Enhanced severity of virus associated lower respiratory tract disease in asthma patients may not be associated with delayed viral clearance and increased viral load in the upper respiratory tract

2.1. Subjects

This prospective study, from September 1998 through August 2000, was approved by the local ethical committee. We recruited 44 people with asthma, aged 18–45 years, from general practices. Asthmatics were diagnosed according to the Dutch asthma guidelines and showed a reversibility in peak expiratory flow (PEF) of ≥15% and/or forced expiratory flow in 1 s (FEV1) of >9% after administration of a β-agonist (Geijer et al., 2003). All patients were treated as usual with various combinations of inhalation therapy (corticosteroids, β-agonists and parasympathicolytics). All asthmatic subjects had a history of atopy. We also enrolled 44 healthy control subjects (controls were not matched to asthmatic subjects) who did not have a history of asthma, allergic rhinitis, eczema or atopy, nor did their family members. Written informed consent was obtained from participants.

2.2. Procedures

During one season, asthmatic and control subjects daily recorded URT and LRT signs and symptoms, rated from 1 (mild) to 3 (severe). URT symptoms that were scored were defined as runny, stuffy or itchy nose, sneezing, watery/sore eyes, sore throat, hoarse voice and headache. LRT symptoms that were scored included coughing, wheezing, difficult breathing and shortness of breath during effort (Johnston et al., 1995). Patients in the asthma group recorded peak flow each morning and afternoon (peak flow score), using their peak flow instrument (Mini Wright meter; Clement Clarke International London, UK) three times in succession, and recording the highest value. The median baseline score for a subject was the mean of the peak flow score in five consecutive time points around the baseline (week 8). Subjects were asked to contact the investigators if the URT symptoms score totaled 4 or more, and/or if the LRT symptoms score exceeded 5, resulting in a home visit within 48 h. Follow-up visits were performed at days 3–4, days 5–6, after 2 weeks and after 8 weeks (baseline). The duration of URT and LRT symptoms was determined by the number of days with a URT or LRT symptoms score of 1 above the individual baseline score, respectively. The severity of the URT and LRT symptoms was defined as the maximum of URT and LRT symptoms scores, respectively. During each visit a nasal wash (NW) was collected from each nostril as described previously (Grunberg et al., 1997b). Briefly, 10 ml of sterile saline was instilled in each nostril and collected in the same tube. The tube was mixed thoroughly and immediately placed on wet ice. One milliliter of NW was used for virus culture. The remainder was mixed with an equal volume of diluted sputolysin reagent (6.5 mM dithiothreitol in 100 mM phosphate buffer) (Calbiochem-Novabiochem Corporation, San Diego, CA). Both aliquots were then clarified by centrifugation (1000 ×  g for 10 min at room temperature). The remaining supernatant fluid was stored at −80 °C until further analysis.

2.3. Respiratory virus detection and quantitation

Conventional viral cultures were performed as described by van Elden et al. (2003). Basically, by inoculating HEP-2C, R-HELA, and tertiary monkey kidney (t-MK) cells (National Institute for Public Health and the Environment (RIVM; Bilthoven, The Netherlands) with 1 ml NW each for the detection of respiratory viruses (parainfluenza viruses 1–3, respiratory syncytial virus (RSV) A and B, influenza viruses, adenovirus and picornaviruses). The cultures were examined twice weekly for 10 days for the development of cytopathic effect (CPE). In CPE-positive cultures, virus was identified by immunofluorescence with monoclonal antibodies (Dako Imagen) for influenza viruses A and B, RSV A and B, adenovirus and parainfluenza viruses 1–3. HRV were distinguished from enteroviruses through acid-lability testing. In shell vial cultures an immunofluorescence test was performed after 2 days of culture, usually before a cytopathic effect was noticed, using the above mentioned monoclonal antibodies (van Elden et al., 2003).

