbers of patients and samples positive for respiratory viruses are shown in Figure 1A ,D, respectively, according to age (pediatric and adult patients) and immune competence.

As for the distribution of the two most represented infections, 188/284 (66.2%) episodes of HRV and RSV infection or co-infection interested immunocompetent infants and young children (15 cases of double infection), while 41 (14.4%) affected young immunocompromised patients (3 cases of double infection), and 55 (19.4%) adult (n ¼ 13 immunocompetent, and n ¼ 42 immunocompromised) patients (Table I) . However, while the absolute number of HRV and RSV infections was significantly higher in young immunocompetent patients compared to young immunocompromised and adult patients, the percentage of HRV infections was comparable in young and adult immunocompetent and immunocompromised patients. On the contrary, the percentage of RSV infections was significantly higher in pediatric immunocompetent patients compared to other groups. In addition, the ratio of HRV/RSV infections was 0.6 (predominant RSV infections) in pediatric immunocompetent patients, whereas it was 1.7 in young immunocompromised patients. In both groups of adults, the ratio was greater than 2.0 in immunocompromised, and greater than 3.0 in immunocompetent patients (Table I) .

The most interesting features of the relationship observed in the four groups of patients between HRV load and clinical symptoms are reported as follows.

On the whole, 40 patients were selected from this group due to the availability of follow-up data and clinical findings. Out of 40 infants and young immunocompetent children with an episode of acute respiratory tract infection undergoing follow-up, 26 were admitted to hospital within a median time of 3 (range 1-5) days after onset of symptoms and examined within 24-48 hr. Of these, 15 showed the presence of an HRV strain only, and 11 had a co-infection by another respiratory virus associated with HRV. The remaining 14 young patients were examined late (10-15 days) with respect to onset of the respiratory episode or during chronic respiratory infection episodes.

Of the 15 patients positive for HRV only 9 showed a fall in viral load from admission through discharge in close association with improvement of clinical symptoms (Table II) . In three patients (#5, #26, and #27), a high viral load was detected in the single sample taken at admission concomitantly with acute respiratory symptoms. However, in three patients (#9, #20, and #22), viral load increased or remained stable from admission to discharge in contrast with improvement of clinical symptoms (Table II) .

In the group of 11 patients with co-infection (Table IV) , HRV was present at a different titer in nasopharyngeal aspirates in association with another respiratory virus (hPIV-3 in 1 patient, RSV-A in 7 patients, and RSV-B in 3 patients). In all patients, the HRV-associated virus was detected at a titer greater than 1.0 Â 10 5 RNA copies/ml nasopharyngeal aspirate at the time of admission to hospital, and a variably lower titer at the time of discharge. As for HRV, the virus was detected at a titer greater than 1.0 Â 10 5 copies/ml nasopharyngeal aspirate in 5/11 patients (#1, #6, #7, #8, and #17), and in four of these patients the HRV titer decreased at discharge with respect to patient admission. However, in the fifth (#6) patient the HRV titer rose from 10 6 to 10 7 , and in two additional patients (#11 and #13) the titer rose from 10 2 to 10 4 during the hospital stay. All 11 patients were clinically affected by bronchiolitis or pneumonia, often in association with wheezing at the time of admission, while at the time of hospital discharge all patients showed clinical improvement including the three patients showing a rise in HRV titer during the hospital stay (Table III) .

As shown in Table IV , in patients with single HRV infection, viral load was significantly higher (P ¼ 0.03) at time of admission to hospital (5.1 Â 10 6 RNA copies/ml) with respect to time of dismission (7.1 Â 10 4 RNA copies/ ml). In patients with a co-infection, the HRV load was comparable (P ¼ ns) between admission (1.3 Â 10 4 copies/ml) and discharge (1.8 Â 10 4 copies/ml), whereas the co-infecting virus showed a viral load in nasopharyngeal aspirates at admission (1.4 Â 10 7 copies/ml) significantly higher (P ¼ 0.015) than that found at discharge (1.0 Â 10 4 copies/ml). On the other hand, at admission, HRV load in patients with a single infection (5.1 Â 10 6 copies/ml) was significantly higher (P ¼ 0.002) than that of patients with a co-infection (1.3 Â 10 4 copies/ml), while HRV viral load at discharge was comparable (P ¼ 0.45) in patients with single infection (7.1 Â 10 4 copies/ml) and with co-infection (1.8 Â 10 4 copies/ml). Co-infection did not entail a greater duration in the hospital stay (8.8 days) as compared to single HRV infection (8.5 days).

