The majority of individuals affected with Fabry disease report profound gastrointestinal symptoms, such as diarrhoea, abdominal pain and early satiety, which can have a profoundly negative effect on their quality of life. Without treatment, the disease is unremitting and progressive, with increasing amounts of lipid-rich materials being stored in a variety of cells, including intrinsic neurones of the intestinal tract. With the advent of enzyme replacement therapy, it is hoped that the pathological changes may be preventable, and even reversible, and recent data suggest that gastrointestinal symptoms are especially responsive to treatment. This supports the case for closer monitoring and evaluation of the gastrointestinal manifestations of Fabry disease, which could provide important insights that may help to optimize treatment strategies.

Introduction

Fabry disease is a rare, X-linked deficiency of α-galactosidase A, affecting approximately 2–5 individuals per 1 million live births. Typical presenting symptoms include acroparaesthesiae, hypohidrosis, cutaneous angiokeratomas and cornea verticillata. In hemizygous males, the condition manifests almost universally, while a proportion of heterozygous females are also affected [1, 2].

Clinical manifestations usually become apparent in childhood, and the diagnosis is frequently suspected on the basis of a positive family history. Approximately 60% of patients with manifest Fabry disease report gastrointestinal symptoms [3]. This compares with a prevalence of 77% for neuropathic pain, 78% for sensorineural deafness, 30% for renal failure and 24% for cerebrovascular disease. Although not usually life threatening, gastrointestinal symptoms can have a major negative impact on quality of life [2, 4–11]. Systematic studies of large numbers of patients, however, are lacking. Early indications are that enzyme replacement therapy (ERT) may have a positive effect on gastrointestinal symptoms and may potentially provide early evidence of overall treatment efficacy [5, 6, 11].

As yet, the pathophysiology of the altered gastrointestinal function in Fabry disease is poorly understood, with detailed investigation having been confined mainly to small patient groups and case studies. Clearly, therefore, further investigation and data collection could have an impact on our understanding of this important aspect of Fabry disease.

Gastrointestinal features of Fabry disease

The most commonly reported gastrointestinal symptom in affected patients is diarrhoea, with frequent loose bowel motions and cramping abdominal pain. Patients often complain of postprandial symptoms, including diarrhoea, urgency and flatulence. Stool frequency in patients with Fabry disease can be as high as 10–12 soft or semi-solid motions per day; however, in contrast to inflammatory bowel diseases, such as Crohn's disease and ulcerative colitis, patients do not typically report rectal passage of mucus or blood. In many patients, episodes of diarrhoea are interspersed with periods of normal, or even reduced, bowel activity, when they may complain of constipation; this alternating pattern is reminiscent of irritable bowel syndrome (IBS) [12].

Patients may also experience a sense of early satiety after meals, epigastric discomfort and abdominal bloating. These symptoms may lead them to avoid meals and to report a reduced appetite. Again, the symptoms of abdominal discomfort and bloating associated with meals are features of IBS. Altered gastric emptying, which is ameliorated by treatment with metoclopramide, has also been reported in Fabry disease [9]. Sharp or stabbing epigastric pain, dyspepsia and heartburn suggest alternative diagnoses, such as peptic ulcer disease or gastrooesophageal reflux, and should prompt appropriate investigation and treatment.

MacDermot et al. described widespread gastrointestinal symptomatology in a cohort of adults with Fabry disease, and suggested that loss of weight and reduced body mass index (BMI) are associated with the disease [3]. Similarly, in a recent case report on the response to ERT, an improvement in gastrointestinal symptoms was accompanied by an increase in BMI [5]. Analysis of the current data in FOS – the Fabry Outcome Survey – on the other hand, does not suggest that reduced BMI is a feature of progressive Fabry disease, although the lack of reliable normative values makes this a difficult area to evaluate. Furthermore, serum protein, albumin, folate, vitamin B₁₂, calcium and phosphate levels are typically found to be normal, suggesting that malabsorption of nutrients is not a major clinical feature; alternative diagnoses should be sought if these parameters are abnormal [1, 4]. Intestinal xylose absorption is unaffected, and endoscopic biopsies usually show normal villous architecture at the light microscopic level [4]. Interestingly, however, anaemia is a recently identified feature of Fabry disease, and may relate to renal and cardiac dysfunction and to systemic inflammation [13]. In clinical practice, physicians may encounter patients with severe gastrointestinal symptoms, a history of recent loss of weight and a reduced BMI. The effects of a reduced appetite, nausea, vomiting and diarrhoea may be implicated in such patients, although other factors, such as cardiac and renal disease, may also be important.

