|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Divisions of Infectious Diseases (Drs. Moorman and Saad) and Allergy and Immunology (Drs. Krishnaswamy and Kosseifi), Department of Internal Medicine, James H. Quillen VAMC and James H. Quillen College of Medicine, East Tennessee State University, Johnson City, TN.
Correspondence to: Jonathan P. Moorman, MD, PhD, Department of Internal Medicine, Division of Infectious Diseases, James H. Quillen College of Medicine, East Tennessee State University, Box 70622, Johnson City, TN 37614; e-mail: moorman{at}mail.etsu.edu
| Abstract |
|---|
|
|
|---|
Key Words: complications hepatitis lung diseases review
| Introduction |
|---|
|
|
|---|
Over the last decade, an increasing number of reports have suggested that chronic HCV infection is also associated with both direct and indirect effects on pulmonary tissue. While not all of these effects have been tightly linked to HCV infection, there are now sufficient studies available that warrant a review of the major pulmonary sequelae associated with chronic HCV infection. We first describe what appear to be primary/direct effects of the virus on the lung and the possible mechanisms underlying these effects. We subsequently outline the secondary effects of chronic HCV infection on lung parenchyma and the pulmonary vasculature, including those related to cirrhosis, cryoglobulinemia, and interferon (IFN) therapy. Finally, we discuss options for future clinical and molecular studies that might broaden our understanding of the mechanisms by which chronic HCV induces pulmonary pathology.
| Effects of Hepatitis C on the Lung |
|---|
|
|
|---|
COPD:
COPD and asthma are chronic inflammatory conditions of the airways and lung parenchyma with differing patterns of airflow obstruction with respect to reversibility, whether spontaneous or in response to treatment.5 Several reports6789 have suggested an important role for latent viral infections, in particular adenovirus and HIV, in the etiology and/or progression of COPD. Based on these reports, investigators have hypothesized that chronic HCV infection might also function as a trigger for inflammation in the lungs, thereby either initiating or exacerbating the development of COPD. These data are limited by sample size but provide intriguing information that should urge investigators to pursue further trials.
The only prospective study10 to address the association between HCV and COPD was carried out by investigators from Japan, who randomly enrolled 30 HCV-positive and 29 HCV-negative patients with COPD and classified them into four groups: 15 HCV-negative ex-smokers, 14 HCV-negative current smokers, 14 HCV-positive ex-smokers, and 16 HCV-positive current smokers. Each patient underwent spirometric measurements and assessments of diffusion capacity of the lung for carbon monoxide (DLCO) every 4 months over a period of 5 years. Linear regression analysis was performed for each patients 5-year data to assess the decline in lung function. The annual rates of decline in FEV1 and DLCO in current smokers and ex-smokers were significantly higher in HCV-positive patients. When the change in FEV1 was assessed in patients who were treated for HCV infection with interferon (IFN)-
, IFN responders exhibited a slower progression of decline in FEV1 than the IFN nonresponders. The authors10 suggested that the airway disease may be related to underlying chronic inflammation with emphasis on the possible effects mediated by HCV-specific T lymphocytes and latent viral infection.
Asthma:
Several studies have now also documented an accelerated decline in lung function in asthmatic patients with chronic HCV infection associated with impaired responses to inhaled ß2-adrenoceptor agonists and corticosteroids and increased responses to inhaled anticholinergic agents. Kanazawa and colleagues11 assessed 40 asthmatic patients with chronic HCV infection for responses to the inhaled corticosteroid beclomethasone dipropionate (BDP) with and without IFN therapy. Patients were randomly selected, all were nonsmokers, and none of them were receiving steroid therapy. All patients received inhaled BDP therapy for 6 weeks, at which point 30 patients received therapy for HCV with IFN while all patients continued BDP therapy. Eleven patients responded to IFN therapy in terms of viral clearance, while 19 patients did not respond. Prebronchodilator and postbronchodilator FEV1 values were obtained after 6 weeks of BDP therapy and at 1-year following the end of IFN therapy.
The study11 showed no significant differences in either prebronchodilator or postbronchodilator FEV1 among all patients at 6 weeks of BDP therapy. At 1 year after IFN treatment, however, these values were significantly higher in the IFN responder group than in the IFN nonresponders and the IFN nontreatment groups. Furthermore, the IFN responder group had significantly higher prebronchodilator and postbronchodilator FEV1 values at 1 year after the end of IFN therapy compared to those obtained after 6 weeks of BDP therapy. As in the study10 examining the association of HCV with COPD, this study11 demonstrated declines in pulmonary function over time in individuals with HCV infection, which were reversible with successful control of viral replication by IFN treatment.
