SEARCH

SEARCH BY CITATION

Keywords:

  • Activity;
  • damage;
  • indices;
  • primary Sjögren’s syndrome

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conflicts of interest
  8. Address
  9. References

Eur J Clin Invest 2010; 40 (7): 636–644

Abstract

Background  Significant differences exist in outcome measures developed to assess patients with primary Sjögren’s Syndrome (pSS). In this review, we have compared proposed indices.

Methods  Three activity – SSDAI (SS Disease Activity Index), SCAI (Sjögren’s Systemic Clinical Activity Index) and ESSDAI (EULAR SS Disease Activity Index) – and two damage indices – SSDDI (SS Disease Damage Index) and SSDI (SS Damage Index) have been analysed. Assessment ‘tools’ for perspectives of outcome (PROFAD – Profile of Fatigue and Discomfort, SF-36 – Medical Outcomes Study Short-Form 36-item questionnaire and ESSPRI – EULAR Sjögren’s Syndrome Patients Reported Index) were also considered.

Results  SSDAI and ESSDAI are global scores. SCAI is a composite score. Validity is a limitation for SSDAI and SCAI. ESSDAI is complemented by ESSPRI for the assessment of subjective features. It is more accurate in detecting changes in activity. Damage indices differ with respect to observation period and external validation but both have low content validity. Main limitations for all indices are: inclusion of patients with mainly mild stable disease and lack of information about the completion time of the forms.

Conclusions  All indices demonstrate a potentially useful benefit but further, larger studies are needed to assess reliability and sensitivity to change, to validate their use in clinical trials. Improvement in our knowledge of pathophysiology and clinical evolution of pSS is important to address unresolved issues: whether to include prognostic factors for an adverse outcome, an agreed definition of flare and most notably the distinction between activity and damage.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conflicts of interest
  8. Address
  9. References

Primary Sjögren’s syndrome (pSS) is a chronic multi-system disease, characterised by immune-mediated inflammation primarily, but not exclusively, of the exocrine glands, resulting in dryness of the mucosal surfaces. The precise aetiology remains obscure with some authors proposing an environmental stimulus (e.g. viral infection) promoting an autoimmune reaction in genetically susceptible persons [1] and others assuming an age-related loss of self-tolerance within epithelia with the possible contribution of hormonal factors [2–4]. The clinical course is usually indolent and benign, with keratoconjunctivitis sicca and xerostomia being the dominant manifestations. However, in approximately 25% of patients, a wide range of systemic involvement features (e.g. pulmonary, neurological, haematological) can occur and an increased risk of lymphoma is also evident [5–8]. Furthermore, reported symptoms like fatigue and arthralgia have consistently been correlated with a reduced health-related quality of life [9]. Evidence-based therapy has mainly been symptomatic (and restricted mostly to sicca complaints). Clinical trials of disease-modifying treatments have been limited by the lack of validated instruments to assess disease outcome, particularly those ‘capturing’ systemic involvement.

Outcome measures are used to evaluate the severity and progression/evolution of disease. They might help to prevent or delay the damage by identifying those patients with active disease potentially amenable to change by therapy and to select subsets with more severe disease. Furthermore, when validated, they can be used as primary endpoints in clinical trials both for new therapies and for agents already in use making inter-study comparison feasible. Assessing outcome in rheumatic disorders in clinical trials requires indices of disease activity and damage, measures that identify the patient’s self-assessment of outcome, economic consequences of the disease, mortality and drug-related toxicity. The validity, reliability and sensitivity to change of these ‘tools’ need to be established [10].

In an attempt to provide tools to assess patients with pSS in clinical practice and therapeutic trials, several outcome measures have been developed. There are significant differences between them. In this review, we compare recent proposals for activity and damage indices to assess outcome in patients with pSS.

Methods

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conflicts of interest
  8. Address
  9. References

Three activity – SSDAI, SCAI and ESSDAI [11–13] – and two damage indices – SSDDI and SSDI [11,14] – were analysed, focusing on: the number of patients included, the basis for the development and the type of index, their validity, reliability and sensitivity to change and the comparative strengths and weaknesses. Proposals for the assessment of perspectives of outcome (PROFAD, SF-36 and ESSPRI) [15,16] were also considered. A review of the literature in relation to pathophysiology, evolution, prognosis and treatment complemented the analysis.

Results

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conflicts of interest
  8. Address
  9. References

The main clinometric characteristics and comparative strengths and weaknesses of the indices analysed are presented in Table 1 for activity and Table 2 for damage.

Table 1.   Clinometric characteristics and comparative strengths and weaknesses of the three activity indices analysed
Name of the indexSSDAISCAIESSDAI
  1. All statistical results are with a 95% confidence interval.

  2. SSDAI, Sjögren’s Syndrome Disease Activity Index; SCAI, Sjögren’s Systemic Clinical Activity Index; ESSDAI, EULAR Sjögren’s Syndrome Disease Activity Index; ECLAM, European Consensus Lupus Activity Measurement; BILAG, British Island Lupus Assessment Group; pSS, primary Sjögren’s Syndrome; PhGA, physician’s global assessment; ESSPRI – EULAR Sjögren’s Syndrome Patients Reported Index); ICC, intraclass correlation coefficient; CT, computed tomography; EMG, electromyography; SRM, standardised response mean.

