- Email: [email protected]
Epidemiology of respiratory viruses in patients hospitalized with near-fatal asthma, acute exacerbations of asthma, or chronic obstructive pulmonary disease
Epidemiology of Respiratory Viruses in Patients Hospitalized with Near-Fatal Asthma, Acute Exacerbations of Asthma, or Chronic Obstructive Pulmonary Disease Wan C. Tan, MD, Xueyu Xiang, MSc, Diwen Qiu, MSc, Tze Pin Ng, MD, Sin F. Lam, MD, Richard G. Hegele, MD, PhD PURPOSE: We compared the prevalence and spectrum of common respiratory viruses among patients with near-fatal asthma, acute exacerbations of asthma, or chronic obstructive pulmonary disease (COPD), and the relation of these findings to acute respiratory symptoms. METHODS: We obtained adequate samples of respiratory secretions from 17 patients hospitalized with near-fatal asthma, 29 with acute asthma, and 14 with COPD. We used a polymerase chain reaction– based method to test for six common respiratory viruses in samples from endotracheal tube aspirates from patients with near-fatal asthma, and from induced sputum specimens from patients with acute asthma or COPD. Respiratory symptoms (runny nose, sore throat, fever, chills, malaise, and cough) were recorded. Quiescent-phase induced sputum specimens were examined from patients who were initially virus positive. RESULTS: Viral nucleic acids were detected in 52% (31/60) of
acute-phase specimens and 7% (2/29) of quiescent-phase specimens examined (P ⬍0.001), with similar proportions of viruspositive patients during the acute phase in the three groups: 59% (10/17) of those with near-fatal asthma, 41% (12/29) with acute asthma, and 64% (9/14) with COPD. Picornavirus (47% [n ⫽ 8]) and adenovirus (24% [n ⫽ 4]) were most commonly identified in near-fatal asthma, whereas influenza virus (36% [n ⫽ 5]) predominated in COPD. Virus-positive patients had a significantly increased frequency of runny nose, sore throat, fever, chills, and malaise (odds ratio ⫽ 4.1 to 18; P ⫽ 0.02 to 0.001). CONCLUSION: Respiratory viruses are associated with hospitalizations for near-fatal asthma, acute asthma, and COPD, with some differences in the spectrum of viruses involved in the different groups of patients. Respiratory viruses are a target for the prevention and perhaps the treatment of these conditions. Am J Med. 2003;115:272–277. ©2003 by Excerpta Medica Inc.
N
fatal attack (7). This suggests that the sensitivity of PCR may identify patients who are carriers of small amounts of viruses that are not pathogenic (8). We prospectively studied the association between near-fatal asthma and common respiratory viral infection in the lower respiratory tract using a PCR-based method (9,10) to detect six common respiratory viruses that have been implicated in acute exacerbations of asthma in children and adults: picornavirus (rhinovirus/ enterovirus), respiratory syncytial virus, parainfluenza virus, influenza A and influenza B viruses, and the adenovirus group (4,5,7). Specimens were collected from patients during hospitalization for an acute exacerbation and when they were outpatients during the stable (quiescent) phase of their disease. Two other groups, consisting of patients who were hospitalized for non-life-threatening acute asthma exacerbations or acute exacerbations of COPD, were also investigated. In all three groups of patients, our objectives were to determine the prevalence and spectrum of viral nucleic acids in lower respiratory tract specimens obtained during the acute phase and subsequent quiescent phase, and the relation between viral PCR results and symptoms of viral respiratory infections.
ear-fatal asthma has been studied intensively because it has the same clinical profile as fatal asthma (1–3). Previous studies have focused on the investigations of case history, clinical and social risk factors, the effects of therapy, and prevention and prognosis using interviews and self-administered questionnaires. More recently, sensitive molecular diagnostic methods, based on amplification of virus-specific nucleic acid sequences by polymerase chain reaction (PCR), have shown that viral infections are common triggers of acute exacerbations of asthma and chronic obstructive pulmonary disease (COPD) (4 – 6). However, a cross-sectional PCR-based study reported a high prevalence of common respiratory viruses, which was not clearly related to the
From the Departments of Medicine (WCT, XX, DQ) and Community Medicine (TPN), National University of Singapore, Singapore; Alexandra Hospital (SFL), Singapore; and the University of British Columbia McDonald Research Laboratories and iCAPTURE Centre (RGH), St. Paul’s Hospital, Vancouver, Canada. This study was supported by grant RP 3960401 from the National Medical Research Council of Singapore. Requests for reprints should be addressed to Wan C. Tan, MD, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, Singapore 119074, Singapore, or [email protected]. Manuscript submitted August 22, 2002, and accepted in revised form April 11, 2003. 272
© 2003 by Excerpta Medica Inc. All rights reserved.
0002-9343/03/$–see front matter doi:10.1016/S0002-9343(03)00353-X
Epidemiology of Respiratory Viruses/Tan et al
METHODS Subjects All patients with asthma had the diagnosis made by a physician, were treated with asthma medication, and had evidence of a forced expiratory volume in 1 second (FEV1) increase of 200 mL and 20% above baseline after 400 g of inhaled beta2-agonist or after a 2-week course of oral corticosteroid (11). Patients with COPD smoked cigarettes heavily (⬎20 pack-years), had chronic cough and dyspnea, and an FEV1% predicted of less than 50%, without beta-agonist reversibility (12). Near-fatal cases consisted of asthmatic patients who required ventilatory support in the Medical Intensive Care Unit (National University Hospital and The Alexandra Hospital, Singapore) because of a severe, life-threatening exacerbation (1–3,13). Acute asthma cases met at least one of the following criteria: failure to show sustained improvement after two administrations of nebulized beta-agonists during a 2-hour observation, or a severe attack based on clinical and blood gas criteria (13). Patients with COPD had been admitted based on clinical judgment.
Study Design We attempted to recruit all patients with near-fatal asthma, and other patients in a ratio of 2 acute asthma patients and 1 COPD patient per case of near-fatal asthma. Care was taken to ensure that the recruitment of patients from each of the three groups occurred within the same week to minimize potential seasonal variations in viral infections. Patients were studied on two occasions: during hospitalization (acute phase), and 4 to 6 weeks after discharge from the hospital, at a time when they were clinically stable (quiescent phase). Patients were questioned about respiratory symptoms (runny nose, sore throat, fever, chills, malaise, increased cough) within the past week during the acute and quiescent phases. Patients were considered to be in the quiescent phase in the absence of these symptoms.
