Vasculitis in Cystic Fibrosis

Introduction

Introduction

Cystic fibrosis (CF), the commonest recessive genetic disorder affecting the caucasian population, is caused by mutations of the cystic fibrosis transmembrane conductance regulator (CFTR) gene (1, 2). The resulting defect in the CFTR protein causes an increased susceptibility to infections, persistent inflammation and progressive airway obstruction. In addition, extrapulmonary manifestations are common (3) In this article we will review the literature regarding the association between CF and vasculitis.

Vasculitis is defined as inflammation of blood vessel walls (4) and may occur de novo or as a result of a variety of insults including infection. Cases of vasculitis not associated with infection can be further classified depending on the size and distribution of the affected vessels. The most commonly used classification is based on the Chapel Hill Consensus Conference (5) and defines three major categories of vasculitis: large-vessel, medium-sized vessel and small-vessel (Table 1) .

Frequency of vasculitis in CF

Vasculitis is a recognised but unusual complication of CF (6-9). The literature mainly comprises of case reports and case series. Whilst vasculitis may present in childhood, the majority of cases occur in patients over 20 years of age (Table 2) . There appears to be no gender predilection. Although the cutaneous form of vasculitis is most frequently described, systemic involvement has been reported (Table 3) .

In a large retrospective study of approximately 500 patients with CF attending St Christopher Hospital for Children in Philadelphia, 16 cases of purpuric rash due to vasculitis were reported over a 10 year period (9). Similar results were found in a study of 430 CF patients attending the Brompton Hospital in London, with 12 cases of vasculitis being reported over a 22 years period (6). In a survey from the CF centre in Copenhagen (over a period of 5 years), 4 of the 156 patients developed hypergammaglobulinemic purpura, a condition characterised by the recurrent purpura of the legs together with an increased serum immunoglobulin concentration (10). In another report, one of 59 adult patients with CF attending St Vincent's Hospital in Dublin had purpuric eruptions that occurred during an acute episode of arthritis (7). The subsequent histological examination revealed leukocytoclastic vasculitis. In a more recent study investigating vasculitis in 66 patients with CF, 6 were reported to have secondary vasculitis (11). However, the authors did not provide data on the total number of CF patients attending the Brompton Hospital at the time of the study. In contrast, none of the 140 children with CF treated over a period of 6 years at Booth Hall Children Hospital in Manchester had vasculitis (12). Based on the above described surveys, frequency of vasculitis in CF can therefore, be estimated in the range of 2-3%. In comparison general population studies estimates incidence rate of vasculitis at 5 - 200 per million (13).

Aetiology of vasculitis in CF

The underlying mechanisms of vasculitis is not fully understood, but is likely to reflect a combination of either an atypical response to infection or drugs on a background of genetic predisposition (4). In CF related vasculitis, persistent bacterial airway colonisation has been suggested as one of the most likely contributory factors (8). The continuous presence of bacteria in the lungs of patients with CF induces a strong immunological response locally and systematically. In a study by Finnegan et al (6) looking at the role of bacteria as a stimulus for vasculitis, stainable staphylococcal antigen could only be found in one patient. It is possible that the presence of prolonged systemic inflammation is of greater relevance than primary infection with airway inflammation being present both chronically and during acute exacerbations (14-18). Whether the initial trigger is the passage of bacterial antigens from the airways into the blood stream, the triggering of inflammatory mediators, or a combination of both, CF related vasculitis is associated with the development of circulating immunocomplexes (8).

Immune complexes have been described in vasculitis associated with suppurative conditions such as bronchiectasis (19). In addition, a number of immune complexes have been described in CF patients without evidence of vasculitis (20). Examples of this include IgA, IgM and four subclasses of IgG, against lipopolysaccharide (LPS), mucoid exo-polysaccharide, and outer membrane proteins (21, 22). More importantly circulating immune complexes have been reported in CF with vasculitis (6, 23). However, it remains unknown, why vasculitis develops in only a small proportion of CF patients, in whom circulating immune complexes are present.

