Skip Navigation


Rheumatology Advance Access originally published online on November 15, 2005
Rheumatology 2006 45(4):470-477; doi:10.1093/rheumatology/kei191
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
45/4/470    most recent
kei191v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by van Woerkom, J. M.
Right arrow Articles by Bijlsma, J. W. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van Woerkom, J. M.
Right arrow Articles by Bijlsma, J. W. J.
Related Collections
Right arrow Diagnostics and Imaging Procedures
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2005. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Clinical significance of quantitative immunohistology in labial salivary glands for diagnosing Sjögren's syndrome

J. M. van Woerkom, A. A. Kruize, P. J. Barendregt1, L. Kater, R. Hené, H. Bootsma2, R. J. H. Custers, J. W. G. Jacobs and J. W. J. Bijlsma

Department of Rheumatology and Clinical Immunology, University Medical Center Utrecht, Utrecht, 1 Department of Rheumatology, Erasmus University Medical Centre, Rotterdam and 2 Department of Rheumatology, University Medical Centre Groningen, Groningen, The Netherlands.

Correspondence to: J. M. van Woerkom, Department of Rheumatology and Clinical Immunology, F02.127 University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, The Netherlands. E-mail: jm.woerkom{at}gelre.nl


   Abstract
Top
 Abstract
Introduction
 Patients and methods
 Results
 Discussion
 References
 
Objectives. Because patients with primary Sjögren's syndrome (pSS) are at risk of developing other autoimmune phenomena and malignant lymphoma, it is important to distinguish pSS from non-Sjögren's (nSS) sicca syndrome. However, this distinction might be difficult because of the lack of a gold standard for pSS. We studied the clinical significance of quantitative immunohistology (QIH) in labial salivary glands for diagnosing pSS.

Methods. In a model mimicking the making of a clinical diagnosis, five experts diagnosed 396 patients as nSS, ‘indefinite’, pSS or secondary SS (sSS) using 25 clinical parameters. Patients were diagnosed twice, namely without (yielding gold-standard diagnoses) and with knowledge of QIH. The numbers of changes in diagnosis from ‘indefinite’ to ‘definite’ (nSS, pSS or sSS) or vice versa were compared. Patient groups with vs without a changed diagnosis in the four gold-standard diagnosis groups were compared regarding objective autoimmune parameters.

Results. Sensitivity, specificity, positive and negative predictive value for abnormal QIH in pSS vs nSS were 93, 86, 76 and 96%, respectively. Changes in diagnosis from ‘indefinite’ to ‘definite’ (31%) were found more often (P = 0.00) than changes from ‘definite’ to ‘indefinite’ (10%). Knowledge of QIH distinguished patient groups within the gold-standard nSS, indefinite and pSS patient group with regard to autoimmune parameters.

Conclusion. In view of the consequences of distinguishing pSS from nSS, these results point to an additional diagnostic role for QIH in clinical practice.

KEY WORDS: Primary Sjögren's syndrome, Sicca syndrome, Quantitative immunohistology


   Introduction
Top
 Abstract
 Introduction
Patients and methods
 Results
 Discussion
 References
 
Sjögren's syndrome (SS) is a generalized autoimmune disease of unknown aetiology characterized by lymphoid infiltration and functional deterioration of exocrine glands, primarily the salivary and lacrimal glands, leading to dry eyes (keratoconjunctivitis sicca) and dry mouth (xerostomia). Sjögren's syndrome may occur in the absence (primary SS; pSS) or presence of another systemic autoimmune disease such as rheumatoid arthritis (RA), systemic lupus erythematosus (SLE) or systemic sclerosis (secondary SS; sSS).

In general, after exclusion of other causes of dryness, a diagnosis of pSS might be easily established based on a combination of clinical, serological, ophthalmological and histopathological parameters. In some patients, however, it may remain difficult to distinguish pSS from non-Sjögren's (nSS) sicca syndrome because of the lack of a gold standard for the diagnosis pSS. Nevertheless, distinguishing pSS from nSS is important, since pSS patients are at risk not only for the development of organ-specific autoimmune disease like autoimmune hypothyroidism and primary biliary cirrhosis but also for the development of malignant lymphoma [1].