Specimen Collection and Handling Lower respiratory tract secretions were collected and consisted of either endotracheal tube aspirates from patients with near-fatal asthma during the acute phase, or induced sputum specimens from patients with acute asthma or COPD during the acute phase, and from all three groups during the quiescent phase. Nasopharyngeal secretions were not sampled because induced sputum provides similar results for viral detection (10). Endotracheal tube aspirates were collected in sterile containers by standard suctioning by a trained intensive care nurse. Induced sputum specimens were obtained following the protocol of Fahy et al (14), after inhalation of nebulized sterile hypertonic saline (3%) from an ultrasonic nebulizer (ULTRA-NEB 99TM; DeVilbiss, Somerset, Pennsyl-
vania). Based on logistical considerations, specimens of induced sputum were collected during the quiescent phase only from patients whose acute samples had tested positive for viral nucleic acid.
Extraction of Deoxyribonucleic Acid and Ribonucleic Acid and Quality Control Deoxyribonucleic acid (DNA) was extracted from the specimens by the QIAamp Tissue Kit method (QIAgen GmbH, Hilden, Germany). Total ribonucleic acid (RNA) was extracted by using trizol reagent (9,15). A fixed amount of RNA (1 g) was used for complementary DNA (cDNA) synthesis. Reverse transcription was performed using random hexamers as primers (7–10). Five L of DNA or cDNA product was used in each tube for subsequent PCR amplification. Polymerase chain reaction for virus-specific nucleic acid sequences was performed on specimens deemed adequate after initial successful amplification of a constitutively expressed human gene, glycerol-3-phosphate dehydrogenase (GAPDH) (9,10). Inadequate specimens, in which GAPDH was not amplified by PCR, were excluded from subsequent PCR analyses for the six respiratory viruses included in the panel.
PCR Amplification, Dot Blot, Filter Hybridization, and Autoradiography Virus-specific nucleic acid sequences were detected based on multiple single-tube PCRs for picornavirus, respiratory syncytial virus, parainfluenza virus, influenza A and influenza B viruses, and adenovirus (9,10). Specimens were analyzed by a nested PCR method for adenovirus and by a reverse-transcription PCR method for the five RNA viruses tested. Positive viral PCR amplification was further confirmed by dot blot, hybridization with 3’-end DIG-labeled oligonucleotide probes, and autoradiography. A PCR result was considered positive when an unequivocal band of predicted size was seen on autoradiographs (or for adenovirus, on ethidium bromide-stained agarose gels) in the presence of appropriate results for known positive and negative cell culture control specimens.
Statistical Analysis Because not every patient contributed paired (acute- and quiescent-phase) specimens, statistical analyses were based on unpaired data. Fisher’s exact tests (16) were used to compare the proportions of virus-positive specimens between groups (SPSS Inc., Chicago, Illinois). To determine the relation between viral PCR results and respiratory symptoms, odds ratios with 95% confidence intervals were calculated for the virus-positive versus virus-negative reference group. In all instances, a twotailed P value ⬍0.05 was considered to be statistically significant.
September 2003
THE AMERICAN JOURNAL OF MEDICINE威
Volume 115 273
Epidemiology of Respiratory Viruses/Tan et al
Table 1. Characteristics of Patients with Asthma or Chronic Obstructive Pulmonary Disease Patient Group Characteristic
Near-Fatal Asthma (n ⫽ 18)
Acute Asthma (n ⫽ 35)
P Value COPD (n ⫽ 15)
All Groups
Near-Fatal vs. Acute Asthma
Number (%) or Mean ⫾ SD Age (years) Male sex Smoking history Ex-smoker Current smoker Amount (pack-years) Asthma history Onset age (years) Duration (years) Family history Admission finding Respiratory rate (beats per minute) Pulse rate (beats per minute) PaCO2 (mm Hg) pH Outpatient spirometry (% predicted) FEV1-pre* FVC-pre* FEV1-post† FVC-post† Medication use during 3 months before index date Inhaled steroids Oral steroids Inhaled beta-agonist Oral beta-agonist Antileukotrienes
42 ⫾ 15 10 (56)
42 ⫾ 15 8 (23)
71 ⫾ 11 13 (87)
0.001 0.001
1.00 0.001
5 (28) 0 15 ⫾ 8
2 (6) 2 (6) 8⫾3
11 (73) 4 (27) 31 ⫾ 5
0.001 0.02 0.001
0.04 0.20 0.22
30 ⫾ 20 12 ⫾ 10 15 (83)
23 ⫾ 16 18 ⫾ 13 18 (51)
– – 1 (7)
– – 0.001
0.34 0.19 0.02
28 ⫾ 7 125 ⫾ 31 57 ⫾ 8 7.23
26 ⫾ 16 95 ⫾ 23 39 ⫾ 2 7.41
27 ⫾ 6 110 ⫾ 22 47 ⫾ 12 7.38
0.80 0.001 0.001 0.001
0.68 0.007 0.001 0.001
66 ⫾ 22 88 ⫾ 24 78 ⫾ 21 102 ⫾ 20
65 ⫾ 26 77 ⫾ 23 65 ⫾ 24 78 ⫾ 23
40 ⫾ 5 63 ⫾ 12 44 ⫾ 7 70 ⫾ 14
0.001 0.005 0.001 0.001
0.94 0.25 0.24 0.03
9 (50) 0 13 (72) 10 (56) 0
20 (57) 4 (11) 21 (60) 9 (23) 2 (5)
8 (53) 0 11 (73) 10 (67) 0
0.88 0.13 0.54 0.006 0.38
0.42 0.18 0.29 0.02 0.43
* Before inhaled beta-agonist (% predicted normal). † After 400 g of inhaled beta-agonist (% predicted normal). COPD ⫽ chronic obstructive pulmonary disease; FEV1 ⫽ forced expiratory volume in 1 second; FVC ⫽ forced vital capacity; Paco2 ⫽ partial pressure of carbon dioxide, arterial.
RESULTS Sixty-eight patients (18 with near-fatal asthma, 35 with acute asthma, and 15 with COPD) were recruited (Table 1). Comorbidity was common in patients with COPD: 4 had heart failure, 7 had hypertension, 1 had tuberculosis, and 1 had diabetes. At follow-up during the quiescent phase, all asthmatic patients were receiving inhaled glucocorticosteroids.