Hyperglobulinemia is common in adult CF patients, and IgG levels rise with age and severity of the disease (24, 25). Thus increased antigen levels as a result of hypergammaglobulinemia could contribute to the aetiology of vasculitis in CF. However, hyperglobulinemia associated with purpura have been described only in a small number of CF patients (10). As hypergammaglobulinemia, is more common in severe disease, its association with purpura may simply represent the immune response of the final stage of CF (26). In fact, half of the patients with hypergammaglobulinemia and purpura were reported to die within 6 months from the appearance of the rash (10). Similarly, it remains difficult to establish the role of cryoglobulins in vasculitis of CF. Purpura due to cryoglobulinemia has been described in one CF patient (27) but cryoglobulin levels have been measured in only a small proportion of patients.

Finally, patients with CF receive frequent courses of antibiotics and other therapies. Thus drug reactions leading to formation of immunocomplexes and vasculitis may be another possible explanation (23). In the series described by Finnegan et al (6) only one patient was thought to have drug related vasculitis. In this particular case vasculitis subsided after cessation of ranitidine.

Antineutrophil cytoplasmic antibodies in CF

Antineutrophil cytoplasmic antibodies (ANCA) are serological markers of vasculitis and are thought to have pathological role in these conditions (28). Indirect immunofluorescence microscopy reveals two staining patterns: cytoplasmic (cANCA) and perinuclear (pANCA). On immunoassay testing the majority of cANCA and pANCA are directed against myeloperoxidase and neutrophil proteinase 3 respectively. Whilst cANCA is strongly associated with Wegener's granulomatosis, pANCA is usually present in microscopic polyangiitis and Churg Strauss syndrome (4). ANCA have been described in a proportion of CF patients with vasculitis (6, 8). More specifically, sera from CF patients with and without vasculitis have been reported to contain ANCA directed against bactercidal permeability increasing protein (BPI) (11, 29, 30). BPI is a 55-kDa membrane associated cationic protein that forms an important part of defence system against gram negative bacteria, as it binds to LPS, it neutralises endotoxin and opsonizes bacteria (31-33).

Zhao et al (11) reported that 60 out of 66 patients with CF had IgA or IgG autoantibodies to the neutrophil granule protein BPI.In the small proportion of patients with secondary vasculitis, the levels of these antibodies were higher. The BPI autoantibody levels correlated with reduction in lung function, and the presence of secondary vasculitis. The association between anti BPI autoantibodies and severity of the lung disease was also observed by Mahadeva et al (29) who reported that 55% and 70% of 148 adults patients with CF, had IgA and IgG anti BPI antibodies respectively. Sediva et al (30) reported that 51 of 71 children with CF were positive for BPI ANCA. In contrast to adult patients, there was no correlation observed between anti-BPI antibodies and severity of the impairment of lung function.

Two types of ANCA relevant to CF have been proposed (30). Firstly there is the anti BPI ANCA which, is thought to be present early in the disease, and therefore presumed to be associated with the defect in CFTR. BPI may be more easily released from the azurophilic granules of the neutrophils, encouraging antibody formation. Consequently these antibodies would block BPI, resulting in ineffective immune response and infection with Pseudomonas species. In fact, Pseudomonas colonisation is associated with higher levels of IgA anti BPI antibodies, which inhibit BPI induced phagocytosis, and therefore may contribute to more severe lung damage (29). Conversely, anti PR 3 ANCA may be a result of recurrent infections occurring over many years (30). BPI expresses a protective activity against lipopolysaccharide induced injury on vascular endothelial cells (34). Thus anti BPI antibodies by interfering with this particular function may conciliate vasculitic inflammation. Whether this particular mechanism contributes to vasculitis in CF remains an issue of speculation.

