Vol. 23 No. 5 437–442, 2006
The Prognostic Value of Nailfold Capillary
Changes for the Development of Connective
Tissue Disease in Children and Adolescents with
Primary Raynaud Phenomenon: A Follow-up
Study of 250 Patients
Slavica Pavlov-Dolijanovic´, M.D., M.Sc.,* Nemanja Damjanov, M.D., Ph.D.,* Predrag Ostojic´,
M.D., M.Sc.,* Gordana Susˇic´, M.D., M.Sc.,* Roksanda Stojanovic´, M.D., Ph.D.,* Dragica
Gacic´, M.D.,* and Aleksandra Grdinic´, M.D., M.Sc.
- Institute of Rheumatology-Belgrade, KBC ‘‘Dragisˇa Misˇovic´-Dedinje’’, Belgrade, Serbia and Montenegro
Abstract: To assess the prognostic value of capillaroscopy findings for
the development of connective tissue disease in children and adolescents
with Raynaud phenomenon, we followed up a group of 250 (mean age
15 years) for 1 to 6 years after the first capillaroscopy was performed. Every
6 months they were screened for signs and symptoms of connective tissue
disease. Analysis was performed on capillary changes registered 6 months
before the development of connective tissue disease. Capillary changes
were classified into three types: normal, nonspecific, and sclerodermatous.
At the end of the follow-up period, 191 (76%) subjects had primary Raynaud
phenomenon, 27 (10.8%) were diagnosed as having undifferentiated con-
nective tissue disease, and 32 (12.8%) fulfilled the criteria for a diagnosis of a
specific connective tissue disease. Systemic lupus erythematosus was
found in nine (3.6%) patients, rheumatoid arthritis in 10 (4%) patients (six of
them with juvenile onset rheumatoid arthritis), and scleroderma spectrum
disorders in 13 (5.2%). The mean time for the evolution of Raynaud phe-
nomenon into undifferentiated connective tissue disease or a form of the
disease was 2 years. Most of the subjects with primary Raynaud phenom-
enon (173/191, 91%), undifferentiated connective tissue disease (22/27, 81%),
juvenile onset rheumatoid arthritis/rheumatoid arthritis (7/10, 70%), and
systemic lupus erythematosus (6/9, 67%) had normal capillary findings.
Nonspecific capillary changes occurred in 3 of 10 (30%) patients with rheu-
matoid arthritis, 2 of 9 (22%) with systemic lupus erythematosus, 4 of 27
(15%) with undifferentiated connective tissue disease, and 18 of 191 (9%)
with primary Raynaud phenomenon. Of all the subjects, only 10 (4%) showed
sclerodermatous disease type capillary changes 6 months before the
Address Correspondence to Slavica Pavlov-Dolijanovic´, M.D., M.Sc., Institute of Rheumatology-Belgrade, Institute of Rheuma- tology-Belgrade, Resavska 69, 11 000 Belgrade, Serbia and Montenegro, or e-mail: email@example.com or
expression of a particular disease: eight (62%) of these developed sclero-
derma spectrum disorders, one expressed systemic lupus erythematosus,
and one had undifferentiated connective tissue disease. We concluded that
there were no specific capillary changes predictive for future development of
systemic lupus erythematosus, juvenile onset rheumatoid arthritis/rheuma-
toid arthritis, and undifferentiated connective tissue disease in children and
adolescents with Raynaud phenomenon. Most of our study subjects with
Raynaud phenomenon who developed these diseases had normal capillary
findings or nonspecific changes. Children and adolescents who developed
scleroderma spectrum disorders showed a sclerodermatous type of capillary
changes 6 months before the expression of the disease, indicating that this
type of capillary changes in children and adolescents with Raynaud phe-
nomenon highly correlated with further development of scleroderma spec-
Raynaud phenomenon (RP) in adults, as well as in children, is a vasospastic disorder characterized by the episodic color changes of blanching, cyanosis, and hyperemia of the digits in response to cold (1). In adults RP can occur in response to emotional stress. However, no evidence of relationship between the occurrence of RP and emotional stress was found in children (2). It can be classiﬁed as a primary phenomenon of unknown cause, or secondary to a number of diﬀerent diseases, such as connective tissue diseases (CTD) (3). Most of the follow- up studies of children with RP noted the development of associated CTD in 25% to 30% of those aﬀected (4,5), but some investigators reported up to 52% transition to CTD and 15% transition to probable CTD (6). The mean time between the onset of RP and development of CTD in children was 2 to 6 years (4–6).