Finally, in 14 additional young patients with delayed admission to hospital with respect to onset of symptoms (not reported), HRV detection at a titer consistently lower than 1.0 Â 10 5 RNA copies/ml nasopharyngeal aspirate at the time of admission to hospital did not provide any useful result in view of the definition of the current respiratory tract infection.

Of 28 episodes of symptomatic HRV respiratory infection (21 single infections and 7 co-infections) concerning as many young oncohematologic patients, nine (with single infection) were followed for at least 6 months. As shown in Figure 2 , an inverse correlation (described by an exponential decay model) was observed between time (days) from transplantation to onset of HRV infection and duration of HRV infection (adjusted R 2 ¼ 0.95). Patients experiencing HRV infection within 60 days from the allograft showed a high viral load in the last nasopharyngeal aspirate sample tested, thus suggesting sustained viral excretion. In patients experiencing HRV infection >80 days after transplantation, HRV disappeared from nasopharyngeal aspirates within 1-2 weeks after onset of infection. However, in these patients, the peak viral load measured during the infection was not different in magnitude from viral load peaks observed in patients undergoing HRV infection in the early posttransplantation period. Three patients of this group were infected by HRV-C strains. One of these patients, who had received two Tcell depleted allografts for relapsed leukemia from the HLA-haploidentical mother, died 6 months after transplantation due to respiratory failure (hemorrhagic alveolitis). The patient had been infected by two different HRV-C strains, one detected in nasopharyngeal secretions upon transplant, and the other detected in bronchoalveolar lavage at the time of death.

As observed in immunocompetent pediatric patients, immunocompetent patients from the adult population, who exhibited levels of HRV load greater than 1 Â 10 5 RNA copies/ml, as a rule, presented with respiratory symptoms interesting the upper or lower respiratory tract, whereas lower viral levels did not appear to correlate with the kinetics of clinical symptoms. This observation was made during follow-up, when it was noted that in some patients low HRV load became negative in association with persistent clinical symptoms, and in other patients low viral load remained unchanged in association with improvement of clinical symptoms.

The pathogenic role of HRV infection was also investigated in 12 lung transplant recipients. During followup, these patients were examined either in nasopharyngeal aspirates (n ¼ 8 positive/24 examined) during acute respiratory tract infection episodes, or in bronchoalveolar lavage (even in the absence of symptoms) following a routine bronchoscopy examination (n ¼ 14 positive/84 examined). In detail, nasopharyngeal aspirates from five episodes of acute upper respiratory tract infections interesting five lung transplant recipients (#1, #2, #3, #4, and #6, Table V ) were found to be associated with a high HRV load (>10 6 RNA copies/ml nasopharyngeal aspirates) in upper respiratory secretions, thus documenting an association between respiratory HRV load and upper respiratory clinical symptoms (no other infectious agents were detected in these patients). On the other hand, of the remaining seven patients, three (#5, #9, and #11) with respiratory tract infection showed a HRV load >10 5 RNA copies/ml bronchoalveolar lavage, while the other four patients (#7, #8, #10, and #12) free of acute respiratory symptoms in the lower respiratory tract had a viral load of 10 3 -10 4 RNA copies/ml bronchoalveolar lavage. This finding documented the HRV presence in the lower respiratory tract in the absence of overt clinical symptoms (Table V) . Clinical follow-up excluded the possibility that the low HRV amounts detected in bronchoalveolar lavage in these patients was related to the end of a previous lower respiratory tract infection episode or the onset of a new respiratory infection episode occurring in the lower respiratory tract.