Laboratory Viral Diagnosis During the acute phase, 60 (17 from patients with nearfatal asthma, 29 with acute asthma, and 14 with COPD) of 68 specimens (88%), were considered adequate for viral PCR based on successful amplification of the host GAPDH gene. In the quiescent phase, the proportion was 29/31 (94%). The PCR panel detected viral nucleic acids in 31 (52%) of the 60 specimens examined, and the overall viral prevalence in the three groups was 274
September 2003
THE AMERICAN JOURNAL OF MEDICINE威
similar (Table 2). In the quiescent phase, viral nucleic acid was identified by PCR in only 2 (7%) of the 29 specimens, and these revealed picornavirus in both instances. During the acute phase, the patients with near-fatal asthma were significantly more likely to have picornavirus or adenovirus than patients in the other two groups (Table 2). The near-fatal asthma group had a significantly lower percentage of combined influenza A and B positivity than the COPD group. Coinfection was found in 3 patients in each of the two groups of patients with asthma: picornavirus and adenovirus in near-fatal asthma, and picornavirus and influenza A in acute asthma.
Relation between PCR Viral Detection and Respiratory Symptoms During the acute phase, there were significant associations between detection of viral DNA and rhinorrhea,
Volume 115
Epidemiology of Respiratory Viruses/Tan et al
Table 2. Spectrum of Viruses Detected by Polymerase Chain Reaction in Patients during Acute Exacerbations Patient Group Near-Fatal Asthma (n ⫽ 17)
Virus
Acute Asthma (n ⫽ 29)
P Values
COPD (n ⫽ 14)
Near-Fatal Asthma vs. Acute Asthma
Near-Fatal Asthma vs. COPD
Near-Fatal Asthma vs. Acute Asthma ⫹ COPD
Number (%) Picornavirus Adenovirus Picornavirus or adenovirus Influenza A virus Influenza B virus Influenza A virus ⫹ influenza B virus Respiratory syncytial virus Parainfluenza virus Any virus
8 (47) 4 (24) 12 (71) 1 (6) 0 1 (6)
8 (28) 1 (3) 9 (31) 5 (17) 1 (3) 6 (21)
3 (21) 1 (7) 4 (28) 5 (36) 0 5 (36)
0.15 0.05 0.01 0.27 0.63 0.18
0.13 0.23 0.02 0.05 0.99 0.05
0.10 0.05 0.005 0.11 0.72 0.08
0 0 10 (59)
0 0 12 (41)
0 0 9 (64)
0.99 0.99 0.36
0.99 0.99 1.00
0.99 0.99 0.57
COPD ⫽ chronic obstructive pulmonary disease.
sore throat, fever, chills, and malaise, but not with cough (Table 3).
DISCUSSION We investigated the prevalence and spectrum of common respiratory viruses in patients hospitalized for life-threatening or non-life-threatening exacerbations of asthma, or for acute exacerbations of COPD. Overall, our results confirmed the high prevalence of nucleic acid sequences from common respiratory viruses in acute exacerbations of asthma (picornavirus and adenovirus) and COPD (influenza viruses) in adults. The detection of picornavirus in endotracheal tube aspirates and induced sputum supports a pathogenic role of these viruses as upper and lower respiratory tract pathogens (17,18). The large number of cases of adenovirus positivity in near-fatal asthma raises the possibility that this virus may be important in this condition.
A previous report of viral prevalence in fatal asthma showed nucleic acids from common respiratory viruses in postmortem specimens obtained from the lower airways, but there did not appear to be any pattern of viruses specific to fatal outcome in this study (7). We included a longitudinal component by performing repeat evaluation during disease quiescence to compare the findings obtained during acute exacerbations. We also studied the association between viral detection and respiratory symptoms. This design also allowed us to investigate whether viral nucleic acids detected by the highly sensitive PCR panel could be pathogenic during acute exacerbations, rather than merely identifying patients who were carriers of small amounts of viral nucleic acids (8). Overall, we consider that the viral PCR results are clinically relevant for several reasons. First, compared with the high rates of viral detection during the acute phase, the viral prevalence during the quiescent phase was low (7%) and within the 3% to 16% range reported in the literature (6,19). Second, the positive viral PCR re-
Table 3. Relation between Viral Results by Polymerase Chain Reaction and Respiratory Symptoms during Acute Exacerbations Respiratory Symptom
PCR-Positive (n ⫽ 32)
PCR-Negative (n ⫽ 28)
Odds Ratio (95% Confidence Interval)
P Value
9.2 (2.8–30) 18.0 (4.8–68) 10.0 (2.8–36) 5.9 (1.7–30) 4.1 (1.2–15) 3.3 (0.6–18)
0.001 0.001 0.001 0.02 0.03 0.16
Number (%) Runny nose Sore throat Fever Chills Malaise Cough
26 (81) 24 (75) 20 (63) 10 (31) 13 (41) 30 (94)
9 (32) 4 (14) 4 (14) 2 (7) 4 (14) 23 (82)
PCR ⫽ polymerase chain reaction. September 2003
THE AMERICAN JOURNAL OF MEDICINE威
Volume 115 275
Epidemiology of Respiratory Viruses/Tan et al
sults were associated with several respiratory symptoms. Third, the pattern of picornavirus dominance in exacerbations of asthma was compatible with the findings reported by others (4,5,19). The frequent detection of influenza viral nucleic acid in the acute asthma group is in apparent contrast to the results of other studies (4,5,19), in which influenza was not implicated as a common trigger of asthma exacerbations. Some of this apparent discrepancy could be related to differences in the severity of the exacerbations (all of our patients required hospitalization) or to differences in the laboratory methods used for the diagnosis of influenza viruses. Alternatively, the comparatively high prevalence of influenza that we observed might be attributable to the annual seasonal peak for the virus that occurs during April to June in Singapore, although our study took place over an entire year. In addition, a seasonal peak of influenza could not explain the lower prevalence of influenza in the near-fatal asthma group. Although other investigators did not find evidence of influenza virus during exacerbations of COPD (6), that study included influenza-vaccinated patients who were managed as outpatients. Our patients had more severe baseline disease, had not received similar vaccination, and required hospitalization for severe exacerbations. It is possible that the disease burden of rhinoviruses in severe COPD exacerbations could be underestimated and overshadowed by the more recognizable epidemic of influenza (20). Further studies using PCR-based viral detection techniques would permit clarification of the relative importance of respiratory viruses in COPD exacerbations. Our study has several limitations. Although the sex distribution was similar for patients with near-fatal asthma, patients with acute asthma were predominantly female and COPD patients were almost all male. In addition, patients hospitalized with acute exacerbations of COPD were significantly older than asthmatic patients. It is unclear whether these differences could be due to possible age- and sex-based susceptibilities for viral infection and type of exacerbation, differences in the natural history of the asthmatic attack (21–23), or perhaps differences in hospitalization (24 –28) or health care utilization (29). Another limitation is that we combined groups of viruses (picornavirus and adenovirus, and influenza A and B viruses) in the data analysis; these were not prespecified hypotheses. In conclusion, we have shown a high prevalence of nucleic acids from common respiratory viruses in specimens obtained from the lower respiratory tract of patients hospitalized for near-fatal asthma, acute asthma, or COPD. Different types of viruses may be responsible, depending on the type of underlying airway disease. The prevention or early treatment of respiratory viral infections in high-risk patients may be an important strategy to reduce hospitalization from asthma and improve 276
September 2003
THE AMERICAN JOURNAL OF MEDICINE威
health-related quality of life and disease control for patients who have difficult-to-manage asthma.