A real-time quantitative RT-PCR was performed on the stored supernatant fluids (nasal wash) for parainfluenza virus 1–4 and picornaviruses (van de Pol et al., 2007). A rhinovirus specific real-time PCR detected all HRV subtypes tested: 1a, 1b, 2–16, 18, 20–24, 26–30, 32–37, 39–67, 70–86, 88–100). Real-time PCR for influenza virus A and B, RSV A and B and human coronaviruses OC43 and 229E were performed as described previously (van Elden et al., 2004, van Elden et al., 2003, van Elden et al., 2001, Nijhuis et al., 2002). The number of viral RNA copies of influenza A and B, RSV A and B and rhinovirus in the clinical samples was determined by extrapolation from a standard curve generated by amplification of serial dilutions of electron microscopically (EM)-counted virus stocks. To estimate the quantity of human coronaviruses OC43 and 229E, for which no EM-counted virus stocks were available, viral copies were expressed as relative units (RU). Every amplification cycle represents a 2-fold increase in the number of viral RNA copies. The viral load was expressed as 2-fold increase per cycle relative to a baseline value of 2 copies at threshold cycle 36 (RU = 2^{36-threshold cycle}).

2.4. Assays for cytokines

IFN-γ, IL-6, IL-8 (CLB, Amsterdam, The Netherlands) and IL-5 (BioSource International, CA, USA) were measured using commercially available human cytokine ELISA kits according to the manufacturer's recommended protocol. The NW supernatants were assayed in duplicate and the results were expressed in pg/ml. The limits of sensitivity of these assays were as follows: IFN-γ, 1 pg/ml; IL-5, 4 pg/ml; IL-6, 0.2 pg/ml; and IL-8, 1 pg/ml.

2.5. Statistical analysis

Analyses were based only on the first episode of infection of a subject. Differences between asthmatic and control groups were analyzed by Mann–Whitney U-test (SPSS 11.0 for Windows). Comparisons of measures of infection and illness (viral load, symptom assessment and peak flow measurements) and cytokine levels between asthmatics and controls were done with the Wilcoxon signed rank test (SPSS 11.0 for Windows). Correlations were determined using the Spearman's rank correlation coefficient (SPSS 11.0 for Windows). An area under the curve (AUC) analysis including both virus load and duration of shedding was performed using the trapezium rule (Matthews et al., 1990). The AUC calculated from the data points from all individuals was then compared with means of the Mann–Whitney test (SPSS 11.0 for Windows).

3.1. Study population and episodes of viral respiratory infection

A total of 44 asthmatics and 44 control subjects were enrolled. Forty-three of these reported 57 episodes of respiratory tract illness. A respiratory virus was detected by PCR in 33 out of 57 (58%) episodes. Viral culture gave no added diagnostic value. There were 14 first episodes each of respiratory virus infection for analysis in asthmatics and controls. Patient characteristics are shown in Table 1 .

No significant differences were seen between the number of respiratory tract illnesses caused by a respiratory virus in asthmatics (17/33, 52%) and controls (16/33, 48%) (Table 2 ). The majority of infections (n  = 20) were caused by HRV.

3.2. Symptoms of viral RTI

The duration and severity of URT symptoms were similar for both the asthmatic and the control group. However, the severity of LRT symptoms was greater in the asthmatics as compared to the controls (p  = 0.02: Table 2). The duration of symptoms of the LRT was also significantly longer in asthmatic patients (p  = 0.03). Subgroup analyses performed in the asthmatic population did not indicate significant clinical differences between HRV and other respiratory virus infections, however, the sample size was small.

3.3. Viral dynamics

Using a quantitative real-time PCR for the detection of respiratory viruses we did not find a significant difference between study groups for either the maximum levels of virus produced or the duration of detection of viral RNA (Table 2). In addition for all subjects, neither the peak virus load nor the duration of viral shedding correlated with the duration or severity of LRT symptoms. This lack of correlation was observed with the subgroup analysis limited to HRV infections. The area under the curve plotting virus load against duration of shedding was not statistically different between the study groups (p  = 0.45). If the dynamics of the virus infections in asthma patients is compared to the peak flow reduction and the LRT symptoms, it seems clear that the actual viral infection in the upper respiratory tract precedes the peak flow reduction and LRT symptoms (Fig. 1 ).

3.4. Inflammatory cytokine response

NW levels of the proinflammatory cytokines IL-6 and IL-8 were significantly increased at days 3–4, as compared to baseline (Table 3 ). The levels of IL-6 paralleled the LRT symptoms and peak flow reduction and peaked at a time when virus load was declining (Fig. 2 ). No significant differences in the IL-6 or IL-8 response could be determined between both study groups during signs and symptoms of RTI. The same results were obtained when a subgroup analysis on just the HRV was performed. IL-5 and IFN-γ were not detected in NW of asthmatics and controls.