The Relationship Between Nailfold Capillaroscopic Assessment and Telangiectasia Score with Severity of Peripheral Vascular Involvement in Systemic Sclerosis Y

Home , Microangiopathy

The relationship between nailfold capillaroscopic assessment and telangiectasia score with severity of peripheral vascular involvement in systemic sclerosis Y. Yalcinkaya, O. Pehlivan, A. Omma, N. Alpay, B. Erer, S. Kamali, L. Ocal, M. Inanc

Division of Rheumatology, ABSTRACT Introduction Department of Internal Medicine, Objective. To determine the association Systemic sclerosis (SSc) is a connective Istanbul University, Istanbul, Turkey. of nailfold video-capillaroscopy (NVC) tissue disease characterised by autoim- Yasemin Yalcinkaya, MD findings and telangiectasia score with mune inflammation, vascular injury Ozlem Pehlivan, MD digital ulcer (DU) history and sever- and fibrosis (1). Vascular injury starts Ahmet Omma, MD ity of peripheral vascular involvement in the early phases of the disease and Nilufer Alpay, MD Burak Erer, MD (PVI) in systemic sclerosis (SSc). can cause digital ischaemia-gangrene Sevil Kamali, MD Methods. Fifty-nine SSc patients fulfill- and pulmonary arterial hypertension Lale Ocal, MD ing Leroy & Medsger criteria were eval- (PAH) which were major factors of Murat Inanc, MD uated including telangiectasia score, high morbidity and mortality in SSc. Please address correspondence to: disease activity and severity scores. (2). Raynaud’s phenomenon (RP) is ac- Dr Yasemin Yalcinkaya, NVC was performed according to quali- cepted as the initial clinical symptom Division of Rheumatology, tative (early, active and late patterns) reflecting the microvascular injury be- Department of Internal Medicine, and semi-quantitative assessments. fore the onset of other manifestations Istanbul University, Results. When DU+ and DU- groups Millet Caddesi - Gapa, (3, 4). Fatih 34093, were compared; the mean score of Microvascular injury induces structural Istanbul, Turkey. capillary number (CN) was 2.0±0.5 changes of capillary array and can be E-mail: [email protected] vs. 1.4±0.7 (p<0.001), irregularly en- established by a non-invasive imag- Received on September 22, 2014; accepted larged capillaries (IEC) was 1.8±0.6 ing tool, nailfold video-capillaroscopy in revised form on January 12, 2015. vs. 1.4±0.7 (p<0.05), microangiopathy (NVC), from the early stages of the di- Clin Exp Rheumatol 2015; 33 (Suppl. 91): evolution score (MES) was 2.5±1.5 vs. asese. Therefore it has been accepted as S92-S97. 1.8±1.0 (p<0.05) and ‘early’ pattern a useful method for differentiation of © Copyright Clinical and was significantly less frequent in DU+ primary RP from secondary, early diag- Experimental Rheumatology 2015. patients (1 vs. 9, p=0.016). The fre- nosis of SSc and follow-up of the pro- quency of severe-PVI (Medsger sever- gression of microangiopathy in SSc (5- Key words: systemic sclerosis, ity score of 2–4) was 22% in females 8). NVC assessment has been shown to capillaroscopy, telangiectasia score, (12/54) and 80% in males (4/5). When be a reliable method for the prediction peripheral vascular involvement severe and non-severe groups were of development of digital ulcers (DUs) compared; the mean score of CN was and severity of the organ involvement 2.1±0.4 vs. 1.5±0.7 (p<0.001), MES (9-13). Improvement in NVC abnor- was 2.8±1.6 vs. 1.8±1.1 (p<0.05) and malities after 12 months of treatment ‘early’ pattern was significantly less with an endotelin receptor antagonist, frequent in patients with severe PVI (0 bosentan, supports that NVC may be vs. 9, p=0.049). The mean values of tel- also used for establishing the efficacy angiectasia score were similar between of treatment (14). groups. RP and chronic digital ischaemia lead- Conclusion. DU history and severe ing to DUs can be seen in all subsets PVI in SSc were associated with capil- of disease but more frequent in dcSSc. lary loss and microangiopathy. ‘Early’ DUs heal slowly especially when com- NVC pattern was very rare in patients plicated with infections or gangrene with DU history and was not found in (15, 16). Early stages of the SSc were severe PVI. Severe PVI in males was shown to have higher risk of DU de- more frequent than females. Telan- velopment in a French cohort; first DU giectasia scores were not found to be occured in 43% within 1 year following related to PVI. NVC may be a helpful the first non-Raynaud’s symptom and method in the assessment of SSc pa- 73% within 5 years (17). Digital lesions tients for PVI prognosis, warranting also have a tendency to recur. In the Competing interests: none declared. prospective studies. study by Nihtyanova et al., 1168 SSc

S-92 NVC, telangiectases and digital lesions in SSc / Y. Yalcinkaya et al. patients analysed, 16.6% had at least if RP requiring vasodilators, 2 if digital disorganisation of the normal capillary one episode of DU and 12% hospital- pitting scars, 3 if DU, 4 if gangrene was array, ramified/bushy capillaries (9). ised at least once for vasodilatory treat- present] (23) were evaluated simultane- Secondly, NVC assessed semi-quanti- ments during 18-month period (18). ously with NVC. tatively and scoring of the capillary ab- Telangiectases are one of the most NVC was performed on all patients af- normalities (CN/IEC/GC/H/CR/CAD) visible clinical manifestations of mi- ter 20 minutes of resting time period in (score:0–3) was performed as follows: crovascular pathology composed of a separate room with a constant tem- 0 = no changes, 1 = less than 33% of vasodilated post-capillary venules in perature of 20–22°C by a video-capil- capillary alterations/reduction, 2 = 33– the papillary and superficial reticular laroscopy (Optilia, Mediscope-Digital 66% of capillary alterations /reduction, dermis of the skin without inflamma- Video Microscope). OptiPix (2010, 3 = more than 66% of capillary altera- tory changes (19). They may be found Version:1.2.0) programme was used tions/reduction, per linear mm at the on hands, face, mucosa, limbs and to store the images. The investigator distal row of the nailfold. Microangi- trunk of both of two subgroups of SSc (YY) who recruited patients into the opathy evolution score (MES) was the patients. In a study of Shah et al., a sig- study and evaluated medical records, sum of CN, CR and CAD scores. The nificant correlation between number of performed NVC studies. After a drop of mean score for each capillary abnor- telangiectases and pulmonary vascular cedar oil on the nailfold, at least 2 rep- mality was calculated from the average disease-PAH was found, but associa- resentative images for each of second to score of two images of 1 mm of each 8 tion with other vascular manifestations fifth fingers of left and right hands were digits (2nd, 3rd, 4th and 5th fingers of both like digital ulcers or gangrene could not captured with the 200x optical probe. hands), added together, and then the fi- be shown (20). All images (16 images for 8 fingers) nal value divided by eight to obtain the In this study we determine the associa- were stored and examined later by a score (10). tion of NVC abnormalities and telangi- second investigator (MI) under blinded The study was approved by local ethics ectasia score (TS) with DU history and conditions. Number of capillaries (CN), committee and informed consent was severity of peripheral vascular involve- irregularly enlarged capillaries (IEC), obtained from all participants. ment (PVI) in SSc. giant capillaries (GC), microhaemor- For statistical analysis of the data ‘SPSS rhages (H), capillary ramifications version 16 for Windows’ programme Patients and methods (CR), capillary array disorganisation was used. Groups were compared with A total of 59 SSc patients with non-RP (CAD) and maximum capillary diam- Pearson chi-square/Fisher’s tests for symptom of <8 years (preferably <2 eter were calculated