ACKNOWLEDGMENT We thank Dr. G. B. Lim of the University of Singapore for her technical assistance; and Velikeloth Sanila and Kelly Ngerng for help in the collection of clinical samples.
REFERENCES 1. Campbell DA, McLennan G, Coates JR, et al. A comparison of asthma deaths and near-fatal asthma attacks in South Australia. Eur Respir J. 1994;7:490 –497. 2. Turner MO, Noertjojo K, Vedal S, et al. Risk factors for near-fatal asthma. A case control study in hospitalized patients with asthma. Am J Respir Crit Care Med. 1998;157:1804 –1809. 3. Siddiqi A, Bandi V. Case discussions on the pathophysiology and clinical features of near-fatal asthma episodes. Curr Opin Pulm Med. 1999;5:47–51. 4. Johnston SL, Pattemore PK, Sanderson G, et al. Community study of role of viral infections in exacerbations of asthma in 9-11 year old children. BMJ. 1995;310:1225–1229. 5. Nicholson KG, Kent J, Ireland DC. Respiratory viruses and exacerbations of asthma in adults. BMJ. 1993;307:982–986. 6. Seemungal TA, Harper-Owen R, Bhowmik A, et al. Detection of rhinovirus in induced sputum at exacerbation of chronic obstructive pulmonary disease. Eur Respir J. 2000;16:677–683. 7. Macek V, Dakhama A, Hogg JC, et al. PCR detection of viral nucleic acid in fatal asthma: is the lower respiratory tract a reservoir for common viruses? Can Respir J. 1999;6:37–43. 8. Hogg JC. Viral infection and exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2001;164:1555– 1556. 9. Xiang X, Qui D, Hegele RD, Tan WC. Comparison of different methods of total RNA extraction for viral detection in sputum. J Virol Methods. 2001;94:129 –135. 10. Xiang X, Qui D, Chan KP, et al. Comparison of three methods of respiratory virus detection between induced sputum and nasopharyngeal aspirate specimens in acute asthma. J Virol Methods. 2002; 101:127–133. 11. American Thoracic Society. Lung function testing: selection of reference values and interpretative strategies. Am Rev Respir Dis. 1991; 144:1202–1218. 12. Pauwels R, Buist S, Calverley P, et al. Global strategy for the diagnosis, management and prevention of chronic obstructive pulmonary disease—NHLBI/WHO Global Initiative for chronic obstructive Lung Disease (GOLD) workshop summary. Am J Respir Crit Care Med. 2001;163:1256 –1276. 13. Global Initiative for Asthma—Global Strategy for Asthma Management and Prevention. Bethesda, MD: National Institutes of Health; National Heart, Lung and Blood Institute; 1995. NHLBI/WHO Workshop Report, March 1993. Publication No. 95-3659. 14. Fahy JV, Liu J, Wong H, Boushey HA. Cellular and biochemical analysis of induced sputum from asthmatic and from healthy subjects. Am Rev Respir Dis. 1993;147:1126 –1131. 15. Chomczynski P, Sacchi N. Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 1987;162:156 –159. 16. Fienberg SE. The Analysis of Cross-Classified Categorical Data. Cambridge, MA: Massachusetts Institute of Technology Press; 1980. 17. Gern JE, Galagan DM, Jarjour NN, et al. Detection of rhinovirus RNA in lower airway cells during experimentally induced infection. Am J Respir Crit Care Med. 1997;155:1159 –1161.
Volume 115
Epidemiology of Respiratory Viruses/Tan et al 18. Papadopoulos NG, Bates PJ, Bardin PG, et al. Rhinoviruses infect the lower airways. J Infect Dis. 2000;181:1875–1884. 19. Gern J. Rhinovirus respiratory infections and asthma. Am J Med. 2002;112:19S–27S. 20. Greenberg SB. Viral respiratory infections in elderly patients and patients with chronic obstructive pulmonary disease. Am J Med. 2002;112:28S–32S. 21. Gustavo J R, Carlos Rodrigo. Rapid-onset asthma attack—a prospective cohort study about characteristics and response to emergency department treatment. Chest. 2000;118:1547–1552. 22. Plaza V, Serrano J, Picado C, Sanchis J, on behalf of the High Risk Asthma Research Group. Frequency and clinical characteristics of rapid-onset fatal and near-fatal asthma. Eur Respir J. 2002;19:846 – 852. 23. Mitchell, I, Tough SC, Semple LSK, et al. Near-fatal asthma. A population-based study of risk factors. Chest. 2002;121:1407–1413.
