- Email: [email protected]
Pleuropulmonary involvement in ankylosing spondylitis
Joint Bone Spine 72 (2005) 496–502 http://france.elsevier.com/direct/BONSOI/
Review
Pleuropulmonary involvement in ankylosing spondylitis Abdellah El Maghraoui * Rheumatology and Rehabilitation Unit, Military Teaching Hospital Mohammed V, Rabat, Morocco Received 8 December 2003; accepted 13 May 2004 Available online 07 August 2004
Abstract Pleuropulmonary involvement was long described as an uncommon and late event in the course of ankylosing spondylitis (AS). This belief was based on studies that relied on symptoms and chest radiographs to evaluate the lungs. However, pleuropulmonary involvement in AS patients is usually asymptomatic, and the early lesions are undetectable on chest radiographs. Apical fibrosis, interstitial infiltrates, and pleural thickening were considered to be the main patterns. However, the introduction of high-resolution computed tomography (HRCT) has led to the description of many pulmonary abnormalities that are clinically silent and undetectable on plain radiographs. These abnormalities mainly affect the interstitium and have no influence on respiratory function, which is dependent on the severity of chest wall inflammation or ankylosis in recent-onset and established AS, respectively. Cytological and histological studies suggest that, in common with uveitis and aortic regurgitation, the structural lung changes shown by HRCT may be specific of AS. © 2005 Elsevier SAS. All rights reserved. Keywords: Ankylosing spondylitis; Computed tomography of the chest; Lungs; Apical fibrosis; Lung function testing
1. Introduction Ankylosing spondylitis (AS) is the archetypal spondyloarthropathy and as such is strongly associated with the HLA B27 antigen. The disease primarily affects the axial joints [1], most notably the sacroiliac joints. Extraarticular manifestations are represented by uveitis, aortic regurgitation, bowel disease, and other systemic disorders [2–4]. Respiratory abnormalities have been reported in 0% to more than 30% of patients with AS [5–8]. Chest wall rigidity related to involvement of the costovertebral joints is responsible for a restrictive ventilatory pattern. In addition, AS can cause pleuropulmonary lesions, of which the most common are apical fibrosis, interstitial infiltrates, and pleural thickening [9–11]. These lesions are usually asymptomatic, and in their early stages they are not visible on chest radiographs. As a result, pleuropulmonary lesions were long believed to be uncommon and delayed manifestations of AS. However, over the last decade high-resolution computed tomography (HRCT) has been found superior over plain radiography in detecting early pulmonary lesions due to various systemic diseases * Tel.: +212-61-54-7190; fax: +212-37-62-6179. E-mail address: [email protected] (A. El Maghraoui). 1297-319X/$ - see front matter © 2005 Elsevier SAS. All rights reserved. doi:10.1016/j.jbspin.2004.05.006
including rheumatoid arthritis, systemic sclerosis, and AS [12–15]. In patients with AS, HRCT has shown not only the classic lung lesions, but also a wide variety of subtle abnormalities undetectable by plain radiography. The relevance of these abnormalities remains to be determined. This review article starts with a description of the steps that led to the recognition of pleuropulmonary abnormalities associated with AS. Then, the frequency and distribution of the abnormalities are discussed, with emphasis on differences related to the detection methods used (plain radiographs, HRCT, lung function tests (LFTs), bronchoalveolar lavage (BAL), and histological studies of transbronchial biopsy or autopsy specimens). The review ends with a presentation of current pathophysiological hypotheses and of the impact of pleuropulmonary manifestations on the everyday management of patients with AS. 2. Historical overview The first report of pleuropulmonary disease in patients with AS was written in 1941 by Dunham and Kautz [16], who described four cases of apical tuberculosis among 20 patients with AS. From a literature review, they concluded that pulmonary tuberculosis was more common among AS patients
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
than in the population at large. In 1949, Hamilton [17] also reported an excess risk of pulmonary tuberculosis in AS patients and suggested that this finding might reflect diagnostic error. Two of his patients had apical fibrosis that was ascribed to recurrent bacterial infection. In 1965, Campbell and Mac Donald [18] described six new cases of apical fibrosis and suggested for the first time that this abnormality might be an extraarticular manifestation of AS, similar to aortic regurgitation and uveitis. They ruled out the other causes of apical fibrosis, including recurrent infection, pulmonary tuberculosis, fungal infections, radiation-induced fibrosis, sarcoidosis, syphilis, pneumoconiosis, and rheumatoid disease. Three years later, Jessamine [19] reported seven additional cases of apical fibrosis in AS patients. Although microbiological studies were negative, five of these seven patients were given antitubercular drugs. Histological specimens were obtained from six patients and showed intraalveolar fibrosis, hyalinization, and elastic tissue degeneration. Jessamine [19] pointed out that these abnormalities resembled those seen in aortic valves from AS patients with aortic regurgitation. The same year, Lauritzen et al. [20] reported a retrospective review of 60 cases of AS: 11 patients had fibrocavitating lesions of the lung apices but only two had pulmonary tuberculosis. In a 1970 study of cardiomyopathy in AS patients, Takkunenn et al. [21] obtained radiographs from 55 AS patients and found apical fibrosis in seven patients. In 1972, Davies [22] described six new cases and found 50 previously published cases of “nontuberculous apical fibrosis” in AS patients. The typical patient was a male with advanced axial AS, onset of pulmonary symptoms several years after the first joint symptoms, distinctive radiological changes in the upper lobes, and a high rate of Aspergillus superinfection. Davies [22] pointed out the differences between the pulmonary abnormalities associated with AS and those seen in patients with rheumatoid arthritis, and he agreed with Campbell and Mac Donald [18] that apical fibrosis was a specific extraarticular manifestation of AS. In a 1974 study of 225 AS patients, Crompton et al. [23] found that 14 (6%) patients had apical fibrosis and that five (2%) had pulmonary tuberculosis. Wolson and Rohwedder [24] reported in 1975 that two of 52 AS patients had apical fibrosis. At the Mayo Clinic, Rosenow et al. [25] retrospectively reviewed the chest radiographs of 2080 AS patients and found 26 (1.2%) cases of apical fibrosis. Subsequently, case-reports were written by Manresa et al. [26] in 1976, Touraine et al. [27] in 1977, Brocard et al. [28] in 1978, Chatel et al. [29] in 1980, Peltier et al. [30] in 1982, Ducolone et al. [31] and Agguilaniu et al. [32] in 1984, and Leménager et al. [33] in 1987.
3. Chest radiographs Rates of pleuropulmonary involvement were low in studies that used chest radiographs as the only imaging method. In the largest case-series, the rate of occurrence was 1.2%
497
Table 1 Reported rates of occurrence of pulmonary abnormalities visible on chest radiographs in patients with AS
Chakera et al. [34] Rosenow et al. [25] Leménager et al. [33] Bouchea and Sundstrom [8] Casserly et al. [37] Turetschek et al. [39] El Maghraoui et al. [35]
Number of patients 42 2080 50 32 26 21 55
Abnormal chest radiograph N (%) 19 (45.2) 28 (1.3) 10 (20) 11 (34.3) 4 (15.3) 0 (0) 2 (3.6)
[25]. The only abnormalities detected in these studies were apical fibrosis, pleural thickening, bronchiectasis, and interstitial disease. Apical fibrosis develops in the upper lung lobes and follows a progressive course. Pulmonary fibrosis associated with AS is consistently located in the upper part of the lungs, in sharp contrast to the preferential involvement of the lung bases at the early stages of pulmonary fibrosis due to other systemic diseases. Thickening of the apical pleura is a common concomitant of apical fibrosis, indicating that the two abnormalities are directly related to each other. Apical pleural thickening can antedate the development of apical fibrosis [33,34]. Apical bronchiectasis has been ascribed to traction on the airways by the fibrotic lung parenchyma. Table 1 reports the rates of occurrence of pleuropulmonary abnormalities found by plain radiography in patients with AS.