Number of patients206 (201 ♀)104 (all ♀)96 real pts (89 ♀) + 702 realistic vignettes
Basis for the developmentECLAMBILAGNot specified
Type of indexGlobal score (8-domain scale)Composite score (9-domain scale)Global score (12-domain scale)
ValidityConstruct validity: good correlation with the gold standard – PhGA (r = 0·872 and 0·817 ); confirmed by analysis of convergent (correlation between most of the index variables and expert ratings of activity with P < 0·0001) and divergent validity (weak or no correlation between activity and damage scores) Content validity: lowConstruct validity: relatively low correlation of the final score with the gold standard - PhGA (r = 0·523 and 0·571 ); reasonable results in the comparison with PROFAD and SF-36 for subjective items (for the different domains: r = 0·431–0·515 and r = −0·420–−0·610, respectively, for all with P < 0·001) Content validity: lowConstruct validity: relatively low correlation of the final score with the gold standard - PhGA (r = 0·58 and 0·61 ) Content validity: high (includes all possible systemic disease involvement)
ReliabilityNot reportedReliability: good correlation between a.m. and p.m. scores among assessors Repeatability: good correlation between a.m. and p.m. domain scores for each assessor (k-scores = 0·71 for both tests)Results for reliability of physician’s VAS: Inter-rater reliability: high variability of rating among experts (ICC = −4·92 to + 4·61; k coefficient = 0·42) Intra-rater reliability: good result for variation of scored activity between visits (ICC = 0·21–0·63, k coefficient = 0·86)
Sensitivity to changeGood results in the correlation between scores of disease activity at time of enrolment and 3 months later of patients with active disease (r = 0·817, P < 0·0001) Reasonable correlation between the results for time variation of scores (observers vs. index) (r = 0·683, P < 0·0001)Modest results for sensitivity and specificity (respectively): correspondence between index- and PhGA-derived flares (74 and 65%) correlation between index flares and dosage changes of systemic therapy (91 and 28%) comparison between changes in PhGA and index scores between visits (52 and 61%)For improved patients: SRMs −1·08 to −1·38 and −0·50 to −0·76 between visits 1 and 2 and visits 2 and 3 respectively. For worsened patients: SRMs (between visits 1 and 2 and visits 2 and 3 respectively) +0·46 and +1·10 for ESSDAI, −0·03 and +0·79 for SSDAI, and +0·17 and +1·02 for SCAI. For patients with stable activity: SRMs, between visits 1 and 2 and visits 2 and 3, were, respectively, 0·00 and −0·13 for ESSDAI, −0·44 and −0·11 for SSDAI, and −0·36 and +0·34 for SCAI.
StrengthsIncludes subjective items (e.g. fatigue) Accuracy in distinguishing patients with active/very active from those with inactive or mildly/moderately active disease using a cut-off ≥ 5 with good sensitivity and specificity (84·5% and 87·6% respectively) Some of the items are defined according to their variation with respect to the previous assessment Final score is global and so simplerIncludes subjective items (e.g. fatigue) Longest follow-up (12 months) Scoring algorithm for each item according to the change since the previous assessment developed by using physician’s intention-to-treat principle Evaluates correlations between serological items and clinical parameters concluding for the absence of usefulness of the first as markers of disease activity in short-term clinical trialsHigh content validity Includes a biological domain (with assessment of serological markers of B-cell activation) Stratifies each domain’s items according to the severity of organ’s involvement (by physician’s intention-to-treat principle) Separates the assessment of objective from subjective manifestations (ESSPRI) Multinational study
WeaknessesMost patients had mild stable disease Includes several subjective items which are multifactorial in nature and hard to valorise in all patients Requires too many complementary exams to score the different items and many of the objective items are defined based on the results of exams which are not readily accessible, cannot be repeated for each visit and/or are invasive (e.g. CT scans, EMG, renal biopsy) Has the shortest follow-up (3 months) National study: with limitations of less common manifestations of pSS and no geographic or ethnic diversity Time to complete the form not evaluatedMost patients had mild stable disease The items included are mainly objective systemic features with low correlations with the gold standard for external validation (PhGA) Osteoarthritis bias in the scoring for the arthritis item National study: with limitations of less common manifestations of pSS and no geographic or ethnic diversity Time to complete the form not evaluatedMost patients had mild stable disease Many complementary exams (which are not readily accessible, cannot be repeated for each visit and/or are invasive, e.g. CT scan, EMG, renal biopsy) to score or define items Includes only objective systemic items, requiring filling in a second form for evaluation of subjective features Many realistic clinical vignettes used which, however, have a significant lower number of organs than the real patients. Maximum theoretical final score (total = 123) is significantly different from the actual maximum scores obtained (only 25% with totals ≥ 13 or ≥ 21 for ‘vignettes’ and real patients respectively) Time to complete the form not evaluated
Table 2.   Clinometric characteristics and comparative strengths and weaknesses of the two damage indices analysed
Name of the indexSSDDISSDI
  1. All statistical results are with a 95% CI.