24. National Heart, Lung, and Blood Institute. Morbidity and Mortality: Chartbook on Cardiovascular, Lung, and Blood Diseases. Bethesda, MD: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health; 1998. 25. Higgins MW, Thom T. Incidence, prevalence, and mortality: intraand inter-country differences. In: Hensley M, Saunders N, eds. Clinical Epidemiology of Chronic Obstructive Pulmonary Disease. New York: Marcel Dekker; 1989:23–43. 26. Alderson M. Trends in morbidity and mortality and asthma. Popul Trends. 1987;49:18 –23. 27. Hyndman SJ, Williams DRR, Merill SL, et al. Rates of admission to hospital for asthma. BMJ. 1994;308:1596 –1600. 28. Harju T, Keistinen T, Tuuponen T, et al. Hospital admissions for asthmatics by age and sex. Allergy. 1996;51:693–696. 29. Kesten S, Chew R, Ilanania N. Health-care utilization after nearfatal asthma. Chest. 1995;107:1564 –1569.
September 2003
THE AMERICAN JOURNAL OF MEDICINE威
Volume 115 277
acute-phase specimens and 7% (2/29) of quiescent-phase specimens examined (P ⬍0.001), with similar proportions of viruspositive patients during the acute phase in the three groups: 59% (10/17) of those with near-fatal asthma, 41% (12/29) with acute asthma, and 64% (9/14) with COPD. Picornavirus (47% [n ⫽ 8]) and adenovirus (24% [n ⫽ 4]) were most commonly identified in near-fatal asthma, whereas influenza virus (36% [n ⫽ 5]) predominated in COPD. Virus-positive patients had a significantly increased frequency of runny nose, sore throat, fever, chills, and malaise (odds ratio ⫽ 4.1 to 18; P ⫽ 0.02 to 0.001). CONCLUSION: Respiratory viruses are associated with hospitalizations for near-fatal asthma, acute asthma, and COPD, with some differences in the spectrum of viruses involved in the different groups of patients. Respiratory viruses are a target for the prevention and perhaps the treatment of these conditions. Am J Med. 2003;115:272–277. ©2003 by Excerpta Medica Inc.
N
fatal attack (7). This suggests that the sensitivity of PCR may identify patients who are carriers of small amounts of viruses that are not pathogenic (8). We prospectively studied the association between near-fatal asthma and common respiratory viral infection in the lower respiratory tract using a PCR-based method (9,10) to detect six common respiratory viruses that have been implicated in acute exacerbations of asthma in children and adults: picornavirus (rhinovirus/ enterovirus), respiratory syncytial virus, parainfluenza virus, influenza A and influenza B viruses, and the adenovirus group (4,5,7). Specimens were collected from patients during hospitalization for an acute exacerbation and when they were outpatients during the stable (quiescent) phase of their disease. Two other groups, consisting of patients who were hospitalized for non-life-threatening acute asthma exacerbations or acute exacerbations of COPD, were also investigated. In all three groups of patients, our objectives were to determine the prevalence and spectrum of viral nucleic acids in lower respiratory tract specimens obtained during the acute phase and subsequent quiescent phase, and the relation between viral PCR results and symptoms of viral respiratory infections.
ear-fatal asthma has been studied intensively because it has the same clinical profile as fatal asthma (1–3). Previous studies have focused on the investigations of case history, clinical and social risk factors, the effects of therapy, and prevention and prognosis using interviews and self-administered questionnaires. More recently, sensitive molecular diagnostic methods, based on amplification of virus-specific nucleic acid sequences by polymerase chain reaction (PCR), have shown that viral infections are common triggers of acute exacerbations of asthma and chronic obstructive pulmonary disease (COPD) (4 – 6). However, a cross-sectional PCR-based study reported a high prevalence of common respiratory viruses, which was not clearly related to the
From the Departments of Medicine (WCT, XX, DQ) and Community Medicine (TPN), National University of Singapore, Singapore; Alexandra Hospital (SFL), Singapore; and the University of British Columbia McDonald Research Laboratories and iCAPTURE Centre (RGH), St. Paul’s Hospital, Vancouver, Canada. This study was supported by grant RP 3960401 from the National Medical Research Council of Singapore. Requests for reprints should be addressed to Wan C. Tan, MD, Department of Medicine, National University Hospital, 5 Lower Kent Ridge Road, Singapore 119074, Singapore, or [email protected]. Manuscript submitted August 22, 2002, and accepted in revised form April 11, 2003. 272
© 2003 by Excerpta Medica Inc. All rights reserved.
0002-9343/03/$–see front matter doi:10.1016/S0002-9343(03)00353-X
Epidemiology of Respiratory Viruses/Tan et al
METHODS Subjects All patients with asthma had the diagnosis made by a physician, were treated with asthma medication, and had evidence of a forced expiratory volume in 1 second (FEV1) increase of 200 mL and 20% above baseline after 400 g of inhaled beta2-agonist or after a 2-week course of oral corticosteroid (11). Patients with COPD smoked cigarettes heavily (⬎20 pack-years), had chronic cough and dyspnea, and an FEV1% predicted of less than 50%, without beta-agonist reversibility (12). Near-fatal cases consisted of asthmatic patients who required ventilatory support in the Medical Intensive Care Unit (National University Hospital and The Alexandra Hospital, Singapore) because of a severe, life-threatening exacerbation (1–3,13). Acute asthma cases met at least one of the following criteria: failure to show sustained improvement after two administrations of nebulized beta-agonists during a 2-hour observation, or a severe attack based on clinical and blood gas criteria (13). Patients with COPD had been admitted based on clinical judgment.
Study Design We attempted to recruit all patients with near-fatal asthma, and other patients in a ratio of 2 acute asthma patients and 1 COPD patient per case of near-fatal asthma. Care was taken to ensure that the recruitment of patients from each of the three groups occurred within the same week to minimize potential seasonal variations in viral infections. Patients were studied on two occasions: during hospitalization (acute phase), and 4 to 6 weeks after discharge from the hospital, at a time when they were clinically stable (quiescent phase). Patients were questioned about respiratory symptoms (runny nose, sore throat, fever, chills, malaise, increased cough) within the past week during the acute and quiescent phases. Patients were considered to be in the quiescent phase in the absence of these symptoms.
Specimen Collection and Handling Lower respiratory tract secretions were collected and consisted of either endotracheal tube aspirates from patients with near-fatal asthma during the acute phase, or induced sputum specimens from patients with acute asthma or COPD during the acute phase, and from all three groups during the quiescent phase. Nasopharyngeal secretions were not sampled because induced sputum provides similar results for viral detection (10). Endotracheal tube aspirates were collected in sterile containers by standard suctioning by a trained intensive care nurse. Induced sputum specimens were obtained following the protocol of Fahy et al (14), after inhalation of nebulized sterile hypertonic saline (3%) from an ultrasonic nebulizer (ULTRA-NEB 99TM; DeVilbiss, Somerset, Pennsyl-
vania). Based on logistical considerations, specimens of induced sputum were collected during the quiescent phase only from patients whose acute samples had tested positive for viral nucleic acid.