4. High-resolution computed tomography Falashi et al. [36] wrote the first description of early pulmonary lesions detected by HRCT, in four patients with AS. Since then, including the report of our preliminary findings [35], only five prospective studies of pleuropulmonary lesions detected by HRCT in AS patients have been published. Table 2 recapitulates the HRCT changes found in these studies. Casserly et al. and Fenlon et al. [37,38] reported HRCT abnormalities in 69% of 26 AS patients, of whom only 15% had abnormal chest radiograph findings. Similarly, Turetschek et al. [39] found pulmonary abnormalities by HRCT in 71% of 21 patients, Senocak et al. [40] in 85% of 20 patients, and Kiris et al. [41] in 64% of 28 patients; in our study, the proportion was 56% (21/55) [42]. Pooling the findings from these studies yields a rate of occurrence of about two in three patients, with the most common abnormalities being pleural thickening (20%), parenchymal bands (19.3%), micronodules (17.3%), and subpleural bands (12.6%). Overall, apical fibrosis was found in 9.3% of patients, bronchiectasis in 11.3%, and emphysema in 9.3%. Bronchiectasis was usually related to traction by fibrotic tissue. Apical fibrosis and bronchiectasis are irreversible, whereas ground-glass densities (6.6% of patients) reflect alveolitis that may be reversible under treatment. Figs. 1–4 show examples of these lesions taken from our study.
498
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
Table 2 Reported rates of occurrence of pulmonary abnormalities visible by computed tomography in patients with AS
Number of patients Abnormal CT, n (%) Apical fibrosis, n (%) Emphysema, n (%) Bronchiectasis, n (%) Ground-glass densities, n (%) Nonspecific interstitial densities, n (%) Micronodules, n (%) Pleural thickening, n (%) Bronchial wall thickening, n (%) Subpleural band, n (%) Parenchymal band, n (%) Parenchymal retraction, n (%) Uneven interface, n (%) Blebs, n (%)
Casserly et al. [37] 26 19 (73) 2 (7.6) 4 (15) 6 (23) 1 (3.8) 11 (42.3) 1 (3.8) 1 (3.8) 4 (15) 6 (23) 8 (30.7) – 3 (11.5) 1 (3.8)
Turetschek et al. [39] 21 15 (71) 4 (19) 2 (9.5) 2 (9.5) 2 (9.5) – 2 (9.5) 6 (28.5) 6 (28.5) – – – – –
Fig. 1. Computed tomography of the chest in a 57-year-old man with an over two decade-long history of AS: retracting fibrosis in the left upper lobe (black arrow) with traction bronchiectasis (white arrow).
Senocak et al. [40] 20 17 (85) 3 (15) 3 (15) 3 (15) 3 (15) – 8 (40) 9 (45) 5 (25) 7 (35) 3 (15) – – –
Kiris et al. [41] 28 18 (64.2) – – – 2 (7.1) – 8 (25) – 2 (7.1) – 5 (17.8) – 1 (3.6) –
El Maghraoui et al. [1,35] 55 31 (56.3) 5 (9) 5 (9) 6 (10.9) 2 (3.6) 26 (47.2) 7 (12.9) 14 (25.4) – 6 (10.9) 13 (23.6) 12 (21.8) 4 (7.2) 7 (12.9)
Total 150 100 (66.6) 14 (9.3) 14 (9.3) 17 (11.3) 10 (6.6) – 26 (17.3) 30 (20.0) 16 (10.6) 19 (12.6) 29 (19.3) 12 (8.0) 8 (5.3) 8 (5.3)
Fig. 2. Computed tomography of the chest in a 66-year-old woman with a 5-year history of AS: interstitial disease with irregular interfaces, micronodules, parenchymal bands (black arrow), and severe primary bronchiectasis (white arrow).
In our study, we found no statistically significant differences regarding the rate of occurrence of HRCT abnormalities between smokers and nonsmokers, in keeping with findings by Turetschek et al. and Senocak et al. The rates of apical fibrosis and bronchiectasis increased with disease duration, whereas the rates of nonspecific abnormalities were not influenced by this factor. This finding suggests that nonspecific abnormalities may progress over time to yield apical fibrosis and bronchiectasis.
5. Lung function testing AS is associated with a pure restrictive pattern due to rigidity of the chest wall [43,44]. Vital capacity (VC) is decreased in proportion to disease severity; in patients with advanced
Fig. 3. Computed tomography of the chest in a 40-year-old man with a 16-year history of AS: apical fibrosis (black arrow), parenchymal bands, and subpleural emphysema (white arrow).
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
Fig. 4. Computed tomography of the chest in the same patient as in Fig. 3: ground-glass densities (black arrow) and parenchymal band (white arrow).
499
striking contrast to experience with rheumatoid arthritis. In a study of 14 AS patients, Kchir et al. [48] found that BAL was normal in 12 patients and showed lymphocytosis in two patients, of whom one was a smoker and the other had alveolar fibrosis. In a 1994 study by Bonnet et al. [49] in 34 patients with spondyloarthropathy but no pulmonary abnormalities (reactive arthritis, n = 4; AS, n = 19; and unclassifiable spondyloarthropathy, n = 11) and nine controls, lymphocytic alveolitis (with a predominance of CD8+ cells) was found in patients with early disease and neutrophilic alveolitis in those with advanced disease. The findings were independent from smoking status, nonsteroidal antiinflammatory drug use, and presence of airborne pollutants. The authors concluded that a specific mechanism caused the apical fibrosis but that upper lobe hypoventilation stimulated the fibrotic process. The same year, Scherak et al. suggested a similar interpretation [50].
diseases, VC is rarely less than 60% of predicted. Residual volume (RV) and functional residual capacity (FRC) are increased or normal, in contrast to findings in other restrictive diseases. Although expiratory muscle performance is enhanced, the chest wall rigidity causes muscle exhaustion, so that complete expiration cannot be achieved. Thus, the decrease in total lung capacity (TLC = VC + RV) is less marked than the decrease in VC. In our study [42], 16 of 55 patients had a restrictive pattern, and 11 of these 16 had abnormalities by HRCT (nonspecific interstitial changes in eight patients, bronchiectasis in two, emphysema in one, and interstitial syndrome in two). Nevertheless, we found no correlation between HRCT changes and LFT results. In contrast, LFT results were correlated with variables reflecting the severity of symptoms and structural damage related to AS, in keeping with earlier evidence that severity of axial disease but not presence of HRCT lesions governed the severity of the restrictive pattern [45,46]. LFT may show obstructive disease in smokers, but this finding is independent from the restrictive pattern (Table 3).
Histological studies have been done on transbronchial or percutaneous biopsies, operative specimens, and autopsy specimens [51,52]. Nonspecific interstitial fibrosis with degeneration of collagen and elastic fibers was found. Focal lymphocytic infiltrates were visible in some cases. In patients with advanced disease, cystic cavities were visible within the fibrous areas. No granulomas or signs of vasculitis were found. Kchir et al. [48] obtained transbronchial biopsies from 12 of their 14 AS patients and found interstitial fibrosis in five, with pleural adhesions. The few histological studies available in the literature indicate a high rate of interstitial disease with a potential for progression to fibrosis.
6. Bronchoalveolar lavage
More than 50 years after the seminal description, pleuropulmonary involvement by AS remains an enigma. Pathophysiological hypotheses are based on two main theories, the mechanical theory and the AS-specific theory.
To elucidate the pathophysiological mechanism underlying pulmonary lesions in patients with AS, Wendling et al. [47] performed BAL in 15 patients and 17 controls. Two patients had interstitial disease in the lung bases; no information is given on the pulmonary status in the remaining 13 patients. BAL showed no evidence of inflammation, in
7. Histological studies
8. Etiological and pathogenic hypotheses
8.1. Mechanical theory The chest wall rigidity seen in patients with AS may induce major ventilatory disturbances, with hyperventilation (Table 3) of the lung bases and severe hypoventilation of the apices.
Table 3 Reported rates of occurrence of lung function testing abnormalities in patients with AS Eulry et al. [43] Casserly et al. [37] Turetschek et al. [39] Senocak et al. [40] El Maghraoui et al. [1,35] a
LFT: lung function testing.
