Skip Navigation


Rheumatology Advance Access originally published online on April 10, 2007
Rheumatology 2007 46(7):1107-1110; doi:10.1093/rheumatology/kel432
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
46/7/1107    most recent
kel432v1
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 Serednicka, K.
Right arrow Articles by Denton, C. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Serednicka, K.
Right arrow Articles by Denton, C. P.
Related Collections
Right arrow Systemic Sclerosis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Using a self-reported functional score to assess disease progression in systemic sclerosis

K. Serednicka1,2, A. E. Smyth3, C. M. Black1 and C. P. Denton1

1Centre for Rheumatology, Royal Free Hospital, London, 2Karolinska University Hospital, Stockholm, Sweden and 3Ulster Hospital, Belfast, UK.

Correspondence to: C. P. Denton, Centre for Rheumatology, Royal Free Hospital, Pond Street, London NW3 2QG, UK. E-mail: c.denton{at}medsch.ucl.ac.uk.


   Abstract
Top
 Abstract
Introduction
 Methods
 Assessment measures
 Data collection and analysis
 Statistical analysis
 Results
 Discussion
 Acknowledgements
 References
 
Objectives. This study compares the scleroderma Functional Score (FS) with the validated Disability Index of the Health Assessment Questionnaire (HAQ-DI) and other outcome measures. The aim is to determine if the FS is useful as an objective assessment tool for longitudinal evaluation of the functional impact of systemic sclerosis (SSc).

Methods. A cohort of 135 patients was studied (M:F, 15:120), with a mean age of 45.7 (S.D. = 13.2) at SSc disease onset. 69 (51%) had diffuse cutaneous scleroderma (dcSSc) and 66 (49%) had limited disease (lcSSc). The mean interval between the two assessments was 1.8 yrs (S.D. = 1.2). Functional impact was determined by evaluating archived self-reported questionnaires (FS, HAQ and scleroderma-VAS). Concurrent evaluation of the disease severity score was derived from clinical data stored in the hospital database and from medical case note reviews.

Results. At baseline, the mean FS was 11.0 (S.D. = 9.0) and at reassessment 12.0 (S.D. = 9.2). The mean absolute change in FS between the two assessments was 4.1 (S.D. = 4.9). With time 49% (n = 66) showed a clinically significant change in their functional ability with regard to FS, of these 29% (n = 39) worsened and 20% (n = 27) improved. There was an excellent cross-sectional correlation between the FS and the HAQ-DI ({rho} = 0.90; P < 0.0001). Also, a strong correlation between longitudinal change in these two outcome measures ({rho} = 0.59, P < 0.0001) was observed.

Conclusions. This is the first longitudinal study of the scleroderma FS. It demonstrates that the FS can capture bidirectional and clinically significant changes in SSc related disability over time. The concurrent validity of the FS is asserted through its strong correlation with the HAQ-DI. The FS is a disease-specific, inexpensive and practical instrument for assessing functional status in SSc. It is a promising self-administered assessment tool for use in evaluating new SSc treatments.

KEY WORDS: Scleroderma, Functional Indices, HAQ-DI, SSc Functional Score


   Introduction
Top
 Abstract
 Introduction
Methods
 Assessment measures
 Data collection and analysis
 Statistical analysis
 Results
 Discussion
 Acknowledgements
 References
 
Systemic sclerosis (SSc, scleroderma) is a complex multisystem autoimmune disease involving vascular damage, inflammation and fibrosis [1]. In recent years there has been a positive trend of a prolonged survival in SSc, and accordingly morbidity has emerged as one of the major clinical aspects of the disease. Validated scleroderma-specific disability outcome measures are needed in clinical practice and for research. At present the Scleroderma Health Assessment Questionnaire (SHAQ) [2] is the most widely used and best characterized outcome measure in SSc. It combines the disability index of the (Health Assessment Questionnaire) HAQ [3], which is widely validated for other rheumatic diseases with a series of visual analogue scales (VAS) [2] specifically related to SSc. The SHAQ has been shown to be both cross-sectionally and longitudinally reliable [4]. An alternative index of function is the Scleroderma Functional Score (FS) [5], which focuses on disability caused by skin tightness in the upper limb and proximal muscle weakness. It was developed using a Delphi technique by a multidisciplinary group of scleroderma experts [5]. Smyth et al. [1] recently assessed cross-sectional validity in a study of more than 100 consecutive patients, where the FS was shown to correlate with the Health Assessment Disability Index (HAQ-DI) and individual outcome measures.