Early detection of CTD in subjects with RP is important, and sometimes crucial for better treatment results, and a better prognosis. Therefore, it is important to assess the prognostic signiﬁcance of various diagnostic methods for early detection of CTD in subjects with RP. One of these methods is nailfold capillaroscopy. This simple, noninvasive, easy-to-perform method is very important as a screening tool for detecting the subjects with RP who are at high risk for the development of CTD as adults (7). However, long-term protocol studies deal- ing with nailfold capillaroscopy changes in large homogeneous groups of children and adolescents with long-standing primary and secondary RP have not been conducted.
Signiﬁcant changes in capillary morphology are pre- sent in 80% of adult patients with scleroderma and related disorders (7,8). These are called scleroderma (SD) type capillary changes and include mainly a decrease in capillary density and widened, giant loops, often sur- rounded by avascular areas. The extent of microangi- opathy detected by nailfold capillaroscopy has been shown to correlate with disease severity and prognosis
(7,9). Furthermore, the presence of SD type capillary changes in adult patients with RP is thought to be indicative of the future development of CTD, even in the absence of other disease symptoms (10,11). Comparable quantitative capillaroscopic data for healthy children, as well as for pediatric rheumatology patients are limited (12–15). Nevertheless, nailfold capillaroscopy ﬁndings have been proposed as a potential marker for more persistent disease in children with juvenile dermatomy- ositis (12) or systemic sclerosis (16). Correlation between the degree of vaculopathy and the clinical course has also been documented (12,13,16).
The aim of this study was to assess the prognostic value of capillaroscopic ﬁndings for the development of CTD in children and adolescents with RP, to detect the spectrum of CTD associated with RP, and to evaluate diﬀerences in duration of RP before the development of a speciﬁc CTD in the same patient population.
PATIENTS AND METHODS Patients
Two hundred ﬁfty children and adolescents with RP (205 girls and 45 boys) aged 10 to 20 years (mean age 15 years) were prospectively followed up in a study approved by the Research Ethic Committee of the Institute of Rheumatology, Belgrade. The median duration of RP was 2 years, ranging from 1 month to 11 years.
Methods Children and adolescents with RP were followed up for 1 to 6 years after the ﬁrst capillaroscopy was done.
Capillaroscopy and a clinical search for criteria of CTD were repeated every 6 months.
Nailfold capillaroscopy was carried out using the semiquantitative methodology described by Damjanov
Vol. 23 No. 5 September ⁄ October 2006 et al (17), which is similar to the technique described by Maricq (18). Brieﬂy, an OPTON microscope with 16· and 100· magniﬁcations was used. Cold light was pro- vided by a Schott Mainz KL 150 ﬁber optic light source (50/60Hz, 2000W, lampe 15V/150W, Germany). Each subject was seated indoors for a minimum of 15 to 20 minutes before the examination, at a room tempera- ture of 20 to 22 C. The nailfolds of eight ﬁngers (the second, third, fourth, and ﬁfth on both hands) were examined in each patient, after a drop of immersion oil was placed on the nailfold bed to improve resolution. Fingers aﬀected by recent local trauma were not ana- lyzed. Capillary changes found 6 months before the cri- teria for a diagnosis of CTD were fulﬁlled were used for analysis. These were classiﬁed into three types (7,8): normal (typical hair pin structure), nonspeciﬁc capillary changes (meandering and crossed capillaries, focal dis- tribution of capillary hemorrhages, capillary thinning, and capillary spasm), and the sclerodermatous type mentioned above.
Clinical search comprised history taking, physical examination, laboratory testing [including serologic tests for antinuclear antibodies (ANA), rheumatoid factor, and antiphospholipid (APL) antibodies], and particular analyses (including radiographies and ultrasound). Ser- ology and speciﬁc analyses were ordered according to clinical impression. None of the patients used beta- blockers.