Clinical presentation of vasculitis in CF

Clinical presentation of vasculitis is frequently non specific, and may include a number of constitutional symptoms such as fever, malaise, arthralgia or myalgia (4). Patients with CF and vasculitis usually present with cutaneous vasculitis (Table 3) . In addition, there have been reports of vasculitis in the joints (6, 7, 35), central nervous system (6), renal (6, 9) and gastrointestinal tract (23, 37, 38). A palpable purpuric rash is considered to be pathognomonic of vasculitis (4) and in CF literature this type of rash has been the most common manifestation of cutaneous vasculitis. The rash is usually in a form of macular eruptions that become non blanching and purpuric (9, 10, 27). It usually appears suddenly, and may be preceded by pruritus or a burning sensation. The rash occurs predominately on the lower limbs, mainly over the dorsa of the feet, ankles, and tibial surfaces, but may extend to the trunk upper limbs or the face (6-9). Whilst in the majority of patients lesions disappear within 1-2 weeks, in some, episodic purpura have been reported (9, 36). Non specific arthralgia and myalgia may be present resulting occasionally in immobility (6). Knee and ankle joints are most commonly affected (39). This overlap between joint disease and cutaneous vasculitis is characteristic of CF with almost a half of patients with purpuric rash reporting joint pains and around 8% of patients with arthralgia having a purpuric rash (39).

Central nervous system vasculitis has been described infrequently (6). In non CF cases, this type of vasculitis has been associated with poor prognosis (4). Finneghan et al (6) described one fatal case of cerebral vasculitis, in whom aggressive therapy with corticosteroids was unsuccessful. This patient had severe CF, and presented with purpuric rash, periorbital oedema and ascites. Computerised tomography (CT) of the head and cerebrospinal fluid examination did not assist in the diagnosis, which was confirmed on the post mortem examination of the brain tissue.

In general, renal vasculitis can be evident by the presence of microscopic haematuria (4). However, whilst haematuria has been a frequent finding in vasculitis of CF, renal biopsies have been rarely undertaken. For example, in the series from Brompton Hospital and St Christopher's Hospital, only two patients had renal biopsies performed, and these showed changes characteristic of Henoch-Schonlein purpura (6, 9). Thus the frequency of renal vasculitis in CF remains a matter of conjuncture.

Although CF patients frequently present with abdominal pain or vomiting, gastrointestinal vasculitis has rarely been reported (Table 3). McFarlane et al (38) described extensive deposits of immunoglobulins and complement immune complexes in gastrointestinal tract of 2 patients with CF. Similarly, Soter et al (23) reported immune complex vasculitis in the post mortem specimen of small intestine in one CF patient. More recently, Parameswran et al (37) described a CF patient with chronic blood loss due to gastrointestinal vasculitis. In this particular case, repeated blood transfusions and numerus investigations including laparotomy were required before the diagnosis was made based on the endoscopic findings of telangiectatic lesions of the gastric mucosa. Once treatment with corticosteroids was commenced, the patient's condition dramatically improved.

There is no clear association between infective exacerbations, or sputum microbiology and vasculitis (Table 2) . Vasculitis may occur with minimal lung disease however, in a large proportion of patients, purpuric rash may present within few months from death or during the final illness (6, 9, 12, 23, 27). In fact, as high as 75% to 90% of patients have been reported to not survive beyond 2 years after having been diagnosed with purpura (10, 12). Whether this association merely reflects or contributes to the final illness is unclear.