  2. SSDDI (SS Disease Damage Index); SSDI, Sjögren’s Syndrome Damage Index; SLICC, Systemic Lupus International Collaborating Clinics; PROFAD-SSI, Profile of Fatigue and Discomfort - Sicca Symptoms Inventory, SF-36, Medical Outcomes Study Short-Form 36-item questionnaire; pSS, primary Sjögren’s Syndrome; PhGA, physician’s global assessment.

Number of patients206 (201 ♀)104 (all ♀)
Basis for the developmentNot specifiedSLICC
Type of indexGlobal scoreGlobal score or 3 separate scales (ocular, oral and systemic damage)
ValidityConstruct validity: Comparison with damage scores of PhGA with close correlation to the index-derived scores (r = 0·760, P < 0·0001) Assessment of convergent (close correlation with the corresponding gold standard – investigator scores of damage) and divergent validity (index variables were not or weakly correlated with the gold standard for activity) Content validity: lowConstruct validity: Used a panel of experts in the 3 main fields involved (Ophthalmology, Oral Medicine and Rheumatology) Correlation analyses of the association between damage totals and domain scores of PROFAD-SSI and SF-36 with statistically significant but weak associations and not consistently present on both visits (ocular damage with PROFAD-SSI ‘eye dry’ domain r = 0·228, T = 0 months; r = 0·365, T = 12 months and total damage with ‘physical functioning’ of SF-36 r = 0·250, T = 0 months; r = 0·261, T = 12 months) Content validity: low
ReliabilityNot evaluatedNot formally evaluated but potential inter-rater variability is admitted
Sensitivity to changeNot evaluatedComparison between the damage scores at time of enrolment and 12 months after with statistically significance for 2 domains and total score (exception for the systemic domain) (z = −3·262, P < 0·01).
StrengthsLittle or no correlation with the gold standard for activity Includes a stratification of the items according to relative importance of the organ affected (with the highest scores for malignancy and systemic involvement)Little or no correlation with the gold standard for activity Longer follow-up (12 months) Expert validation with collective approach from the 3 major fields involved in the care of patients with pSS Differentiation between the damage secondary to pSS and that derived from co-morbidities and/or therapy Includes an additional number of uncommon but well-recognised cardiovascular, gastrointestinal and musculoskeletal items
WeaknessesSingle national study with low content validity Short follow-up period (3 months) Individual judgement for external validation Time to complete the form not evaluatedSingle national study with low content validity Items are not stratified according to relative importance of the organ involved Time to complete the form not evaluated

Activity indices

Clinometric properties.  The demographic details of patients were similar with a significant preponderance of Caucasian women, fulfilling American–European Consensus Group criteria [17]. The size of cohorts was, however, somewhat different making comparison of the results difficult. Furthermore, the total number of patients studied was relatively small. The immediate consequence is a lack of representation of the whole range of manifestations of pSS. To overcome this problem, ESSDAI generated a large number of clinical vignettes to capture as many types of systemic disease involvement as possible. Data were abstracted from the histories of real patients.

Most of the indices used already validated tools for the assessment of activity in other autoimmune diseases (mainly for Systemic Lupus Erythematosus - SLE), as a model. This approach seems reasonable given the partial resemblance between the two diseases, notably their systemic involvement. Most of these indices, e.g. ECLAM (European Consensus Lupus Activity Measurement), are global scores and unidimensional in which they generate a total score representing disease activity at a particular moment. SSDAI is a global index and was based in ECLAM. A global score provides more simplicity. Its disadvantage is that features are scored if present but no distinction is made between problems that are getting worse or better or staying the same. In contrast, BILAG (British Island Lupus Assessment Group) is multi-dimensional and assesses each domain separately according to the change in clinical features since the previous assessment. It was based on the principle of physician’s ‘intention-to-treat’ as a marker for clinically important change. SCAI was based on this approach and is a composite score. ESSDAI is a global score and is based on appreciation of the level of activity, defined by consensus of 39 experts, of the different domains affected by the disease. However, the same parameter was also used as the independent variable for external validity.

Construct (external) validity was measured for the three indices mainly by the strength of correlation between the index-derived score and the physician’s global assessment (PhGA) of the disease activity as the gold standard. However,the results are generally lower than those from other studies evaluating disease activity scores for systemic disorders, with SSDAI performing better than the others. Various reasons are proposed to justify this fact. The polymorphous nature of pSS and the variable extent to which each organ can be involved, added to the preponderant influence that subjective symptoms, like fatigue and pain, have on physician’s evaluation, make assessment of disease activity more challenging. An ‘intention-to-treat’ approach is also not as clear-cut with respect to systemic pSS compared with other autoimmune diseases, because there is little consensus about what treatment is to be used and when, for systemic involvement. In addition, the multi-systemic nature of the condition, particularly in more complicated cases, makes therapeutic decisions more variable among physicians. Construct validity was further confirmed, for SSDAI, by the analysis of convergent and divergent validity and the scores correlation within a 3-month interval. SCAI obtained reasonable correlations between the subjective items of the scale and the PROFAD and SF-36 scores.

SSDAI and SCAI were based on national studies from a limited geographical area, resulting in limitations in content validity. In addition, most patients had inactive or weakly active disease, increasing the difficulty to ‘capture’ adequately the items related to organs less frequently involved (e.g. liver, kidney). The large number of realistic clinical vignettes generated for ESSDAI ensured its high content validity.