A prospective study12 with a 6-year follow-up was designed to determine whether chronic HCV infection affects declines in lung function and airway responses to the ß2-adrenoceptor agonist salbutamol in nonsmoking asthmatic patients. All HCV-positive patients received IFN for 6 months. One year after IFN therapy, patients were administered either inhaled salbutamol or oxitropium bromide in a double-blind manner. FEV1 values were then recorded, and measurements were repeated at 3 and 6 years after IFN therapy; 55 HCV-positive and 20 HCV-negative asthmatic patients completed the 6-year follow-up. Of the 55 HCV-positive patients, 18 were IFN responders and 37 were IFN nonresponders. Notably, all patients received rescue inhaled ß2-agonists for acute symptoms as well as inhaled corticosteroids, and their use was stable in all groups during the 6-year follow-up period.
The prebronchodilator and postbronchodilator FEV1 as well as the reversibility with salbutamol were significantly lower in the IFN nonresponder group when compared to the IFN responders and the HCV-negative group. Moreover, the study12 showed a steep decline in reversibility with salbutamol during the 6-year follow-up only in the IFN nonresponders, whereas there was a steep improvement in reversibility with the anticholinergic agent, oxitropium. These results suggested that chronic HCV infection is associated with an accelerated decline in lung function and impaired responses to salbutamol but not oxitropium in asthmatic patients. The authors12 suggested that chronic HCV infection might induce CD8+ T lymphocytes that cause asthma with a COPD-like inflammation, an effect that would explain the increased responses noted with the anticholinergic agent oxitropium.
This superior response to anticholinergic agents was further assessed in a recent study13 in which 36 HCV-positive asthmatic patients were administered IFN therapy followed by oxitropium bromide and compared to a group of 16 HCV-negative asthmatics. The study once again found significant increases in FEV1 and forced expiratory flow between 25% and 75% of FVC after oxitropium bromide administration in IFN nonresponders (ie, those with active HCV infection) when compared to the HCV-negative and the IFN responder groups. The authors suggested the possibility that patients with asthma and HCV infection respond differently to the various bronchodilator therapies than patients without HCV infection, and that HCV might modulate acetylcholine-mediated airway responses.
Mechanisms Regulating Declining Lung Function in HCV Infection:
These clinical studies support an association between chronic HCV infection and obstructive airway diseases, but the exact role of HCV in the pathogenesis of declining pulmonary function is not well understood. Several mechanisms could be hypothesized (Fig 1
), but the chronic immune activation and inflammation induced by HCV infection may play an important role. This has been shown with latent adenoviral infection in emphysematous smokers with COPD, who exhibit increased lung inflammation associated with increased expression of adenoviral EA1 protein in alveolar epithelial cells9; patients with more severe emphysema were found to have absolute increases in neutrophils, macrophages, and CD4+ and CD8+ lymphocytes.
|
One candidate for a role in pulmonary inflammation may be the T lymphocyte, in particular the CD8+ T cell. During viral infections, cytotoxic CD8+ T lymphocytes are in general up-regulated and activate a cascade of inflammatory pathways leading to the release of inflammatory mediators.17 CD8+ cells are also believed to play a key role in the development of airway inflammation associated with COPD, being overrepresented in the lungs of patients with COPD in an inverse relationship to lung function.18 Cytotoxic CD8+ T lymphocytes were also found to contribute to the pathology of severe or persistent asthma (which is generally a CD4+ cell-predominant process).19 CD8+ T lymphocytes also contribute to dysregulation of muscarinic M2 receptors, the general function of which is to inhibit acetylcholine release and thereby limit airway bronchoconstriction.20 CD8+ lymphocyte expression of IFN-
, for example, down-regulates M2 receptor expression in airway parasympathetic neurons and so exacerbates airway hyperreactivity.2021 Thus far, however, no study has confirmed the presence of HCV-specific cytotoxic CD8+ (or CD4+) T lymphocytes in BALF. Future studies of cellular responses in the lung in the settings of chronic HCV infection are needed to clarify this issue.
