Fabry disease, respiratory symptoms, and airway limitation – a systematic review

Background Fabry disease is an X-linked disorder caused by a deficiency of the lysosomal enzyme α-galactosidase A, resulting in accumulation of glycosphingolipids in multiple organs, primarily heart, kidneys, skin, CNS, and lungs. Materials and method A systematic literature search was performed using the PubMed database, leading to a total number of 154 hits. Due to language restriction, this number was reduced to 135; 53 papers did not concern Fabry disease, 19 were either animal studies or gene therapy studies, and 36 papers did not have lung involvement in Fabry disease as a topic. The remaining 27 articles were relevant for this review. Results The current literature concerning lung manifestations describes various respiratory symptoms such as dyspnoea or shortness of breath, wheezing, and dry cough. These symptoms are often related to cardiac involvement in Fabry disease as respiratory examinations are seldom performed. Pulmonary function tests primarily show obstructive airway limitation, but a few articles also report of patients with restrictive limitation and a mixture of both. No significant association has been found between smoking and the development of symptoms or spirometry abnormalities in patients with Fabry disease. Electron microscopy of lung biopsy and induced sputum show lamellar inclusion bodies (Zebra bodies) in the cytoplasm of cells in the airway wall. X-ray and CT scan have shown patchy ground-glass pulmonary infiltrations, fibrosis, and air trapping. Fibrosis diagnosed by high-resolution CT has not been significantly correlated with lung spirometry. Conclusion Consistent findings have not been shown in the current literature. Pulmonary function tests and registration of symptoms showed various results; however, there is a trend towards obstructive airway limitation in patients with Fabry disease. Further studies are needed to evaluate pathogenesis, progression, and the effects of treatment.

*Correspondence to: Camilla Kara Svensson, Fogedmarken 12, 3. th., 2200 Copenhagen N, Denmark, Email: c.k.s@live.dk F abry disease is an X-linked disorder caused by a deficiency of the lysosomal enzyme a-galactosidase A. This deficiency causes a progressive accumulation of glycosphingolipids (primarily globotriaosylceramid) in multiple organs, resulting in decreased perfusion, cellular dysfunction, tissue remodelling, development of fibrosis and, consequently, organ damage (1).
The early manifestations of Fabry disease appear in childhood and are usually acroparesthesia, angiokeratomas, and hypohidrosis. Complications such as left ventricular hypertrophy (LVH), kidney failure, visual disturbances, hearing loss, tinnitus, cerebrovascular involvement, and gastrointestinal disturbances develop later in life (1,2).
The general assumption has been that hemizygous men have more pronounced manifestations as well as earlier onset than found among heterozygous women, but newer theories suggest that women can be as severely affected as men. The most frequent mutations of the a-galactosidase A (GLA) gene are point mutations. The classic forms of Fabry disease are associated with mutations resulting in complete loss of function of the enzyme, whereas milder phenotypes and late variants might be associated with a substitution of a single amino acid (2,3). The observed different phenotypes in women might also be explained by variable X-inactivation (4).
According to Danish guidelines, Fabry disease is diagnosed by 1) low activity of the enzyme a-galactosidase , allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license.
A in leukocytes and increased excretion of globotriasylceramid in urine and 2) detection of a mutation in the GLA gene. Apart from general supportive treatments such as angiotensin converting enzyme (ACE) inhibitor and angiotensin receptor blockers, painkillers, antidepressants, acetylsalicylic acid, clopidogrel or similar, the current treatment of Fabry disease is enzyme replacement therapy (ERT). ERT of patients with Fabry disease might limit the tissue damage and slow down the development of end-stage illness of the involved organs, but this treatment has only been possible for the last decade, and a large number of patients might already have been at or close to end-stage organ failure before treatment became available.
Involvement of the lungs in Fabry disease, and the severity of this, is still an issue of dispute, as very few publications exist on this subject. The hypothesis is that patients with Fabry disease develop interstitial lung diseases due to tissue remodelling in the alveolar surroundings causing fibrosis induced by accumulation of glycosphingolipids. However, if the remodelling includes the bronchial tree, one could hypothesise that chronic airway limitation might develop, resulting in the development of progressive airway obstruction.
In this review, the existing literature is compared and the pulmonary manifestations in Fabry disease elucidated and summarised.

Materials and method
A systematic search was performed using the PubMed database on 2 July 2014, with the use of MeSH terms. To include literature to which MeSH terms have not yet been attached, words in text were also used in the literature search. Figure 1 demonstrates the search profile.