per 1 millimeter categorical variables or student’s t- and years) and fulfilling Leroy & Medsger (mm) area of the images. Mann Witney U-tests for continuous classification criteria (21) were included Firstly, NVC assessed qualitatively. variables. Results were presented as into the study. Patients with age of <18, Sclerodermic NVC pattern (early or mean ± standard deviation (±SD). P- non-RP symptom of >8 years and not active or late pattern) was investigated values less than 0.05 were considered eligible for performing NVC were ex- and decided as follows: early NVC pat- significant. The distribution and median cluded. Demographics, onset of symp- tern, few giant capillaries (homogene- levels of NVC abnormalities and TS in toms, organ involvement, treatment ously enlarged loop with a diameter patient groups according to DU history details, laboratory and imaging findings >50 μm), few capillary haemorrhages and severity of vascular involvement including serology, echocardiography, (dark mass due to haemosiderin depos- are summarised in boxplot graphics. respiratory function tests, high-resolu- it), regular capillary distribution and no tion computed tomography (HRCT) of significant loss of capillaries (at least Results chest were noted into a pre-defined pro- 9–10 capillaries per linear mm counted Demographics and clinical characteris- tocol. TS (For each body area including at the distal row of the nailfold); active tics of SSc patients were summarised face, hands, arms, chest and abdomen, NVC pattern, frequent giant capillar- in Table I. back, legs and feet; telangiectases were ies, frequent capillary haemorrhages, The mean age of the patients was 45.6 scored as 0 if no telangiectasia was pre- moderate loss of capillaries, mild dis- and 91.5% were females. The mean sent, 1 if there were fewer than 10 tel- organisation of the capillary architec- duration of RP, non-RP symptoms angiectases, 2 if there were 10 or more ture (irregular capillary distribution (NRP) and skin involvement were telangiectases, the total possible telan- and orientation with respect to the 6.1±6.5, 3.1±2.0 and 3.0±2.0 years, giectasia score was 22 (20), modified nailfold along with shape heterogene- respectively. Of the patients 34% were Rodnan skin score (MRSS), Valentine ity of the loops), absent or mild rami- diffuse, 66% were limited cutaneous; activity score (VAS) (different scores fied capillaries (branching, bushy or 22% were anti-centromere(+) and 49% of 10 parameters, the total possible coiled capillary often originating from were anti-Scl70(+). Smoking habitus VAS score was 10) (22) and Medsger a single normal sized capillary); late (active or ex-smoker) was 46%. RP severity score (SS) [Score of 0 to 4 for NVC pattern, irregular enlargement of (97%), gastrointestinal involvement 9 involvements, the total possible SS capillaries, few or absent giant capillar- (72%), telangiectases (58%), arthritis score was 36. PVI was scored as 0 if ies and capillary haemorrhages, severe (49%) and DU (46%) were frequent no RP or not requiring vasodilators, 1 loss of capillaries with avascular areas, manifestations in SSc cohort. Calcium

S-93 NVC, telangiectases and digital lesions in SSc / Y. Yalcinkaya et al.

Table I. Demographics and characteristics of SSc patients. were frequent in all group but similar between DU+ and DU- patients. The SSc Patients (n=59) mean scores of CN, IEC and MES were Age (mean ± SD) 45.6 ± 11.5 significantly higher in DU+ patients Female sex (n) (%) 54 (91%) (2.0±0.5 vs. 1.4±0.7, p<0.001, 1.8±0.6 Duration of RP (year) (mean ± SD) 6.1 ± 6.5 vs. 1.4±0.7, p=0.029 and 2.5±1.5 vs. Duration of non-RP symptom (year) (mean ± SD) 3.1 ± 2.0 Duration of skin involvement (year) (mean ± SD) 3.0 ± 2.0 1.8±1.0, p=0.041, respectively). Smoking history(n) (%) 27 (46%) Patients were grouped according to Skin Diffuse cutaneous SSc (dcSSc) (n) (%) 20 (34%) their current PVI status as severe