Reliability is a particularly important feature of an index, both for its broad use in clinical practice and even more for its use in clinical trials, where much of the efficacy of performance relies on the proper application of outcome measures by different investigators. For SCAI, satisfactory results were obtained for both, reliability and repeatability. For ESSDAI, only the reliability of physician’s VAS has been assessed so far, which revealed a significant variability of rating among experts, although with better results in the intra-rater reliability. Additional tests for reliability and variability are currently underway. Tests for reliability have not been reported for SSDAI.

Sensitivity to change analysis for SSDAI and SCAI generally resulted in modest results, with the former performing better. ESSDAI has showed sensitivity to change similar to that of the other two indices for patients with improved disease activity, but it detects changes in activity more accurately than those indices for patients with stable disease activity (Table 1) [18].

Comparative strengths and weaknesses.  Both SSDAI and SCAI include fatigue in the score as it often influences the decision to initiate systemic therapy. In contrast, the BILAG group removed fatigue from their new lupus activity index (BILAG-2004), as they perceived significant difficulty in being sure if it was really because of activity [19].

SSDAI’s final score demonstrated accuracy in distinguishing patients with more active disease from those with a milder form using a cut-off ≥ 5.

SCAI was developed from the study with the longest follow-up (12 months). Its item definitions are based on the physician’s ‘intention-to-treat’ principle to differentiate clinically important change in activity. Serological items were not considered useful markers of disease activity in short-term clinical trials, given the weak correlation with clinical parameters.

ESSDAI is a European consensus disease activity index, initiated by EULAR, which took into account the limitations of the former SSDAI and SCAI and involved 39 experts of the field, including the authors of these earlier indices.

Unlike the others, ESSDAI has a high content validity (comprising all possible systemic and some rarer manifestations of pSS not included in the others). It is the only index to include a biological domain, with assessment of serological markers of B-cell activation (addressing some of the factors identified as predictors of an adverse outcome separately). It is unique in separating assessment of objective systemic involvement from subjective items, (assessed by a distinct scale). It also uses physician’s ‘intention-to-treat’ to stratify the items of each domain.

Globally the three indices have two major limitations. First, most of the patients only had mild stable disease and even in the case of ESSDAI, for which patients were selected because of their prominent systemic involvement, the index showed only low activity scores. Second, although being less prominent in the SCAI index, generally the necessity for a significant number of complementary exams to score the different items is apparent and the definition of many relies on the results of tests which are not readily accessible, cannot be repeated at each visit (e.g. computed tomography scans) and/or are invasive (e.g. renal biopsy). These difficulties may make the indices of less certain value in routine clinical evaluation and clinical trials. In addition, given their limitations in content validity, SSDAI and SCAI, lack the necessary diversity to be sure of their universal applicability.

The short follow-up used to develop SSDAI limits the relative levels of different activity manifestations. It includes several subjective items (e.g. fatigue and pain), which may be multifactorial in nature and hard to validate in all patients.

For SCAI, the arthritis domain was found to correlate closely with the presence of osteoarthritis, which may bias the scoring.

ESSDAI requires a separate scale for the assessment of subjective features (ESSPRI), which implies filling in a second form for all patients. The vignettes used to develop the score have a significantly lower number of organs involved when compared with real patients. Finally, there is a significant difference between the ESSDAI maximum theoretical final score (total = 123) and the actual maximum scores obtained with the patients (75% of the totals were below 13 and 21 in the ‘vignettes’ and real patients respectively) (Table 1). This might be secondary to the low activity level of patients included but may also reveal that several items included in the score are not the optimal markers of disease activity.

Damage indices

Clinometric properties.  SSDDI was based on a larger cohort (n = 206) than SSDI (n = 104) but with similar demographic characteristics.

SSDDI was developed using the same methodology as that used by international consensus groups to develop scales for SLE, but the exact model index is not specified. A modified version of SLICC (Systemic Lupus International Collaborating Clinics) damage index for SLE was used to develop SSDI. SSDDI is a global score including exocrine and nonexocrine features. SSDI comprises a three-domain assessment (ocular, oral and systemic damage), which, in view of their weak correlations, may be more useful if considered as three separate rating scales.

External validation was assessed quite differently. SSDDI used PhGA of the disease damage as gold standard (with close correlation to the index-derived scores) and further confirmed it by the assessment of convergent and divergent validity. SSDI used the input of a panel of experts from three major medical areas (Ophthalmology, Oral Medicine and Rheumatology). Further assessment revealed statistically significant but modest associations between the index-derived scores and PROFAD-SSI (Profile of Fatigue and Discomfort - Sicca Symptoms Inventory) and SF-36 domains.

Reliability was not formally evaluated for either index but, for SSDI, some inter-rater variability is likely based on the lacking of broadly accepted strict definitions for many of the damage items.

SSDI was considered sensitive to change over 12 months. Sensitivity to change was not evaluated for SSDDI.

Comparative strengths and weaknesses.  Both damage indices demonstrated low or no correlation with the activity scores.