Results of the present study focusing on the analysis of the correlation of HRV load in respiratory secretions and presence/absence of clinical symptoms are limited to the winter-spring season and allow the formulation of some general conclusions: (i) viral load greater than 10 5 HRV RNA copies/ml nasopharyngeal aspirate appears to be mostly associated with presence of clinical symptoms both in upper and lower respiratory infections; (ii) in immunocompetent infants and young children, in the case of HRV þ RSV co-infection, the quantification of viral load allows to hypothesize the viral agent responsible for the current respiratory episode and, if required, to adopt the adequate therapeutic measures; (iii) in young stem cell transplantation recipients, a 6month follow-up allowed the detection of an inverse correlation between time from transplantation and duration of HRV infection, while the peak viral load was not influenced by the time elapsed from transplantation to the occurrence of viral infection; (iv) in adult lung transplant recipients examined in bronchoalveolar lavage during post-transplant follow-up, HRV load lower than Fig. 2 . Inverse correlation between the duration (days) of HRV infection and time (days) elapsed after stem cell transplantation prior to onset of HRV infection in nine young patients. For each patient, vertical arrow (") indicates HRV persistence at time of last viral control, whereas HRV disappearance is indicated by (>). Over each patient, the peak HRV load detected in respiratory secretions for each patient is reported.

10 5 RNA copies/ml did not appear to be associated with the presence of clinical symptoms, which were detected with larger viral loads.

There has been a lot of debate regarding the presence of HRV in the respiratory tract of asymptomatic children and adults Blomqvist et al., 2002; Nokso-Koivisto et al., 2002; Kusel et al., 2006; Rawlinson et al., 2003; Van Benten et al., 2003; Winther et al., 2006] . Using molecular assays, given the 1-to 3-week duration of HRV shedding around each infection episode, incidental HRV detection during an independent hospitalization may occur frequently [Turner, 2007] . This event may occur in individuals recently experiencing respiratory symptoms or having contact with household members with an acute respiratory tract infection, or in patients who develop symptoms after sampling. However, even excluding these individuals, it is estimated that overall 5% of apparently healthy children are picornavirus-RNA-positive in the absence of previous or incipient respiratory symptoms [Nokso-Koivisto et al., 2002] .

In order to address the clinical significance of HRV present in the respiratory tract, it was decided to quantify HRV load in respiratory secretions and to correlate viral load with clinical symptoms. We found that a viral load greater than 10 5 RNA copies/ml in nasopharyngeal aspirate appeared to be associated with the presence of clinical symptoms either interesting the upper or the lower respiratory tract, while a viral load lower than 10 5 copies/ml was infrequently associated with clinical symptoms. Under experimental conditions, it has been shown that HRVs are able to grow in the lower respiratory tract, as demonstrated by their ability to grow both in vitro in primary human bronchial epithelial cells [Schroth et al., 1999; Papadopoulos et al., 2000] and in vivo in lower airway secretions following experimental infection by intranasal instillation and aerosols [Gern et al., 1997; Papadopoulos et al., 2000 ]. In addition, it was shown that in bronchoscopy samples, virus titers were similar to those observed in nasal brush samples [Halperin et al., 1983] . These findings support the assumption that virus quantitation in nasopharyngeal aspirates in lower respiratory tract infections mirror the virus amount present in the lower respiratory tract. Along this line of thought, it appears reasonable to attribute the etiologic role of a lower respiratory episode to HRVs, whenever a high viral load is detected in nasopharyngeal aspirates from patients not shedding other viruses in respiratory secretions.

As reported by Deffernez et al. [2004] , all HRV serotypes tested (n ¼ 29) and clinical isolates are detectable by their assay. However, it is not possible to exclude a lower sensitivity for detection of other serotypes, including those belonging to the recently identified group C. In this study, some of the amplicons were sequenced, thus showing the ability of the assay to detect strains belonging to A, B, and C species HRV.

As many as 21/62 episodes (33.9%) of dual HRV þ RSV (in three episodes a third virus was detected) co-infection were observed. In previous studies on bronchiolitis, no greater disease severity was reported on admission of patients with co-infection compared to patients with single infection [Andreoletti et al., 2000; Papadopoulos et al., 2002] . In our study, HRV þ RSV co-infections were predominant among the other combinations of co-infections. Virus quantification allowed to hypothesize that RSV was the etiologic agent of the acute respiratory episode, while HRV, present in lower amounts, was more likely to be considered a bystander virus with respect to the current respiratory infection episode. Although it is relatively infrequent that two viruses are detected at a high concentration in respiratory secretions (>10 5 RNA copies/ml NPA), our experience indicates that more than the absolute value of viral load in the admission sample, the kinetics of viral load drop in association with improvement of clinical symptoms may discriminate, among two or three co-infecting viruses, the etiologic agent of a respiratory syndrome Gerna et al., 2008] .