Investigations of patients with CF and vasculitis

The diagnosis of vasculitis frequently depends on clinical suspicion. Whilst there is no specific diagnostic test, tissue or blood eosinophilia may be present (4). Frequently a histopathological examination of the affected tissue is required. In CF where cutaneous vasculitis is most common, skin biopsies may aid in the diagnosis. These have frequently demonstrated leukocytoclastic vasculitis (6). Skin biopsy can also reveal cutaneous necrotizing venulitis, a condition that can mimic vasculitis. For example, Soter et la (23). demonstrated on skin biopsy of palpable purpura from two patients with CF the presence of perivenular infiltrate composed of neutrophilic leukocytes, fibrin, hypo-granulated mast cells, and endothelial cell necrosis, that was characteristic of necrotizing venulitis rather than vasculitis. Microscopic haematuria and proteinuria, raised serum urea or creatinine can suggest renal vasculitis. Screening for faecal blood, or endoscopy may be helpful in the diagnosing of the gastrointestinal vasculitis. Chest and sinus radiograph or CT scans may assist in detecting respiratory tract involvement. Laboratory assessment for antinuclear antibodies, rheumatoid factor, ANCA, complement, cryoglobulins or antibodies to hepatitis B and C may be of assistance in investigating patients with vasculitis in general (4). In CF vasculitis the levels of immunoglobulins, and circulating immunocomplexes have frequently been characteristically increased (6, 8, 9, 27). Complement levels may be normal raised or decreased. In addition, rheumatoid factor may occasionally be elevated, mainly in patients with associated arthropathy (39). Rarely purpura may be associated with cryoglobulinemia (27). In approximately 40% of patients with cutaneous vasculitis complicating CF and none of the controls with CF without vasculitis have been reported ANCA positive (6). In other studies ANCA were reported in CF patients without vasculitis (11, 29, 30, 40). ANCA therefore, are markers of vasculitis in CF albeit without great sensitivity.

Treatment and prognosis

Corticosteroids, immunosuppressive or cytotoxic therapy have been used in treatment of vasculitis (4). In the context of CF, corticosteroids are mandatory in vasculitis of central nervous and gastrointestinal systems, although the prognosis in central nervous system vasculitis has been perceived as poor (6, 37). In cases of cutaneous vasculitis, the duration of the rash can vary. In a proportion of patients the rash subsides spontaneously within a few days to a few weeks (6, 9, 36). Whilst the majority of cases would simply require supportive measures such as nonsteroidal anti-inflammatory and antihistaminic agents to relief pain and pruritus, in some cases corticosteroids may be necessary (9, 36). There has also been one case report of patients with recurrent purpuric rash that fully recovered in response to methotrexate (41). Hypergammaglobulinemic purpura has been generally associated with poor prognosis with the majority of patients not surviving 18 months from the diagnosis (10). In some plasmapheresis, may be indicated (10). Purpuric skin lesions of cutaneous necrotizing venulitis may resolve spontaneously or with corticosteroids (23). Renal vasculitis, as in the case of Henoch Schonlein purpura, may require corticosteroids and azathioprine (6, 9). Although the data on vasculitis due to cryoglobulinemia in CF are limited, the response to corticosteroids is likely to be favourable in this condition (27). In cases of drug induced vasculitis, the suspected agent should be withdrawn (6).

Conclusion

Vasculitis in CF has been associated with more advanced disease. Circulating immune complexes have been frequently reported suggesting underlying immune mechanism. Anti BPI antibodies of ANCA type have been described suggesting that these may contribute to the immunological and clinical aspects of CF. Whilst many reports described poor long-term survival of patients with purpura, this may be related to the severity of CF rather than deleterious effects of vasculitis. As evident from the case series described in this review, vasculitis is a rare complication of CF. However the true prevalence of vasculitis in CF may be confounded by the possible under reporting of the cases. Since more stringent data collection on CF and its complications have been introduced in the UK, Europe and North America, the information acquired may contribute to better understanding of vasculitis in CF.

Reference

1 Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z, et al. Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science 1989;245:1066-1073.

2 Rommens JM, Iannuzzi MC, Kerem B, Drumm ML, Melmer G, Dean M, et al. Identification of the cystic fibrosis gene: chromosome walking and jumping. Science 1989;245:1059-1065.