SSDDI includes a stratification of the items according to the relative importance of the organ affected. SSDI was based in a study with a much longer follow-up (12 months) and its validation is improved by the collective approach with diverse expert input. It differentiates between the damage secondary to pSS and that derived from co-morbidities and/or therapy, considering that damage secondary to these last ones should be recorded alongside the scale but not included or scored in the index. It also includes an additional number of uncommon but well-recognised cardiovascular, gastrointestinal and musculoskeletal items.

Both indices have limitations in content validity, as they were based on relatively small single national-cohorts.

SSDDI was developed from a short-term study (3 months) which limits the evaluation of damage acquisition over time and probably contributed to the lower range of damage involvement recorded. SSDI items are not stratified for severity according to the organ impaired, but simply rated as present or absent.

The time it takes to complete activity and damage assessments will limit their widespread use in clinical practice. However, the completion time for these has not been reported for any of the indices (activity or damage). Given the large number of items to be completed on each form, a significant proportion of which are unlikely to be rated at each consultation, we anticipate that few centres will use these indices routinely.

Patient’s outcome self-assessment

Several tools for assessment of patient perception of disease and evaluation of subjective features of autoimmune diseases have been developed and validated. These generally evaluate the psychosocial impact of disease in life and functional disabling, particularly with respect to disease related and global functional impairment, overall quality of life (QOL), economic consequences, mortality and drug-related toxicity. Sicca features are the hallmark of pSS and a major cause of disability and reduced QOL. In addition, approximately 75% of patients report fatigue as a particularly prominent and disabling feature further reducing QOL. Other key patient-reported extra-glandular symptoms include arthralgia, myalgia and Raynaud’s phenomenon [9,20–22]. Finally, Sutcliffe et al. demonstrated that the degree of functional disability in patients with pSS was as great as in those with SLE [23].

Sicca features can be assessed using objective tests [e.g. Schirmer-I test of tear production or the unstimulated salivary flow rate (USFR)]. Despite being crude measures, they are easily performed in a routine clinical setting, which is their great advantage over other, more precise, ways of assessing gland function, and they appear to be sensitive to change. However, the effects of a reduction in tear/saliva flow on symptom severity vary among individuals, most probably because of differences in corneal/oral sensitivity and/or psychological processes [24]. Thus, these symptoms might be better assessed if measured separately from tests of physical signs of disease.

SF-36 is an eight-domain questionnaire that globally evaluates health-related QOL and includes a ‘vitality’ domain, which correlates reasonably well with fatigue questionnaires.

PROFAD was specifically designed for patients with pSS to evaluate fatigue and discomfort [15]. Bowman et al. developed and validated Sicca Symptoms Inventory (SSI index) for sicca severity [22]. The two scales have been used together (PROFAD-SSI) in various studies, as a profile that captures the symptomatic component of pSS. It includes 64 questions in eight domains (somatic and mental fatigue, arthralgia, vascular dysfunction and oral, ocular, cutaneous and vaginal dryness). A short form (19 questions) of PROFAD-SSI or a briefer version (PROFAD-SSI-brev) comprising a series of visual analogue scales (VAS) were both found to correlate well with the long form and so may be preferred in clinical trials [21]. The study that supported SCAI found reasonable correlations between fatigue, arthritis and Raynaud’s domains of the index and comparable domains of SF-36 and PROFAD.

Recently, ESSPRI was proposed to complement the ESSDAI to assess the subjective items and was evaluated experimentally along with the new activity index. It is a short, VAS-based, questionnaire comprising the items of the PROFAD-SSI-brev with some additional test items (study awaiting publication).

Discussion

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conflicts of interest
  8. Address
  9. References

Unresolved issues

The generally accepted concept that potentially reversible inflammatory activity leads over time to permanent damage and consequent functional impairment has been essential to the development of outcome measures in autoimmune diseases. In conditions such as Rheumatoid Arthritis (RA) and SLE, this approach has proven to be of value in routine clinical assessment. However, in pSS differentiating between features determined by disease activity and those attributed to damage, is more challenging.

Impairment of glandular function may reflect active inflammation or damage to the glands (acinar damage or cell loss) or auto-antibody-mediated functional suppression in the absence of much local inflammation. The correlation between the degree of inflammation or damage on biopsy of the exocrine glands and the degree of their functional impairment leading to symptoms and signs of dryness is, however, modest [25]. Clinical measurement of glandular function, e.g. Schirmer-I test or USFR cannot therefore be used to distinguish, sensitively or reliably, between the inflammatory activity and damage in the gland. In their studies using B-cell depletion in pSS, Pijpe and Devauchelle did find some benefit for treating dryness but restricted to patients with early disease [26,27], and Meijer et al. had showed efficacy in the improvement of salivary function in patients with residual secretory function [28], indicating that these symptoms might be, at least partially, reversible during the clinical course. In all the indices analysed, however, sicca complaints are considered to be because of damage. Routine repeat biopsies are probably unacceptable to most patients, emphasising the importance of other less invasive techniques to ‘capture’ the consequences of Sjögren’s pathophysiology.