Other strong candidates for a role in pulmonary inflammation in the setting of HCV include inflammatory cytokines. In COPD, increased levels of interleukin (IL)-1ß, IL-6, IL-8, and tumor necrosis factor-
have been found and increase further with exacerbations.22 Patients with persistent asthma and COPD have an influx of neutrophils and increased local pulmonary IL-8 levels, and both asthma and COPD are characterized by increased nuclear factor-
B and 15-lipoxygenase expression.232425 In COPD, the bronchiolar epithelium also overexpresses monocyte chemoattractant protein-1 and IL-8, which act as leukocyte chemoattractants and thereby may contribute to the elevated neutrophil levels found in sputum.26
IL-8, because of its well-known chemotactic effects mediated by interaction with its receptor (CXCR 1 or 2) present on inflammatory cells such as neutrophils, can mediate cellular recruitment and propagate pulmonary inflammation (Fig 1).27 IL-8 has been shown to directly provoke bronchoconstriction28 and may contribute to the establishment of chronic reactive airway disease directly and indirectly by stimulating neutrophil recruitment and activation. Interestingly, studies2930 in patients with chronic HCV infection have demonstrated increased levels of both serum and intrahepatic cytokines, in particular IL-8. Expression of IL-8 may inhibit the antiviral activity of IFN-
and correlates with the degree of hepatic fibrosis and portal inflammation during HCV infection.3132
It remains unclear if HCV is acting to exacerbate underlying pulmonary disease, to initiate disease, or both. Further clinical and basic studies are clearly needed to examine in particular the cellular and cytokine responses occurring within pulmonary tissues in individuals with HCV infection.
HCV Infection and Interstitial Lung Disease:
Since HCV is well known to induce chronic inflammation and fibrosis in the liver, it was thought that HCV may play a similar role in the lung and be involved in the pathogenesis of pulmonary fibrosis. This idea put forth by investigators33 from Japan tests the presence of HCV antibodies in a cohort of patients with idiopathic pulmonary fibrosis (IPF); to their surprise, they found a higher prevalence of serum antibodies to HCV in patients with IPF (28.8%) than in age-matched control subjects (3.6%), which was statistically significant (p < 0.05). Although the findings of Irving and his colleagues34 failed to confirm this linkage, more recent studies and case reports continue to point toward an association between chronic HCV infection and interstitial lung disease. Meliconi et al35 found a high incidence of HCV infection in Italian patients with IPF (13%), as well as an increased incidence of noninterstitial lung disease in HCV patients, suggesting that chronic HCV infection might affect the lungs through different mechanisms and lead to a spectrum of clinical presentations.
Similarly, Ferri et al36 screened 300 patients with chronic HCV infection for the presence of lung disease by means of clinical symptoms and chest radiographs. Eight patients had evidence of interstitial lung involvement, which was further confirmed by high-resolution CT (HRCT). No patient had any obvious predisposing factors for pulmonary fibrosis. Four patients had severe interstitial lung fibrosis, while the other four patients had mild-to-moderate involvement. All of the patients had different degrees of DLCO reduction, and BALF showed an increased percentage of neutrophils in four of four patients. Lung involvement worsened in two patients slowly over time and remained stable in five patients; one patient died with rapidly progressive respiratory failure. The HCV genome was demonstrated in the lung biopsy specimen of one of the patients, a finding that might support a more direct pathogenic role for HCV in pulmonary fibrosis. Anecdotal reports3738394041 have supported these findings as well.
Investigators have studied the association between HCV and pulmonary fibrosis in patients with and without known lung disease. In a recent prospective study in individuals with no known pulmonary disease, Okutan et al42 compared the results of pulmonary function tests (PFTs) and HRCT in 34 patients with chronic HCV infection and 10 healthy control subjects and found a trend toward decreased DLCO in patients with HCV. While the differences in DLCO were not statistically significant, patients with HCV exhibited statistically significant interstitial lung involvement as seen on the HRCT; this involvement did not correlate to the degree of liver impairment. The small size of the patient sample in this study likely influenced the statistical power of the results. In a retrospective study43 of 81 liver transplant candidates with hepatitis C-induced cirrhosis, the results of echocardiography, arterial blood gas analysis, and PFTs were reviewed. Pulmonary changes were found to be frequent in this cohort, with reduced DLCO being the most common (found in 43% of patients), followed by restrictive lung impairment (17%) and obstructive airway disease (11%).