In the present study, we assess the utility of the FS, as an objective assessment tool for use in longitudinal evaluation of SSc. We also compare changes in FS with other outcome measures including the SHAQ.


   Methods
Top
 Abstract
 Introduction
 Methods
Assessment measures
 Data collection and analysis
 Statistical analysis
 Results
 Discussion
 Acknowledgements
 References
 
Patient characteristics
One hundred and thirty-five scleroderma patients from the Royal Free cohort were studied. All patients fulfilled the 1980 American College of Rheumatology Criteria [6]. Limited cutaneous SSc (lcSSc) was present in 49%, and diffuse cutaneous sclerosis (dcSSc) in 51%. Demographic features including sex, age, the duration of Raynaud's phenomenon (RP) and the duration of scleroderma were documented. Consecutive cases were identified through their willingness to participate and complete the FS-and SHAQ questionnaires and through the availability of at least one previous assessment. Local ethical approval was obtained for this analysis.

Patients were predominantly female (88.9%) and had a mean age of 45.7 yrs (S.D. = 13.2, range 12.6–80.9) at SSc disease onset for both male and female cases. The mean disease duration at baseline was 7.8 yrs (S.D. = 7.4, range 0–35.3) and they had experienced RP for a mean of 11.8 yrs (S.D. = 12.5, range 0.3–64). For more detailed demographic data, refer to Table 1.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Demographics and clinical features of the cohort

 

   Assessment measures
Top
 Abstract
 Introduction
 Methods
 Assessment measures
Data collection and analysis
 Statistical analysis
 Results
 Discussion
 Acknowledgements
 References
 
The FS [5] is a self-administered 11-item functional questionnaire, scored from 0 (normal) to 4 (impossible to achieve), with an overall score between 0–33. The HAQ [7], measures physical disability in 8 domains of activity. The individual scores are aggregated to yield a single index of disability HAQ-DI, ranging from 0 (normal) to 3 (severe). The use of aids and devices were not taken into account. The SHAQ includes the disability and pain scale of the HAQ plus five scleroderma-specific VAS [2], scaled from 0–3. In addition the composite Hand HAQ Disability Index [8] (Hand HAQ-DI) was evaluated. The validated Medsger Severity Scale (SS) [9] was used to determine organ specific disease severity, scaled from 0–36. The modified Rodnan Skin Score [10] quantifying skin thickness was recorded, scaled from 0–51.


   Data collection and analysis
Top
 Abstract
 Introduction
 Methods
 Assessment measures
 Data collection and analysis
Statistical analysis
 Results
 Discussion
 Acknowledgements
 References
 
Two or more different assessment times were compared using self-reported questionnaires. The mean duration between the first and second assessment was 1.8 yrs (S.D. = 1.2, range 0.2–5.7). Concurrent evaluation of the disease severity score was derived from the departmental database and medical case note reviews. Likewise, specific clinical variables such as antibodies, lung function, forced vital capacity (FVC%) and carbon monoxide transfer factor (TLCO%) were recorded. Skin scoring and clinical data collection were performed by five rheumatologists, each trained in assessment of SSc patients by the same senior investigator (CPD).

When evaluating the change in the score between two assessments, patients were classified according to their condition being stable, improved or worsened, in relation to each assessment tool (FS, HAQ-DI and SS). A change in three units for FS and two units for the SS, was judged as a clinically significant change in functional impairment and disease severity, respectively. The minimal clinically important difference (MCID) for a change in the HAQ-DI has previously been shown to be 0.22 units in rheumatoid arthritis [11,12].