Number of patients 49 26 21 20 55
Normal LFT a, n (%) 42 (85.7) 16 (61.5) 9 (42.8) 12 (60) 37 (67.2)
Restrictive pattern, n (%) 4 (8.1) 7 (26.9) 12 (57.1) 6 (30) 16 (29)
Obstructive pattern, n (%) 3 (6.1) 3 (11.5) 0 (0) 2 (10) 2 (3.6)
500
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
According to Jessamine [19], apical hypoventilation may result in poor airway clearance, which in turn may cause chronic inflammation. However, studies of mucociliary function fail to support this hypothesis [53]. In addition, exposure to cigarette smoke and particulate matter undoubtedly promotes the development of fibrosis. The main piece of evidence supporting the mechanical theory is that most of the AS patients with pleuropulmonary disease had predominantly axial joint disease. Pleuropulmonary disease is rare in patients with peripheral joint disease but no axial involvement.
8.2. Disease-specific theory Several case-reports support the involvement of an ASspecific process. Thus, Campbell and Mac Donald [18] reported a case in which apical fibrosis developed 2 years before the first joint symptoms, well before the chest wall had lost its flexibility. Mucociliary function was studied by Bouvier et al. [54] in 10 patients with AS and normal chest radiograph findings. Mucociliary clearance was decreased in the lung apices. The authors suggested that apical fibrosis might be secondary to recurrent infections related to apical hypoventilation. They found no statistically significant differences in mucociliary function between the patients and the controls or between the upper and the lower lungs, and they concluded that apical fibrosis was produced by a specific mechanism related to the nature of AS. Parkin et al. [55] compared regional ventilation in 27 AS patients, including three with apical fibrosis, and 18 controls. Hypoventilation was found only in the apices of the patients with apical fibrosis. In contrast to the mechanical theory, the disease-specific theory is consistent with the differences commonly seen between the right and left lung apices of individual patients. HRCT studies also support a role for a disease-specific mechanism. In our study, HRCT abnormalities were found even in patients with mild symptoms and minimal structural damage that did not affect spinal alignment or mobility. Smoking is widely believed to promote apical fibrosis and interstitial inflammation [56]. Exposure to tobacco smoke is associated with increased counts of macrophages and neutrophils in the deep lung parenchyma. Thus a role for smoking in precipitating or worsening lung disease in AS patients is plausible. However, HRCT abnormalities were found in similar proportions of smokers and nonsmokers in our study. Similarly, Turetschek et al. studied only nonsmokers, yet found a high rate of HRCT abnormalities. Thus, the most likely explanation to lung abnormalities in AS patients is a disease-specific inflammatory process whose course is parallel to that of the joint manifestations. This hypothesis is in line with the results of BAL studies and other investigations.
9. Evaluation of pulmonary involvement in clinical practice A careful physical examination is crucial to the evaluation of the lungs in patients with AS. Chest expansion should be measured, and finger clubbing or other signs denoting hypoventilation should be sought. In patients who are free of symptoms, auscultation of the lung fields may reveal crackling rales that suggest fibrosis and indicate a need for further investigations. LFTs should be obtained to look for a restrictive pattern, most notably in patients with severe or longlasting AS. A chest radiograph should be obtained routinely, and HRCT should be considered in patients with normal radiographic findings but with abnormal physical findings. The HRCT abnormalities reported in studies of AS patients were subtle and showed no correlations with clinical symptoms or lung function. Therefore, it seems unreasonable to recommend routine HRCT in patients with AS. The high cost of HRCT is an additional reason for restricting the use of this investigation. Nevertheless, our results suggest that HRCT may be useful in selected patients, including those with a restrictive pattern that cannot be explained solely by mechanical factors, symptoms, or abnormal chest radiograph findings. In addition, with the increasing use of anti-TNF alpha agents in patients with AS, both HRCT and BAL may prove useful to rule out tuberculosis before treatment initiation in patients with chest radiograph abnormalities. Physicians should be thoroughly familiar with the broad array of abnormalities associated with AS so that they can confidently distinguish them from changes caused by tuberculosis or other infections.
10. Treatment and follow-up The presence of pleuropulmonary abnormalities in a patient with AS has no impact on the treatment strategy. Prophylactic measures including smoking cessation and physiotherapy to improve overall and respiratory function may be useful, however, in conjunction with disease-modifying antirheumatic drugs and nonsteroidal antiinflammatory agents. Whether early and appropriate treatment of AS decreases the risk of pleuropulmonary involvement is not known. Longitudinal studies of patients with recent-onset AS and normal lungs would be needed to address this issue. Close monitoring to ensure early detection of complications is in order in every patient with AS. The main complications associated with apical fibrosis are infections, most notably due to Aspergillus and atypical mycobacteria, and pneumothorax. Life-threatening pulmonary complications include massive hemoptysis due to Aspergillus infection and cardiorespiratory failure due to bacterial infection. Functional symptoms and physical signs denoting lung disease should be sought carefully at each visit.
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
References [1]
[2]
[3] [4]
[5] [6] [7]
[8] [9]
[10]
[11] [12] [13] [14]
[15]
[16] [17] [18] [19] [20]
[21] [22] [23]
[24]
El Maghraoui A, Bensabbah R, Bahiri R, Bezza A, Guedira N, HajjajHassouni N. Cervical spine involvement in ankylosing spondylitis. Clin Rheumatol 2003;22:94–8. El Maghraoui A, Dougados M, Freneaux E, Chaussade S, Amor B, Breban M. Concordance between abdominal scintigraphy using technetium-99m hexamethylpropylene amine oxime-labelled leucocytes and ileocolonoscopy in patients with spondyloarthropathies and without clinical evidence of inflammatory bowel disease. Rheumatology 1999;38:543–6. El Maghraoui A. Osteoporosis and ankylosing spondylitis. Joint Bone Spine 2003; [in press]. Bezza A, El Maghraoui A, Ghadouane M, Tabache F, Abouzahir A, Abbar M. Idiopathic retroperitoneal fibrosis and ankylosing spondylitis. A new case-report. Jt Bone Spine 2002;69:502–5. Richards AJ. Ankylosing spondylitis and pulmonary apical fibrosis. Proc R Soc Med 1974;67:45–6. Appelrouth D, Gottlieb NL. Pulmonary manifestations of ankylosing spondylitis. J Rheumatol 1975;2:446–53. Feltelius N, Hedenstrom H, Hillerdal G, Hallgren R. Pulmonary involvement in ankylosing spondylitis. Ann Rheum Dis 1986;45:736– 40. Bouchea DK, Sundstrom WR. The pleuropulmonary manifestations of ankylosing spondylitis. Semin Arthritis Rheum 1989;18:277–81. Rumancik WM, Firooznia H, Davis Jr MS, Leitman BS, Golimbu C, Rafii M. Fibrobullous disease of the upper lobes: an extraskeletal manifestation of ankylosing spondylitis. J Comput Tomogr 1984;8: 225–9. Ferdoutsis M, Bouros D, Meletis G, Patrsourakis G, Siafakas NM. Diffuse interstitial lung disease as an early manifestation of ankylosing spondylitis. Respiration 1995;62:286–9. Hurwitz SS, Conlan AA, Krige LP. Fibrocavitating pulmonary lesions in ankylosing spondylitis. S Afr Med J 1982;61:168–70. Hansell DM, Kerr IH. The role of high-resolution tomography in the diagnosis of interstitial lung disease. Thorax 1991;46:77–84. Gamsu G, Klein JS. High-resolution computed tomography of diffuse lung disease. Clin Radiol 1989;40:554–6. Cortet B, Flipo RM, Rémy-Jardin M, Coquerelle P, Duquesnoy B, Rémy J. Use of high-resolution computed tomography of the lungs in patients with rheumatoid arthritis. Ann Rheum Dis 1995;54:815–9. Wells AU, Hansell DM, Corrin B, Harrison NK, Golstraw P, Black CM. High-resolution computed tomography as a predictor of lung histology in systemic sclerosis. Thorax 1992;53:738–42. Dunham CL, Kautz FG. Spondylarthritis ankylopoietica: a review of and report of twenty cases. Am J Med Sci 1941;201:232–50. Hamilton KA. Pulmonary disease manifestations of ankylosing spondylarthritis. Ann Intern Med 1949;31:216–27. Campbell AH, Mac Donald CB. Upper lobe fibrosis associated with ankylosing spondylitis. Br J Chest 1965;59:90–101. Jessamine AG. Upper lung lobe fibrosis in ankylosing spondylitis. Can Med Assoc J 1968;98:25–9. Lauritzen H, Medina G, Loken MD, Mac Donald F. Pulmonary disease in patients with ankylosing spondylitis. Am Rev Respir Dis 1968;98:126. Takkunenn J, Vuopala U, Isomaki H. Cardiomyopathy in ankylosing spondylitis. Ann Clin Res 1970;2:106–12. Davies D. Ankylosing spondylitis and lung fibrosis. Q J Med 1972;41: 395–417. Crompton GK, Cameron SJ, Langlands AO. Pulmonary fibrosis, pulmonary tuberculosis and ankylosing spondylitis. Br J Chest 1974;68: 51–6. Wolson AH, Rohwedder JJ. Upper lobe fibrosis in ankylosing spondylitis. Am J Roentgenol 1975;124:466–71.