Involvement Limited cutaneous SSc (lcSSc) (n) (%) 39 (66%) (SS;2-4) (n=16) or non-severe (SS;0- Serology (n) (%) ANA (+) 54 (92%) Anti-centromere (+) 13 (22%) 1) (n=43). Demographics, duration of Anti-Scl70 (+) 29 (49%) symptoms, serology, organ involve- Organ(n) (%) Raynaud’s 57 (97%) ments and treatments were similar be- Involvement Digital ulcer history 27 (46%) Telangiectases 34 (58%) tween groups. Severe PVI was found Arthritis 29 (49%) to be more frequent in males (4/5, 80% Renal crisis 4 (7%) vs. 12/54, 22%, p=0.017). When we Flexion contracture-t.friction rubs 9-5 (15-9%) compared severe and non-severe PVI Dysphagia-reflux-diarrhoea 43 (72%) FVC<%80-DLCO<%80 14-23 (24-39%) groups; early sclerodermic pattern was Pulmonary fibrosis 10 (17%) significantly less frequent in patients Pulmonary hypertension 4 (7%) with severe PVI (0 vs. 9, p=0.049) (Ta- Treatment(n) (%) No treatment (recent diagnosis) 11 (19%) ble II). Active and late patterns were Calcium channel blockers-PPI 46-46 (78-78%) Acetylsalicylic acid 37 (63%) frequent but similar between groups. Ilioprost 3 (5%) The mean scores of CN and MES Immunosuppressives 36 (61%) were significantly higher in patients Steroids 29 (49%) with severe-PVI ( 2.1±0.4 vs. 1.5±0.7, p<0.001 and 2.8±1.6 vs. 1.8±1.1, Table II. NVC Patterns between different groups of peripheral vascular involvement. p=0.038) (Table III). Distribution and Peripheral vascular involvement median levels of scores significantly different between groups according to NVC patterns DU history+ DU history- PVI-severe PVI-non-severe DU history and severity of RP were (n=27) (n=32) (n=16) (n=43) summarised in Figure 1. Early 1* 9 0* 10 The mean values of TS was not statis- Active 14 16 9 21 tically different between groups (Table Late 12 7 7 12 Total 27 32 16 43 IV; Fig. 2). TS was correlated only with duration of non-RP sypmtom (r=0.407, *p<0.05, when DU+ and DU- groups were compared, when severe and non-severe PVI groups were p=0.021) in DU- group. compared with Chi-square tests. The mean values of MRSS, VAS and SS were not statistically different be- Table III. Semi-quantitative scoring of NVC parameters between different groups of periph- tween groups (Table IV; Fig. 2). In DU+ eral vascular involvement. group; scores of ramification and disor- Peripheral vascular involvement (PVI) ganisation were correlated with activity and severity (r=0.396/p=0.041 and Scores of NVC parameters DU history+ DU history- PVI-severe PVI-non-severe (n=27) (n=32) (n=16) (n=43) r=0.702/p<0.001, r=0.451 /p=0.018 and r=0.712/p<0.001, respectively), capil- Capillary reduction 2.0 ± 0.5** 1.4 ± 0.7 2.1 ± 0.4** 1.5 ± 0.7 lary loss and microangiopathy were cor- Irregularly enlarged capillaries 1.8 ± 0.6* 1.4 ± 0.7 1.6 ± 0.6 1.6 ± 0.7 related with severity (r=0.530/p=0.004 Microangiopathy evolution score 2.5 ± 1.5* 1.8 ± 1.0 2.8 ± 1.6* 1.8 ± 1.1 and r=0.611/p=0.001). In DU- group; *p<0.05, **p<0.001 when DU+ and DU- groups were compared, when severe and non-severe PVI scores of capillary loss, ramifications, groups were compared with Mann Whitney U and student’s t-tests disorganisation and microangiopathy were correlated with activity (r=0.427/ channel blockers (78%), acetylsalicylic and treatments were similar between p=0.015, r=0.750/p<0.001, r=0.370/ acid (63%) and ilioprost (5%) were the groups. NVC patterns and semi-quanti- p=0.037 and r=0.673/p<0.001) and treatments for PVI (Table I). tative scoring of NVC parameters were scores of capillary loss, giants, ramifi- When we compared the patients accord- summarised in Table II and Table III. cation and microangiopathy were cor- ing to presence of DU history (DU+ or Early NVC pattern was significantly related with severity (r=0.400/p=0.023, DU-); demographics, duration of symp- less frequent in DU+ patients (1 vs. r=0.357/p=0.045, r=0.779/p<0.001 and toms, serology, organ involvements 9, p=0.016). Active and late patterns r=0.696/p<0.001).