When evaluating a patient with α-galactosidase A deficiency and gastrointestinal symptoms, a careful history and examination provide good preliminary evidence for deciding on the likely aetiology. Care should be taken to document the severity of symptoms. For example, with diarrhoea, what is the average daily frequency of bowel movements, do they relate to meals, and is early satiety after meals a feature? Simple investigations, such as a full blood count, urea, electrolytes, serum albumin, liver enzymes, C-reactive protein and erythrocyte sedimentation rate, provide important information about the general nutritional state, extent of renal involvement and presence of a systemic inflammatory response. Serological testing for coeliac disease and estimation of haematological parameters, such as levels of iron, ferritin, saturation of transferrin, folic acid and vitamin B₁₂, should be performed routinely to exclude gastrointestinal disease causing malabsorption or occult blood loss [14]. Chronic pancreatitis is another important cause of diarrhoea, abdominal pain and weight loss, and should be excluded by careful clinical evaluation and appropriate imaging – for example, with abdominal radiography or ultrasound scanning, or with computed tomography or magnetic resonance imaging if the clinical suspicion is high.

In Fabry disease, endoscopic and colonoscopic appearances are typically normal, although microscopic examination of biopsies may demonstrate accumulation of lipid within the cytoplasm of enteric neurones [4]. In many cases, upper endoscopy and colonoscopy will be indicated to eliminate diagnostic uncertainty.

Functional testing of intestinal physiology may ultimately prove highly valuable in the evaluation of gastrointestinal manifestations of Fabry disease. For example, altered motility, secretion and bacterial overgrowth within portions of the small bowel may all contribute to the pathogenesis of diarrhoea; hence, tests to determine the rates of gastric emptying and colonic transit, and the presence of bacterial overgrowth in the small intestine, may yield important information. The appropriate tests, such as scintigraphic gastric emptying scans, and the relevant hydrogen breath tests, are not universally available; if possible, specialists with access to suitable facilities should evaluate patients with Fabry disease and gastrointestinal symptoms.

Pathophysiology

Deficiency of α-galactosidase A results in progressive intracellular accumulation of the cell membrane-derived glycolipid, globotriaosylceramide (Gb₃). Gb₃ accumulates within the cytoplasm and lysosomes of affected cells, and may be detected microscopically as a foamy deposit within the cell or ultrastructurally as electron-dense intralysosomal striped 'zebra-like' 0.5–0.75 μm bodies [4]. Gb₃ accumulates most prominently in endothelial, perithelial and perineural cells, cardiomyocytes, renal glomerular cells, and neurones. Small unmyelinated neurones, such as those responsible for peripheral pain perception and those in the enteric nervous system, are most affected [15, 16].

The pathogenesis of gastrointestinal symptoms remains unknown, although it is plausible that accumulation of Gb₃ in neurones of the submucosal and myenteric nerve plexuses causes enteric neuropathy. Gb₃ also accumulates in intestinal smooth muscle, and a direct myopathic effect, or combined myopathy and neuropathy, may be important [17, 18]. Enteropathy affecting the sympathetic and parasympathetic divisions of the autonomic nervous system is also possible, and autonomic nerve involvement has been reported in Fabry disease [19]. There is evidence of delayed gastric emptying, as assessed by radionuclide studies, which improves after metoclopramide therapy [9]. This abnormality may arise from dysfunction of the intrinsic enteric nervous system, as well as from parasympathetic dysfunction, and is seen, for example, in diabetes mellitus and following surgical vagotomy [20]. Other reported changes include formation of diverticula and saccules within the intestinal wall, which may also be related to disruption of muscle and nerve function [17, 18].