Extraction of Deoxyribonucleic Acid and Ribonucleic Acid and Quality Control Deoxyribonucleic acid (DNA) was extracted from the specimens by the QIAamp Tissue Kit method (QIAgen GmbH, Hilden, Germany). Total ribonucleic acid (RNA) was extracted by using trizol reagent (9,15). A fixed amount of RNA (1 g) was used for complementary DNA (cDNA) synthesis. Reverse transcription was performed using random hexamers as primers (7–10). Five L of DNA or cDNA product was used in each tube for subsequent PCR amplification. Polymerase chain reaction for virus-specific nucleic acid sequences was performed on specimens deemed adequate after initial successful amplification of a constitutively expressed human gene, glycerol-3-phosphate dehydrogenase (GAPDH) (9,10). Inadequate specimens, in which GAPDH was not amplified by PCR, were excluded from subsequent PCR analyses for the six respiratory viruses included in the panel.
PCR Amplification, Dot Blot, Filter Hybridization, and Autoradiography Virus-specific nucleic acid sequences were detected based on multiple single-tube PCRs for picornavirus, respiratory syncytial virus, parainfluenza virus, influenza A and influenza B viruses, and adenovirus (9,10). Specimens were analyzed by a nested PCR method for adenovirus and by a reverse-transcription PCR method for the five RNA viruses tested. Positive viral PCR amplification was further confirmed by dot blot, hybridization with 3’-end DIG-labeled oligonucleotide probes, and autoradiography. A PCR result was considered positive when an unequivocal band of predicted size was seen on autoradiographs (or for adenovirus, on ethidium bromide-stained agarose gels) in the presence of appropriate results for known positive and negative cell culture control specimens.
Statistical Analysis Because not every patient contributed paired (acute- and quiescent-phase) specimens, statistical analyses were based on unpaired data. Fisher’s exact tests (16) were used to compare the proportions of virus-positive specimens between groups (SPSS Inc., Chicago, Illinois). To determine the relation between viral PCR results and respiratory symptoms, odds ratios with 95% confidence intervals were calculated for the virus-positive versus virus-negative reference group. In all instances, a twotailed P value ⬍0.05 was considered to be statistically significant.
September 2003
THE AMERICAN JOURNAL OF MEDICINE威
Volume 115 273
Epidemiology of Respiratory Viruses/Tan et al
Table 1. Characteristics of Patients with Asthma or Chronic Obstructive Pulmonary Disease Patient Group Characteristic
Near-Fatal Asthma (n ⫽ 18)
Acute Asthma (n ⫽ 35)
P Value COPD (n ⫽ 15)
All Groups
Near-Fatal vs. Acute Asthma
Number (%) or Mean ⫾ SD Age (years) Male sex Smoking history Ex-smoker Current smoker Amount (pack-years) Asthma history Onset age (years) Duration (years) Family history Admission finding Respiratory rate (beats per minute) Pulse rate (beats per minute) PaCO2 (mm Hg) pH Outpatient spirometry (% predicted) FEV1-pre* FVC-pre* FEV1-post† FVC-post† Medication use during 3 months before index date Inhaled steroids Oral steroids Inhaled beta-agonist Oral beta-agonist Antileukotrienes
42 ⫾ 15 10 (56)
42 ⫾ 15 8 (23)
71 ⫾ 11 13 (87)
0.001 0.001
1.00 0.001
5 (28) 0 15 ⫾ 8
2 (6) 2 (6) 8⫾3
11 (73) 4 (27) 31 ⫾ 5
0.001 0.02 0.001
0.04 0.20 0.22
30 ⫾ 20 12 ⫾ 10 15 (83)
23 ⫾ 16 18 ⫾ 13 18 (51)
– – 1 (7)
– – 0.001
0.34 0.19 0.02
28 ⫾ 7 125 ⫾ 31 57 ⫾ 8 7.23
26 ⫾ 16 95 ⫾ 23 39 ⫾ 2 7.41
27 ⫾ 6 110 ⫾ 22 47 ⫾ 12 7.38
0.80 0.001 0.001 0.001
0.68 0.007 0.001 0.001
66 ⫾ 22 88 ⫾ 24 78 ⫾ 21 102 ⫾ 20
65 ⫾ 26 77 ⫾ 23 65 ⫾ 24 78 ⫾ 23
40 ⫾ 5 63 ⫾ 12 44 ⫾ 7 70 ⫾ 14
0.001 0.005 0.001 0.001
0.94 0.25 0.24 0.03
9 (50) 0 13 (72) 10 (56) 0
20 (57) 4 (11) 21 (60) 9 (23) 2 (5)
8 (53) 0 11 (73) 10 (67) 0
0.88 0.13 0.54 0.006 0.38
0.42 0.18 0.29 0.02 0.43
* Before inhaled beta-agonist (% predicted normal). † After 400 g of inhaled beta-agonist (% predicted normal). COPD ⫽ chronic obstructive pulmonary disease; FEV1 ⫽ forced expiratory volume in 1 second; FVC ⫽ forced vital capacity; Paco2 ⫽ partial pressure of carbon dioxide, arterial.
RESULTS Sixty-eight patients (18 with near-fatal asthma, 35 with acute asthma, and 15 with COPD) were recruited (Table 1). Comorbidity was common in patients with COPD: 4 had heart failure, 7 had hypertension, 1 had tuberculosis, and 1 had diabetes. At follow-up during the quiescent phase, all asthmatic patients were receiving inhaled glucocorticosteroids.