Focal sialadenitis of grade IV [2], corresponding to a lymphocytic focus score of >one focus/4 mm2 in a labial salivary gland (LSG) biopsy specimen is a widely accepted tool for histopathological confirmation of pSS [3]. However, a lymphocytic focus score (LFS) of >1 has also been found in other diseases (RA, SLE, AIDS, myasthenia gravis) and even in healthy subjects [4], consequently decreasing its specificity. On the other hand, smoking and treatment with corticosteroids have been shown to decrease the number of foci in the LSG of SS patients, decreasing the sensitivity of LFS [5, 6].

In the search for better diagnostic tools, attention has been paid to plasma cells in LSG. Since the 1970s it has been known that in the LSG lymphocytic infiltrate of SS patients, large amounts of immunoglobulin (Ig) G- and IgM-containing plasma cells are present, in contrast to healthy individuals whose LSG plasma cells almost exclusively produce IgA [7–10]. In 1989, de Wilde and co-workers further evaluated differences in distribution patterns of LSG plasma cells between SS and controls; they introduced quantitative immunohistology (QIH) as a new and sensitive criterion for the diagnosis of SS. Quantitative immunohistology is a histological method for detecting and calculating IgA-, IgG- and IgM-producing plasma cells in LSG biopsy specimens, using monoclonal antibodies against IgA, IgG and IgM. Using a formula in which the percentages of IgA- and IgG-containing plasma cells are used, they were able to reduce the number of false-positive diagnoses by approximately 50%, yielding a specificity of 95.4% and sensitivity of 100% in pSS [11]. In this paper, pSS was defined as the combination of xerostomia, ophthalmologically diagnosed keratoconjunctivitis sicca and an LFS >1.

In a more recent report from the same group, the high sensitivity and specificity of the QIH formula in pSS was confirmed. Moreover, the univariate ‘IgA <70%’ criterion was introduced, which means that the increase in IgG- and IgM-containing plasma cells leads to a decrease of IgA-containing plasma cells below 70% [12]. A comparison of the LFS with this ‘IgA <70%’ provided greater disease sensitivity and specificity in pSS in favour of the latter, using serological parameters as gold standard [13].

So far, changed plasma cell distributions in LSG biopsy specimens (‘IgA <70%’) appear to be quite a promising diagnostic tool for pSS and might elegantly reflect ‘the degree of autoimmunity’, being a marker for clonal B-cell expansion. However, whether QIH is helpful in making a clinical diagnosis and is able to distinguish between nSS and pSS in difficult diagnostic cases has been poorly investigated.

In order to study this role for QIH, a model mimicking the making of a clinical diagnosis in daily practice was introduced: in this model an expert panel classified sicca patients based on a data set with relevant clinical parameters. Patients were classified twice, namely without and with knowledge of QIH, enabling us to observe changes in diagnoses. After these classification procedures were performed, the following aims were addressed: firstly, to calculate the sensitivity, specificity and positive and negative predictive value in pSS vs nSS patients in our tertiary referral centre cohort of sicca patients. Secondly, to investigate the extent to which knowledge of QIH leads to changes in diagnoses and whether these changes are clinically relevant. Finally, to investigate whether QIH (IgA <70%) compared with LFS (LFS ≥1) has a superior association with signs of autoimmunity.


   Patients and methods
Top
 Abstract
 Introduction
 Patients and methods
Results
 Discussion
 References
 
Model mimicking the making of a clinical diagnosis
In the model used to mimic clinical decision-making, a data set with 25 relevant clinical parameters was put at the disposal of five experts. These experts classified 396 sicca patients twice; in first instance with knowledge of the 25-parameter data set but without knowledge of QIH and in second instance with knowledge of the 25-parameter data set as well as knowledge of QIH.