The lack of α-galactosidase A, which catalyses the breakdown of specific glycosylceramides, may also affect the accumulation of other lipids, such as isoglobotriaosylceramide (iGb₃), which was recently identified as an endogenous ligand for the CD1d molecule [21]. CD1d is homologous to major histocompatability complex class I molecules and is considered critical for the function of natural killer T lymphocytes (NKT cells), which bear a restricted subset of invariant T cell receptor molecules and recognize glycolipid antigens [22]. Until recently, the likely endogenous or exogenous ligand for NKT cells and CD1d had not been identified, and a synthetic form of a marine sponge-derived glycolipid, α-galactosylceramide, was the most widely used experimental ligand. Recent research, however, has identified endogenous iGb₃ and glycolipids derived from the cell walls of Gram-negative bacteria as the more likely physiological and pathophysiological ligands [21, 23]. The CD1d-dependent pathway of NKT cell activation is considered highly important in the maintenance of host defence in the intestine, and it is possible that some aspects of gastrointestinal symptomatology in Fabry disease may relate to as yet unidentified and subtle immunological alterations [24].

From the preceding discussion, it is apparent that almost all the cells and systems that contribute to intestinal structure and function may be affected by Fabry disease; the various targets are illustrated in Figure 1. The inset photomicrograph demonstrates the accumulation of glycolipid within enlarged ganglion cells of the myenteric plexus. Less readily detected accumulation within epithelial cells, smooth muscle, myofibroblasts, vascular endothelial cells and immune cells may also have important effects that have yet to be defined.

Figure 1

Main features of the hollow gastrointestinal viscera that may be affected by Fabry disease, showing extrinsic nerve and blood vessel supply, and intrinsic structures including epithelial cells, immune cells, smooth muscle cells and the myenteric (Auerbach's) (more...)

Natural course

As with other manifestations of Fabry disease, gastrointestinal symptoms follow an unremitting progressive course. Data from FOS reveal that the occurrence of typical symptoms, such as diarrhoea and abdominal pain, increases progressively with time (Figure 2). Interestingly, heterozygous females are sometimes affected as severely as males, although there is a time lag in the development of symptoms in the majority of cases [1, 2, 8]. Thus, the curves charting the occurrence of symptoms for males and females diverge with time, although the final prevalence reached may be identical.

Figure 2

Reported occurrence of (a) diarrhoea and (b) abdominal pain with age among patients entered in FOS – the Fabry Outcome Survey. The curves for males and females diverge as a result of the more rapid onset of symptoms in males. Nonetheless, it is (more...)

It is potentially difficult to determine whether common symptoms, such as diarrhoea and abdominal pain, which have a high prevalence in the general population and may arise from many causes, are attributable to Fabry disease. FOS data provide a potential diagnostic aid in this regard. Examination of the occurrence of diarrhoea and abdominal pain in patients enrolled in FOS, for example, shows that the proportion of patients affected increases with age and that the rate of accumulation of cases is greater for men than for women (Figure 2). This divergence of the occurrence curves would be predicted for symptoms associated with progressive Fabry disease. By contrast, for a clinical feature such as haemorrhoids, which, a priori, is unlikely to be related to progressive Fabry disease, no difference between the sexes is seen and the curves do not diverge appreciably (Figure 3). Interestingly, such analysis suggests that constipation, which shows an equivalent prevalence in males and females, is also not related to progressive Fabry disease (Figure 3). More extensive and detailed collection of data is necessary to establish these relationships unequivocally, and to determine exactly which gastrointestinal symptoms and signs relate to Fabry disease, in order to use these more effectively to chart the progress of disease and the response to treatment.

Figure 3

Reported occurrence of (a) haemorrhoids and (b) constipation with age among subjects entered in FOS – the Fabry Outcome Survey. In contrast to Figure 2, there is a much lower overall occurrence of these symptoms and little divergence in the curves (more...)