Laboratory Viral Diagnosis During the acute phase, 60 (17 from patients with nearfatal asthma, 29 with acute asthma, and 14 with COPD) of 68 specimens (88%), were considered adequate for viral PCR based on successful amplification of the host GAPDH gene. In the quiescent phase, the proportion was 29/31 (94%). The PCR panel detected viral nucleic acids in 31 (52%) of the 60 specimens examined, and the overall viral prevalence in the three groups was 274
September 2003
THE AMERICAN JOURNAL OF MEDICINE威
similar (Table 2). In the quiescent phase, viral nucleic acid was identified by PCR in only 2 (7%) of the 29 specimens, and these revealed picornavirus in both instances. During the acute phase, the patients with near-fatal asthma were significantly more likely to have picornavirus or adenovirus than patients in the other two groups (Table 2). The near-fatal asthma group had a significantly lower percentage of combined influenza A and B positivity than the COPD group. Coinfection was found in 3 patients in each of the two groups of patients with asthma: picornavirus and adenovirus in near-fatal asthma, and picornavirus and influenza A in acute asthma.
Relation between PCR Viral Detection and Respiratory Symptoms During the acute phase, there were significant associations between detection of viral DNA and rhinorrhea,
Volume 115
Epidemiology of Respiratory Viruses/Tan et al
Table 2. Spectrum of Viruses Detected by Polymerase Chain Reaction in Patients during Acute Exacerbations Patient Group Near-Fatal Asthma (n ⫽ 17)
Virus
Acute Asthma (n ⫽ 29)
P Values
COPD (n ⫽ 14)
Near-Fatal Asthma vs. Acute Asthma
Near-Fatal Asthma vs. COPD
Near-Fatal Asthma vs. Acute Asthma ⫹ COPD
Number (%) Picornavirus Adenovirus Picornavirus or adenovirus Influenza A virus Influenza B virus Influenza A virus ⫹ influenza B virus Respiratory syncytial virus Parainfluenza virus Any virus
8 (47) 4 (24) 12 (71) 1 (6) 0 1 (6)
8 (28) 1 (3) 9 (31) 5 (17) 1 (3) 6 (21)
3 (21) 1 (7) 4 (28) 5 (36) 0 5 (36)
0.15 0.05 0.01 0.27 0.63 0.18
0.13 0.23 0.02 0.05 0.99 0.05
0.10 0.05 0.005 0.11 0.72 0.08
0 0 10 (59)
0 0 12 (41)
0 0 9 (64)
0.99 0.99 0.36
0.99 0.99 1.00
0.99 0.99 0.57
COPD ⫽ chronic obstructive pulmonary disease.
sore throat, fever, chills, and malaise, but not with cough (Table 3).
DISCUSSION We investigated the prevalence and spectrum of common respiratory viruses in patients hospitalized for life-threatening or non-life-threatening exacerbations of asthma, or for acute exacerbations of COPD. Overall, our results confirmed the high prevalence of nucleic acid sequences from common respiratory viruses in acute exacerbations of asthma (picornavirus and adenovirus) and COPD (influenza viruses) in adults. The detection of picornavirus in endotracheal tube aspirates and induced sputum supports a pathogenic role of these viruses as upper and lower respiratory tract pathogens (17,18). The large number of cases of adenovirus positivity in near-fatal asthma raises the possibility that this virus may be important in this condition.
A previous report of viral prevalence in fatal asthma showed nucleic acids from common respiratory viruses in postmortem specimens obtained from the lower airways, but there did not appear to be any pattern of viruses specific to fatal outcome in this study (7). We included a longitudinal component by performing repeat evaluation during disease quiescence to compare the findings obtained during acute exacerbations. We also studied the association between viral detection and respiratory symptoms. This design also allowed us to investigate whether viral nucleic acids detected by the highly sensitive PCR panel could be pathogenic during acute exacerbations, rather than merely identifying patients who were carriers of small amounts of viral nucleic acids (8). Overall, we consider that the viral PCR results are clinically relevant for several reasons. First, compared with the high rates of viral detection during the acute phase, the viral prevalence during the quiescent phase was low (7%) and within the 3% to 16% range reported in the literature (6,19). Second, the positive viral PCR re-
Table 3. Relation between Viral Results by Polymerase Chain Reaction and Respiratory Symptoms during Acute Exacerbations Respiratory Symptom
PCR-Positive (n ⫽ 32)
PCR-Negative (n ⫽ 28)
Odds Ratio (95% Confidence Interval)
P Value
9.2 (2.8–30) 18.0 (4.8–68) 10.0 (2.8–36) 5.9 (1.7–30) 4.1 (1.2–15) 3.3 (0.6–18)
0.001 0.001 0.001 0.02 0.03 0.16
Number (%) Runny nose Sore throat Fever Chills Malaise Cough
26 (81) 24 (75) 20 (63) 10 (31) 13 (41) 30 (94)
9 (32) 4 (14) 4 (14) 2 (7) 4 (14) 23 (82)
PCR ⫽ polymerase chain reaction. September 2003
THE AMERICAN JOURNAL OF MEDICINE威
Volume 115 275
Epidemiology of Respiratory Viruses/Tan et al
sults were associated with several respiratory symptoms. Third, the pattern of picornavirus dominance in exacerbations of asthma was compatible with the findings reported by others (4,5,19). The frequent detection of influenza viral nucleic acid in the acute asthma group is in apparent contrast to the results of other studies (4,5,19), in which influenza was not implicated as a common trigger of asthma exacerbations. Some of this apparent discrepancy could be related to differences in the severity of the exacerbations (all of our patients required hospitalization) or to differences in the laboratory methods used for the diagnosis of influenza viruses. Alternatively, the comparatively high prevalence of influenza that we observed might be attributable to the annual seasonal peak for the virus that occurs during April to June in Singapore, although our study took place over an entire year. In addition, a seasonal peak of influenza could not explain the lower prevalence of influenza in the near-fatal asthma group. Although other investigators did not find evidence of influenza virus during exacerbations of COPD (6), that study included influenza-vaccinated patients who were managed as outpatients. Our patients had more severe baseline disease, had not received similar vaccination, and required hospitalization for severe exacerbations. It is possible that the disease burden of rhinoviruses in severe COPD exacerbations could be underestimated and overshadowed by the more recognizable epidemic of influenza (20). Further studies using PCR-based viral detection techniques would permit clarification of the relative importance of respiratory viruses in COPD exacerbations. Our study has several limitations. Although the sex distribution was similar for patients with near-fatal asthma, patients with acute asthma were predominantly female and COPD patients were almost all male. In addition, patients hospitalized with acute exacerbations of COPD were significantly older than asthmatic patients. It is unclear whether these differences could be due to possible age- and sex-based susceptibilities for viral infection and type of exacerbation, differences in the natural history of the asthmatic attack (21–23), or perhaps differences in hospitalization (24 –28) or health care utilization (29). Another limitation is that we combined groups of viruses (picornavirus and adenovirus, and influenza A and B viruses) in the data analysis; these were not prespecified hypotheses. In conclusion, we have shown a high prevalence of nucleic acids from common respiratory viruses in specimens obtained from the lower respiratory tract of patients hospitalized for near-fatal asthma, acute asthma, or COPD. Different types of viruses may be responsible, depending on the type of underlying airway disease. The prevention or early treatment of respiratory viral infections in high-risk patients may be an important strategy to reduce hospitalization from asthma and improve 276
September 2003
THE AMERICAN JOURNAL OF MEDICINE威
health-related quality of life and disease control for patients who have difficult-to-manage asthma.