501
[25] Rosenow CE, Strimlan CV, Muhm JR, Ferguson RH. Pleuropulmonary manifestations of ankylosing spondylitis. Mayo Clin Proc 1977; 52:641–9. [26] Manresa F, Lopez Munoz JA, Manresa FG, Rotes QGD. Fibrose pulmonaire cavitaire au cours de la spondylarthrite ankylosante. Nouv Presse Méd 1976;5:2467. [27] Touraine R, Bonnamour CH, Vignon G. Fibrose pulmonaire apicale extensive compliquant une spondylarthrite ankylosante. Nouv Presse Méd 1977;6:2319–22. [28] Brocard H, Gallouadec CH, Blanchon F, Chamouard JL, Grunenwald D. Fibrose pulmonaire destructrice et progressive associée à une spondylarthrite ankylosante. Ann Med Int 1978;129:273–7. [29] Chatel A, Blanchon F, Bigot JM, Cremniter D, Hélénon CH, Brocard H. Fibrose pulmonaire bi-apicale au cours de la spondylarthrite ankylosante. J Radiol 1980;61:451–5. [30] Peltier P, Germaud P, Martin M, Barrier J, Cottin S, Grolleau JY. Fibrose pulmonaire et pelvispondylite rhumatismale. Sem Hop Paris 1982;58:596–600. [31] Ducolone A, Vandevenne A, Fraisse P, Burghard G. Fibrose pulmonaire et spondylarthrite ankylosante. Rev Pulm Clin 1984;40:145–9. [32] Agguilaniu B, Blanc-Jouvan F, Lévy P, Geraads A, Brambilla C, Paramelle B. Pelvispondylite rhumatismale et lésions scléreuses non tuberculeuses des sommets. Rev Pneumol Clin 1984;40:71–3. [33] Leménager J, L’Hirondel JL, Spauvajon H, Sesboue B. Manifestations respiratoires de la spondylarthrite ankylosante (50 observations). Rev Pneumol Clin 1987;43:8–12. [34] Chakera TMH, Howarth FH, Kendall MJ, Lawrence DS, Whitfield AGW. The chest radiograph in ankylosing spondylitis. Clin Radiol 1975;26:455–60. [35] El Maghraoui A, Chaouir S, Bezza A, Tabache F, Abouzahir A, Ghafir D. Thoracic high-resolution CT in patients with ankylosing spondylitis and without respiratory symptoms. Ann Rheum Dis 2003; 62:185–6. [36] Falashi F, Sbragia P, Trippi D, Neri E, Cambi L, Federiconi E. Study of bullous fibrosis of the upper lobes in ankylosing spondylitis with high-resolution computerized tomography. Radiol Med 1996;92:358– 62. [37] Casserly IP, Fenlon HM, Breatnach E, Sant SM. Lung findings on high-resolution computed tomography in idiopathic ankylosing spondylitis. Correlation with clinical findings, pulmonary function testing and plain radiography. Br J Rheumatol 1997;36:677–82. [38] Fenlon HM, Casserly I, Sant SM, Breatnach E. Plain radiographs and thoracic high-resolution computed tomography in patients with ankylosing spondylitis. Am J Roentgenol 1997;168:1067–72. [39] Turetschek K, Ebner W, Fleischmann D, Wunderbaldinger P, Erlacher L, Zontsich T. Early pulmonary involvement in ankylosing spondylitis: assessment with thin-section CT. Clin Radiol 2000;55: 632–6. [40] Senocak O, Manisaly M, Ozaksoy D, Sevinc C, Akalyn E. Lung parenchyma changes in ankylosing spondylitis: demonstration with high-resolution CT and correlation with disease duration. Eur J Radiol 2003;45:117–22. [41] Kiris A, Ozgocmen S, Kocakoc E, Ardicoglu O, Ogur E. Lung findings on high-resolution CT in early ankylosing spondylitis. Eur J Radiol 2003;47:71–6. [42] El Maghraoui A, Chaouir S, Abid A, Bezza A, Tabache F, Achemlal L. Lung findings on thoracic high-resolution computed tomography in patients with ankylosing spondylitis. Correlations with disease duration, clinical findings, and pulmonary function testing. Clin Rheumatol 2004;23:123–8. [43] Eulry F, Marotel C, Chevalier D, Crozes P, Magnin J. Atteinte respiratoire de la spondylarthrite ankylosante: intérêt de l’exploration fonctionnelle respiratoire précoce. Sem Hop Paris 1995;71:770–6. [44] Fournié B, Boutes A, Dromer C, Sixou L, Le Guennec P, Granel J. Prospective study of anterior chest wall involvement in ankylosing spondylitis and psoriatic arthritis. Rev Rhum Engl Ed 1997;64:22–5.
502
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
[45] Cerraholugu L, Unlu Z, Can M, Goktan C, Celik P. Lumbar stiffness but not radiographic changes relate to alteration of lung function tests in ankylosing spondylitis. Clin Rheumatol 2002;21:275–9. [46] Girard J, Louyot P, Sadoul P, Graimprey J. Les conséquences ventilatoires de la rigidité thoracique secondaire à la spondylarthrite ankylosante. Sem Hop Paris 1956;40:2300–10. [47] Wendling D, Dalphin JC, Toson B, Gibey R, Racadot E, Fellman D. Etude du lavage broncho-alvéolaire au cours de la spondylarthrite ankylosante. Rev Rhum 1990;57:393–5. [48] Kchir MM, Mtimet S, Kochbati S, Zouari R, Ayed M, Gharbi T. Bronchoalveolar lavage and transbronchial biopsy in spondylarthropathies. J Rheumatol 1992;19:913–6. [49] Bonnet D, Jeandel P, Chouc PY, Molinier S, Raphenon G, Martet G. Etude cytologique du poumon profond au cours des spondylarthropathies. Rev Pneumol Clin 1993;49:100–5.
[50] Scherak O, Kolartz G, Popp W, Wottawa A, Ritschka L, Braun O. Lung involvement in rheumatoid factor-negative arthritis. Scand J Rheumatol 1993;22:225–8. [51] Zorab PA. The lungs in ankylosing spondylitis. Q J Med 1962;123: 267–77. [52] Cohen AA, Natelson EA, Fechner RE. Fibrising interstitial pneumonitis. Chest 1971;59:369–71. [53] Stewart MR, Ridyard JB, Pearson JD. Regional lung function in ankylosing spondylitis. Thorax 1976;31:433–7. [54] Bouvier M, Guyonnet G, Tebib J, Massonet B, Azzar D. Fibrose apicale de la spondylarthrite ankylosante, apport de la clairance mucociliaire. Presse Med 1988;17:1493–4. [55] Parkin A, Phil D, Robinson PJ, Hickling P. Regional lung ventilation in ankylosing spondylitis. Br J Radiol 1982;55:833–6. [56] Jeandel P, Bonnet D, Chouc PY, Molinier S, Raphenon G, Martet G. Mise en évidence d’une alvéolite pulmonaire infraclinique au cours des spondylarthropathies. Rev Rhum 1994;61:301–9.