Exclusion criteria
With a language restriction using only papers written in English, Danish, Swedish, and Norwegian, the result was reduced to 135 articles.
A total of 53 out of the 135 articles did not concern Fabry disease but were present because the authors were named Fabry, they mentioned a Fabry-perot filter, or they simply mentioned Fabry disease in a passing remark, and they were therefore excluded.
Additionally, 11 animal studies and 8 laboratory studies or gene therapy studies were excluded.
A total of 36 papers that did not have lung involvement in Fabry disease as a main topic (or did not mention it in abstracts) were excluded. These involved heart, kidney, craniofacial involvement, priapism, quality of life, anaesthesia, and the administration and availability of drugs in different countries.

Final material
Of 135 papers, 108 were excluded due to the abovementioned criteria, whereas the remaining 27 articles were relevant and have been used in this review. None of the 27 papers were written in a Nordic language. Figure 2 illustrates the literature search.

Respiratory symptoms
The majority of the literature describes symptoms and signs of pulmonary involvement such as dyspnoea, wheezing, and dry cough in Fabry disease (5Á9).
A sleep study showed that 5 out of 22 patients with Fabry disease experienced Cheyne-Stokes respiration during sleep with no significant correlation to heart failure but a possible association with structural cerebral damage. Excessive daytime sleepiness in a patient with Fabry disease has also been described (16).
A study by Brown et al. (9) found 25 patients with dyspnoea, wheezing, and/or cough without simultaneous cardiac problems. Also, the study did not find any patients with symptoms under the age of 26 in the group of 25 patients (pB0.05).
Pulmonary function test A greater part of the literature described patients with an obstructive airflow limitation defined by forced expiratory volume in 1 sec/forced vital capacity (FEV1/FVC) B 70% (4, 7, 8, 17Á23). Furthermore, a review of Franzen et al. mentioned that obstructive lung disease was up to 10 times more frequent in patients with Fabry disease compared to the background population 24). Moreover, evidence has suggested that the pulmonary obstruction is age-dependent and increasing with age (9,20).
Magage et al. (19) defined obstruction of the small airways as only FEV1 ! 80% of that predicted and forced expiratory flow at 25Á75% of the FVC curve (%FEF25Á 75) B 70% of that predicted. Several patients met these criteria (19,21,22). The study showed that the values of %FEF25Á75 decreased with the same speed in males and There has been no significant responsiveness towards methacholine challenge testing (9) in patients with Fabry disease, and reversibility test with beta2-agonist was not significant (9,20).
A few cases focusing on spirometry results of patients with Fabry disease showed restrictive ventilatory impairment and/or a mixture of restrictive and obstructive airflow limitation in patients with Fabry disease (5,9,21,25).
A case report from 1961 described spirometry results suggesting mild emphysema (11).
A study by Koskenvou et al. (26) of 17 patients showed a significant association between dyspnoea and low ejection fraction (p B0.001) and a long QRS interval (p 00.04), while the symptoms did not have any association with the pulmonary function tests.
The study by Brown et al. (9) showed that dyspnoeic patients were significantly more likely to exhibit obstructive ventilatory impairment (pB0.005), but there was no correlation between cough and obstruction (p00.10).
One study of 39 patients with Fabry disease found a significantly lower% expected FVC in males than in females (p00.04) (22).
In a cohort of 15 patients undergoing a cardiopulmonary exercise test, 13 of them achieved maximal oxygen consumption (V O2 max ) less than the predicted range (21). Other studies (4,22) also showed patients with abnormal V O2 max .

Tobacco consumption
In seven patients with Fabry disease, airway obstruction was worse in patients who smoked compared to nonsmoking patients, but the difference was estimated as greater than what could be explained by smoking alone (10).
Koskenvou et al. (26) found no correlation between fibrosis (diagnosed by high-resolution CT) and lung spirometry.
Studies hypothesise that obstructive lung disease might be due to a constriction of airways caused by accumulation of glycosphingolipids in bronchial cells (24), smooth muscle (5), fibrosis, or hyperplasia of epithelia. Rosenberg et al. (10) suggest that obstruction may be due to loss of elastic recoil.
Enzyme replacement therapy Infusion of agalsidase beta (1 mg/kg) biweekly has shown some effect on symptoms and pulmonary function test (8). However, one case report (5) described worsening of total lung capacity in pulmonary function tests, and worsening of fibrosis found on CT and X-ray 49 months into treatment. It was, therefore, suggested that the progression of fibrosis cannot be reversed if a certain 'point of no return' is reached.
An interventional study of six patients receiving either ERT or placebo showed an increased exercise tolerance in patients receiving ERT (21). Tables (1Á5) summarize the results of the included studies and case reports.