3 Koch C, Hoiby N. Diagnosis and treatment of cystic fibrosis. Respiration 2000;67:239-247.

4 Jennette JC, Falk RJ. Small-vessel vasculitis. N Engl J Med 1997;337 :1512-1523.

5 Jennette JC, Falk RJ, Andrassy K, Bacon PA, Churg J, Gross WL, et al. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum 1994;37:187-192.

6 Finnegan MJ, Hinchcliffe J, Russell-Jones D, Neill S, Sheffield E, Jayne D, et al. Vasculitis complicating cystic fibrosis. Q J M 1989;72:609-621.

7 Bourke S, Rooney M, Fitzgerald M, Bresnihan B. Episodic arthropathy in adult cystic fibrosis. Q J M 1987;64:651-659.

8 Hodson ME. Vasculitis and arthropathy in cystic fibrosis. J R Soc Med 1992;85:38-40.

9 Schidlow DV, Panitch HB, Zaeri N, Zenel J, Alpert BE. Purpuric rashes in cystic fibrosis. Am J Dis Child 1989;143:1030-1032.

10 Nielsen HE, Lundh S, Jacobsen SV, Hoiby N. Hypergammaglobulinemic purpura in cystic fibrosis. Acta Paediatr 1978;67:443-447.

11 Zhao MH, Jayne DR, Ardiles LG, Culley F, Hodson ME, Lockwood CM. Autoantibodies against bactericidal/permeability-increasing protein in patients with cystic fibrosis. Q J M 1996;89:259-265.

12 Phillips BM, David TJ. Pathogenesis and management of arthropathy in cystic fibrosis. J R Soc Med 1986;79:44-50.

13 Savage CO, Harper L, Cockwell P, Adu D, Howie AJ. ABC of arterial and vascular disease: vasculitis. B M J 2000;320:1325-1328.

14 Khan TZ, Wagener JS, Bost T, Martinez J, Accurso FJ, Riches DW. Early pulmonary inflammation in infants with cystic fibrosis. Am J Respir Crit Care Med 1995;151:1075-1082.

15 Armstrong DS, Grimwood K, Carzino R, Carlin JB, Olinsky, Phelan PD. Lower respiratory infection and inflammation in infants with newly diagnosed cystic fibrosis. B M J 1995;310:1571-1572.

16 Konstan MW, Hillard K, Norvell TM, Berger M. Bronchoalveolar lavage findings in cystic fibrosis patients with stable, clinically mild lung disease suggest ongoing infection and inflammation. Am J Respir Crit Care Med 1994;150:448-454.

17 Birrer P, McElvaney NG, Rudeberg A, Sommer CW, Liechti-Gallati S, Kraemer R, et al. Protease-antiprotease imbalance in the lungs of children with cystic fibrosis. Am J Respir Crit Care Med 1994;150:207-213.

18 Cantin A. Cystic fibrosis lung inflammation: early, sustained, and severe. Am J Respir Crit Care Med 1995;151:939-941.

19 Hilton AM, Hasleton PS, Bradlow A, Leahy BC, Cooper KM, Moore M. Cutaneous vasculitis and immune complexes in severe bronchiectasis. Thorax 1984;39:185-191.

20 Moss RB, Hsu YP. Isolation and characterization of circulating immune complexes in cystic fibrosis. Clin Exp Immunol 1982;47:301-308.

21 Kronborg G. Lipopolysaccharide (LPS), LPS-immune complexes and cytokines as inducers of pulmonary inflammation in patients with cystic fibrosis and chronic Pseudomonas aeruginosa lung infection. APMIS 1995;50:1-30.

22 Pier GB, Saunders JM, Ames P, Edwards MS, Auerbach H, Goldfarb J, et al. Opsonophagocytic killing antibody to Pseudomonas aeruginosa mucoid exopolysaccharide in older noncolonized patients with cystic fibrosis. N Engl J Med 1987;317:793-798.

23 Soter NA, Mihm MC, Colten HR. Cutaneous necrotizing venulitis in patients with cystic fibrosis. J Pediatr 1979;95:197-201.