Review
Pleuropulmonary involvement in ankylosing spondylitis Abdellah El Maghraoui * Rheumatology and Rehabilitation Unit, Military Teaching Hospital Mohammed V, Rabat, Morocco Received 8 December 2003; accepted 13 May 2004 Available online 07 August 2004
Abstract Pleuropulmonary involvement was long described as an uncommon and late event in the course of ankylosing spondylitis (AS). This belief was based on studies that relied on symptoms and chest radiographs to evaluate the lungs. However, pleuropulmonary involvement in AS patients is usually asymptomatic, and the early lesions are undetectable on chest radiographs. Apical fibrosis, interstitial infiltrates, and pleural thickening were considered to be the main patterns. However, the introduction of high-resolution computed tomography (HRCT) has led to the description of many pulmonary abnormalities that are clinically silent and undetectable on plain radiographs. These abnormalities mainly affect the interstitium and have no influence on respiratory function, which is dependent on the severity of chest wall inflammation or ankylosis in recent-onset and established AS, respectively. Cytological and histological studies suggest that, in common with uveitis and aortic regurgitation, the structural lung changes shown by HRCT may be specific of AS. © 2005 Elsevier SAS. All rights reserved. Keywords: Ankylosing spondylitis; Computed tomography of the chest; Lungs; Apical fibrosis; Lung function testing
1. Introduction Ankylosing spondylitis (AS) is the archetypal spondyloarthropathy and as such is strongly associated with the HLA B27 antigen. The disease primarily affects the axial joints [1], most notably the sacroiliac joints. Extraarticular manifestations are represented by uveitis, aortic regurgitation, bowel disease, and other systemic disorders [2–4]. Respiratory abnormalities have been reported in 0% to more than 30% of patients with AS [5–8]. Chest wall rigidity related to involvement of the costovertebral joints is responsible for a restrictive ventilatory pattern. In addition, AS can cause pleuropulmonary lesions, of which the most common are apical fibrosis, interstitial infiltrates, and pleural thickening [9–11]. These lesions are usually asymptomatic, and in their early stages they are not visible on chest radiographs. As a result, pleuropulmonary lesions were long believed to be uncommon and delayed manifestations of AS. However, over the last decade high-resolution computed tomography (HRCT) has been found superior over plain radiography in detecting early pulmonary lesions due to various systemic diseases * Tel.: +212-61-54-7190; fax: +212-37-62-6179. E-mail address: [email protected] (A. El Maghraoui). 1297-319X/$ - see front matter © 2005 Elsevier SAS. All rights reserved. doi:10.1016/j.jbspin.2004.05.006
including rheumatoid arthritis, systemic sclerosis, and AS [12–15]. In patients with AS, HRCT has shown not only the classic lung lesions, but also a wide variety of subtle abnormalities undetectable by plain radiography. The relevance of these abnormalities remains to be determined. This review article starts with a description of the steps that led to the recognition of pleuropulmonary abnormalities associated with AS. Then, the frequency and distribution of the abnormalities are discussed, with emphasis on differences related to the detection methods used (plain radiographs, HRCT, lung function tests (LFTs), bronchoalveolar lavage (BAL), and histological studies of transbronchial biopsy or autopsy specimens). The review ends with a presentation of current pathophysiological hypotheses and of the impact of pleuropulmonary manifestations on the everyday management of patients with AS. 2. Historical overview The first report of pleuropulmonary disease in patients with AS was written in 1941 by Dunham and Kautz [16], who described four cases of apical tuberculosis among 20 patients with AS. From a literature review, they concluded that pulmonary tuberculosis was more common among AS patients
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
than in the population at large. In 1949, Hamilton [17] also reported an excess risk of pulmonary tuberculosis in AS patients and suggested that this finding might reflect diagnostic error. Two of his patients had apical fibrosis that was ascribed to recurrent bacterial infection. In 1965, Campbell and Mac Donald [18] described six new cases of apical fibrosis and suggested for the first time that this abnormality might be an extraarticular manifestation of AS, similar to aortic regurgitation and uveitis. They ruled out the other causes of apical fibrosis, including recurrent infection, pulmonary tuberculosis, fungal infections, radiation-induced fibrosis, sarcoidosis, syphilis, pneumoconiosis, and rheumatoid disease. Three years later, Jessamine [19] reported seven additional cases of apical fibrosis in AS patients. Although microbiological studies were negative, five of these seven patients were given antitubercular drugs. Histological specimens were obtained from six patients and showed intraalveolar fibrosis, hyalinization, and elastic tissue degeneration. Jessamine [19] pointed out that these abnormalities resembled those seen in aortic valves from AS patients with aortic regurgitation. The same year, Lauritzen et al. [20] reported a retrospective review of 60 cases of AS: 11 patients had fibrocavitating lesions of the lung apices but only two had pulmonary tuberculosis. In a 1970 study of cardiomyopathy in AS patients, Takkunenn et al. [21] obtained radiographs from 55 AS patients and found apical fibrosis in seven patients. In 1972, Davies [22] described six new cases and found 50 previously published cases of “nontuberculous apical fibrosis” in AS patients. The typical patient was a male with advanced axial AS, onset of pulmonary symptoms several years after the first joint symptoms, distinctive radiological changes in the upper lobes, and a high rate of Aspergillus superinfection. Davies [22] pointed out the differences between the pulmonary abnormalities associated with AS and those seen in patients with rheumatoid arthritis, and he agreed with Campbell and Mac Donald [18] that apical fibrosis was a specific extraarticular manifestation of AS. In a 1974 study of 225 AS patients, Crompton et al. [23] found that 14 (6%) patients had apical fibrosis and that five (2%) had pulmonary tuberculosis. Wolson and Rohwedder [24] reported in 1975 that two of 52 AS patients had apical fibrosis. At the Mayo Clinic, Rosenow et al. [25] retrospectively reviewed the chest radiographs of 2080 AS patients and found 26 (1.2%) cases of apical fibrosis. Subsequently, case-reports were written by Manresa et al. [26] in 1976, Touraine et al. [27] in 1977, Brocard et al. [28] in 1978, Chatel et al. [29] in 1980, Peltier et al. [30] in 1982, Ducolone et al. [31] and Agguilaniu et al. [32] in 1984, and Leménager et al. [33] in 1987.
3. Chest radiographs Rates of pleuropulmonary involvement were low in studies that used chest radiographs as the only imaging method. In the largest case-series, the rate of occurrence was 1.2%
497
Table 1 Reported rates of occurrence of pulmonary abnormalities visible on chest radiographs in patients with AS
Chakera et al. [34] Rosenow et al. [25] Leménager et al. [33] Bouchea and Sundstrom [8] Casserly et al. [37] Turetschek et al. [39] El Maghraoui et al. [35]
Number of patients 42 2080 50 32 26 21 55
Abnormal chest radiograph N (%) 19 (45.2) 28 (1.3) 10 (20) 11 (34.3) 4 (15.3) 0 (0) 2 (3.6)
[25]. The only abnormalities detected in these studies were apical fibrosis, pleural thickening, bronchiectasis, and interstitial disease. Apical fibrosis develops in the upper lung lobes and follows a progressive course. Pulmonary fibrosis associated with AS is consistently located in the upper part of the lungs, in sharp contrast to the preferential involvement of the lung bases at the early stages of pulmonary fibrosis due to other systemic diseases. Thickening of the apical pleura is a common concomitant of apical fibrosis, indicating that the two abnormalities are directly related to each other. Apical pleural thickening can antedate the development of apical fibrosis [33,34]. Apical bronchiectasis has been ascribed to traction on the airways by the fibrotic lung parenchyma. Table 1 reports the rates of occurrence of pleuropulmonary abnormalities found by plain radiography in patients with AS.