In the present study, HRV was the second most common respiratory virus causing hospitalization of immunocompetent children after RSV. In children with single HRV infection, the almost constant association with wheezing and/or pneumonia appears to confirm recently published data on the frequency of HRV-associated hospitalizations in young children [Miller et al., 2007] . This study reported a rate of HRV-associated hospitalizations of 5/1,000 children <5 years of age. This rate increased from 5 months to 5 years of age and was higher in children with a history of wheezing/asthma. However, whether HRV infection predisposes to asthma or patients with asthma are more susceptible to HRV infection remains to be defined. The current study confirmed that children with a predisposition to wheezing/asthma are at risk of hospitalization when affected by HRV infection.

The median duration of HRV illness is 7 days, some patients, however, experience a 2-week duration . In two stem cell transplant recipients repeat bronchoalveolar lavage samples were HRVpositive at intervals of 31 and 44 days after first virus detection. This finding raised the possibility of prolonged lower respiratory tract replication [Ison et al., 2003 ]. In the present study, we were able to detect an inverse correlation between duration of HRV infection and time (days) elapsed after transplantation in a group of nine young stem cell transplant recipients who were followed for at least 6 months. This finding supports the recommendation, when feasible, to delay transplantation by at least 1 month, if HRV is detected in nasopharyngeal aspirates in the period preceding the beginning of the preparative regimen to the allograft. On the other hand, peak viral loads were not influenced by the time elapsed from transplantation. In the past, cases of respiratory failure along with HRV respiratory infection were either associated with a co-pathogen or attributed to chemotherapy/radiotherapy [Ghosh et al., 1999; Ison et al., 2003; Van Kraaij et al., 2005] . However, two cases of fatal lower respiratory tract infections in adult stem cell transplant recipients likely caused by HRV have been recently reported [Peck et al., 2007] . In our study, all patients, but one, survived HRV infection during follow-up. The case of a transplanted patient who succumbed to pneumonia following HRV-C infection underscores the potential severe pathogenic role of this newly identified group of HRVs [Lau et al., 2007; Lee et al., 2007] .

Finally, in a group of lung transplant recipients examined during a long-term follow-up, it was documented that, even in lower respiratory tract secretions, the presence of low HRV amounts was not associated with overt clinical symptoms, which, by contrast, were consistently observed in the presence of higher viral load. Data previously reported on the prognostic impact of HRV present in bronchoalveolar lavage of lung transplant recipients are contradictory [Holt et al., 1997; Malcolm et al., 2001] . These findings may help to clarify this issue.

In conclusion, this study shows that: (i) HRV quantitation in nasopharyngeal aspirates appears to correlate with clinical symptoms associated with lower respiratory tract infections; (ii) HRV þ RSV co-infections do not appear to worsen the prognosis of lower respiratory tract infection episodes; (iii) in young stem cell transplant recipients, an inverse correlation between duration of HRV infection and time at which the infection occurs after transplantation is observed; (iv) in lung transplant recipients HRV load in bronchoalveolar lavage appears to correlate with clinical symptoms.

Viral pneumonia in the first month of life

Differential detection of rhinovirus and enterovirus RNA sequences associated with classical immunofluorescence assay detection of respiratory virus antigens in nasopharyngeal swabs from infants with bronchiolitis

Frequency and natural history of rhinovirus infection in adults during autumn

Virological and serological analysis of rhinovirus infections during the first two years of life in a cohort of children

Human respiratory syncytial virus (RSV) RNA quantification in nasopharyngeal secretions identifies the RSV etiologic role in acute respiratory tract infections of hospitalized infants

Effect of upper respiratory tract infections in patients with cystic fibrosis

Amplicon sequencing and improved detection of human rhinovirus in respiratory samples

A prospective hospital-based study of the clinical impact of non-severe acute respiratory syndrome (non-SARS)-related human coronavirus infection

Detection of rhinovirus RNA in lower airway cells during experimentally induced infection

Genetic variability of human coronavirus OC43-, 229E-, and NL63-like strains and their association with lower respiratory tract infections of hospitalized infants and immunocompromised patients