   Statistical analysis
Top
 Abstract
 Introduction
 Methods
 Assessment measures
 Data collection and analysis
 Statistical analysis
Results
 Discussion
 Acknowledgements
 References
 
Spearman's {rho} correlation coefficients estimated the correlation between the functional indices. Chi-squared tests explored association between clinically significant changes in FS/SS and FS/HAQ-DI. As functional ability might be related to age, disease duration and the interval between the two assessments, correlation between FS and these variables was also examined. Cases with missing data were excluded from the relevant analysis. The data were statistically analysed by MINITAB TM software.


   Results
Top
 Abstract
 Introduction
 Methods
 Assessment measures
 Data collection and analysis
 Statistical analysis
 Results
Discussion
 Acknowledgements
 References
 
Baseline characteristics
At the first assessment the mean FS was 11 (S.D. = 9.0) and at reassessment 12 (S.D. = 9.2). The corresponding values for the HAQ-DI was a mean of 1.0 (S.D. = 0.8) at baseline and 1.1 (S.D. = 0.8) at reassessment. The mean total SS score at baseline and the second assessments was in both cases 7.2 (S.D. = 3.0 and S.D. = 2.8, respectively) Among the VAS scales the highest baseline values were observed for overall disease 1.3 (S.D. = 0.9), pain-VAS 1.1 (S.D. = 0.9) and RP 1.0 (S.D. = 0.9).

Change in disease
Diagnostic measures with their accompanying basic statistics and the correlation coefficients for change in FS vs change in outcome measures are detailed in Table 2. The mean absolute change was 4.1 (S.D. = 4.9) between the two evaluations. The HAQ-DI had a mean absolute change of 0.4 (S.D. = 0.4) and corresponding change in SS was 1.6 (S.D. = 1.6). All of the VAS scales were responsive to change ranging between 0.5–0.6, for the mean absolute change. During follow-up 49% of the patients (n = 66) showed a clinically significant change in FS (dFS = ±3), of these 29% (n = 39) worsened in their functional ability and 20% (n = 27) improved. Change in FS was not dependent of the patient's age ({rho} = 0.137 P = 0.113) disease duration ({rho} = –0.036, P = 0.680) and the time interval between the two assessments ({rho} = 0.141, P = 0.104).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Functional scores and clinical features at baseline and reassessment, with corresponding change over time together with correlation coefficients for change in FS vs change in outcome measure and CI. Bold figures indicate comparably very low P-values

 
The change in the score between assessments was also evaluated by classifying the patients according to their condition being stable, improved or worsened in relation to the assessment tools (FS, HAQ-DI, and SS). Although opposite direction of change in FS and SS (worse/improved) was infrequent (n = 9, 7%), the number of subjects showing complete agreement in assessment of change was also relatively low (n = 54, 40%). Overall, there was not a significant association between change in FS and SS (P > 1 Chi-squared test). When comparing the change in FS and change in HAQ-DI there was a significant association between change in the two measures (P ≤ 0.001 Chi-squared test) and 56% (n = 76) of the subjects showed complete agreement in assessment.

Correlation between variables
As previously shown by Smyth et al. [1], there is an excellent cross-sectional correlation between FS and HAQ-DI (Pearson's {rho} = 0.89, P < 0.0001, baseline). In this study the correlation was also analysed using Spearman's correlation coefficient, as it is more robust for outliers [13]. Correlation between FS and HAQ-DI was confirmed for both assessments (Spearman's {rho} = 0.90 baseline, {rho} = 0.89 second assessment, both P < 0.0001). Similarly, the FS also showed high cross-sectional correlation with the Hand HAQ-DI ({rho} = 0.85, P < 0.0001, second assessment). Likewise, the correlations between FS and other outcome measures were congruent with the cross sectional study by Smyth et al. [1]. When comparing change in the functional measures over time, the FS and HAQ-DI also showed strong correlation (Spearman's {rho} = 0.59, P < 0.0001), as did also the correlation of change in FS with change in Hand HAQ-DI, (Spearman's {rho} = 0.58, P < 0.0001) (Fig. 1).