Patients
A total of 135 clinical, serological, ophthalmological and histopathological parameters from 597 patients with sicca symptoms, obtained from 1989 to 2000 during routine diagnostic work-up for SS at our tertiary referral out-patient clinic, were enclosed in a database. Although ethical approval was not required for anonymous storage of data obtained from the patients’ charts, patients gave informed consent for anonymous storage of their data. In our centre, objective salivary gland dysfunction is rarely investigated. In the majority of patients, however, tests for lacrimal gland dysfunction (mainly the Schirmer test) as well as LSG biopsy (for LFS and QIH) are performed. Nevertheless, in 124 patients, no LSG biopsy was performed, mainly because of refusal by the patient or insufficient ‘pre-LSG biopsy probability’ of pSS or for ‘unknown’ reasons. In 77 patients LFS was established but QIH was not, mainly because QIH has only been a standard procedure in our hospital since 1992. In the remaining 396 patients, both LFS and QIH were simultaneously determined; consequently these patients were studied.

First classification procedure without QIH: gold-standard diagnoses
The 396 patients were independently classified by an expert panel, consisting of three rheumatologists (PJB, AAK and HB), one internist–immunologist (LK) and one immunologist–nephrologist (RH). The experts were recruited from three university hospitals in The Netherlands and have been involved in patient care and research regarding SS for many years. Two of the five experts do not use QIH for diagnosing SS in their daily clinical practice.

In first instance the experts classified the 396 patients as nSS, ‘possible SS’ (posSS), pSS or sSS, based on the aforementioned data set consisting of 25 parameters (given in Table 1), selected from the original database. The choice of which parameters to include in the data set was influenced by their clinical relevance and by their availability (number of missing values). During this first classification procedure QIH was unavailable. If there was agreement by at least four experts on a diagnosis in a given patient this was considered the gold-standard diagnosis (nSS, pSS or sSS); these diagnoses were considered ‘definite’. In all other patients (agreement by three experts or less or posSS patients by agreement of at least four experts) the diagnosis was considered ‘indefinite’. Sensitivity, specificity and positive and negative predictive values of abnormal QIH (IgA <70%) were calculated in pSS vs nSS patients according to these gold-standard diagnoses. Moreover, classification of these 396 patients according to the criteria of the American–European Consensus Group (AECG) [14] (nSS, pSS and sSS) was performed.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Twenty-five parameters at the disposal of the expert panel on which classification into gold-standard diagnoses nSS, indefinite, pSS and sSS was established [without knowledge of quantitative immunohistology (QIH)]

 
Second classification procedure with QIH
In the second instance, to observe changes in diagnoses without and with knowledge of QIH, all 396 patients were classified again by the expert panel based on the same 25 parameters but now also with knowledge of QIH (28 parameters, see Table 1). By changing identity codes and patient record sequences during this second procedure, individual patients could not be retrieved or compared with the first classification procedure. The numbers of changes in diagnosis from ‘definite’ to ‘indefinite’ or from ‘indefinite’ to ‘definite’ were calculated and compared using Fisher's exact test (see Figs 1A–D).


Figure 1
View larger version (12K):
[in this window]
[in a new window]
  		FIG. 1A–D. Changes in diagnoses with knowledge of QIH: the four gold-standard diagnoses nSS, indefinite, pSS and sSS (classification by experts without knowledge of QIH) are depicted on the left of each figure. On the right in each figure the diagnoses are depicted after classification by the expert panel with knowledge of QIH. Rounds indicate indefinite diagnoses whereas squares indicate definite diagnoses.

 
Intra-expert variability
Four duplicate patients were included to test intra-expert variability during both classification procedures. So, experts actually classified 400 patients during each classification procedure.

Autoimmune parameters and statistical analysis
In the following sections, descriptions are given for several patient groups being tested for differences in ‘objective autoimmune parameters’. The following autoimmune parameters were tested: erythrocyte sedimentation rate (ESR), serum IgG, LFS and IgA-producing plasma cell percentage (QIH) in labial salivary gland biopsy. Differences in mean values were calculated using the Mann–Whitney U-test. Moreover, differences in presence of the following parameters were calculated using the Fisher's exact test: absence or presence of ESR >20 mm/1st h, serum IgG >15 g/l, an LFS ≥1, IgA-producing plasma cell percentage <70%, anti-nuclear antibodies (ANA), anti-Ro/SS-A and/or anti-La/SS-B antibodies and serum rheumatoid factor (RF). The Statistical Package for the Social Sciences (SPSS) version 11.5 was used for statistical analyses.

