The impact of skin calcinosis on digital ulcers in patients with SSc: clinical and prognostic stratification using the “wound bed score”

Abstract

Digital ulcers (DUs) represent one of the major burdens for patients with systemic sclerosis (SSc), especially when associated with skin calcinosis (SC). The aim of this work is to evaluate the impact of SC in DUs of patients with SSc for clinical characteristics and prognosis assessed by the wound bed score (WBS). We prospectively enrolled 55 patients with DUs and SSc followed in our dedicated wound care clinic. For all the patients we collected clinical and anthropometric data and characteristics of the DU, and we calculated the WBS for each DU. Ninety‐nine DUs were evaluated (24 with SC). SC was prevalent in limited cutaneous SSc (75%) and in patients with longer disease duration (P = 0.02). SC‐DUs were prevalent at the fingertip (P = 0.04). The healing time was significantly higher in patients with SC (10.4 ± 7.9 weeks) compared with non‐SC (7.0 ± 5.7 weeks) P = 0.03. The WBS negatively correlated with the time to achieve complete healing (r = −0.237 P = 0.023) and the correlation was maintained in the non‐SC (r = −0.46, P = 0.033). DUs in SSc patients with SC are common and difficult to heal. When DUs are treated in dedicated centres, the prognosis is good. The WBS is fast and easy and maybe commonly applied in clinical practice.

1. BACKGROUND

Systemic sclerosis (SSc) is a chronic autoimmune disease characterised by three cardinal pathogenetic features: microvascular involvement, activation of the immune system, and increase of extracellular matrix deposition in the skin and internal organs. ¹ The prevalence of skin ulcers represents one of the major burdens for patients with SSc as they are painful, difficult to heal, and lead to substantial functional disability. About 30% of SSc patients develop digital ulcers (DUs) yearly, and up to 50% of the subjects develop this complication during the natural course of the disease. ² , ³

The pathogenesis of DUs in SSc is complex and involves vascular injury with Raynaud's phenomenon (RP)4 with vascular ischaemic damage, ⁵ and skin thickening of the hands that may increase the risk of micro‐traumatisms and local injuries. In addition, patients with SSc may develop skin calcinosis (SC) with a prevalence ranging from 18% to 49%. ⁶ Although the pathogenesis of SC is not completely understood, several mechanisms, including skin inflammation, hypoxia, recurrent trauma, and elevated levels of different cytokines (interleukins 1β and 6, tumour necrosis factor α) may be involved.

The prognostic evaluation of DUs in SSc patients is not well established. An interesting score, named “wound bed score” (WBS) proposed by Falanga and colleagues, ⁷ maybe a feasible and simple method for the prognostic stratification of DU seven although it has never been applied to SSc. Although the presence of SC may have a negative impact on the prevalence of DUs, ⁸ the impact of this typical aspect on clinical features, prognostic stratification, and outcome of DUs has not yet been extensively investigated.

The aim of this work is to evaluate the impact of the presence of subcutaneous SC in the DUs of patients with SSc in terms of DU characteristics, therapeutic response, and prognosis assessed by the WBS.

2. PATIENTS AND METHODS

This is an observational study in which we prospectively enrolled consecutive patients with SSc according to the European League Against Rheumatisms/American College of Rheumatology (EULAR/ACR) criteria ⁹ presenting in our dedicated wound care rheumatologic centre with DUs from October 2018 to August 2019. Systemic treatment was maintained stable during the course of the study. This study was approved by the Local Ethics Committee and each patient signed a written informed consent before participating in this study.

2.1. Clinical variables

For all the patients we collected:

• anthropometric data: age, gender, ethnicity, body weight, and height to compute the body mass index (BMI).

• cardiovascular risk factors, including smoking habits, pack/years, presence of diabetes mellitus, hypercholesterolemia, arterial systemic hypertension

• disease data: disease duration, an early disease defined as disease duration <5 years, and age at DU onset, disease subsets according to (REF) for diffuse cutaneous (dc‐), limited cutaneous (lc‐) and sine scleroderma (ss‐) SSc, nailfold videocapillaroscopy pattern for early, active or late disease (REF), positivity for antitopohysomerase I (Scl‐70), and anticentromere (ACA)

• internal organ involvement: interstitial lung disease and/or pulmonary arterial hypertension, heart involvement, scleroderma renal crisis, oesophagus and bowel involvement, myositis

2.2. Digital ulcer variables

The evaluation of DUs was performed at the first evaluation and then every 2 weeks until the DUs completely healed. The classification of the DU was made according to Amanzi et al ¹⁰ and in accordance with the principles of wound bed preparation:

• number of lesions

• localization: fingertips, nail area, extensor area of proximal/distal interphalangeal joint (PIP – DIP), metacarpophalangeal joint (MCP)

• spontaneous or post‐traumatic

• new‐onset lesion or recurrent DU

• dimensions (area in mm²)

• exudate (absent, low, high)

• borders of the lesion (regular or irregular)

• perilesional skin (normal – oedematous ‐ inflamed)

• local oedema

• depth of the lesion: superficial (confined only to epidermidis), intermediate (extended to the subcutaneous tissue and down to the muscular fascia) and deep (extended through the fascia with the explosion of the tendon, joint capsule, and/or bone)

In addition, we further grouped skin ulcers according to Giuggioli et al ¹¹ in Group 1 DUs of the hands (fingertip or close to the nails), Group 2 DUs of bony prominence (PIP, DIP, MCP), and Group 3 DUs in calcinosis.