Response to ERT

It is appropriate that most attention has focused on dermatological, neurological, renal and cardiac manifestations of Fabry disease, which are either most characteristic or have the most serious and life-threatening sequelae. However, it is now recognized that there is a large burden of gastrointestinal dysfunction among patients with Fabry disease, and a number of reports confirm that gastrointestinal symptoms are ameliorated by ERT [5, 6, 11]. Furthermore, debilitating gastrointestinal symptoms seem to respond rapidly to treatment, and the response may precede measurable changes in other parameters such as creatinine clearance and cardiac function, particularly for patients with severe established Fabry disease [25].

These preliminary data on the response of gastrointestinal manifestations to treatment are encouraging for the following reasons.

• Any demonstration that the established symptoms of Fabry disease can be reversed by therapy is highly welcome.
• If, as is surmised, the gastrointestinal manifestations are secondary to enteric neuronal damage, symptomatic improvement suggests that neuronal damage may be reversible.
• As the clinical response appears to be relatively rapid, it could be used in the evaluation and optimization of treatment for individual patients in a realistic time-frame, as well as to facilitate innovative treatment protocols (e.g. for heterozygous females or for paediatric patients) early in the course of the disease.

If gastrointestinal symptomatology can be ameliorated by treatment, this also presents a challenge to physicians to devise and evaluate specific, quantitative and reliable measures of Fabry-related gastrointestinal dysfunction that can be used in clinical studies. The FOS database, for example, already collects important relevant information, and the FOS Gastrointestinal Working Group is endeavouring to formulate a standardized, comprehensive and practical data collection tool that could be used as the basis of a 'Fabry gastrointestinal disease activity index', modelled, in part, on the successful and widely used Crohn's disease activity index [26].

Outstanding questions

Although Fabry disease is relatively rare, its study may yield insights that have broad relevance. Furthermore, with the recognition of clinical disease among heterozygotes, the true incidence of Fabry-related pathology may be higher than previously suspected. For example, in the context of cerebrovascular disease, heterozygosity for α-galactosidase A is now mooted as a potential risk factor for stroke at a young age [27]. In addition, diarrhoea and abdominal pain are two important symptoms in gastrointestinal practice that often, when they are not caused by identifiable organic pathology, indicate a form of IBS. This syndrome has an overall population prevalence of up to 15% in some communities [12]. The striking similarity in clinical findings between a typical case of Fabry-related gastrointestinal disease and diarrhoea-predominant IBS is summarized in Table 1. Is it possible that a proportion of patients with diarrhoea-predominant IBS are deficient in α-galactosidase A?

Table 1

Clinical features of the gastrointestinal manifestions of Fabry disease and diarrhoea-predominant irritable bowel syndrome.

The pathophysiology of IBS is not understood, although there are indications that visceral nerve and smooth muscle function may be disrupted, possibly mainly via alterations in serotoninergic signalling. If it emerges that Fabry symptoms do relate to enteric neuronal dysfunction, might this reveal a general mechanism by which IBS arises in the general population, with enteric neuropathic changes caused by other distinct disease processes? Such questions can be addressed only by further clinical research.

Reversibility of pathological changes, particularly those affecting non-renewable cells, such as cardiomyocytes, renal glomerular cells and neurones, is a key determinant of the potential benefit of ERT and will, in part, guide the timing of therapy for at-risk groups. Because gastrointestinal symptoms may arise from damage to enteric neurones, the rapid response of such symptoms to therapy is intriguing. Enteric neurones can be studied in tissue sections from endoscopic biopsies, and the response to therapy can potentially be quantified at the pathological level. Although this involves invasive testing, serious consideration should be given to incorporating such measurements in clinical trials and surveys such as FOS. Antibodies to Gb₃ may allow quantitative or semi-quantitative evaluation of lipid loading, and the effect of ERT on this parameter in a tissue that, unlike the heart and kidney, can be accessed without undue risk to the patient, may prove a valuable adjunct.