Changes in diagnosis with knowledge of QIH
Patient groups with and without a change in diagnosis (due to knowledge of QIH) in the gold-standard diagnosis groups (Figs 1A–D) were compared with regard to autoimmune parameters:

  1. In the gold-standard nSS patient group, a new nSS as well as a new indefinite patient group were formed due to the second classification procedure. Differences in autoimmune parameters were calculated between these new nSS and new indefinite patient groups (see Fig. 2A).
  2. In the same way, in the gold-standard pSS patient group differences in autoimmune parameters were calculated between the new pSS and new indefinite patient groups (see Fig. 2C). In the gold-standard indefinite patient group, new nSS, indefinite and pSS patient groups were formed due to the second classification procedure. Differences in autoimmune parameters were calculated between both the new indefinite and new nSS groups as well as between the new indefinite and new pSS patient groups (see Fig. 2B).


Figure 2
View larger version (10K):
[in this window]
[in a new window]
  		FIG. 2A. Mean values of IgA%, LFS, ESR, sIgG as well as presence of IgA <70%, LFS ≥1, ESR >20 and sIgG >15, ANA, anti-Ro or anti-La and RF in the distinct ‘indefinite’ (n = 21) and nSS (n = 165) patient groups as classified by the expert panel with knowledge of QIH. In first instance (without knowledge of QIH) all patients were classified as nSS (n = 186). Statistical significant differences between ‘indefinite’ and nSS patient groups are marked with *. *1 means using Mann–Whitney U statistics, *2 means using Fisher' exact test. See text for abreviations.

 

Figure 3
View larger version (13K):
[in this window]
[in a new window]
  		FIG. 2B. Mean values of IgA% LFS, ESR, sIgG as well as presence of IgA ≤70%, LFS ≥1, ESR >20 and sIgG >15, ANA, anti-Ro or anti-La and RF in the distinct nSS (n = 12), ‘indefinite’ (n = 74) and pSS (n = 19) patient groups as classified by the expert panel with knowledge of QIH. In first instance (without knowledge of QIH) all patients were classified as ‘indefinite’ (n = 107). Statistical significant differences between nSS and ‘indefinite’ patient groups are marked with *. *1 means using Mann–Whitney U statistics, *2 means using Fisher' exact test. Statistical significant differences between ‘indefinite’ and pSS patient groups are marked with #. #1 means using Mann-Whitney U statistics, #2 means using Fisher' exact test. See text for abreviations.

 

Figure 4
View larger version (10K):
[in this window]
[in a new window]
  		FIG. 2C. Mean values of IgA%, LFS, ESR, sIgG as well as presence of IgA <70%, LFS ≥1, ESR >20 and sIgG >15, ANA, anti-Ro or anti-La and RF in the distinct ‘indefinite’ (n = 7) and pSS (n =82) patient groups as classified by the expert panel with knowledge of QIH. In first instance (without knowledge of QIH) all patients were classified as pSS (n = 89). Statistical significant differences between ‘indefinite’ and pSS patient groups are marked with *. *1 means using Mann–Whitney U statistics, *2 means using Fisher' exact test. See text for abreviations.

 
LFS vs QIH and autoimmune parameters in ‘mismatch groups’
To study the differences between LFS and QIH with regard to autoimmune parameters we selected so-called ‘mismatch’ groups. The first mismatch group (group 1) consisted of patients with an abnormal LFS (≥1), but normal QIH (IgA ≥70%). The second mismatch group (group 2) consisted of patients with a normal LFS (<1), but abnormal QIH (IgA <70%). Differences in appropriate autoimmune parameters (see above) were calculated.