S-94 NVC, telangiectases and digital lesions in SSc / Y. Yalcinkaya et al.

Fig. 1. The distribution and median scores of NVC in groups according to presence of DU and severity of PVI.

Fig. 2. The distribution and median levels of telangiectasia score in groups according to presence of DU and severity of PVI.

Discussion in SSc have to be the goal to prevent cluded RP, DU, PAH, interstitial lung In this prospective study, we evalu- organ damage and progression in such disease, telangiectasia, puffy fingers/ ated the relationship between differ- a chronic disease with high morbidity sclerodactily, specific antibodies and ent microvascular manifestations of and mortality. The period of the pro- sclerodermic capillaroscopic patterns SSc; telangiectases, ischaemic digital gression of RP to the skin and organ to capture early SSc and SSc without lesions and capillaroscopic changes involvements was found to be varying significant skin involvement to allow reflecting microangiopathy to identify 1.9 to 4.8 years in diffuse and limited earlier approach (25). the tools and/or markers for earlier ap- cutaneous subgroups of SSc (24). The DUs (46%) and telangiectases (58%) proach. Early diagnosis and treatment new classification criteria for SSc- in were frequent manifestations in our patient group as in reported large co- Table IV. The mean values of Telangiectasia, Modified Rodnan Skin score (MRSS), Valen- horts (26). Increasing number of DUs tini Activity and Medsger Severity scores between different groups of peripheral vascular (especially ≥3) were found to have involvement. higher proportions of diffuse cutaneous involvement and lung fibrosis, more Peripheral vascular involvement impairment in work and functional ac- SSc DU+ DU- PVI-severe PVI tivities in SSc (27). SSc patients were (n=59) (n=27) (n=32) (n=16) non-severe (n=43) grouped according to DU history and severity of current PVI, NVC was as- Telangiectasia score 2.3±3.5 2.7±4.7 1.9±2.1 p=0.78 3.0±5.5 2.0±2.4 p=0.85 sessed qualitatively and semi-quanti- MRSS 8.8±7.9 9.2±7.3 8.4±8.5 p=0.11 9.9±9.1 8.3±7.5 p=0.69 tatively. The frequency of severe PVI Valentini Activity score 1.5±2.1 1.9±2.1 1.2±2.0 p=0.11 2.1±2.4 1.3±1.9 p=0.19 was higher in male patients (80% vs. Medsger Severity score 4.6±2.7 4.7±2.2 4.5±3.0 p=0.09 5.1±2.4 4.4±2.7 p=0.07 22%, p=0.017), who were represented

S-95 NVC, telangiectases and digital lesions in SSc / Y. Yalcinkaya et al. with active DUs at time of diagnosis. parameters showed that the mean Shah, et al., was found to be associated Increased severity of peripheral vas- scores of CN and MES were higher with the presence of pulmonary vascu- culopathy in males has been reported in patients with DU+ (p<0.001 and lar disease (20). In this study, TS was previously in SSc (28). p=0.041) and severe-PVI (p<0.001 and correlated with only duration of non- When NVC patterns were evaluated p=0.038). Score of capillary loss (CN) Raynaud sypmtom in patients without in the subgroups; early sclerodermic and microangiopathy were related with DU history. We did not find any rela- pattern was significantly less frequent presence of DU and severity of PVI in tion between TS with positivity of DU in patients with DU+ and severe PVI. our cohort. Different scoring methods or severity of PVI. ‘Early’ pattern was the less related pat- of capillaroscopic parameters identify- Scores of certain capillaroscopic ab- tern to the presence of DU and severe ing scleroderma patterns were used to normalities including ramification and PVI, which was similar to reported evaluate vascular disease in SSc. In the disorganisation significantly correlated qualitative NVC assessments before. study evaluating the prediction of the with disease activity and severity. Cap- Late sclerodermic pattern was found onset of new DU by NVC, capillaro- illary loss and microangiopathy