Primary RP 173 18 191 Secondary RP
9 2.1 (1 mos–11 yrs) JRA/RA 10 2.7 (1 mos–11 yrs) SSD 13 2.4 (2 mos–6 yrs) Total 59 2.4 (1 mos–11 yrs)
CTD (years), mean (range) UCTD 27 2.1 (8 mos–8 yrs) SLE
29 10 250 TABLE 3. Distribution of Patients who Developed Connective Tissue Diseases (CTD) during the 1- to 6-year Follow-up Period, and Duration of RP before Diagnosis of CTD Diagnosis Number of patients Duration of RP before diagnosis of
59 Total 211
TABLE 2. Distribution of the Types of Capillary Changes in Children and Adolescents with Primary and Secondary Raynaud Phenomenon (RP) Diagnosis Type of capillary changes (patient numbers) Total Normal Nonspeciﬁc SD type
Statistical Analysis Chi-squared test with Yate’s correction for continuity was used. P values were calculated by two-sided Fisher’s exact test.
67 Gender Boys
3 11–12 14 13–14 35 15–16 58 17–18 73 19–20
TABLE 1. Characteristics of the Study Population No. of subjects Age range (years)
Distribution of Subjects Who Developed CTD After 2 years of follow-up, 27 of 250 (10.8%) subjects had some signs and symptoms of CTD, but did not fulﬁll the criteria for CTD, and they were considered to have undiﬀerentiated CTD (Table 3). Thirty-two (12.8%) of
Distribution of the Types of Capillary Changes Two hundred eleven children and adolescents with RP (84.4%) were classiﬁed as having normal type, 29 (11.6%) patients were classiﬁed as having nonspeciﬁc capillary changes, and 10 (4%) patients had the SD type (Table 2). The frequency of normal capillary ﬁndings in the subjects with primary RP was signiﬁcantly higher than that in the patients with secondary RP (p < 0.01). On the other hand, the SD type was noted only among the patients who had secondary RP. Nonspeciﬁc capil- lary changes had a similar distribution in both primary and secondary RP subgroups.
Follow-up Data At the end of the follow-up period, 191 (76%) patients with RP had no signs and symptoms of CTD, and they were considered to have primary RP. Fifty-nine patients (23.6%) developed CTD, and were considered to have secondary RP (Table 2).
RESULTS Among 250 subjects included in the study, there were 110 (44%) children between 10 and 16 years of age, and 140 (56%) adolescents aged 17 to 20 years. Characteristics of the study population are shown in Table 1. Eighty-two percent of the subjects were female children, the fema- le:male ratio was 5:1.
Pavlov-Dolijanovic´ et al: Nailfold Changes in Primary Raynaud Phenomenon the 250 subjects fulﬁlled the criteria for the diagnosis of a particular CTD. The mean time between the beginning of RP until diagnosis of a particular CTD was 2.4 years. Among 32 patients with criteria suﬃcient for diagnosis of a particular CTD, nine (3.6%) patients had systemic lupus erythematosus (SLE), 10 (4%) patients had rheu- matoid arthritis (RA) [six of these were with juvenile onset RA (JRA)], and 13 (5.2%) subjects had sclero- derma spectrum disorders (ﬁve with dermatomyositis/ polymyositis, ﬁve with systemic sclerosis, and three with sclerodermatomyositis). The average duration of RP until diagnosis of CTD in patients who fulﬁlled the cri- teria for JRA/RA was not signiﬁcantly longer than in those with other CTD (p ¼ 0.8).
The distribution of diﬀerent types of capillary changes in patients who developed CTD is shown in Table 4. Normal capillary ﬁndings were not signiﬁcantly diﬀer- ently distributed among the patients with undiﬀerenti- ated connective tissue disease (UCTD), JRA/RA, and SLE. The frequency of normal capillary ﬁndings in the patients with UCTD was signiﬁcantly higher than in those in the scleroderma spectrum disorders (SSD) sub- group (p ¼ 0.0013). Nonspeciﬁc capillary changes were distributed equally in the diﬀerent subgroups of RP. SD- type capillary changes were present in 8 of 13 (61.5%) patients with SSD, in 1 of 9 (11.1%) patients with SLE, and 1 of 27 (3.7%) patients with UCTD. No JRA/RA patient showed any signiﬁcant abnormality of capillary morphology. Sclerodermatous type capillary changes were signiﬁcantly more frequent among the patients with SSD than among those with SLE and UCTD (p < 0.01).