4. High-resolution computed tomography Falashi et al. [36] wrote the first description of early pulmonary lesions detected by HRCT, in four patients with AS. Since then, including the report of our preliminary findings [35], only five prospective studies of pleuropulmonary lesions detected by HRCT in AS patients have been published. Table 2 recapitulates the HRCT changes found in these studies. Casserly et al. and Fenlon et al. [37,38] reported HRCT abnormalities in 69% of 26 AS patients, of whom only 15% had abnormal chest radiograph findings. Similarly, Turetschek et al. [39] found pulmonary abnormalities by HRCT in 71% of 21 patients, Senocak et al. [40] in 85% of 20 patients, and Kiris et al. [41] in 64% of 28 patients; in our study, the proportion was 56% (21/55) [42]. Pooling the findings from these studies yields a rate of occurrence of about two in three patients, with the most common abnormalities being pleural thickening (20%), parenchymal bands (19.3%), micronodules (17.3%), and subpleural bands (12.6%). Overall, apical fibrosis was found in 9.3% of patients, bronchiectasis in 11.3%, and emphysema in 9.3%. Bronchiectasis was usually related to traction by fibrotic tissue. Apical fibrosis and bronchiectasis are irreversible, whereas ground-glass densities (6.6% of patients) reflect alveolitis that may be reversible under treatment. Figs. 1–4 show examples of these lesions taken from our study.
498
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
Table 2 Reported rates of occurrence of pulmonary abnormalities visible by computed tomography in patients with AS
Number of patients Abnormal CT, n (%) Apical fibrosis, n (%) Emphysema, n (%) Bronchiectasis, n (%) Ground-glass densities, n (%) Nonspecific interstitial densities, n (%) Micronodules, n (%) Pleural thickening, n (%) Bronchial wall thickening, n (%) Subpleural band, n (%) Parenchymal band, n (%) Parenchymal retraction, n (%) Uneven interface, n (%) Blebs, n (%)
Casserly et al. [37] 26 19 (73) 2 (7.6) 4 (15) 6 (23) 1 (3.8) 11 (42.3) 1 (3.8) 1 (3.8) 4 (15) 6 (23) 8 (30.7) – 3 (11.5) 1 (3.8)
Turetschek et al. [39] 21 15 (71) 4 (19) 2 (9.5) 2 (9.5) 2 (9.5) – 2 (9.5) 6 (28.5) 6 (28.5) – – – – –
Fig. 1. Computed tomography of the chest in a 57-year-old man with an over two decade-long history of AS: retracting fibrosis in the left upper lobe (black arrow) with traction bronchiectasis (white arrow).
Senocak et al. [40] 20 17 (85) 3 (15) 3 (15) 3 (15) 3 (15) – 8 (40) 9 (45) 5 (25) 7 (35) 3 (15) – – –
Kiris et al. [41] 28 18 (64.2) – – – 2 (7.1) – 8 (25) – 2 (7.1) – 5 (17.8) – 1 (3.6) –
El Maghraoui et al. [1,35] 55 31 (56.3) 5 (9) 5 (9) 6 (10.9) 2 (3.6) 26 (47.2) 7 (12.9) 14 (25.4) – 6 (10.9) 13 (23.6) 12 (21.8) 4 (7.2) 7 (12.9)
Total 150 100 (66.6) 14 (9.3) 14 (9.3) 17 (11.3) 10 (6.6) – 26 (17.3) 30 (20.0) 16 (10.6) 19 (12.6) 29 (19.3) 12 (8.0) 8 (5.3) 8 (5.3)
Fig. 2. Computed tomography of the chest in a 66-year-old woman with a 5-year history of AS: interstitial disease with irregular interfaces, micronodules, parenchymal bands (black arrow), and severe primary bronchiectasis (white arrow).
In our study, we found no statistically significant differences regarding the rate of occurrence of HRCT abnormalities between smokers and nonsmokers, in keeping with findings by Turetschek et al. and Senocak et al. The rates of apical fibrosis and bronchiectasis increased with disease duration, whereas the rates of nonspecific abnormalities were not influenced by this factor. This finding suggests that nonspecific abnormalities may progress over time to yield apical fibrosis and bronchiectasis.
5. Lung function testing AS is associated with a pure restrictive pattern due to rigidity of the chest wall [43,44]. Vital capacity (VC) is decreased in proportion to disease severity; in patients with advanced
Fig. 3. Computed tomography of the chest in a 40-year-old man with a 16-year history of AS: apical fibrosis (black arrow), parenchymal bands, and subpleural emphysema (white arrow).
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
Fig. 4. Computed tomography of the chest in the same patient as in Fig. 3: ground-glass densities (black arrow) and parenchymal band (white arrow).
499
striking contrast to experience with rheumatoid arthritis. In a study of 14 AS patients, Kchir et al. [48] found that BAL was normal in 12 patients and showed lymphocytosis in two patients, of whom one was a smoker and the other had alveolar fibrosis. In a 1994 study by Bonnet et al. [49] in 34 patients with spondyloarthropathy but no pulmonary abnormalities (reactive arthritis, n = 4; AS, n = 19; and unclassifiable spondyloarthropathy, n = 11) and nine controls, lymphocytic alveolitis (with a predominance of CD8+ cells) was found in patients with early disease and neutrophilic alveolitis in those with advanced disease. The findings were independent from smoking status, nonsteroidal antiinflammatory drug use, and presence of airborne pollutants. The authors concluded that a specific mechanism caused the apical fibrosis but that upper lobe hypoventilation stimulated the fibrotic process. The same year, Scherak et al. suggested a similar interpretation [50].
diseases, VC is rarely less than 60% of predicted. Residual volume (RV) and functional residual capacity (FRC) are increased or normal, in contrast to findings in other restrictive diseases. Although expiratory muscle performance is enhanced, the chest wall rigidity causes muscle exhaustion, so that complete expiration cannot be achieved. Thus, the decrease in total lung capacity (TLC = VC + RV) is less marked than the decrease in VC. In our study [42], 16 of 55 patients had a restrictive pattern, and 11 of these 16 had abnormalities by HRCT (nonspecific interstitial changes in eight patients, bronchiectasis in two, emphysema in one, and interstitial syndrome in two). Nevertheless, we found no correlation between HRCT changes and LFT results. In contrast, LFT results were correlated with variables reflecting the severity of symptoms and structural damage related to AS, in keeping with earlier evidence that severity of axial disease but not presence of HRCT lesions governed the severity of the restrictive pattern [45,46]. LFT may show obstructive disease in smokers, but this finding is independent from the restrictive pattern (Table 3).
Histological studies have been done on transbronchial or percutaneous biopsies, operative specimens, and autopsy specimens [51,52]. Nonspecific interstitial fibrosis with degeneration of collagen and elastic fibers was found. Focal lymphocytic infiltrates were visible in some cases. In patients with advanced disease, cystic cavities were visible within the fibrous areas. No granulomas or signs of vasculitis were found. Kchir et al. [48] obtained transbronchial biopsies from 12 of their 14 AS patients and found interstitial fibrosis in five, with pleural adhesions. The few histological studies available in the literature indicate a high rate of interstitial disease with a potential for progression to fibrosis.
6. Bronchoalveolar lavage
More than 50 years after the seminal description, pleuropulmonary involvement by AS remains an enigma. Pathophysiological hypotheses are based on two main theories, the mechanical theory and the AS-specific theory.
To elucidate the pathophysiological mechanism underlying pulmonary lesions in patients with AS, Wendling et al. [47] performed BAL in 15 patients and 17 controls. Two patients had interstitial disease in the lung bases; no information is given on the pulmonary status in the remaining 13 patients. BAL showed no evidence of inflammation, in
7. Histological studies
8. Etiological and pathogenic hypotheses
8.1. Mechanical theory The chest wall rigidity seen in patients with AS may induce major ventilatory disturbances, with hyperventilation (Table 3) of the lung bases and severe hypoventilation of the apices.
Table 3 Reported rates of occurrence of lung function testing abnormalities in patients with AS Eulry et al. [43] Casserly et al. [37] Turetschek et al. [39] Senocak et al. [40] El Maghraoui et al. [1,35] a
LFT: lung function testing.