ACKNOWLEDGMENT We thank Dr. G. B. Lim of the University of Singapore for her technical assistance; and Velikeloth Sanila and Kelly Ngerng for help in the collection of clinical samples.
REFERENCES 1. Campbell DA, McLennan G, Coates JR, et al. A comparison of asthma deaths and near-fatal asthma attacks in South Australia. Eur Respir J. 1994;7:490 –497. 2. Turner MO, Noertjojo K, Vedal S, et al. Risk factors for near-fatal asthma. A case control study in hospitalized patients with asthma. Am J Respir Crit Care Med. 1998;157:1804 –1809. 3. Siddiqi A, Bandi V. Case discussions on the pathophysiology and clinical features of near-fatal asthma episodes. Curr Opin Pulm Med. 1999;5:47–51. 4. Johnston SL, Pattemore PK, Sanderson G, et al. Community study of role of viral infections in exacerbations of asthma in 9-11 year old children. BMJ. 1995;310:1225–1229. 5. Nicholson KG, Kent J, Ireland DC. Respiratory viruses and exacerbations of asthma in adults. BMJ. 1993;307:982–986. 6. Seemungal TA, Harper-Owen R, Bhowmik A, et al. Detection of rhinovirus in induced sputum at exacerbation of chronic obstructive pulmonary disease. Eur Respir J. 2000;16:677–683. 7. Macek V, Dakhama A, Hogg JC, et al. PCR detection of viral nucleic acid in fatal asthma: is the lower respiratory tract a reservoir for common viruses? Can Respir J. 1999;6:37–43. 8. Hogg JC. Viral infection and exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2001;164:1555– 1556. 9. Xiang X, Qui D, Hegele RD, Tan WC. Comparison of different methods of total RNA extraction for viral detection in sputum. J Virol Methods. 2001;94:129 –135. 10. Xiang X, Qui D, Chan KP, et al. Comparison of three methods of respiratory virus detection between induced sputum and nasopharyngeal aspirate specimens in acute asthma. J Virol Methods. 2002; 101:127–133. 11. American Thoracic Society. Lung function testing: selection of reference values and interpretative strategies. Am Rev Respir Dis. 1991; 144:1202–1218. 12. Pauwels R, Buist S, Calverley P, et al. Global strategy for the diagnosis, management and prevention of chronic obstructive pulmonary disease—NHLBI/WHO Global Initiative for chronic obstructive Lung Disease (GOLD) workshop summary. Am J Respir Crit Care Med. 2001;163:1256 –1276. 13. Global Initiative for Asthma—Global Strategy for Asthma Management and Prevention. Bethesda, MD: National Institutes of Health; National Heart, Lung and Blood Institute; 1995. NHLBI/WHO Workshop Report, March 1993. Publication No. 95-3659. 14. Fahy JV, Liu J, Wong H, Boushey HA. Cellular and biochemical analysis of induced sputum from asthmatic and from healthy subjects. Am Rev Respir Dis. 1993;147:1126 –1131. 15. Chomczynski P, Sacchi N. Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 1987;162:156 –159. 16. Fienberg SE. The Analysis of Cross-Classified Categorical Data. Cambridge, MA: Massachusetts Institute of Technology Press; 1980. 17. Gern JE, Galagan DM, Jarjour NN, et al. Detection of rhinovirus RNA in lower airway cells during experimentally induced infection. Am J Respir Crit Care Med. 1997;155:1159 –1161.
Volume 115
Epidemiology of Respiratory Viruses/Tan et al 18. Papadopoulos NG, Bates PJ, Bardin PG, et al. Rhinoviruses infect the lower airways. J Infect Dis. 2000;181:1875–1884. 19. Gern J. Rhinovirus respiratory infections and asthma. Am J Med. 2002;112:19S–27S. 20. Greenberg SB. Viral respiratory infections in elderly patients and patients with chronic obstructive pulmonary disease. Am J Med. 2002;112:28S–32S. 21. Gustavo J R, Carlos Rodrigo. Rapid-onset asthma attack—a prospective cohort study about characteristics and response to emergency department treatment. Chest. 2000;118:1547–1552. 22. Plaza V, Serrano J, Picado C, Sanchis J, on behalf of the High Risk Asthma Research Group. Frequency and clinical characteristics of rapid-onset fatal and near-fatal asthma. Eur Respir J. 2002;19:846 – 852. 23. Mitchell, I, Tough SC, Semple LSK, et al. Near-fatal asthma. A population-based study of risk factors. Chest. 2002;121:1407–1413.
24. National Heart, Lung, and Blood Institute. Morbidity and Mortality: Chartbook on Cardiovascular, Lung, and Blood Diseases. Bethesda, MD: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health; 1998. 25. Higgins MW, Thom T. Incidence, prevalence, and mortality: intraand inter-country differences. In: Hensley M, Saunders N, eds. Clinical Epidemiology of Chronic Obstructive Pulmonary Disease. New York: Marcel Dekker; 1989:23–43. 26. Alderson M. Trends in morbidity and mortality and asthma. Popul Trends. 1987;49:18 –23. 27. Hyndman SJ, Williams DRR, Merill SL, et al. Rates of admission to hospital for asthma. BMJ. 1994;308:1596 –1600. 28. Harju T, Keistinen T, Tuuponen T, et al. Hospital admissions for asthmatics by age and sex. Allergy. 1996;51:693–696. 29. Kesten S, Chew R, Ilanania N. Health-care utilization after nearfatal asthma. Chest. 1995;107:1564 –1569.
September 2003
THE AMERICAN JOURNAL OF MEDICINE威
Volume 115 277