59 Vol. 23 No. 5 September ⁄ October 2006
Antiphospholipid antibodies were found in ﬁve chil- dren who had digital pits. Two of them were APL anti- body-positive, while others were negative. An association of APL antibodies and RP is reported in adults (27) and children (5). The potential pathogenic mechanism for APL antibodies in RP is binding and activating endothelium (28). Endothelial damage and/or thrombosis in the digits might result. We investigated for APL antibodies in ﬁve children who had digital pits. The test was positive in two of them. All ﬁve patients ﬁnally developed SLE. In the study by Nigrovic et al (5), of children with secondary RP who were found to be APL antibody-positive (7/23; 30%), the majority (6/7) had SLE. They also reported a high prevalence of APL antibodies in children with primary RP (18/50; 36%).
Normal Nonspeciﬁc SD type Total UCTD
TABLE 4. Distribution of Diﬀerent Types of Capillary Changes in Patients with UCTD and a Articular CTD Diagnosis Type of capillary changes (no. of patients)
In our study, the greatest number of children and adolescents during the follow-up period developed UCTD (approximately 11%) or scleroderma spectrum disease (2% of patients had systemic sclerosis, 2% of patients had dermatomyositis/polymyositis, and 1.2% had sclerodermatomyositis), and a lesser number of patients developed JRA/RA (4%) or SLE (3.6%). Sim- ilar to our results, Nigrovic et al noted mixed CTD in 1.9% patients, scleroderma in 1.9% patients, and myo- sitis in 1.9% patients. These authors found arthritis in a large number of patients (28.8%). Rheumatoid arthritis is usually not reported as CTD underlying RP, i.e., sec- ondary RP. The data indicate that RP is not rare in adult patients with RA (20,21). Saraux et al (21) registered RP in 17.2% (54/332) RA adult patients. This group had a slightly higher prevalence of vasculitis than RA patients without RP. Among adults with RP, Grassi et al (22) found that 5% (38/781) of patients had RA. Cassidy reported 7% of RP adult patients had RA (23). However, JRA is thought to represent only 1% (23) or less than 1% (24) of the causes of RP. In our study, 6 of 250 (2.5%) children with RP developed JRA during 1 to 6 years of follow-up. Two of them were ANA positive (systemic onset JRA), and others were ANA negative (oligo- and polyarticular JRA). The later group is still being followed
niﬁcant diﬀerence in RP duration before development of any CTD.
We found that the average RP duration before the development of CTD in children and adolescents was 2 years. The mean duration of RP till development of
Our study conﬁrms that, in most patients, RP is not associated with the future development of CTD, but in one of four children or adolescents it could be the ﬁrst symptom of CTD, such as UCTD, SSD, JRA/RA, and SLE.
JRA/RA showed tendency to be longer than that for other CTD (p ¼ 0.88). We were not able to see any sig-
DISCUSSION The results of our study are comparable with those of other authors, indicating that RP in children and ado- and in most cases, is not associated with the future development of CTD (5). In our study, for both primary and secondary RP, 82% of patients were female children. Approximately 76% of children and adolescents did not have any recognized underlying CTD (primary RP) and 24% of children and adolescents developed CTD (sec- ondary RP). The proportion of secondary RP in our study population was lower than that in other studies. Bethhencourt et al (19) found secondary RP in 10 of 32 (31.2%) children, Nigrovic et al found it in 19 of 52 (36.5%), and Duﬀy et al reported it in 14 of 27 (52%) children. One reason for these diﬀerences was probably the fact that we investigated children and adolescents together (up to 20 years of age), whereas the other authors included subjects only up to 18 years of age.