Further evidence of interstitial involvement with chronic HCV infection was provided by a study44 that assessed lung function by measurement of epithelial permeability with 99mTc-labeled diethylenetriaminepentaacetic acid aerosol scintigraphy. In this study,44 26 HCV-positive patients with no clinical pulmonary symptoms were compared to 31 normal control subjects; significantly increased epithelial permeability was found in HCV-positive patients compared to control subjects, a finding that generally suggests early interstitial lung disease.4546
In summary, several lines of evidence support a pathogenic role for chronic HCV infection in interstitial lung disease, but all are limited by sample size and the association remains controversial. Further larger, prospective trials are clearly needed to define the role of HCV in this process.
Secondary Effects of HCV Infection on the Lung
Table 1
lists the various other mechanisms by which the lung may be involved in HCV infection. Cirrhosis of the liver (due to HCV) with the added complications of portopulmonary hypertension (PPHTN) and hepatopulmonary syndrome (HPS), cryoglobulinemia, Sicca-like syndrome, malignant lymphomas, autoimmune thyroid disease, polymyositis, and hypocomplementemic urticarial vasculitis have all been reported in response to HCV infection and may indirectly affect the lung. The following sections discuss some of these secondary effects in greater detail.
|
In addition, two clinically distinct syndromes that represent a continuum of pulmonary vasculopathy have been defined in association with liver cirrhosis: HPS, representing extreme vasodilatation, and PPHTN, representing vasoconstriction. HPS is defined as the presence of intrapulmonary vasodilatations in conjunction with hypoxemia and chronic liver disease. PPHTN is defined by a mean pulmonary artery pressure > 25 mm Hg with a normal pulmonary capillary wedge pressure in the setting of portal hypertension. Although the two syndromes are the extremes of pulmonary vasculopathy, they may occasionally coexist in the same patient.58 They are discussed in the sections that follow.
HPS:
HPS is characterized by the clinical triad of hepatic dysfunction, hypoxemia (PaO2 < 70 mm Hg with an inspiratory fraction of oxygen of 0.21), and intrapulmonary vasodilatations.59 The term HPS was first suggested by Kennedy and Knudson60 in 1977, and its prevalence is widely variable in the literature due to differences in the definition of hypoxemia and the use of different standards. With the use of more stringent criteria, the prevalence of HPS in patients with chronic liver disease is likely from 10 to 15%.6162
The spectrum of clinical abnormalities is wide: impaired oxygenation may be subclinical, and patients may present with symptoms of liver disease rather than respiratory symptoms.5559 A subset of patients present with typical respiratory symptoms that include exertional dyspnea and platypnea (dyspnea that occurs on arising from a supine to a standing position), cyanosis, finger clubbing, spider nevi, hypoxemia, and orthodeoxia (a decrease in PaO2 > 3 mm Hg when a patient arises from a recumbent to a standing position).5562636465 Platypnea and orthodeoxia are common in patients with HPS because the intrapulmonary vascular dilatations that underlie these two manifestations are predominantly found in the lower lung fields, where blood pools due to the effect of gravity on standing.56
Intrapulmonary vascular dilatations are the major cause of hypoxemia in HPS. These occur as vascular dilatations at the precapillary or capillary levels, or as larger arteriovenous communications. Ventilation/perfusion mismatch then follows due to increased perfusion, while ventilation remains the same. This mismatch is thought to be due to inability of oxygen molecules to diffuse from the alveolar space to the center of these pathologically dilated capillaries to oxygenate the hemoglobin in the center.55565763 Finally, impaired hypoxic vasoconstriction in patients with chronic liver disease and the increased pulmonary blood flow may add to the ventilation/perfusion impairment486366 The exact cause of the pulmonary vascular dilatations remains poorly understood.
