Standardized nailfold capillaroscopy in children with rheumatic diseases: a worldwide study

Abstract

Objectives: To standardly assess and describe nailfold videocapillaroscopy (NVC) assessment in children and adolescents with juvenile rheumatic and musculoskeletal diseases (jRMD) vs healthy controls (HCs).

Material and methods: In consecutive jRMD children and matched HCs from 13 centres worldwide, 16 NVC images per patient were acquired locally and read centrally per international consensus standard evaluation of the EULAR Study Group on Microcirculation in Rheumatic Diseases. A total of 95 patients with JIA, 22 with JDM, 20 with childhood-onset SLE (cSLE), 13 with juvenile SSc (jSSc), 21 with localized scleroderma (lSc), 18 with MCTD and 20 with primary RP (PRP) were included. NVC differences between juvenile subgroups and HCs were calculated through multivariable regression analysis.

Results: A total of 6474 images were assessed from 413 subjects (mean age 12.1 years, 70.9% female). The quantitative NVC characteristics were significantly lower or higher in the following subgroups compared with HCs: for density: lower in jSSc, JDM, MCTD, cSLE and lSc; for dilations: higher in jSSc, MCTD and JDM; for abnormal shapes: higher in JDM and MCTD; for haemorrhages: higher in jSSc, MCTD, JDM and cSLE. The qualitative NVC assessment of JIA, lSc and PRP did not differ from HCs, whereas the cSLE and jSSc, MCTD, JDM and cSLE subgroups showed more non-specific and scleroderma patterns, respectively.

Conclusions: This analysis resulted from a pioneering registry of NVC in jRMD. The NVC assessment in jRMD differed significantly from HCs. Future prospective follow-up will further elucidate the role of NVC in jRMD.

Keywords: children; juvenile rheumatic and musculoskeletal diseases; microcirculation; nailfold capillaroscopy; scleroderma pattern.

Graphical Abstract

Figure 1.

Standardized NVC protocol. Images of the second, third, fourth and fifth digits of each hand are assessed according to the standard format to report on capillaroscopic characteristics, including the quantitative (per linear millimetre) and qualitative (the capillary pattern) assessment (image level). To report at the subject level, the capillaroscopic parameters are deduced from all the obtained NVC images from an individual subject by calculating the mean of the capillary density, the mean of the number of dilations and the mean of the number of abnormal shapes (the sum of each NVC parameter divided by the number of assessed images per subject); by describing the presence or absence of the parameters ‘giants’ and ‘microhaemorrhages’ in a dichotomous way and by deriving the overall qualitative assessment (details in section 2.2)

Figure 2.

Standardised assessment of NVC images according to the international consensus EULAR Study Group on Microcirculation in Rheumatic Diseases definitions. (A) An example of a stereotypical ‘normal’ pattern. Density: 8 capillaries/linear mm (↓). Dimension: no dilations, no giants. Morphology: no abnormal shapes. Microhaemorrhages: absent. Interpretation: normal pattern (non-scleroderma pattern). (B) An example of a ‘non-specific’ pattern. Density: 8 capillaries/linear mm. Dimension: presence of 3 dilations/linear mm, no giants. Morphology: presence of two abnormal shapes (§). Microhaemorrhages: present. Interpretation: non-specific abnormalities (non-scleroderma pattern). (C) An example of a ‘scleroderma’ pattern. Density: 5 capillaries/linear mm (↓). Dimension: presence of a giant (⇓). Morphology: no abnormal shapes. Microhaemorrhages: present. Interpretation: (active) scleroderma pattern. Adapted from Smith V et al. Standardization of nailfold capillaroscopy for the assessment of patients with Raynaud’s phenomenon and systemic sclerosis. Autoimmunity Reviews 2020;19:102458 [7]

Figure 3.

Quantitative NVC assessment per subgroup compared with the HCs. (A) Box plot of the mean capillary density per linear millimetre. (B) Box plot of the mean number of dilations per linear millimetre. (C) Box plot of the mean number of abnormal shapes per linear millimetre. Variations in the subgroups are evidenced by the wide blue boxes and by the presence of outliers (°1.5 × interquartile range) and extreme values (*3 × interquartile range). (D) Bar graph of the proportion of subjects with the presence of nailfold microhaemorrhages. The statistically significant differences (by linear regression analysis) between each subgroup and the overall HC subgroup are indicate with an asterisk (*)

Figure 4.

Exploratory analysis of the quantitative NVC assessment in relation to age in the healthy subset. Data are represented in scatter plots and fitted regression lines. (A) Capillary density in the total healthy subset (n = 204). (B) Capillary density in the healthy subset without the youngest group (n = 197). (C) Mean number of capillary dilations per linear millimetre (n = 204). (D) Mean number of abnormal shapes per linear millimetre (n = 204)