Figure 1
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 1. Correlation of change in FS and change in HAQ-DI or Hand HAQ-DI. A. There was a highly statistically significant correlation between change in the FS and the change in HAQ-DI between the two assessments in our study. Spearman's correlation coefficient ({rho} = 0.59), Pearson's ({rho} = 0.58), both P < 0.0001. B. Interval change in a derived score (Hand HAQ-DI) that has been suggested to reflect upper limb function that is a composite of the dressing/grooming, hygiene and grip domains of the HAQ-DI also correlated significantly with the change in functional score for the scleroderma subjects examined in this study. Spearman's correlation coefficient ({rho} = 0.58), Pearson's ({rho} = 0.55) P < 0.0001.

 
Table 2, shows correlation coefficients for change in FS vs change in outcome measures. Change in VAS for overall disease severity reflected change in FS ({rho} = 0.38, P < 0.0001). Similarly, change in FS correlated with Pain-VAS ({rho} = 0.24, P = 0.006).


   Discussion
Top
 Abstract
 Introduction
 Methods
 Assessment measures
 Data collection and analysis
 Statistical analysis
 Results
 Discussion
Acknowledgements
 References
 
In this study we show that the scleroderma FS can capture bidirectional and clinically meaningful changes in functional disability over time in dSSc and lcSSc. In addition, longitudinal correlation between the FS and the HAQ-DI has been confirmed. Moreover, the excellent cross-sectional correlation between the FS and HAQ-DI, as previously established by Smyth et al. [1] is reproduced. However, when evaluating change in assessments (FS/HAQ-DI) in relation to condition being stable, improved or worsened, only 56% (n = 76) of the subjects showed complete agreement in assessment. This emphasizes that the two tools may assess different disease facets and may be complementary, though agreeing overall.

The mean FS score in the cohort ranged from 11.0–12.0 between the first and the second assessment. This is consistent with the mean FS of 11.9, reported previously [1]. Mean HAQ-DI in the population ranged from 1.0–1.1, for the two evaluations. This corresponds well with a formerly reported range of 0.83–1.2 for scleroderma patients [14], and is also similar to the recorded measures for RA and SLE. Despite not being scleroderma specific, HAQ-DI draws its strength from being a well established, valid and reliable tool for assessing functional ability. Also, it allows for comparison across different diseases and provides an overall evaluation of functional ability. Nonetheless, it may not be sensitive to skin tightness and muscle weakness, which are prominent in SSc. The FS, on the other hand, specifically addresses areas that have been identified by SSc patients and therapists as particularly important in limiting their functional ability [5]. Moreover, the FS is an inexpensive and effective way of measuring disability; it is rapid, easy to use and readily understood by patients. It can be completed in approximately 5 min, well below the recommended standard of feasibility [15]. Reassuringly, the FS correlated with the Hand HAQ-DI. However, at present it is not a validated outcome measure. It only includes three out of six HAQ components related to hand function and is thus significantly less precise and comprehensive than the FS.

The Scleroderma VAS captures disability secondary to internal organ involvement. The scales are responsive to change over time; the absolute mean change in this study ranged from 0.5–0.6 units between the different VAS scales, which is somewhat greater than the change of 0.4 units, observed for the HAQ-DI. There was a poor longitudinal correlation between the FS and the VAS scales, only the Pain-VAS and the VAS for overall disease correlating with FS.

Earlier studies suggest that VAS scales are less reliable than the HAQ-DI [1,16] possibly reflecting patient subjectivity and conceptual difficulty. Nonetheless, VAS scales are valuable in that they are distinct to scleroderma and may be useful outcome measures in clinical trials to assess the effect of interventions.