Diagnosis can be established noninvasively by contrast echocardiography or 99mTc-labeled macroaggregated albumin scanning.5556 Pulmonary angiography should be reserved for patients with severe hypoxemia and a poor response to 100% inspired oxygen, in whom vascular embolotherapy to obliterate arteriovenous communications (and eliminate the anatomic shunting) may be a therapeutic option.57
Several pharmacologic agents have been used to treat HPS, but the results have been disappointing. Plasma exchange and mechanical occlusion of the intrapulmonary vascular dilatations have also failed. Liver transplantation remains the only curative option, with resolution of the syndrome described to occur within days of transplant and up to 15 months after transplantation.555657
PPHTN:
PPHTN is characterized by a tetrad of elevated pulmonary artery pressure (> 25 mm Hg at rest), increased pulmonary vascular resistance (> 120 dyne·cm5), a normal wedge pressure (< 15 mm Hg), and underlying portal hypertension (> 10 mm Hg).67 It was first described in 1951 by Mantz and Craige,68 and its prevalence in patients with chronic liver disease is estimated to be between 1% and 5% in different studies.62697071 From 12 to 20% of patients undergoing orthotopic liver transplantation and those with decompensated cirrhosis may acquire this syndrome.496472
In the majority of patients with PPHTN, portal hypertension precedes pulmonary hypertension by an average of 4 to 7 years.6973 The pathogenesis of the structural changes in PPHTN is poorly understood, but the pathologic changes include pulmonary vasoconstriction, remodeling of muscular pulmonary artery walls, and in situ microthrombosis and/or thromboembolic lesions.63747576 Although the pathologic changes in PPHTN are similar to primary pulmonary hypertension, PPHTN is associated with a greatly increased cardiac output.77
The mean age at diagnosis is the fifth decade with a similar distribution in both sexes.6377 The most common symptom on presentation is exertional dyspnea, but other less frequent symptoms include syncope, chest pain, orthopnea, fatigue, palpitations, and hemoptysis. In addition, a large proportion of patients with PPHTN may be asymptomatic. Physical signs of PPHTN include increased intensity of the pulmonary component of the second heart sound, and murmurs of tricuspid and pulmonic regurgitation. Arterial blood gases usually reveal mild hypoxemia and exaggerated respiratory alkalosis, while PFTs may show a mild restrictive pattern with reduction in DLCO.636467737778
Echocardiography is a very helpful noninvasive tool to screen patients with suspected PPHTN. It may show right ventricular enlargement and signs of tricuspid and pulmonary regurgitation.7779 In addition, Doppler echocardiography can give an indirect estimate of the pulmonary artery pressure. The diagnosis is usually established by right-heart catheterization with the direct measurement of the pulmonary artery and right ventricular pressures. Vasodilator responsiveness should be assessed at the time of catheterization to help guide future therapy.
Treatment with vasodilator therapy (prostacyclin or prostacyclin analogues) has been shown to improve survival in a subset of patients with a positive vasodilator response.63678081 Other pharmacologic agents have been used with variable results and include phosphodiesterase inhibitors, inhaled nitric oxide, nitrates, and ß-blockers.508182838485 In contrast to HPS, the role of liver transplantation in PPHTN is not clear because of the increased intraoperative and perioperative death, and reports57626367 of worsening pulmonary hypertension after transplantation. Prognosis is poor overall in the absence of an intervention, with a mean survival period of 15 months and a median survival of 6 months.7374
Essential Mixed Cryoglobulinemia:
Essential mixed cryoglobulinemia (EMC) is a vasculitis characterized by the deposition of circulating immune complexes in small and medium-sized blood vessels and characteristically presents with a triad of arthralgias, purpura, and weakness.86 As the name implies, cryoglobulins are immune complexes that have the tendency to precipitate at cold temperatures. The link between EMC and chronic HCV infection is well established.87 It is estimated that approximately one third of patients with chronic hepatitis C infection have mixed cryoglobulinemia.47868889 In addition, cryoprecipitates were found to contain 10-fold and 1,000-fold levels of HCV antibody and RNA, respectively.8689 EMC may present as a systemic vasculitis that can involve different organs, with renal and neurologic involvement being more commonly reported.
A large number of rheumatologic disorders and vasculitic syndromes can present with a range of pulmonary manifestations by means of immune-mediated injury and autoimmune mechanisms. Likewise, the immune-mediated vasculitic lesions are responsible for the clinical manifestations of EMC, including cutaneous and visceral organ involvement, and particularly pulmonary involvement. Fortunately, pulmonary involvement is usually mild and probably slowly progressive.919293 Bombardieri and colleagues91 evaluated 23 patients with EMC for lung involvement and found that pulmonary symptoms were generally absent or moderate with the exception of 3 patients, who presented with asthma, hemoptysis, or pleurisy. Tests of small airway disease were markedly altered. Radiographic signs of interstitial lung involvement were present, albeit moderate, in 18 of 23 patients and were associated with inhomogeneities of regional blood flow on perfusion lung scanning.