Fatigue, being as prominent in patients with pSS as in those with SLE and RA [8,16,22,29], is considered multifactorial in nature in pSS, with biological and psychosocial causes. Depression and fibromyalgia appear to play only a minor role [15]. In all indices analysed, fatigue is considered an activity item and therapeutic trials with hydroxychloroquine and B-cell depletion, with some of these using it as a primary end-point, have reported significant benefit [30–33]. However, the clinical course of fatigue is not linear with other manifestations of pSS and is probably not completely understood yet. Walker et al. demonstrated that urological symptoms (primarily urgency) and daytime sleepiness are more severe in patients with pSS and a trend towards increased fatigue was simultaneously noted. It was hypothesised that fatigue may, in some cases, be secondary to an underlying sleep disorder leading to day-time sleepiness, which in turn might result from repeated waking ocular/oral discomfort, or possibly from obstructive sleep apnoea associated with altered surface tension in dry upper airway mucosal surfaces [34]. This claim emphasises that, apart being a prominent feature that might justify systemic therapy per se, other causes for fatigue in these patients must always be considered.

A stable serological profile is usually reported in pSS, with the exception of those cases where the long-standing B-cell hyperreactivity transforms into malignant lymphomas. The risk of lymphoma is estimated to be increased by 16- to 40-fold and carries an increased mortality rate [5–8,35,36]. Several groups [6,37–42] have independently characterised some prognostic factors at diagnosis (biomarkers), to identify high-risk patients for an adverse outcome, notably the development of lymphoma (Table 3). These features identify a subset of patients in whom a closer follow-up, and probably an earlier and more aggressive therapeutic management, should be mandatory. A better knowledge of the natural history of pSS is also needed to improve the follow-up of these patients. There is still no consensus as to whether lymphoma should be considered as activity or damage. In the indices analysed, only ESSDAI regarded lymphoma as an activity feature, whilst the other two indices considered it as a damage item and we agree with this ‘position’.

Table 3.   Proposed prognostic factors for an adverse outcome in patients with primary Sjögren’s syndrome
Prognostic factors for an adverse outcome (Lymphoma)Proposing Authors (Year) [6,37–42]
Skin vasculitis/ palpable purpuraSkopouli (2000), Ioannidis (2002), Brito-Zéron (2007)
Hypocomplementaemia (C3/C4)Skopouli (2000), Ioannidis (2002), Ramos-Casals (2005), Theander (2006)
Mixed cryoglobulinaemiaTzioufas (1996), Skopouli (2000), Brito-Zéron (2007)
LymphadenopathyTalal (1967), Voulgarelis (1999)
T CD4+ LymphopeniaTheander (2006), Voulgarelis (1999)
Persistent parotid gland enlargement or severe involvement in scintigraphyTalal (1967), Brito-Zéron (2007) respectively

With the exception of oral damage, end organ damage was demonstrated to be uncommon in pSS [23], and most of the longitudinal studies have reported modest deterioration over time in organ-related symptomatology and function. However, all the unresolved issues described above have made it difficult to define what characterises active disease in pSS and to separate clinical disease manifestations into features of activity and damage/chronicity and partially explains the different approaches taken by the indices analysed.

Although generally considered a stable or slowly progressive condition, some patients may have a ‘flare’ of their disease. Few studies have addressed this issue in pSS. Stevens et al. report a prevalence of 15% of systemic ‘flares’ (corresponding to 3–7% per year) in a small cohort (n = 59) of patients with pSS. The main clinical features were an increase in fatigue (58%) and in musculoskeletal symptoms (58%). All flares settled within 9 months with modest therapy. No exacerbations involving internal organs were seen. However, this was a small retrospective study [43]. It is interesting that several groups (e.g. the UK Newcastle’s cohort – SICCA – and the French ASSESS cohort) are now undertaking the badly needed prospective studies.

Summary

A growing number of clinical trials have emphasised pSS as a potentially treatable disorder. Lack of assessment tools in these trials has hampered comparison between studies. Several proposals for outcome measures have been published, but no consensus on a core set of criteria for monitoring this important autoimmune disorder has been achieved. Three indices for activity and two for damage in pSS were recently proposed. While there is some uniformity about the most significant features to be included, important differences exist, particularly in relation as to whether to include exocrine and nonexocrine features in a single index or to assess the subjective items in a separate scale.

All indices analysed demonstrated a potentially useful benefit in addition to the panel of exocrine function and subjective outcome measures currently available for patients with pSS. Further studies, preferentially with the inclusion of a larger number of patients and with an increased follow-up period, are needed to assess reliability and sensitivity to change, particularly in response to therapy, to validate their use in clinical trials for new therapeutic approaches. Improvement in our knowledge of the pathophysiology and clinical evolution of pSS is important to address some unresolved issues – the most important being the distinction between activity and damage.

Conflicts of interest

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conflicts of interest
  8. Address
  9. References

Ana Campar and David A. Isenberg declare no conflicts of interest.

Address

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conflicts of interest
  8. Address
  9. References

Santo António Hospital, Internal Medicine Department, Porto, Portugal (A. Campar); ARC Diamond Jubilee Professor Rheumatology at University College London, Centre for Rheumatology, University College London, London, UK (D. A. Isenberg).

