**5. Immunodiagnostic approach**

Primary systemic vasculitides were reclassified based on ANCA serology, the presence of immune deposits *in situ*, and the size of the vessels involved. WG, MPA and CSS were subsumed in the group of ANCA‐associated vasculitides, which are characterized clinically by a WG, CSS or non‐granulomatous MPA inflammation commonly involving the respiratory tract and ear–nose–throat (ENT) region and by a necrotizing pauci‐immune (= no or minimal immune deposits) vasculitis typically affecting small‐ to medium‐sized vessels.

ANCA are a heterogeneous group of autoantibodies that can be subdivided by indirect immunofluorescence tests (IFTs) and by enzyme‐linked immunosorbent assays (ELISAs). IFTs can distinguish two major fluorescence patterns on ethanol‐fixed human granulocytes: one of these patterns, classic cANCA, is highly specific for WG, while the other, perinuclear pANCA, is commonly seen in MPA (rarely in WG), but may be detected in a wide variety of other autoimmune conditions (e.g. systemic lupus erythematosus, rheumatoid arthritis, Felty's syndrome and chronic inflammatory bowel diseases with associated disorders).

The clinical utility of cANCA as a diagnostic marker for WG was recently confirmed in a large prospective European study undertaken with sera from vasculitis patients (sensitivity 60%, specificity 95%) [6]. However, when employed as a routine screening method for WG (defined according to ACR criteria) in patients with suspected vasculitis, the sensitivity of cANCA in a recent prospective single‐centre study on 346 consecutive patients was only 28% (specificity for WG: 98%). The sensitivity rose to 83% if only biopsy‐proven WG was considered [7]. A meta‐analysis of 15 studies comprising 13 652 patients (including 736 cases of WG) yielded a pooled sensitivity of 66% and a specificity of 98% [8]. Taken together, these data show that the value of cANCA testing is limited by a rather low sensitivity; the greatest utility of cANCA testing may be in patients with suspected, but not yet proven, WG. This view is supported by a recent analysis of ANCA results in a large routine laboratory (Regional Immunology Laboratory, Belfast, UK). The overall positive predictive value for primary systemic vasculi‐ tides was 38% for all cANCA and only 20% for all pANCA. Specificity improved when only antinuclear antibody (ANA)‐negative samples with a high ANCA titre were considered (cANCA 90%, pANCA 60%) [9]. In most, but not all cases, titres correlated with disease activity. Rising titres should alert the clinician to an increased risk of exacerbation, but are generally not regarded as an indication for intensifying therapy [10].

ELISAs are used to specify further the target antigens of ANCA, namely proteinase 3 (PR3; cANCA‐positive samples have a 99% specificity for WG), myeloperoxidase (MPO; 80% specificity for MPA) [6], as well as less important target antigens such as cathepsin G, lacto‐ ferrin, lysozyme and human leucocyte elastase, which are not specific for any particular vasculitic disorder. Whether anti‐bactericidal permeability increasing protein (BPI) ANCA offer further diagnostic perspectives in vasculitis is still unclear [11-13].

So, ANCA are not specific for ANCA‐associated vasculitides and despite the high specificity of cANCA/PR3‐ANCA for WG and of MPO‐pANCA for MPA, an increasing number of 'false‐ positive' PR3‐/MPO‐ANCA have been described [10]. More recently, we and others have observed PR3‐ANCA in subacute bacterial endocarditis, a condition sometimes associated with vasculitis [14]. Still, because ANCA test results are usually available before histological analyses are completed, ANCA serology remains the most important tool in the diagnostic repertoire for ANCA‐associated vasculitides, especially in seriously ill patients suspected of having vasculitis. Under life‐threatening conditions, therefore, therapy should be commenced based on clinical and serological findings! An overview of predominant immune phenomena in systemic vasculitides associated with the hypersensitivity reaction types (and the serological markers) is given in Table 2.

The incidence of ANCA in CSS is much lower than in WG and MPA, and their immunodiag‐ nostic significance is limited [6]. However, active CSS is characterized by increased eosinophils in conjunction with strongly elevated IgE and eosinophil cationic protein values [15].

Furthermore, endothelial cell damage in active AAV is indicated by markedly increased serum thrombomodulin (sTM) values [15]: in CSS, high levels of sTM correlate closely with the soluble interleukin‐2 receptor, which has been shown to be a promising seromarker of disease activity in WG [16].

Because intermittent infections are a major differential diagnostic problem in seriously ill AAV patients (Table 3), a marker that distinguishes between the two conditions is urgently needed. Procalcitonin was recently shown to be normal in active autoimmune rheumatic disorders, but strongly elevated in concomitant bacterial infections and sepsis [17]. However, these findings have yet to be confirmed [18].

congestive heart failure and myocardial abscesses. IE also produces a wide variety of systemic signs and symptoms through several mechanisms, including both sterile and infected emboli

Infectious Causes of Vasculitis http://dx.doi.org/10.5772/55189 139

IE develops most commonly on the mitral valve, closely followed in descending order of frequency by the aortic valve, the combined mitral and aortic valve, the tricuspid valve, and, rarely, the pulmonic valve. Mechanical prosthetic and bioprosthetic valves exhibit equal rates

**1.** Bacteremia (nosocomial or spontaneous) that delivers the organisms to the surface of the

and various immunological phenomena [19-21].

In about 30% of patients MPO-ANCA or PR3-ANCA were detected.

**Table 2.** Differential diagnostic features of small vessel vasculitides

All cases of IE develop from a commonly shared process, as follows:

of infection.

a

c

bOr in the cryoprecipitates.

Modified from [16].

In particular in the respiratory tract.

valve