Number of patients 49 26 21 20 55
Normal LFT a, n (%) 42 (85.7) 16 (61.5) 9 (42.8) 12 (60) 37 (67.2)
Restrictive pattern, n (%) 4 (8.1) 7 (26.9) 12 (57.1) 6 (30) 16 (29)
Obstructive pattern, n (%) 3 (6.1) 3 (11.5) 0 (0) 2 (10) 2 (3.6)
500
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
According to Jessamine [19], apical hypoventilation may result in poor airway clearance, which in turn may cause chronic inflammation. However, studies of mucociliary function fail to support this hypothesis [53]. In addition, exposure to cigarette smoke and particulate matter undoubtedly promotes the development of fibrosis. The main piece of evidence supporting the mechanical theory is that most of the AS patients with pleuropulmonary disease had predominantly axial joint disease. Pleuropulmonary disease is rare in patients with peripheral joint disease but no axial involvement.
8.2. Disease-specific theory Several case-reports support the involvement of an ASspecific process. Thus, Campbell and Mac Donald [18] reported a case in which apical fibrosis developed 2 years before the first joint symptoms, well before the chest wall had lost its flexibility. Mucociliary function was studied by Bouvier et al. [54] in 10 patients with AS and normal chest radiograph findings. Mucociliary clearance was decreased in the lung apices. The authors suggested that apical fibrosis might be secondary to recurrent infections related to apical hypoventilation. They found no statistically significant differences in mucociliary function between the patients and the controls or between the upper and the lower lungs, and they concluded that apical fibrosis was produced by a specific mechanism related to the nature of AS. Parkin et al. [55] compared regional ventilation in 27 AS patients, including three with apical fibrosis, and 18 controls. Hypoventilation was found only in the apices of the patients with apical fibrosis. In contrast to the mechanical theory, the disease-specific theory is consistent with the differences commonly seen between the right and left lung apices of individual patients. HRCT studies also support a role for a disease-specific mechanism. In our study, HRCT abnormalities were found even in patients with mild symptoms and minimal structural damage that did not affect spinal alignment or mobility. Smoking is widely believed to promote apical fibrosis and interstitial inflammation [56]. Exposure to tobacco smoke is associated with increased counts of macrophages and neutrophils in the deep lung parenchyma. Thus a role for smoking in precipitating or worsening lung disease in AS patients is plausible. However, HRCT abnormalities were found in similar proportions of smokers and nonsmokers in our study. Similarly, Turetschek et al. studied only nonsmokers, yet found a high rate of HRCT abnormalities. Thus, the most likely explanation to lung abnormalities in AS patients is a disease-specific inflammatory process whose course is parallel to that of the joint manifestations. This hypothesis is in line with the results of BAL studies and other investigations.
9. Evaluation of pulmonary involvement in clinical practice A careful physical examination is crucial to the evaluation of the lungs in patients with AS. Chest expansion should be measured, and finger clubbing or other signs denoting hypoventilation should be sought. In patients who are free of symptoms, auscultation of the lung fields may reveal crackling rales that suggest fibrosis and indicate a need for further investigations. LFTs should be obtained to look for a restrictive pattern, most notably in patients with severe or longlasting AS. A chest radiograph should be obtained routinely, and HRCT should be considered in patients with normal radiographic findings but with abnormal physical findings. The HRCT abnormalities reported in studies of AS patients were subtle and showed no correlations with clinical symptoms or lung function. Therefore, it seems unreasonable to recommend routine HRCT in patients with AS. The high cost of HRCT is an additional reason for restricting the use of this investigation. Nevertheless, our results suggest that HRCT may be useful in selected patients, including those with a restrictive pattern that cannot be explained solely by mechanical factors, symptoms, or abnormal chest radiograph findings. In addition, with the increasing use of anti-TNF alpha agents in patients with AS, both HRCT and BAL may prove useful to rule out tuberculosis before treatment initiation in patients with chest radiograph abnormalities. Physicians should be thoroughly familiar with the broad array of abnormalities associated with AS so that they can confidently distinguish them from changes caused by tuberculosis or other infections.
10. Treatment and follow-up The presence of pleuropulmonary abnormalities in a patient with AS has no impact on the treatment strategy. Prophylactic measures including smoking cessation and physiotherapy to improve overall and respiratory function may be useful, however, in conjunction with disease-modifying antirheumatic drugs and nonsteroidal antiinflammatory agents. Whether early and appropriate treatment of AS decreases the risk of pleuropulmonary involvement is not known. Longitudinal studies of patients with recent-onset AS and normal lungs would be needed to address this issue. Close monitoring to ensure early detection of complications is in order in every patient with AS. The main complications associated with apical fibrosis are infections, most notably due to Aspergillus and atypical mycobacteria, and pneumothorax. Life-threatening pulmonary complications include massive hemoptysis due to Aspergillus infection and cardiorespiratory failure due to bacterial infection. Functional symptoms and physical signs denoting lung disease should be sought carefully at each visit.
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
References [1]
[2]
[3] [4]
[5] [6] [7]
[8] [9]
[10]
[11] [12] [13] [14]
[15]
[16] [17] [18] [19] [20]
[21] [22] [23]
[24]
El Maghraoui A, Bensabbah R, Bahiri R, Bezza A, Guedira N, HajjajHassouni N. Cervical spine involvement in ankylosing spondylitis. Clin Rheumatol 2003;22:94–8. El Maghraoui A, Dougados M, Freneaux E, Chaussade S, Amor B, Breban M. Concordance between abdominal scintigraphy using technetium-99m hexamethylpropylene amine oxime-labelled leucocytes and ileocolonoscopy in patients with spondyloarthropathies and without clinical evidence of inflammatory bowel disease. Rheumatology 1999;38:543–6. El Maghraoui A. Osteoporosis and ankylosing spondylitis. Joint Bone Spine 2003; [in press]. Bezza A, El Maghraoui A, Ghadouane M, Tabache F, Abouzahir A, Abbar M. Idiopathic retroperitoneal fibrosis and ankylosing spondylitis. A new case-report. Jt Bone Spine 2002;69:502–5. Richards AJ. Ankylosing spondylitis and pulmonary apical fibrosis. Proc R Soc Med 1974;67:45–6. Appelrouth D, Gottlieb NL. Pulmonary manifestations of ankylosing spondylitis. J Rheumatol 1975;2:446–53. Feltelius N, Hedenstrom H, Hillerdal G, Hallgren R. Pulmonary involvement in ankylosing spondylitis. Ann Rheum Dis 1986;45:736– 40. Bouchea DK, Sundstrom WR. The pleuropulmonary manifestations of ankylosing spondylitis. Semin Arthritis Rheum 1989;18:277–81. Rumancik WM, Firooznia H, Davis Jr MS, Leitman BS, Golimbu C, Rafii M. Fibrobullous disease of the upper lobes: an extraskeletal manifestation of ankylosing spondylitis. J Comput Tomogr 1984;8: 225–9. Ferdoutsis M, Bouros D, Meletis G, Patrsourakis G, Siafakas NM. Diffuse interstitial lung disease as an early manifestation of ankylosing spondylitis. Respiration 1995;62:286–9. Hurwitz SS, Conlan AA, Krige LP. Fibrocavitating pulmonary lesions in ankylosing spondylitis. S Afr Med J 1982;61:168–70. Hansell DM, Kerr IH. The role of high-resolution tomography in the diagnosis of interstitial lung disease. Thorax 1991;46:77–84. Gamsu G, Klein JS. High-resolution computed tomography of diffuse lung disease. Clin Radiol 1989;40:554–6. Cortet B, Flipo RM, Rémy-Jardin M, Coquerelle P, Duquesnoy B, Rémy J. Use of high-resolution computed tomography of the lungs in patients with rheumatoid arthritis. Ann Rheum Dis 1995;54:815–9. Wells AU, Hansell DM, Corrin B, Harrison NK, Golstraw P, Black CM. High-resolution computed tomography as a predictor of lung histology in systemic sclerosis. Thorax 1992;53:738–42. Dunham CL, Kautz FG. Spondylarthritis ankylopoietica: a review of and report of twenty cases. Am J Med Sci 1941;201:232–50. Hamilton KA. Pulmonary disease manifestations of ankylosing spondylarthritis. Ann Intern Med 1949;31:216–27. Campbell AH, Mac Donald CB. Upper lobe fibrosis associated with ankylosing spondylitis. Br J Chest 1965;59:90–101. Jessamine AG. Upper lung lobe fibrosis in ankylosing spondylitis. Can Med Assoc J 1968;98:25–9. Lauritzen H, Medina G, Loken MD, Mac Donald F. Pulmonary disease in patients with ankylosing spondylitis. Am Rev Respir Dis 1968;98:126. Takkunenn J, Vuopala U, Isomaki H. Cardiomyopathy in ankylosing spondylitis. Ann Clin Res 1970;2:106–12. Davies D. Ankylosing spondylitis and lung fibrosis. Q J Med 1972;41: 395–417. Crompton GK, Cameron SJ, Langlands AO. Pulmonary fibrosis, pulmonary tuberculosis and ankylosing spondylitis. Br J Chest 1974;68: 51–6. Wolson AH, Rohwedder JJ. Upper lobe fibrosis in ankylosing spondylitis. Am J Roentgenol 1975;124:466–71.