Modified Rodnan Skin Score is an established outcome measure in SSc, but in recent years some important limitations have become evident. It is valid only when performed by an experienced clinician who has undergone standardized training [17]. Besides, it is mainly applicable to dcSSc [18] and has limited value when applied to lcSSc. Furthermore, in diffuse disease the Skin Score seems to improve spontaneously with time [19]. Indeed, other studies have recorded apparent discrepancy between improvement in Skin Score and worsening of internal organ involvement [20,21]. Similarly, the lack of correlation between Skin Score and FS may also suggest that although Skin Score ameliorates with time, this does not necessarily mean hand strength and grip function improves. In our cohort there was no correlation between change in FS and change in Skin Score, as has similarly been demonstrated by Herrick et al. [22]. Similarly there was a lack of longitudinal correlation between FS and Severity Scale ({rho} = 0.18, P = 0.033) as well as lung parameters (FVC% {rho} = –0.03, P = 0.741, TLCO% {rho} = –0.15, P = 0.094). Although the SS is widely accepted as a useful clinical tool [10]; it was not designed to give a meaningful composite score, and any interpretation should be done with caution. Nevertheless, it is worthy of notice that the FS did not readily correlate with the clinical parameters collected in this study.

In conclusion, we report the first longitudinal evaluation of the scleroderma FS and demonstrate its considerable capacity for capturing bidirectional and clinically significant changes in functional impairment over time. The FS showed strong cross-sectional correlation with the validated HAQ-DI and significant correlation between longitudinal changes in these two measures.

C.P.D. has received lecture fees, grant support or acted as a consultant to Actelion Pharmaceuticals, Genzyme Inc. and Aspreva Pharmaceuticals. C.M.B. has received honoraria, consultancy fees and research grants from Actelion Pharmaceuticals, Genzyme Inc. and Encysive Inc. K.S. has received grants from Fredirika-Bremer Förbundets Stipendiestiftelse and Längmanska Kulturfonden.


   Acknowledgements
Top
 Abstract
 Introduction
 Methods
 Assessment measures
 Data collection and analysis
 Statistical analysis
 Results
 Discussion
 Acknowledgements
References
 
The authors offer sincere thanks to statistician Øystein Evandt, from DNV Norway, for valuable advice regarding the statistical analysis. Financial support for clinical research in the department comes from Arthritis Research Campaign (arc), The Raynaud's and Scleroderma Association (UK) and The Scleroderma Society (UK).


   References
Top
 Abstract
 Introduction
 Methods
 Assessment measures
 Data collection and analysis
 Statistical analysis
 Results
 Discussion
 Acknowledgements
 References
 