corre- to have significant relationship with scopic skin ulcer risk index (CSURI) lated with only disease severity in the digital ulcers and other organ involve- calculated with capillary numbers, gi- DU+ group. Scores of capillary loss, ments. Active/late patterns were more ant capillary numbers and capillary ramification and microangiopathy sig- frequent in the clusters with organ in- diameter parameters, was found to nificantly correlated with disease activ- volvement and moderate/severe MRSS have high specificity and sensitivity to ity and severity even in patients without in a large SSc group of EUSTAR study predict the risk of DU (12). In another a history of DU confirmed that these (29). In the study by Smith et al; sclero- study by Riccieri et al., it was reported parameters should be carefully evaluat- dermic patterns were related with future that patients with PAH had frequently ed in follow-up period to recognise the severe PVI (p=0.003) and pulmonary active or late patterns, NVC score of >1 patients with higher risk of developing (p=0.001) disease development at 18– and avascular area staging >1 (p=0.03, severe SSc. 24.months. Worsening capillaroscopic p=0.03 and p=0.003, respectively) than There are limitations of our study that patterns; early, active and late, were those without PAH. In addition, these should be mentioned. This is a cross- found to predict future PVI and pulmo- scores were found to be correlated with sectional study in a relatively small nary involvement when compared with mean PAP levels (31). PAH was ob- group of SSc patients and a longitudinal normal pattern (odds ratios were 2.5, 6.4 served in only 4 patients of our study follow-up to evaluate the progress of and 16.1 for PVI and 2.3, 5.4 and 12.7 which was insufficient to analyse. Kay- peripheral vascular and other involve- for pulmonary involvement, respective- ser C et al. studied the NVC abnormali- ments by NVC is necessary. We did not ly) (13). Active/late patterns were asso- ties as a predictor of mortality in SSc perform any analysis according to num- ciated with DU and severe-PVI in our and showed the association between ber DUs. For this study, interobserver cohort but we can not conclude about increased risk of death with avascu- aggreement for the analysis of capil- the predictive value of NVC for future lar score higher than 1.5 (HR:2.265). laroscopic parameters have not been DU and severity of PVI due to cross- Survival rates at 1,5 and 10 years were evaluated by statistical analysis rather sectional design of our study. found to be 97%, 86% and 59% in pa- we have decided on the final scoring by Previously, NVC was used for the tients with avascular score higher than consensus. evaluation of the efficacy of treatment. 1.5 and 97%, 97% and 91% in patients In conclusion; in this study we evalu- Late sclerodermic patterns, including with avascular score ≤1.5, respectively. ated the relationship between telan- large avascular areas, severe capillary This study revealed the decreased sur- giectases, ischaemic digital lesions loss and capillary array disorganisation vival in patients with severe capillaro- and capillaroscopic changes reflecting changed to active pattern with disap- scopic changes analysed by semi-quan- microangiopathy in SSc patients with peared avascular areas and new mi- titative assesment of NVC (32). relatively early disease. Severe PVI in crohaemorrhages, in seven patients at Telangiectasia, a non-inflammatory males was more frequent than females. 