Simultaneous detection and typing of human metapneumovirus strains in nasopharyngeal secretions and cell cultures by monoclonal antibodies

Human respiratory coronavirus HKU1 vs other coronavirus infections in Italian hospitalized patients

Correlation of viral load as determined by realtime RT-PCR and clinical characteristics of respiratory syncytial virus lower respiratory tract infections in early infancy

Rhinovirus infections in myelosuppressed adult blood and marrow transplant recipients

Pathogenesis of lower respiratory tract symptoms in experimental rhinovirus infection

Efficacy and safety of oral pleconaril for treatment of colds due to picornaviruses in adults: Results of 2 double-blind, randomized, placebo-controlled trials

Incidence and significance of noncytomegalovirus viral respiratory infection after adult lung transplantation

Rhinovirus infections in recipients of hematopoietic stem cell transplantation with pneumonia

Community study of role of viral infections in exacerbations of asthma in 9-11 year old children

Etiology of communityacquired pneumonia in 254 hospitalized children

Chronic rhinoviral infection in lung transplant recipients

Role of respiratory viruses in acute upper and lower respiratory tract illness in the first year of life

Detection and quantification of human metapneumovirus in pediatric specimens by realtime RT-PCR

Clinical features and complete genome characterization of a distinct human rhinovirus (HRV) genetic cluster, probably representing a previously undetected HRV species, HRV-C, associated with acute respiratory illness in children

A diverse group of previously unrecognized human rhinoviruses are common causes of respiratory illnesses in infants

Rhinovirus illnesses during infancy predict subsequent childhood wheezing

Real-time PCR assays for detection of bocavirus in human specimens

Clinical features of patients with acute respiratory illness and rhinovirus in their bronchoalveolar lavages

Rhinovirus-associated hospitalizations in young children

A novel pancoronavirus RT-PCR assay: Frequent detection of human coronavirus NL63 in children hospitalized with respiratory tract infections in Belgium

Respiratory viruses and exacerbations of asthma in adults

Risk factors for lower respiratory complications of rhinovirus infections in elderly people living in the community: Prospective cohort study

Human picornavirus and coronavirus RNA in nasopharynx of children without concurrent respiratory symptoms

A comparison of nested polymerase chain reaction and immunofluorescence for the diagnosis of respiratory infections in children with bronchiolitis, and the implications for a cohorting strategy

Rhinoviruses infect the lower airways

Association of rhinovirus infection with increased disease severity in acute bronchiolitis

Rhinovirus as a cause of fatal lower respiratory tract infection in adult stem cell transplantation patients: A report of two cases

Rapid detection of human metapneumovirus strains in nasopharyngeal aspirates and shell vial cultures by monoclonal antibodies

Comparison of a real-time reverse transcription PCR assay and a culture technique for quantitative assessment of viral load in children naturally infected with respiratory syncytial virus

Detection of rhinovirus in sinus brushings of patients with acute communityacquired sinusitis by reverse transcription-PCR

Detection of rhinovirus, respiratory syncytial virus and coronavirus infections in acute otitis media by reverse transcription-PCR

Asthma exacerbations in children are associated with rhinovirus but not human metapneumovirus infection

Monoclonal antibodies versus reverse-transcription-PCR for detection of respiratory viruses in a patient population with respiratory tract infections admitted to hospital

Detection and pathogenicity of human metapneumovirus respiratory infections in pediatric Italian patients during a winter-spring season

Rhinovirus replication causes RANTES production in primary bronchial epithelial cells

Respiratory viruses, symptoms, and inflammatory markers in acute exacerbations and stable chronic obstructive pulmonary disease

Early detection of acute rhinovirus infections by a rapid reverse transcription-PCR assay

Rhinovirus: More than just a common cold virus

Predominance of rhinovirus in the nose of symptomatic and asymptomatic infants

Frequent detection of respiratory viruses in adult recipients of stem cell transplants with the use of real-time polymerase chain reaction, compared with viral culture

Picornavirus infections in children diagnosed by RT-PCR during longitudinal surveillance with weekly sampling: Association with symptomatic illness and effect of season

We would like to thank all the technical staff of the Servizio di Virologia for performing the assays. We are also indebted to Daniela Sartori for preparing the manuscript, and Laurene Kelly for revision of the English.