  1. Smyth AE, MacGregor AJ, Mukerjee D, Brough GM, Black CM, Denton CP. A cross-sectional comparison of three self-reported functional indices in scleroderma. Rheumatology (2003) 42:732–8.[Abstract/Free Full Text]
  2. Steen VD, Medsger TA Jr. The value of the Health Assessment Questionnaire and special patient-generated scales to demonstrate change in systemic sclerosis patients over time. Arthritis Rheum (1997) 40:1984–91.[ISI][Medline]
  3. Fries JF, Spitz P, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis Rheum (1980) 23:137–45.[ISI][Medline]
  4. Clements PJ, Wong WK, Hurwitz EL, et al. The Disability Index of the Health Assessment Questionnaire is a predictor and correlate of outcome in the high-dose versus low-dose penicillamine in systemic sclerosis trial. Arthritis Rheum (2001) 44:653–61.[CrossRef][ISI][Medline]
  5. Silman A, Akesson A, Newman J, et al. Assessment of functional ability in patients with scleroderma: a proposed new disability assessment instrument. J Rheumatol (1998) 25:79–83.[ISI][Medline]
  6. Subcommittee for scleroderma criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum (1980) 23:581–90.[ISI][Medline]
  7. Fries JF, Spitz P, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis Rheum (1980) 23:137–45.[ISI][Medline]
  8. Korn JH, Mayes M, Matucci Cerinic M, et al. Digital ulcers in systemic sclerosis: prevention by treatment with bosentan, an oral endothelin receptor antagonist. Arthritis Rheum (2004) 50:3985–93.[CrossRef][ISI][Medline]
  9. Medsger TA Jr, Silman AJ, Steen VD, et al. A disease severity scale for systemic sclerosis: development and testing. J Rheumatol (1999) 26:2159–67.[ISI][Medline]
  10. Clements P, Lachenbruch P, Siebold J, et al. Inter and intraobserver variability of total skin thickness score (modified Rodnan TSS) in systemic sclerosis. J Rheumatol (1995) 22:1281–5.[ISI][Medline]
  11. Kosinski M, Zhao SZ, Dedhiya S, Osterhaus JT, Ware JE Jr. Determining minimally important changes in generic and disease-specific health-related quality of life questionnaires in clinical trials of rheumatoid arthritis. Arthritis Rheum (2000) 43:1478–87.[CrossRef][ISI][Medline]
  12. Guzman J, Maetzel A, Peloso P, Yeung N, Bombardier C. Disability scores in DMARD trials: what is a clinically important change? Arthritis Rheum (1996) 39((Suppl. 9)):S208.
  13. Evandt O, Coleman S, Ramalhoto MF, van Lottum C. A Little-known robust estimator of the correlation coefficient and its use in a robust graphical test for bivariate normality with applications in the aluminium industry. Quality and Reliability Engineering International (2004) 20:433–56.[CrossRef][ISI]
  14. Poole JL, Williams CA, Bloch DA, Hollak B, Spitz P. Concurrent validity of the Health Assessment Questionnaire Disability Index in Scleroderma. Arthritis Care Res (1995) 8:189–93.[CrossRef][Medline]
  15. McHorney CA, Tarlov AR. Individual-patient monitoring in clinical practice: are available health status surveys adequate? Qual Life Res (1995) 4:293–307.[CrossRef][ISI][Medline]
  16. Merkel PA, Herlyn K, Martin RW, et al. Measuring disease activity and functional status in patients with scleroderma and raynaud's phenomenon. Arthritis Rheum (2002) 46:2410–20.[CrossRef][ISI][Medline]
  17. Clements P, Lachenbruch P, Siebold J, et al. Inter and intraobserver variability of total skin thickness score (modified Rodnan TSS) in systemic sclerosis. J Rheumatol (1995) 22:1281–5.[ISI][Medline]
  18. Denton CP, Black CM. Scleroderma–clinical and pathological advances. Best Pract Res Clin Rheumatol (2004) 18:271–90.[Medline]
  19. Seibold JR. Clinical trials: types, design, and end-points. Curr Opin Rheumatol (2001) 13:512–5.[CrossRef][ISI][Medline]
  20. Geirsson AJ, Wollheim FA, Akesson A. Disease severity of 100 patients with systemic sclerosis over a period of 14 years: using a modified Medsger scale. Ann Rheum Dis (2001) 60:1117–22.[Abstract/Free Full Text]
  21. Medsger TA Jr. Epidemiology of systemic sclerosis. Clin Dermatol (1994) 12:207–216.[CrossRef][ISI][Medline]
  22. Herrick A, Rooney B, Finn J, Silman A. Lack of relationship between functional ability and skin score in patients with systemic sclerosis. J Rheumatol (2001) 28:29–5.[ISI][Medline]
Submitted 23 May 2006; revised version accepted 5 December 2006.
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 Full Text (PDF)
Right arrow All Versions of this Article:
46/7/1107    most recent
kel432v1
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 Serednicka, K.
Right arrow Articles by Denton, C. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Serednicka, K.
Right arrow Articles by Denton, C. P.
Related Collections
Right arrow Systemic Sclerosis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?