While it is evident that heterozygous females may manifest clinical features of Fabry disease, the phenomenon is unexplained, as circulating levels of α-galactosidase A are usually adequate. One possible explanation is that the abnormal enzyme exerts a dominant-negative effect. Alternatively, random inactivation of X chromosomes, which may be unbalanced, might provide an explanation, with markedly reduced tissue expression of α-galactosidase A in a proportion of critically important cells. This explanation would predict tissue mosaicism in the expression of α-galactosidase A and in the accumulation of Gb₃, which could potentially be investigated in intestinal tissue, where patches of epithelial cells in crypts and villi are derived from single stem cells [28]. Once more, an important pathophysiological question could be addressed by further research.

Finally, the intriguing possibility that α-galactosidase A deficiency results in the accumulation not only of glycolipids, such as Gb₃, but functionally critical species, such as iGb₃, which may have profound immunological effects, deserves further research. Alterations in CD1d-dependent antigen presentation and NKT cell function could have important effects in the gastrointestinal tract, and research in this area could provide important insights into gastrointestinal dysfunction, with implications beyond Fabry disease.

References

1.

Mehta A, Ricci R, Widmer U, Dehout F, Garcia de Lorenzo A, Kampmann C. et al. Fabry disease defined: baseline clinical manifestations of 366 patients in the Fabry Outcome Survey. Eur J Clin Invest. 2004;34:236–42. [PubMed: 15025684]

2.

Whybra C, Kampmann C, Willers I, Davies J, Winchester B, Kriegsmann J. et al. Anderson–Fabry disease: clinical manifestations of disease in female heterozygotes. J Inherit Metab Dis. 2001;24:715–24. [PubMed: 11804208]

3.

MacDermot KD, Holmes A, Miners AH. Anderson–Fabry disease: clinical manifestations and impact of disease in a cohort of 98 hemizygous males. J Med Genet. 2001;38:750–60. [PMC free article: PMC1734761] [PubMed: 11694547]

4.

O'Brien BD, Shnitka TK, McDougall R, Walker K, Costopoulos L, Lentle B. et al. Pathophysiologic and ultrastructural basis for intestinal symptoms in Fabry's disease. Gastroenterology. 1982;82:957–62. [PubMed: 6800874]

5.

Hoffmann B, Reinhardt D, Koletzko B. Effect of enzyme-replacement therapy on gastrointestinal symptoms in Fabry disease. Eur J Gastroenterol Hepatol. 2004;16:1067–9. [PubMed: 15371935]

6.

Dehout F, Roland D, Treille de Granseigne S, Guillaume B, Van Maldergem L. Relief of gastrointestinal symptoms under enzyme replacement therapy in patients with Fabry disease. J Inherit Metab Dis. 2004;27:499–505. [PubMed: 15303007]

7.

Ries M, Ramaswami U, Parini R, Lindblad B, Whybra C, Willers I. et al. The early clinical phenotype of Fabry disease: a study on 35 European children and adolescents. Eur J Pediatr. 2003;162:767–72. Epub 2003 Sep 20. [PubMed: 14505049]

8.

MacDermot KD, Holmes A, Miners AH. Natural history of Fabry disease in affected males and obligate carrier females. J Inherit Metab Dis. 2001;24:13–14. 11–12. [PubMed: 11758673]

9.

Argoff CE, Barton NW, Brady RO, Ziessman HA. Gastrointestinal symptoms and delayed gastric emptying in Fabry's disease: response to metoclopramide. Nucl Med Commun. 1998;19:887–91. [PubMed: 10581595]

10.

Nelis GF, Jacobs GJ. Anorexia, weight loss, and diarrhea as presenting symptoms of angiokeratoma corporis diffusum (Fabry–Anderson's disease). Dig Dis Sci. 1989;34:1798–1800. [PubMed: 2510982]

11.

Banikazemi M, Ullman T, Desnick RJ. Gastrointestinal manifestations of Fabry disease: clinical response to enzyme replacement therapy. Mol Genet Metab. 2005;85:255–9. [PubMed: 15939645]

12.

Cremonini F, Talley NJ. Irritable bowel syndrome: epidemiology, natural history, health care seeking and emerging risk factors. Gastroenterol Clin North Am. 2005;34:189–204. [PubMed: 15862929]

13.