Viegi et al92 found similar results and confirmed the findings documented by Bombardieri et al.91 A study93 of BALF in patients with EMC and HCV infection provided evidence for a subclinical T-lymphocyte alveolitis; patients with EMC were found to have a significantly lower percentage of alveolar macrophages but significantly higher percentages of CD3+ cells in their BALF than the control group. Also, PFTs done on the same patients showed significantly lower forced expiratory flow between 25% and 75% of FVC and DLCO in the EMC group than the control group. Following therapy with IFN, BALF analysis revealed a significant decrease in the percentage of lymphocytes.94 Whether the presence of T-lymphocyte alveolitis in this study is related to the EMC or to chronic HCV infection remains speculative.
Although the lung involvement in EMC is usually mild, several cases with severe lung involvement have been reported. Roithinger et al95 reported a case of EMC complicated by immunologically mediated pulmonary vasculitis. The patient died from the progressive lung involvement, and autopsy revealed diffuse pulmonary vasculitis. Several other reported cases with EMC presented with diffuse alveolar hemorrhage,51529697 severe lung involvement,9899 and bronchiolitis obliterans/organizing pneumonia (BOOP).53 IFN is now the treatment of choice for patients with EMC and probably the EMC-related pulmonary manifestations.53 Other lines of therapy include steroids and cytotoxic medications.
Miscellaneous Complications:
Other disorders described in HCV infection that may contribute to pulmonary problems are listed in Table 1. Sicca-like syndrome,36 malignant lymphomas,37 autoimmune thyroid disease,87 polymyositis,41 and hypocomplementemic urticarial vasculitis54100 have been described in HCV infection and may contribute to pulmonary disease. Appropriate testing and evaluation can lead to the diagnosis, affording specific therapies in addition to treatments directed against the virus.
| Pulmonary Complications Related to IFN Therapy |
|---|
|
|
|---|
was documented to successfully treat chronic HCV infection very early after HCV was first isolated.101102 This discovery was soon followed by reports of cases of IFN-associated pulmonary complications. Most of these were case reports, making it difficult to accurately estimate the incidence of such complications. However, interstitial pneumonitis, BOOP, ARDS, pulmonary hypertension, exacerbation of asthma, and sarcoid-like disease have been described in patients with hepatitis C undergoing treatment with IFN. Okanoue and his colleagues103 evaluated the complications of IFN in 987 patients, 3 of whom acquired interstitial pneumonia related to IFN therapy. In one report,104 the incidence of interstitial pneumonia due to IFN therapy was thought to be approximately 0.2%. Table 2 summarizes the spectrum of pulmonary complications induced by IFN therapy as documented in the literature.103105106107108109110111112113114115116117118119120121122123124125126127128129130131132 While interstitial pneumonia and sarcoidosis are well-reported complications, the remainder represent rarely associated complications of IFN therapy.
|
| Conclusions |
|---|
|
|
|---|
| Footnotes |
|---|
Received for publication March 2, 2005. Accepted for publication April 13, 2005.
| References |
|---|
|
|
|---|
. Thorax 1997;52,33-37
B, in asthma. Am J Respir Crit Care Med 1998;158,1585-1592
/PBSF) mRNA expression and histological activity index in patients with chronic hepatitis C. Hepatology 1998;28,108-115[CrossRef][ISI][Medline]
-interferon on subclinical lymphocyte alveolitis. Clin Exp Rheumatol 1996;14,219-220[ISI][Medline]
: a multicenter randomized, controlled trial. Hepatitis Interventional Therapy Group. N Engl J Med 1989;321,1501-1506[Abstract]
therapy for chronic hepatitis C: a randomized, double-blind, placebo-controlled trial. N Engl J Med 1989;321,1506-1510[Abstract]
2b [in French]. Gastroenterol Clin Biol 2002;26,405-408[ISI][Medline]
therapy for chronic hepatitis C. Chest 1994;105,939-941[Medline]
-interferon administration. Intern Med 1994;33,337-341[ISI][Medline]
and sho-saiko-to therapy [in Japanese]. Arerugi 1995;44,711-714[Medline]