12.month of bosentan treatment used microvascular change commonly seen ‘Early’ NVC pattern was very rare in for PAH (14). Another study with a in SSc, related to vascular injury and/ patients with DU history and was not small number of SSc patients reported or inadequate vascular repair. In SSc, found in patients with severe PVI. DU some capillaroscopic changes after ilo- to study the relationship between tel- history and severe PVI were associated prost used for PVI (30). These studies angiectases and the severity of PVI with capillary loss and microangiopa- revealed that transition of patterns can might be important because of possible thy. Correlations between severe cap- be observed by specific treatments. We common pathogenetic mechanisms. In illaroscopic changes such as capillary could not identify any capillaroscopic the study of Ennis et al., patients re- loss, ramifications and microangiopa- changes related with treatment due to porting telangiectases were shown to thy with disease activity and severity in small number of patients treated with be more likely to have severe digital all patients with or without peripheral specific therapies for PVI and cross- ischaemia than those not (33). A score vascular involvement show that NVC sectional study design. described for evaluating telangiectases, abnormalities may help clinicians to The semi-quantitative scoring of NVC telangiectasia score (TS) in the study of predict the patients with higher risk of

S-96 NVC, telangiectases and digital lesions in SSc / Y. Yalcinkaya et al. severe disease to allow early treatment. 10. SULLI A, SECCHI ME, PIZZORNI C, CUTOLO Rheumatol 2003; 21 (Suppl. 29): S39-41. Telangiectasia score was not found to M: Scoring the nailfold microvascular chang- 23. MEDSGER TA Jr, BOMBARDIERI S, CZIRJAK es during the capillaroscopic analysis in L, SCORZA R, Della ROSSA A, BENCIVELLI be related to PVI and possibly not help- systemic sclerosis patients. Ann Rheum Dis W: Assessment of disease severity and prog- ful in the prediction of PVI. NVC was 2008; 67: 885-87. nosis. Clin Exp Rheumatol 2003; 21 (Suppl. found to be a useful tool in the assess- 11. SMITH V, CARMEN P, FILIP De K et al.: 29): S42-6. ment of SSc patients for PVI prognosis Reliability of the qualitative and semiquan- 24. MATUCCI-CERINIC M, BELLANDO-RAN- titative nailfold videocapillaroscopy assess- DONE S, LEPRI G, BRUNI C, GUIDUCCI S: and an essential method for research of ment in a systemic sclerosis cohort: a two- Very early versus early disease: the evolving microvascular pathology, warranting centre study. Ann Rheum Dis 2010; 69: 1092- definition of the ‘many faces’ of systemic prospective studies in a large number of 6. sclerosis. Ann Rheum Dis 2013; 72: 319-21. 12. SEBASTIANI M, MANFREDI A, VUKATANA G 25. Van den HOOGEN F, KHANNA D, FRANSEN J patients. et al.: Predictive role of capillaroscopic skin et al.: 2013 classification criteria for system- ulcer risk index in systemic sclerosis: a multi- ic sclerosis: an American College of Rheu- References centre validation study. Ann Rheum Dis 2012; matology/European League against Rheuma- 1. VARGA J, ABRAHAM D: Systemic sclerosis: 71: 67-70. tism collaborative initiative. Arthritis Rheum a prototypic multisystem fibrotic disorder. J 13. SMITH V, DECUMAN S, SULLI A et al.: Do 2013; 65: 2737-47. Clin Invest 2007; 117: 557-67. worsening scleroderma capillaroscopic pat- 26. WALKER UA, TYNDAL A, CZIRJAK L et al.: 2. WIGLEY FM: Vascular disease in scleroderma. terns predict future severe organ involve- Clinical risk assessment of organ manifesta Clin Rev Allerg Immunol 2009; 36: 150-75. ment? a pilot study. Ann Rheum Dis 2012; 71: tions in systemic sclerosis: a report from the 3. DISTLER JH, GAY S, DISTLER O: Angiogene- 1636-9. EULAR Scleroderma Trials And Research sis and vasculogenesis in systemic sclerosis. 14. GUIDUCCI S, BELLANDO RANDONE S, group database. Ann Rheum Dis 2007; 66: Rheumatology (Oxford) 2006; 45: 26-7. BRUNI C et al.: Bosentan fosters microvascu- 754-63. 