In our study, subjects with primary RP and most of children and adolescents with RP who developed SLE, JRA/RA, and UCTD had normal capillary ﬁnding or nonspeciﬁc capillary changes. Only one subject who developed SLE (1/10, i.e., 11%) and one subject who developed UCTD (1/27, i.e., 3.7%) during the follow- up period had SD type capillary changes 6 months before the full expression of the disease. On the other hand, 8 of 13 (61%) children and adolescents who developed SSD (systemic sclerosis, dermatomyositis and sclerodermatomyositis) exhibited SD type capillary changes 6 months before the full expression of the disease.
8. Maricq HR, LeRoy EC, D Angelo WA, et al. Diagnostic potential of in vivo capillary microscopy in scleroderma and related disorders. Arthritis Rheum 1980;23:183–189.
15. Herrick AL, Moore T, Hollis S et al. The inﬂuence of age on nailfold capillary dimensions in childhood. J Rheu- matol 2000;27:797–800.
14. Terreri MTRA, Andrade LEC, Pucinelli ML, et al. Nail fold capillaroscopy: normal ﬁndings in children and adolescents. Semin Arthritis Rheum 1999;29:36–42.
13. Nussbaum AL, Silver RM, Maricq HR. Serial changes in nailfold capillary morphology in childhood dermatomyo- sitis. Arthritis Rheum 1983;26:1169–1172.
12. Spencer-Green G, Crowe WE, Levinson JE. Nailfold capillary abnormalities and clinical outcome in childhood dermatomyositis. Arthritis Rheum 1982;25:954–958.
11. Harper FE, Maricq HR, Turner RE, et al. A prospect- ive study of Raynaud phenomenon and early connective tissue disease. A ﬁve year report. Am J Med 1982;72:883– 888.
10. Maricq HR, Weiberger AB, LeRoy EC. Early detection of scleroderma spectrum disorders by in vivo capillary microscopy: a prospective study of patients with Raynaud phenomenon. J Rheumatol 1982;9:289–291.
Decreased nailfold capillary density in Raynaud phenom- enon: a reﬂection of immunologically mediated local and systemic vascular disease. Ann Rheum Dis 1985;44:603– 609.
9. Houtman PM, Kallenberg CGM, Wouda AA et al.
7. Maricq HR, Spencer-Green G, LeRoy EC. Skin capillary abnormalities as indicators of organ involvement in scleroderma (systemic sclerosis), Raynaud syndrome and dermatomyositis. Am J Med 1976;61:862–870.
In our 250 children and adolescents, we found that SD-type capillary changes signiﬁcantly correlated with the future development of scleroderma spectrum dis- orders. This ﬁnding was similar to that reported by Duﬀy et al. Also, Dolezalova et al found a strong correlation between lower capillary density, enlarged dilated capillary loops, abnormal capillaries, avascular areas, and nailfold vascular disarrangement (SD-type changes), and CTD (juvenile dermatomyositis, systemic sclerosis, and Mixed CTD) in children. Spencer-Green et al found SD-type capillary changes in 11 of 19 (58%) children with dermatomyositis. In contrast, children and adolescents with primary RP and most of the subjects who developed SLE, JRA/RA, and UCTD had normal capillary ﬁndings or nonspeciﬁc capillary changes. We did not ﬁnd any speciﬁc capillary changes predictive for future development of SLE, JRA/RA, and UCTD in children and adolescents with RP. These ﬁndings were similar to those of other studies in chil- dren, as well as in adults (16,29,30). Scleroderma type capillary changes in children and adolescents with RP strongly suggest the development of scleroderma spec- trum disorders in the near future. This ﬁnding gives us a chance to deﬁne, among a large number of children and adolescents with RP, a small subgroup with a high risk of development of SSD. Early detection of patients with SSD allows for an early start of proper treatment,
6. Duﬀy CM, Laxer RM, Lee P, et al. Raynaud syndrome in childhood. J Pediatr 1989;114:73–78.
5. Nigrovic PA, Fuhlbrigge RC, Sundel RP. Raynaud phenomenon in children: a retrospective review of 123 patients. Pediatrics 2003;111:715–721.
4. Gonzalez Pascual E, Esquinas Rychen G, et al. Raynaud phenomenon in childhood: review and progress control in eight cases. An Esp Pediatr 1998;48:603–607.