Discussion
There is limited consensus regarding lung manifestations among patients suffering from Fabry disease. Few publications have investigated lung involvement in patients with Fabry disease, partly because Fabry disease is a rare disease, which has resulted in studies of small cohorts. Also, none of the publications mentioned in this review have evaluated the power of their respective studies. Therefore, there is a great risk of type-2 errors. Larger multicentre surveys would certainly improve our knowledge in this area. Moreover, the majority of the publications mentioned in this review were case reports, also with a low level of evidence.
The majority of the literature agreed that the following respiratory symptoms might develop: dyspnoea, wheezing, and dry cough. These findings were found independently of airflow limitation. There are divergent opinions regarding whether the symptoms are caused by pulmonary involvement or heart disease (9,25,26).
Pulmonary function tests of patients with Fabry disease also showed different results. The majority of the studies and case reports showed obstructive airflow limitation, but emphysema, restrictive ventilatory impairment, and/ or a mixture of restrictive and obstructive airflow limitation have also been described. However, one should be aware that the definition of obstructive, restrictive, and mixed airflow limitation was not homogenous throughout the literature, and not all of the articles have mentioned the individual spirometry values. In this review, obstruction of airways is defined as FEV1/FVC B 70%, and as the patients with Fabry disease are middle-aged patients, this seems to be acceptable and not leading to either overor under-estimation of chronic obstructive pulmonary disease (COPD).
There were no signs of variable airway disease such as those found in patients with asthma, as no airway hyperresponsiveness (9) or reversibility towards beta2agonist was found (9,20). This argues against airway   inflammation generally found in Th2 diseases such as asthma in Fabry disease. However, there is a need for further studies with a greater statistical power to decline beta2-agonist as a way of treatment in Fabry disease.
The presence of small airway disease among patients with Fabry disease was found in several studies (19,21,22). The study of Magage et al. showed that the values of %FEF25Á75 decreased with the same speed in males and females and in younger and older patients, which could indicate that an airway obstruction starts in the peripheral airways, and along with the progression of Fabry disease, the larger airways gradually become affected (19).
In their study population of 25 with respiratory symptoms, Brown et al. reported no patients under the age of 26, and only one article in the remaining literature in this review has reported a young patient with lung involvement (9-year-old with recurrent respiratory infections). The suggestion that pulmonary involvement is age-dependent and increasing with age is, therefore, strengthened. On the other hand, the onset of lung involvement in Fabry disease (the age with which the symptoms occur or whether or not they occur at all) might be dependent on phenotype. To date, no study comparing respiratory symptoms and pulmonary function test with phenotypes has been performed.
Development of emphysema or lung fibrosis with reduced gas exchange (DLco) has only been found in a few cases (4,5,7,8), which indicated limited statistical support for destruction of the lung parenchyma in patients with Fabry disease. Lung involvement in patients with Fabry disease might occur because of constriction of airways caused by smooth-muscle hyperplasia or accumulation of glycosphingolipids in bronchial cells. Rosenberg et al. (10) suggested that obstruction may be due to loss of elastic recoil, but up till now no study has been able to find the mechanisms behind lung involvement in Fabry disease.
It is well known that tobacco consumption is associated with the development of respiratory symptoms and COPD, but this review did not find support for increased effects of smoking on the development of lung symptoms in patients with Fabry disease. There seems to be no relationship between smoking and the degree of obstruction (9,20,22), and smokers with obstruction were not younger than non-smokers with obstruction (p00.86) (9). This lack of association might be a question of statistical power in the different studies.
ERT is used in patients with other organ manifestations, such as kidney failure and LVH, and has also been evaluated in a few patients with Fabry disease with lung involvement. ERT might reduce or stabilise symptoms and lung manifestations (5,8), but the current literature is not convincing. It was suggested (5) that the progression of fibrosis cannot be reversed if a certain 'point of no return' is reached.
The current patients receiving ERT had lived with Fabry disease for decades before treatment was initiated. If the hypothesis about a 'point of no return' where fibrosis, or maybe other manifestations, cannot be reversed, clinicians should consider the benefits of initiating treatment early in the disease development.

Conclusion
Examination of existing literature concerning lung involvement in patients with Fabry disease shows little consensus. Predominantly, the literature reports of patients having dyspnoea, wheezing, dry cough, obstructive airway limitation, and histological abnormalities such as lamellar inclusion bodies in epithelial cells and smoothmuscle hyperplasia. Smoking has not been significantly associated with increased obstructive airway limitation or respiratory symptoms. It is still debated whether or not cardiac involvement causes respiratory symptoms seen in patients with Fabry disease.
ERT has shown improvement in some aspects of lung involvement.
There is a need for further studies investigating lung involvement, progression, effects of smoking, and effects of ERT as well as treatment with beta2-agonists in patients with Fabry disease.