Kleinert J, Dehout F, Schwarting A, de Lorenzo AG, Ricci R, Kampmann C. et al. Anemia is a new complication in Fabry disease: data from the Fabry Outcome Survey. Kidney Int. 2005;67:1955–60. [PubMed: 15840043]

14.

Tumer L, Ezgu FS, Hasanoglu A, Dalgic B, Bakkaloglu SA, Memis L, Dursun A. The co-existence of Fabry and celiac diseases: a case report. Pediatr Nephrol. 2004;19:679–81. [PubMed: 15085421]

15.

Dutsch M, Marthol H, Stemper B, Brys M, Haendl T, Hilz MJ. Small fiber dysfunction predominates in Fabry neuropathy. J Clin Neurophysiol. 2002;19:575–86. [PubMed: 12488789]

16.

Schiffmann R, Floeter MK, Dambrosia JM, Gupta S, Moore DF, Sharabi Y. et al. Enzyme replacement therapy improves peripheral nerve and sweat function in Fabry disease. Muscle Nerve. 2003;28:703–10. [PubMed: 14639584]

17.

Friedman LS, Kirkham SE, Thistlethwaite JR, Platika D, Kolodny EH, Schuffler MD. Jejunal diverticulosis with perforation as a complication of Fabry's disease. Gastroenterology. 1984;86:558–63. [PubMed: 6420224]

18.

Jack CI, Morris AI, Nasmyth DG, Carroll N. Colonic involvement in Fabry's disease. Postgrad Med J. 1991;67:584–5. [PMC free article: PMC2398882] [PubMed: 1924035]

19.

Sung JH. Autonomic neurons affected by lipid storage in the spinal cord in Fabry's disease: distribution of autonomic neurons in the sacral cord. J Neuropathol Exp Neurol. 1979;38:87–98. [PubMed: 122226]

20.

Stacher G, Lenglinger J, Bergmann H, Schneider C, Brannath W, Festa A. et al. Impaired gastric emptying and altered intragastric meal distribution in diabetes mellitus related to autonomic neuropathy? Dig Dis Sci. 2003;48:1027–34. [PubMed: 12822859]

21.

Zhou D, Mattner J, Cantu C 3rd,, Schrantz N, Yin N, Gao Y. et al. Lysosomal glycosphingolipid recognition by NKT cells. Science. 2004;306:1786–9. [PubMed: 15539565]

22.

Kawano T, Cui J, Koezuka Y, Toura I, Kaneko Y, Motoki K. et al. CD1d-restricted and TCR-mediated activation of Valpha14 NKT cells by glycosylceramides. Science. 1997;278:1626–9. [PubMed: 9374463]

23.

Mattner J, Debord KL, Ismail N, Goff RD, Cantu C 3rd,, Zhou D. et al. Exogenous and endogenous glycolipid antigens activate NKT cells during microbial infections. Nature. 2005;434:525–9. [PubMed: 15791258]

24.

van de Wal Y, Corazza N, Allez M, Mayer LF, Iijima H, Ryan M. et al. Delineation of a CD1d-restricted antigen presentation pathway associated with human and mouse intestinal epithelial cells. Gastroenterology. 2003;124:1420–31. [PubMed: 12730881]

25.

Tsambaos D, Chroni E, Manolis A, Monastirli A, Pasmatzi E, Sakkis T. et al. Enzyme replacement therapy in severe Fabry disease with renal failure: a 1-year follow-up. Acta Derm Venereol. 2004;84:389–92. [PubMed: 15370707]

26.

Best WR, Becktel JM, Singleton JW, Kern F Jr. Development of a Crohn's disease activity index. National Cooperative Crohn's Disease Study. Gastroenterology. 1976;70:439–44. [PubMed: 1248701]

27.

Giacomini PS, Shannon PT, Clarke JT, Jaigobin C. Fabry's disease presenting as stroke in a young female. Can J Neurol Sci. 2004;31:112–14. [PubMed: 15038481]

28.

Novelli MR, Williamson JA, Tomlinson IP, Elia G, Hodgson SV, Talbot IC. et al. Polyclonal origin of colonic adenomas in an XO/XY patient with FAP. Science. 1996;272:1187–90. [PubMed: 8638166]