   Results
Top
 Abstract
 Introduction
 Patients and methods
 Results
Discussion
 References
 
Patient characteristics and first classification by the expert panel
During the first classification procedure by the expert panel (knowledge of 25 parameters but no QIH; yielding gold-standard diagnoses) agreement by at least four experts on the diagnosis was achieved in 289 (73%) patients; 186 patients were diagnosed as nSS, 89 as pSS and 14 as sSS. Consequently, 107 patients (27%) were diagnosed as indefinite. Table 1 shows the 25 parameters (with mean values or frequencies in each gold-standard group) at the disposal of the expert panel. Moreover, according to the AECG criteria, 262 (66%) nSS, 124 (31%) pSS and 11 (3%) sSS patients were classified as such. The distribution of the AECG diagnoses in the gold-standard diagnoses nSS, pSS and sSS are depicted in Table 1.

Sensitivity, specificity and positive and negative predictive value of QIH
Eighty-three pSS patients and 26 nSS patients had abnormal QIH, leading to a sensitivity of 93%, specificity of 86%, positive predictive value of 76% and negative predictive value of 96% of abnormal QIH in pSS vs nSS (see Table 2).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Sensitivity, specificity, positive predictive value and negative predictive value of QIH in pSS vs nSS

 
Second classification by expert panel and changes in diagnoses with knowledge of QIH
During the second classification procedure by the expert panel (knowledge of the 25 parameters plus QIH) agreement by at least four experts on the diagnosis was achieved in 294 (74%) patients; 177 patients were classified as nSS, 101 as pSS and 16 as sSS. Consequently, 102 (26%) patients were classified as indefinite.

Figures 1A–D depict the changes in diagnoses after the second classification procedure by the expert panel: on the left side in each figure the gold-standard diagnosis (without knowledge of QIH) is depicted whereas the arrows point to the diagnoses obtained at the second classification procedure (depicted right in each panel). Taking all changes in diagnoses together, changes from definite diagnoses (nSS, pSS or sSS; indicated with squares in the figures) to indefinite diagnoses (agreement by three experts or less or posSS patients; indicated with rounds in the figures) occurred in 28 of 289 patients (10%; see Figs 1A, C and D). On the other hand, changes from indefinite to definite diagnoses occurred in 33 of 107 patients (31%; see Fig. 1B). So, knowledge of QIH more often changed a diagnosis from indefinite to definite than vice versa (P = 0, using Fisher's exact test).

Intra-expert correlation
All five experts scored four duplicate patients in both classification procedures, yielding 40 diagnoses of which 35 were scored concordantly (87.5%). Without knowledge of QIH, 2 of 20 diagnoses were scored discordantly; with knowledge of QIH, 3 of 20 diagnoses were scored discordantly.

Changes from definite to indefinite diagnoses (Figs 2A and C)
Figure 2A shows that in 21 of 186 gold-standard nSS patients (11%), the diagnosis changed towards indefinite with knowledge of QIH. All tested autoimmune parameters differed significantly between the new nSS and indefinite patient groups, with more autoimmune phenomena being found in the new indefinite group. It is remarkable that in the new indefinite group not all patients had abnormal QIH (16 of 21) and that in the new nSS group not all patients had normal QIH (155 of 165). Nevertheless, knowledge of QIH is able to distinguish a subgroup with more autoimmune phenomena within the gold-standard nSS group.

In the same way, Fig. 2C shows that in 7 of 89 gold-standard pSS patients (8%) the diagnosis changed towards indefinite. All tested autoimmune parameters (except LFS ≥1) differed significantly between the new pSS and indefinite patient groups, with more autoimmune phenomena being found in the new pSS group. Although not all indefinite patients had normal QIH (three of seven) and not all new pSS patients had abnormal QIH (79 of 82), also in the gold-standard pSS group knowledge of QIH is able to distinguish a subgroup with fewer autoimmune parameters.

Changes from indefinite to definite diagnoses
Without knowledge of QIH, in 107 patients the diagnosis was indefinite (see Fig. 1B). With knowledge of QIH, 74 patients were still classified as indefinite, consequently in 33 patients (31%) a definite diagnosis was established, as mentioned before. Twelve patients (11%) were classified as nSS, 19 (18%) were classified as pSS and 2 (2%) were classified as sSS.