24 Pedersen SS. Lung infection with alginate-producing, mucoid Pseudomonas aeruginosa in cystic fibrosis. APMIS 1992;28:1-79.

25 Wheeler WB, Williams M, Matthews WJJ, Colten HR. Progression of cystic fibrosis lung disease as a function of serum immunoglobulin G levels: a 5-year longitudinal study. J Pediatr 1984;104:695-699.

26 Matthews WJJ, Williams M, Oliphint B, Geha R, Colten HR. Hypogammaglobulinemia in patients with cystic fibrosis. N Engl J Med 1980;302:245-249.

27 Garty BZ, Scanlin T, Goldsmith DP, Grunstein M. Cutaneous manifestations of cystic fibrosis: possible role of cryoglobulins. Br J Dermatol 1989;121:655-658.

28 Kallenberg CG, Brouwer E, Mulder AH, Stegeman CA, Weening JJ, Tervaert JW. ANCA pathophysiology revisited. Clin Exp Immunol 1995;100:1-3.

29 Mahadeva R, Dunn AC, Westerbeek RC, Sharples L, Whitehouse DB, Carroll NR, et al. Anti-neutrophil cytoplasmic antibodies (ANCA) against bactericidal/permeability-increasing protein (BPI) and cystic fibrosis lung disease. Clin Exp Immunol 1999;117:561-567.

30 Sediva A, Bartunkova J, Kolarova I, Hrusak O, Vavrova V, Macek, M. et al. Antineutrophil cytoplasmic autoantibodies (ANCA) in children with cystic fibrosis. J Autoimmun 1998;11:185-190.

31 Weiss J, Olsson I. Cellular and subcellular localization of the bactericidal/permeability-increasing protein of neutrophils. Blood 1987;69:652-659.

32 Ooi CE, Weiss J, Elsbach P, Frangione B, Mannion B. A 25-kDa NH2-terminal fragment carries all the antibacterial activities of the human neutrophil 60-kDa bactericidal/permeability-increasing protein. J Biol Chem 1987;262:14891-14894.

33 Iovine NM, Elsbach P, Weiss J. An opsonic function of the neutrophil bactericidal/permeability-increasing protein depends on both its N- and C-terminal domains. Proc Natl Acad Sci USA 1997; 94:10973-10978.

34 Arditi M, Zhou J, Huang SH, Luckett PM, Marra MN, Kim KS. Bactericidal/permeability-increasing protein protects vascular endothelial cells from lipopolysaccharide-induced activation and injury. Infect Immun 1994;62:3930-3936.

35 Summers GD, Webley M. Episodic arthritis in cystic fibrosis: a case report. Br J Rheumatology 1986;25:393-395.

36 John EG, Medenis R, Rao S. Cutaneous necrotizing venulitis in patients with cystic fibrosis. J Pediatr 1980;97:505.

37 Parameswaran K, Keaney NP, Veale D. A case of chronic gastrointestinal blood loss in cystic fibrosis. Respir Med 1995;89:577-579.

38 McFarlane H, Holzel A, Brenchley P, Macfarlane R, Rose D, Weston V, et al. Immune complexes in cystic fibrosis. B M J 1975;1:423-428.

39 Turner MA, Baildam E, Patel L, David TJ. Joint disorders in cystic fibrosis. J R Soc Med 1997;90:13-20.

40 Efthimiou J, Spickett G, Lane D, Thompson A. Antineutrophil cytoplasmic antibodies, cystic fibrosis, and infection. Lancet 1991;337:1037-1038.

41 Quattrucci P, Cannuzzi M, Pigna M, Antonelli M. Methotrexate in the treatment of vasculitis in a Cystic Fibrosis patient. 20th European Cystic Fibrosis Conference, Brussels June 1995 1995;I005(Abstract)

42 Fradin MS, Kalb RE, Grossman ME. Recurrent cutaneous vasculitis in cystic fibrosis. Pediatr Dermatol 1987;4:108-111.