4. WALKER UA, TYNDALL A, CZIRJÁK L et al.: lar de-remodelling in systemic sclerosis. Clin 27. GUILLEVIN L, HUNSCHE E, DENTON CP Clinical risk assessment of organ manifesta- Rheumatol 2012; 31: 1723-25. et al.: Functional impairment of system- tions in systemic sclerosis: a report from the 15. ALIVERNINI S, De SANTIS M, TOLUSSO B et ic scleroderma patients with digital ulcera- EULAR Scleroderma Trials And Research al.: Skin ulcers in systemic sclerosis: deter- tions: results from the DUO Registry. Clin group database. Ann Rheum Dis 2007; 66: minants of presence and predictive factors of Exp Rheumatol 2013; 31 (Suppl. 76): 71-80. 754-63. healing. J Am Acad Dermatol 2009; 60: 426- 28. PANOPOULOS ST, BOURNIA VK, SFIKAKIS 5. CUTOLO M, SULLI A, SMITH V: Assess- 35. PP: Is vasculopathy associated with systemic ing microvascular changes in systemic scle- 16. AMANZI L, BRASCHI F, FIORI G et al.: sclerosis more severe in men? J Rheumatol rosis diagnosis and management. Nat Rev Digital ulcers in scleroderma: staging, char- 2013; 40: 46-51. Rheumatol 2010; 6: 578-87. acteristics and sub-setting through observa- 29. INGEGNOLI F, ARDOINO I, BORACCHI P; 6. INGEGNOLI F, BORACCHI P, GUALTIEROTTI tion of 1614 digital lesions. Rheumatology EUSTAR co-authors: Nailfold capillaroscopy R et al.: Prognostic model based on nailfold (Oxford) 2010; 49: 1374-82. in systemic sclerosis: Data from the EULAR capillaroscopy for identifying Raynaud’s 17. HACHULLA E, CLERSON P, LAUNAY D et al.: scleroderma trials and research (EUSTAR) phenomenon patients at high risk for the Natural history of ischemic digital ulcers in database. Microvasc Res 2013; 89: 122-8. development of a scleroderma spectrum dis- systemic sclerosis: single center retrospec- 30. SHAH P, MURRAY AK, MOORE TL, HERRICK order: PRINCE (prognostic index for nail- tive longitudinal study. J Rheumatol 2007; AL: Effects of iloprost on microvascular fold capillaroscopic examination). Arthritis 34: 2423-30. structure assessed by nailfold. J Rheumatol Rheum 2008; 58: 2174-82. 18. NIHTYANOVA SI, BROUGH GM, BLACK CM, 2011; 38; 2079-80. 7. PAVLOV-DOLIJANOVIC S, DAMJANOV NS, DENTON CP: Clinical burden of digital vas- 31. RICCIERI V, VASILE M, LANNACE N et al.: STOJANOVIC RM, VUJASINOVIC STUPAR culopathy in limited and diffuse cutaneous Systemic sclerosis patients with and without NZ, STANISAVLJEVIC DM: Scleroderma systemic sclerosis. Ann Rheum Dis 2008; 67: pulmonary arterial hypertension: a nailfold pattern of nailfold capillary changes as pre- 120-3. capillaroscopy study. Rheumatology (Ox- dictive value for the development of a con- 19. WALKER JG, STIRLING J, BEROUKAS D et ford) 2013; 52: 1525-28. nective tissue disease: a follow-up study of al.: Histopathological and ultrastructural fea- 32. KAYSER C, SEKIYAMA JY, PRÓSPERO 3,029 patients with primary Raynaud’s phe- tures of dermal telangiectasias in systemic LC, CAMARGO CZ, ANDRADE LE: Nailfold nomenon. Rheumatol Int 2012; 32: 3039-45. sclerosis. Pathology 2005; 37: 220-5. capillaroscopy abnormalities as predictors of 8. HERRICK AL, CUTOLO M: Clinical implica- 20. SHAH AA, WIGLEY FM, HUMMERS LK: mortality in patients with systemic sclerosis. tions from capillaroscopic analysis in pa- Telangiectases in scleroderma: A potential Clin Exp Rheumatol 2013; 31 (Suppl. 76): tients with Raynaud’s phenomenon and sys- clinical marker of pulmonary arterial hyper- 103-8. temic sclerosis. Arthritis Rheum 2010; 62: tension. J Rheumatol 2010; 37; 98-104. 33. ENNIS H, HERRICK AL, CASSIDY C, GRIF- 2595-604. 21. LeROY EC, MEDSGER TA Jr.: Criteria for FITHS CE, RICHARDS HL: A pilot study of 9. CUTOLO M, SULLI A, PIZZORNI C, ACCARDO the classification of early systemic sclerosis. body image dissatisfaction and the psycho- S: Nailfold videocapillaroscopy assessment J Rheumatol 2001; 28: 1573-6. logical impact of systemic sclerosis-related of microvascular damage in systemic sclero- 22. VALENTINI G, SILMAN AJ, VEALE D: telangiectases. Clin Exp Rheumatol 2013; 31 sis. J Rheumatol 2000; 27: 155-60. Assessment of disease activity. Clin Exp (Suppl. 76): 12-7.

S-97