Comparing the new nSS (n = 12) with the new indefinite patient group (n = 74), most autoimmune parameters differed significantly, with more autoimmune phenomena found in the indefinite group (Fig. 2B).

Comparing the new pSS group (n = 19) with the new indefinite patient group (n = 74), they only differed statistically significant with regard to mean IgA plasma cell percentage (45 vs 56), the presence of IgA <70% (19/19 vs 55/74) and the presence of antinuclear antibodies (ANA). Moreover, comparing the 55 patients with IgA <70% in the new indefinite group with the 19 patients in the new pSS group, no differences with regard to autoimmune parameters were found (data not shown).

Secondary SS patients
Regarding the sSS patients, all 14 sSS patients without knowledge of QIH were classified as such with knowledge of QIH (see Fig. 1D). Two patients in the indefinite patient group without knowledge of QIH were classified as sSS with knowledge of QIH (see Fig. 1B). Because of the small numbers, the sSS patient group was left out of further consideration.

LFS vs QIH and ‘parameters of autoimmunity’ in mismatch groups
Mismatch group 1 (abnormal LFS, normal QIH) and group 2 (normal LFS, abnormal QIH) consisted of 29 and 32 patients, respectively. Table 3 shows the tested autoimmune parameters; more autoimmune phenomena were found in mismatch group 2 (abnormal QIH), except for the presence of ANA. Statistical significance was neared for the difference in presences of anti-Ro/SS-A and/or anti-La/SS-B antibodies and was established for the difference in presences of RF.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Autoimmune parameters in the two ‘mismatch’ patient groups: abnormal LFS (LFS ≥1) but normal QIH (IgA ≥70) (n = 29) vs normal LFS (LFS <1) but abnormal QIH (IgA <70) (n = 32)

 

   Discussion
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
References
 
In search of better diagnostic tools for Sjögren's syndrome, focus is put in this study on the clinical relevance of abnormal plasma cell distributions (‘IgA <70%’) in labial salivary glands, the site of the autoimmune response. In order to study this clinical relevance of QIH, a model mimicking the making of a clinical diagnosis was introduced, in which an expert panel classified sicca patients twice, based on a data set of 25 clinical parameters (without and with knowledge of QIH). Gold-standard diagnoses were defined by agreement on a certain diagnosis by at least four of the five experts. Credits given to a certain test might vary remarkably between clinicians and may also be influenced by ‘local prejudice and belief’. We tried to justify this phenomenon by recruiting five experts from three different university hospitals, two of them actually not working with QIH in their clinical practice. Therefore it seemed relevant to define a gold-standard diagnosis if agreement by at least four experts was reached, which was the case in approximately 75% of patients in both expert classification procedures.

A remarkable conformity between the expert diagnoses and diagnoses based on AECG criteria was observed. The expert panel, being aware of the validated AECG criteria, will probably have been influenced by these criteria in classifying the study patients.

This study confirms the high sensitivity (93%) for abnormal QIH found in the study by Bodeutsch et al. (95%) [12]. However, the high specificity found in the Bodeutsch study (99%) could not be confirmed in this study (86%) due to relative large numbers of false-positive cases.

When QIH is compared with other diagnostic tools for pSS, similar values are found: the sensitivity of anti-Ro/SS-A antibodies varies between 70 and 97% (depending on detection procedures), while their specificity is reported to be 87%. The sensitivity of anti-La/SS-B antibodies varies between 70 and 95%, while their specificity is reported to be 94% [15–18]. In the first paper dealing with anti-{alpha}-fodrin antibodies, sensitivity of 95% in pSS and 63% in sSS and a specificity 100% in pSS was reported [19]. However, in subsequent studies much lower values for sensitivity and specificity of anti-{alpha}-fodrin antibodies were reported [20, 21].