References

  1. Top of page
  2. Abstract
  3. Background
  4. Methods
  5. Results
  6. Discussion
  7. Conflicts of interest
  8. Address
  9. References
  • 1
    Fox RI. Sjögren’s syndrome. Lancet 2005;366:32131.
  • 2
    Konttinen YT, Käsna-Ronkainen L. Sjögren’s syndrome: viewpoint on pathogenesis. Scand J Rheumatol 2002;116(31 Suppl):1522.
  • 3
    Delaleu N, Jonsson MV, Appel S, Jonsson R. New concepts in the pathogenesis of Sjögren’s syndrome. Rheum Dis Clin N Am 2008;34:83345.
  • 4
    Nikolov NP, Illei GG. Pathogenesis of Sjögren’s syndrome. Curr Opin Rheumatol 2009;21:46570.
  • 5
    Lazarus MN, Robinson D, Mak V, Møller H, Isenberg DA. Incidence of cancer in a cohort of patients with primary Sjögren’s syndrome. Rheumatology 2006;45:10125.
  • 6
    Ioannidis JP, Vassiliou VA, Moutsopoulos HM. Long-term risk of mortality and lymphoproliferative disease and predictive classification of primary Sjögren’s syndrome. Arthritis Rheum 2002;46:7417.
  • 7
    Theander E, Manthorpe R, Jacobsson LT. Mortality and causes of death in primary Sjögren’s syndrome: a prospective cohort study. Arthritis Rheum 2004;50:12629.
  • 8
    Kassan SS, Moutsopoulos HM. Clinical manifestations and early diagnosis of Sjögren’s syndrome. Arch Intern Med 2004;164:127584.
  • 9
    Bowman SJ. Patient-reported outcomes including fatigue in primary Sjögren’s syndrome. Rheum Dis Clin N Am 2008;34:94962.
  • 10
    Vitali C. Measurement of disease activity and damage in Sjögren’s syndrome. Rheum Dis N Am 2008;34:96371.
  • 11
    Vitali C, Palombi G, Baldini C, Benucci M, Bombardieri S, Covelli M et al. Sjögren’s Syndrome Disease Damage Index and Disease Activity Index. Arthritis Rheum 2007;56:222331.
  • 12
    Bowman SJ, Sutcliffe N, Isenberg DA, Goldblatt F, Adler M, Price E et al. Sjögren’s Systemic Clinical Activity Index (SCAI) – a systemic disease activity measure for use in clinical trials in primary Sjögren’s syndrome. Rheumatology (Oxford) 2007;46:184551.
  • 13
    Seror R, Ravaud P, Bowman S, Baron G, Tzioufas A, Theander E et al. EULAR Sjögren’s Syndrome Disease Activity Index (ESSDAI): development of a consensus systemic disease index in primary Sjögren’s syndrome. Ann Rheum Dis 2009; doi:10.1136/ard.2009.110619 [Epub ahead of print].
  • 14
    Barry RJ, Sutcliffe N, Isenberg DA, Price E, Goldblatt F, Adler M et al. The Sjögren’s Syndrome Damage Index – a damage index for use in clinical trials and observational studies in primary Sjögren’s syndrome. Rheumatology (Oxford) 2008;47:11938.
  • 15
    Bowman SJ, Booth DA, Platts RG, UK Sjögren’s Interest Group. Measurement of fatigue and discomfort in primary Sjögren’s syndrome using a new questionnaire tool. Rheumatology (Oxford) 2004;43:75864.
  • 16
    Strombeck B, Ekdahl C, Manthorpe R, Wikstrom I, Jacobsson L. Health-related quality of life in primary Sjögren’s syndrome, rheumatoid arthritis and fibromyalgia compared to normal population data using SF-36. Scand J Rheumatol 2000;29:208.
  • 17
    Vitali C, Bombardieri S, Jonsson R, Moutsopoulos HM, Alexander EL, Carsons SE 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;61:5548.
  • 18
    Seror R, Ravaud P, Bowman SJ, Baron G, Tzioufas AG, Theander E et al. Development and preliminary validation of the EULAR Sjögren’s syndrome disease activity index (ESSDAI): a new consensual disease activity index that detects changes accurately. Presented at the American College Rheumatology 2009 Annual Meeting, Pennsylvania, October 18.
  • 19
    Yee CS, Farewell V, Isenberg DA, Griffiths B, Teh LS, Bruce IN et al. The BILAG-2004 index is sensitive to change for assessment of SLE disease activity. Rheumatology (Oxford) 2009;48:6915.
  • 20
    Segal B, Bowman SJ, Fox PC, Vivino FB, Murukutla N, Brodscholl J et al. Primary Sjögren’s Syndrome: health experiences and predictors of health quality among patients in the United States. Health and Quality of Life Outcomes 2009;7:46. doi:10.1186/1477-7525-7-46.
  • 21
    Bowman SJ, Hamburger J, Richards A, Barry RJ, Rauz S. Patient-reported outcomes in primary Sjögren’s syndrome: comparison of the long and short versions of the Profile of Fatigue and Discomfort-Sicca Symptoms Inventory. Rheumatology (Oxford) 2009;48:1403.
  • 22
    Bowman SJ, Booth DA, Platts RG, Field A, Rostron J, UK Sjögren’s Interest Group. Validation of the Sicca Symptoms Inventory for clinical studies of Sjögren syndrome. J Rheumatol 2003;30:125966.