3. Jayanetti S, Smith CP, Moore T, et al. Thermography and nailfold capillaroscopy as noninvasive measures of circu- lation in children with Raynaud phenomenon. J Rheuma- tol 1998; 25:997–999.
Prevalence and association with other common symptoms. Arthritis Rheum 2003;48:3518–3521.
2. Gareth TJ, Ariane LH, Sarah EW et al. Occurrence of Raynaud phenomenon in children ages 12–15 years.
1. Bolster MB, Maricq HR, Leﬀ RL. Oﬃce evaluation and treatment of Raynaud phenomenon. Cleve Clin J Med 1995;62:51–61.
ACKNOWLEDGMENTS We thank Dr Zorica Sˇporcˇic´ for helpful comments on the manuscript.
16. Dolezalova P, Young SP, Bacon PA et al. Nailfold capillary microscopy in healthy children and in childhood rheumatic diseases: a prospective single blind observational study. Ann Rheum Dis 2003;62:444–449. Pavlov-Dolijanovic´ et al: Nailfold Changes in Primary Raynaud Phenomenon
17. Damjanov N, Pavlov S, Radosavljevic V, et al. Prognostic signiﬁcance of capillaroscopic ﬁndings in subjects with Raynaud phenomenon. Acta Rheumatol Belgradensia 1998;28:12–17.
18. Maricq HR. Wideﬁeld capillary microscopy: technique and rating scale for abnormalities seen in scleroderma and related disorders. Arthritis Rheum 1981;24:1159–1164.
19. Bethhencourt J, Bustabad S, Alvarez A, et al. Immunolo- gical features and nail-fold capillary patterns in 32 cases of Raynaud phenomenon in childhood. Ann Rheum Dis 2003;62:ABO 472.
20. Greager MA, Dzau VJ. Vascular diseases of the extremities
- – Raynaud phenomenon. In: Fauci AS, Braunwald E, Isselbacher KJ, et al. eds. Harrisons principles of internal medicine, 14th edn. New York: McGraw-Hill, 1998:1401– 1402.
21. Saraux A, Allain J, Guedes C, et al. Raynaud phenomenon in rheumatoid arthritis. Br J Rheumatol 1996;35:752–754.
22. Grassi W, De Angelis R, Lapadula G, et al. Clinical diag- nosis found in patients with Raynaud phenomenon: a multicentre study. Rheumatol Int 1998;18:17–20.
23. Cassidy JT. The systemic sclerodermas and related disor- ders. In: Cassidy JT, Petty RE eds. Textbook of pediatric rheumatology, 4th edn. New York: Churchill Livingstone 2001:505–533.
24. Hossny E, Hady HA, Mabrouk R. Anti-centromere antibodies as a marker of Raynaud phenomenon in pediatric rheumatologic diseases. Pediatr Allergy Immunol 2000;11:250–255.
25. Takei S, Maeno N, Shigemori M, et al. Two cases with SLE and MCTD developed after a long period of chronic arthritis that was initially diagnosed as JRA. Ryumachi 1997;37:702–708.
26. Vuilleumier C, Sauvain MJ, Aebi C, et al. Systemic lupus erythematosus initially presenting as idiopathic juvenile arthritis with positive antinuclear antibodies. Acta Paediatr 2003;92:512–513.
27. Vayssairat M, Abuaf N, Baudot N, et al. Abnormal IgG cardiolipin antibody titers in patients with Raynaud phenomenon and/or related disorders: prevalence and clinical signiﬁcance. J Am Acad Dermatol 1998; 38:555–558.
28. Meroni PL, Raschi E, Testoni C, et al. Endothelial activation by APL: a potential pathogenic mechanism for the clinical manifestations of the syndrome. J Autoimmun 2000;15:237–240.
29. Leﬀord F, Edwards JCW. Nailfold capillary microscopy in connective tissue: a quantitative morphological analysis.
Ann Rheum Dis 1986;45:741–749.
30. Kabasacal Y, Elvins DM, Ring EFJ et al. Quantitative nailfold capillaroscopy ﬁndings in a population with connective tissue disease and in normal healthy controls. Ann Rheum Dis 1996;55:507–512. Vol. 23 No. 5 September ⁄ October 2006