It may be of clinical relevance whether a test is able to contribute to establishing a diagnosis in ‘difficult cases’. In this study, the gold-standard indefinite group can be considered as a ‘difficult cases’ group. In nearly one-third of these patients, a definite diagnosis (nSS or pSS) could be established with knowledge of QIH, pointing to an additional role for QIH in these cases in particular. This is also of clinical importance because definite diagnoses provide clarity for patients and clinicians. Moreover, with knowledge of QIH, pSS was diagnosed in 12 additional patients (89 pSS patients without and 101 with knowledge of QIH), which has consequences for follow-up of these patients.

In patient groups with gold-standard definite diagnosis (nSS, pSS or sSS), changes in diagnoses occurred from nSS to indefinite or from pSS to indefinite. After knowledge of QIH, no changes from nSS to pSS or vice versa occurred in these groups. One can conclude that in clear cases (nSS, pSS or sSS definite patient groups) no clinically relevant contribution is found for QIH. Nevertheless, QIH is able to distinguish a subgroup in the gold-standard nSS group with significantly more autoimmune phenomena and to distinguish a subgroup in the gold-standard pSS group with significantly fewer autoimmune phenomena.

According to our hypothesis, abnormal QIH shows a better association with signs of autoimmunity than LFS when autoimmune parameters were compared in the two mismatch groups. Although all parameters (except ANA) occurred more frequently in mismatch group 2 (abnormal QIH), differences in autoimmune parameters between mismatch groups are not great, reflected by statistical significant difference in only one parameter (RF). However, in the indefinite patient group with knowledge of QIH (n = 102), QIH significantly correlated negatively with ESR and serum IgG whereas LFS did not (data not shown), which is in agreement with aforementioned hypothesis.

In summary, in our model of making a clinical diagnosis, abnormal QIH shows equivalent or superior diagnostic value compared with other reported diagnostic tools. Moreover, knowledge of QIH more often leads to definite than to indefinite diagnoses; in nearly one-third of indefinite patients the diagnosis changes to a definite diagnosis. Finally, abnormal QIH is more related to autoimmune parameters than abnormal LFS. In view of the consequences of distinguishing pSS from nSS, these data point to an additional diagnostic role for QIH in clinical practice.
Figure 5


   Acknowledgments
 
The authors thank A. Stoffel, B. Hajian and M. Noordzij for assistance in completing the original database.

The authors have declared no conflicts of interest.