  • 23
    Sutcliffe N, Stoll T, Pyke S, Isenberg DA. Functional disability and end organ damage in patients with Systemic Lupus Erythematosus (SLE) and Sjögren’s Syndrome (SS), and primary SS. J Rheumatol 1999;26:5047.
  • 24
    Bowman SJ. Collaborative research into outcome measures in Sjögren’s syndrome. Update on disease assessment. Scand J Rheumatol 2002;116(31 Suppl):237.
  • 25
    Hay EM, Thomas E, Pal B, Hajeer A, Chambers H, Silman AJ. Weak association between subjective symptoms and objective testing for dry eyes and dry mouth: results from a population based study. Ann Rheum Dis 1998;57:204.
  • 26
    Pijpe J, Van Imhoff GW, Spijkervet FK, Roodenburg JL, Wolbink GJ, Mansour K et al. Rituximab treatment in patients with primary Sjögren’s syndrome: an open-label phase II study. Arthritis Rheum 2005;52:274050.
  • 27
    Devauchelle-Pensec V, Pennec Y, Morvan J, Pers JO, Daridon C, Jousse-Joulin S et al. Improvement of Sjögren’s syndrome after two infusions of rituximab (anti-CD20). Arthritis Rheum 2007;57:3107.
  • 28
    Meijer JM, Meiners PM, Vissink A, Spijkervet FK, Abdulahad W, Kamminga N et al. Effective rituximab treatment in primary Sjögren’s syndrome: a randomised, double-blind, placebo-controlled trial. Arthritis Rheum 2010;62:9608.
  • 29
    Barendregt PJ, Visser MR, Smets EM, Tulen JH, Van Den Meiracker AH, Boomsma F et al. Fatigue in primary Sjögren’s syndrome. Ann Rheum Dis 1998;57:2915.
  • 30
    Dass S, Bowman SJ, Vital EM, Ikeda K, Pease CT, Hamburger J et al. Reduction of fatigue in Sjögren’s syndrome with rituximab: results of a randomised, double-blind, placebo controlled pilot study. Ann Rheum Dis 2008;67:15414.
  • 31
    Ramos-Casals M, Tzioufas AG, Font J. Primary Sjögren’s syndrome: new clinical and therapeutic concepts. Ann Rheum Dis 2005;64:34754.
  • 32
    Mariette X. Therapeutic potential for B-cell modulation in Sjögren’s syndrome. Rheum Dis Clin N Am 2008;34:102533.
  • 33
    Isaksen K, Jonsson R, Omdal R. Anti-CD20 treatment in primary Sjögren’s syndrome. Scand J Immunol 2008;68:55464.
  • 34
    Walker J, Gordon T, Lester S, Downie-Doyle S, McEvoy D, Pile K et al. Increased severity of lower urinary tract symptoms and daytime somnolence in primary Sjögren’s syndrome. J Rheumatol 2003;30:240612.
  • 35
    Zintzaras E, Voulgarelis M, Moutsopoulos HM. The risk of lymphoma development in autoimmune diseases. Arch Intern Med 2005;165:233744.
  • 36
    Skopouli FN, Dafni U, Ioannidis JP, Moutsopoulos HM. Clinical evolution and morbidity and mortality pf primary Sjögren’s syndrome. Semin Arthritis Rheum 2000;29:296304.
  • 37
    Brito-Zerón P, Ramos-Casals M, Bove A, Sentis J, Font J. Predicting adverse outcomes in primary Sjögren’s syndrome: identification of prognostic factors. Rheumatology (Oxford) 2007;46:135962.
  • 38
    Ramos-Casals M, Brito-Zerón P, Yagüe J, Akasbi M, Bautista R, Ruano M et al. Hypocomplementaemia as an immunological marker of morbidity and mortality in patients with primary Sjögren’s syndrome. Rheumatology (Oxford) 2005;44:8994.
  • 39
    Theander E, Henriksson G, Ljungberg O, Mandl T, Manthorpe R, Jacobsson LT. Lymphoma and other malignancies in primary Sjögren’s syndrome: a cohort study on cancer incidence and lymphoma predictors. Ann Rheum Dis 2006;65:796803.
  • 40
    Tzioufas AG, Boumba DS, Skopouli FN, Moutsopoulos HM. Mixed monoclonal cryoglobulinemia and monoclonal rheumatoid factor cross-reactive idiotypes as predictive factors for the development of lymphoma in patients with primary Sjögren’s syndrome. Arthritis Rheum 1996;39:76772.
  • 41
    Talal N, Sokoloff L, Barth WF. Extrasalivary lymphoid abnormalities in Sjögren’s syndrome (reticulum-cell sarcoma, “pseudolymphoma”, macroglobulinemia). Am J Med 1967;43:5065.
  • 42
    Voulgarelis M, Dafni UG, Isenberg DA, Moutsopoulos HM, Members of the European Concerted Action on Sjögren’s Syndrome. Malignant lymphoma in primary Sjögren’s syndrome: a multicenter, retrospective, clinical study by the European Concerted Action on Sjögren’s Syndrome. Arthritis Rheum 1999;42:176572.
  • 43
    Stevens RJ, Hamburger J, Ainsworth JR, Holmes G, Bowman SJ. Flares of systemic disease in primary Sjögren’s syndrome. Rheumatology (Oxford) 2005;44:4023.