501
[25] Rosenow CE, Strimlan CV, Muhm JR, Ferguson RH. Pleuropulmonary manifestations of ankylosing spondylitis. Mayo Clin Proc 1977; 52:641–9. [26] Manresa F, Lopez Munoz JA, Manresa FG, Rotes QGD. Fibrose pulmonaire cavitaire au cours de la spondylarthrite ankylosante. Nouv Presse Méd 1976;5:2467. [27] Touraine R, Bonnamour CH, Vignon G. Fibrose pulmonaire apicale extensive compliquant une spondylarthrite ankylosante. Nouv Presse Méd 1977;6:2319–22. [28] Brocard H, Gallouadec CH, Blanchon F, Chamouard JL, Grunenwald D. Fibrose pulmonaire destructrice et progressive associée à une spondylarthrite ankylosante. Ann Med Int 1978;129:273–7. [29] Chatel A, Blanchon F, Bigot JM, Cremniter D, Hélénon CH, Brocard H. Fibrose pulmonaire bi-apicale au cours de la spondylarthrite ankylosante. J Radiol 1980;61:451–5. [30] Peltier P, Germaud P, Martin M, Barrier J, Cottin S, Grolleau JY. Fibrose pulmonaire et pelvispondylite rhumatismale. Sem Hop Paris 1982;58:596–600. [31] Ducolone A, Vandevenne A, Fraisse P, Burghard G. Fibrose pulmonaire et spondylarthrite ankylosante. Rev Pulm Clin 1984;40:145–9. [32] Agguilaniu B, Blanc-Jouvan F, Lévy P, Geraads A, Brambilla C, Paramelle B. Pelvispondylite rhumatismale et lésions scléreuses non tuberculeuses des sommets. Rev Pneumol Clin 1984;40:71–3. [33] Leménager J, L’Hirondel JL, Spauvajon H, Sesboue B. Manifestations respiratoires de la spondylarthrite ankylosante (50 observations). Rev Pneumol Clin 1987;43:8–12. [34] Chakera TMH, Howarth FH, Kendall MJ, Lawrence DS, Whitfield AGW. The chest radiograph in ankylosing spondylitis. Clin Radiol 1975;26:455–60. [35] El Maghraoui A, Chaouir S, Bezza A, Tabache F, Abouzahir A, Ghafir D. Thoracic high-resolution CT in patients with ankylosing spondylitis and without respiratory symptoms. Ann Rheum Dis 2003; 62:185–6. [36] Falashi F, Sbragia P, Trippi D, Neri E, Cambi L, Federiconi E. Study of bullous fibrosis of the upper lobes in ankylosing spondylitis with high-resolution computerized tomography. Radiol Med 1996;92:358– 62. [37] Casserly IP, Fenlon HM, Breatnach E, Sant SM. Lung findings on high-resolution computed tomography in idiopathic ankylosing spondylitis. Correlation with clinical findings, pulmonary function testing and plain radiography. Br J Rheumatol 1997;36:677–82. [38] Fenlon HM, Casserly I, Sant SM, Breatnach E. Plain radiographs and thoracic high-resolution computed tomography in patients with ankylosing spondylitis. Am J Roentgenol 1997;168:1067–72. [39] Turetschek K, Ebner W, Fleischmann D, Wunderbaldinger P, Erlacher L, Zontsich T. Early pulmonary involvement in ankylosing spondylitis: assessment with thin-section CT. Clin Radiol 2000;55: 632–6. [40] Senocak O, Manisaly M, Ozaksoy D, Sevinc C, Akalyn E. Lung parenchyma changes in ankylosing spondylitis: demonstration with high-resolution CT and correlation with disease duration. Eur J Radiol 2003;45:117–22. [41] Kiris A, Ozgocmen S, Kocakoc E, Ardicoglu O, Ogur E. Lung findings on high-resolution CT in early ankylosing spondylitis. Eur J Radiol 2003;47:71–6. [42] El Maghraoui A, Chaouir S, Abid A, Bezza A, Tabache F, Achemlal L. Lung findings on thoracic high-resolution computed tomography in patients with ankylosing spondylitis. Correlations with disease duration, clinical findings, and pulmonary function testing. Clin Rheumatol 2004;23:123–8. [43] Eulry F, Marotel C, Chevalier D, Crozes P, Magnin J. Atteinte respiratoire de la spondylarthrite ankylosante: intérêt de l’exploration fonctionnelle respiratoire précoce. Sem Hop Paris 1995;71:770–6. [44] Fournié B, Boutes A, Dromer C, Sixou L, Le Guennec P, Granel J. Prospective study of anterior chest wall involvement in ankylosing spondylitis and psoriatic arthritis. Rev Rhum Engl Ed 1997;64:22–5.
502
A. El Maghraoui / Joint Bone Spine 72 (2005) 496–502
[45] Cerraholugu L, Unlu Z, Can M, Goktan C, Celik P. Lumbar stiffness but not radiographic changes relate to alteration of lung function tests in ankylosing spondylitis. Clin Rheumatol 2002;21:275–9. [46] Girard J, Louyot P, Sadoul P, Graimprey J. Les conséquences ventilatoires de la rigidité thoracique secondaire à la spondylarthrite ankylosante. Sem Hop Paris 1956;40:2300–10. [47] Wendling D, Dalphin JC, Toson B, Gibey R, Racadot E, Fellman D. Etude du lavage broncho-alvéolaire au cours de la spondylarthrite ankylosante. Rev Rhum 1990;57:393–5. [48] Kchir MM, Mtimet S, Kochbati S, Zouari R, Ayed M, Gharbi T. Bronchoalveolar lavage and transbronchial biopsy in spondylarthropathies. J Rheumatol 1992;19:913–6. [49] Bonnet D, Jeandel P, Chouc PY, Molinier S, Raphenon G, Martet G. Etude cytologique du poumon profond au cours des spondylarthropathies. Rev Pneumol Clin 1993;49:100–5.
[50] Scherak O, Kolartz G, Popp W, Wottawa A, Ritschka L, Braun O. Lung involvement in rheumatoid factor-negative arthritis. Scand J Rheumatol 1993;22:225–8. [51] Zorab PA. The lungs in ankylosing spondylitis. Q J Med 1962;123: 267–77. [52] Cohen AA, Natelson EA, Fechner RE. Fibrising interstitial pneumonitis. Chest 1971;59:369–71. [53] Stewart MR, Ridyard JB, Pearson JD. Regional lung function in ankylosing spondylitis. Thorax 1976;31:433–7. [54] Bouvier M, Guyonnet G, Tebib J, Massonet B, Azzar D. Fibrose apicale de la spondylarthrite ankylosante, apport de la clairance mucociliaire. Presse Med 1988;17:1493–4. [55] Parkin A, Phil D, Robinson PJ, Hickling P. Regional lung ventilation in ankylosing spondylitis. Br J Radiol 1982;55:833–6. [56] Jeandel P, Bonnet D, Chouc PY, Molinier S, Raphenon G, Martet G. Mise en évidence d’une alvéolite pulmonaire infraclinique au cours des spondylarthropathies. Rev Rhum 1994;61:301–9.