   References
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 

  1. Kruize AA, Hene RJ, van der Heide A et al. Long-term followup of patients with Sjögren's syndrome. Arthritis Rheum 1996;39:297–303.[Web of Science][Medline]
  2. Chisholm DM, Mason DK. Labial salivary gland biopsy in Sjögren's disease. J Clin Pathol 1968;21:656–60.[Abstract/Free Full Text]
  3. Greenspan JS, Daniels TE, Talal N, Sylvester RA. The histopathology of Sjögren's syndrome in labial salivary gland biopsies. Oral Surg Oral Med Oral Pathol 1974;37:217–29.[CrossRef][Web of Science][Medline]
  4. Takeda Y, Komori A. Focal lymphocytic infiltration in the human labial salivary glands: a postmortem study. J Oral Pathol 1986;15:83–6.[Medline]
  5. Manthorpe R, Benoni C, Jacobsson L et al. Lower frequency of focal lip sialadenitis (focus score) in smoking patients. Can tobacco diminish the salivary gland involvement as judged by histological examination and anti-SSA/Ro and anti-SSB/La antibodies in Sjögren's syndrome? Ann Rheum Dis 2000;59:54–60.[Abstract/Free Full Text]
  6. Zandbelt MM, van den Hoogen FH, de Wilde PC, van den Berg PJ, Schneider HG, van de Putte LB. Reversibility of histological and immunohistological abnormalities in sublabial salivary gland biopsy specimens following treatment with corticosteroids in Sjögren's syndrome. Ann Rheum Dis 2001;60:511–3.[Abstract/Free Full Text]
  7. Anderson LG, Cummings NA, Asofsky R et al. Salivary gland immunoglobulin and rheumatoid factor synthesis in Sjögren's syndrome. Natural history and response to treatment. Am J Med 1972;53:456–63.[CrossRef][Web of Science][Medline]
  8. Talal N, Asofsky R, Lightbody P. Immunoglobulin synthesis by salivary gland lymphoid cells in Sjögren's syndrome. J Clin Invest 1970;49:49–54.[Medline]
  9. Matthews JB, Deacon EM, Wilson C, Potts AJ, Hamburger J. Plasma cell populations in labial salivary glands from patients with and without Sjögren's syndrome. Histopathology 1993;23:399–407.[Medline]
  10. Speight PM, Cruchley A, Williams DM. Quantification of plasma cells in labial salivary glands: increased expression of IgM in Sjögren's syndrome. J Oral Pathol Med 1990;19:126–30.[CrossRef][Medline]
  11. de Wilde PC, Kater L, Baak JP, van Houwelingen JC, Hene RJ, Slootweg PJ. A new and highly sensitive immunohistologic diagnostic criterion for Sjögren's syndrome. Arthritis Rheum 1989;32:1214–20.[Web of Science][Medline]
  12. Bodeutsch C, de Wilde PC, Kater L et al. Quantitative immunohistologic criteria are superior to the lymphocytic focus score criterion for the diagnosis of Sjögren's syndrome. Arthritis Rheum 1992;35:1075–87.[Web of Science][Medline]
  13. Zandbelt MM, Wentink JR, de Wilde PC, van Damme PA, van de Putte LB, van den Hoogen FH. The synergistic value of focus score and IgA% score of sublabial salivary gland biopsy for the accuracy of the diagnosis of Sjögren's syndrome: a 10-year comparison. Rheumatology 2002;41:819–23.[Abstract/Free Full Text]
  14. Vitali C, Bombardieri S, Jonsson R et al. Classification criteria for Sjögren's syndrome: a revised version of the European criteria proposed by the American-European Consensus Group. Ann Rheum Dis 2002;61:554–8.[Abstract/Free Full Text]
  15. Reichlin M. Clinical significance of anti-Ro/SS-A and anti-La/SS-B antibodies. Up to date online version 13.1. 25 April 2005.
  16. Harley JB, Alexander EL, Bias WB et al. Anti-Ro (SS-A) and anti-La (SS-B) in patients with Sjögren's syndrome. Arthritis Rheum 1986;29:196–206.[Web of Science][Medline]
  17. Phan TG, Wong RC, Adelstein S. Autoantibodies to extractable nuclear antigens: making detection and interpretation more meaningful. Clin Diagn Lab Immunol 2002;9:1–7.
  18. Peene I, Meheus L, Veys EM, De Keyser F. Diagnostic associations in a large and consecutively identified population positive for anti-SSA and/or anti-SSB: the range of associated diseases differs according to the detailed serotype. Ann Rheum Dis 2002;61:1090–4.[Abstract/Free Full Text]
  19. Haneji N, Nakamura T, Takio K et al. Identification of alpha-fodrin as a candidate autoantigen in primary Sjögren's syndrome. Science 1997;276:604–7.[Abstract/Free Full Text]
  20. Nordmark G, Rorsman F, Ronnblom L et al. Autoantibodies to alpha-fodrin in primary Sjögren's syndrome and SLE detected by an in vitro transcription and translation assay. Clin Exp Rheumatol 2003;21:49–56.[Medline]
  21. Zandbelt MM, Vogelzangs J, van de Putte LB, van Venrooij WJ, van den Hoogen FH. Anti-alpha-fodrin antibodies do not add much to the diagnosis of Sjögren's syndrome. Arthritis Res Ther 2004;6:R33–8.[CrossRef][Web of Science][Medline]
Submitted 4 June 2005; revised version accepted 6 October 2005.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
45/4/470    most recent
kei191v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by van Woerkom, J. M.
Right arrow Articles by Bijlsma, J. W. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van Woerkom, J. M.
Right arrow Articles by Bijlsma, J. W. J.
Related Collections
Right arrow Diagnostics and Imaging Procedures
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?