We collected the time to DU complete healing, presence of signs of infection (defined as severe/sustained pain, redness or swelling, pus discharge, bad odour), collected DU‐related pain according to the patients' visual analogue scale (VAS) 0 to 10 cm at baseline.

Finally, in each DU the WBS was computed by VV, a trained nurse experienced in wound care. Briefly, the WBS is calculated assigning a score from 0 (worst) to 2 (best) to 8 parameters of the lesion and perilesional skin: the presence of black eschar, eczema/dermatitis, maximum depth of the lesion, presence of scarring, colour of the wound bed, oedema and/or swelling, resurfacing epithelium, and the amount of exudate. The total score ranges from 0 to a maximum of 16 which represents the best score and it has been associated with a better prognosis. ⁷

The presence of SC has been evaluated according to the presence/absence of calcium deposits, evaluation of the type of calcium deposits according to Bartoli et al for the presence of nets, plate, stone, and mousse. ¹²

In all the DUs a standardised local treatment has been used according to the principles of TIME as published in Barsotti et al. ¹³ Each ulcer was treated with wound cleaning with sterile saline solution, disinfection with sodium hypochlorite 0.5%, application of anaesthetic solutions (lidocaine), surgical debridement with a sterile scalpel, application of the appropriate moist wound dressing with antimicrobial activity (silver‐containing dressings or dialkylcarbomoyl chloride‐coated dressings).

2.3. Statistical analysis

As the values were normally distributed, the data were expressed as mean ± standard deviation (SD) and analysed with a parametric test for means (t student, t student for independent samples, one‐way ANOVA); Bonferroni posthoc test was used to test differences between the groups. Correlations between continuous variables were assessed by Pearson's correlation (r). Fisher's Exact test was used in contingency tables 2 × 2 for categoric variables. P values <0.005 were considered statistically significant. Statistical analyses were assessed using dedicated software (SPSS 21, IBM, US).

3. RESULTS

3.1. Patients characteristics

A total of 55 patients with SSc‐related DUs were enrolled, 37 (68.3%) with lcSS and 18 with dcSSc (32.7%). Of the cohort the male: female ratio was 5:50 with a mean age at an evaluation of 62.33 ± 17.2 years and a mean disease duration of 15.4 ± 9.7 years. Fifty‐three (96%) patients were Caucasian, while 2 (4%) were from Latin America. Eighteen patients had a DU with calcinosis (32.7%). The comparisons between patients with and without calcinosis are reported in Table 1. Although the mean disease duration was not significantly different between the two groups, only one patient with the early disease was identified.

TABLE 1.

Epidemiologic/anthropometric data and internal organ involvement in patients with and without calcinosis‐related DUs

Parameter	Calcinosis (18 Patients)	Non‐calcinosis (37 Patients)	P
Gender male	1 (5.5%)	17 (15.3%)	n.s.
Age	67.1 ± 16.9	59.3 ± 16.4	0.047
Disease duration	16.6 ± 8.4	13.4 ± 11.0	n.s.
Early disease (<5 years)	1 (5.5)	12 (32.5%)	0.025
ldSSc	14 (77.8%)	23 (62.2%)	n.s.
Anticentromere autoantibodies	13 (72.2%)	5 (27.8%)	0.05
Caucasian	18 (100%)	35 (94.6%)	n.s.
Organ involvement
Oesophagus	12 (66.6%)	19 (51.4%)	n.s.
Bowel	3 (16.6%)	4 (10.8%)	n.s.
Heart	1 (5.5%)	4 (10.8%)	n.s.
Ischaemic cardiopathy	0 (0%)	4 (10.8%)	n.s.
Interstitial lung disease	8 (44.4%)	20 (54.0%)	n.s.
Pulmonary arterial hypertension	6 (33.3%)	3 (8.1%)	0.035
Myositis	0 (0%)	2 (5.4%)	n.s.

Among the cardiovascular risk factors: 3 were smokers, 15 ex‐smokers with a mean of 12.0 ± 15.1 pack/years. The mean BMI was 23.6 ± 3.87 kg/m². Thirteen patients were affected by systemic arterial hypertension, one by diabetes mellitus, six by hypercholesterolemia. No differences in cardiovascular risk factors were identified between patients with and without calcinosis.

The NVC pattern identified an SSc pattern in 53 patients. Among them, one presented an early pattern (1.9%), while the active and late pattern was present in 21 (39.6%) and 25 (47.2%) of the patients, respectively. No significant difference was identified in the NVC pattern between the two groups. Internal organ involvement is reported in Table 1.

3.2. DUs characteristics

A total of 99 DUs were evaluated, among them 24 (24.2%) were associated with calcinosis, while 75 (76.8%) were not. DUs associated with calcinosis were more frequent in patients with lcSSc (18, 75%) compared with dcSSc (6, 25%) P = 0.023, while no significant differences were identified in the NVC pattern. Only one DU with calcinosis was identified during the early disease course (4.2%), while non‐calcinosis DUs were more common in early patients (21 28.0%) P = 0.02.

No significant difference was found with regard to the pathogenesis as two patients with calcinosis (8.3%) and four without calcinosis (5.3%) had a post‐traumatic onset. Calcinosis DUs were more frequently recurrent (6, 33.3%) compared to non‐calcinosis DUs (5.3%) P = 0.003.

The different localizations of DUs is reported in Figure 1. Fingertip localization is the only one that is prevalent in DU associated with calcinosis (16, 67%) compared to DUs not associated with SC (33, 44%), P = 0.04. We did not find any differences between right and left hands, or prevalence in one finger compared with the others.

FIGURE 1.

Localization of digital ulcers in patients with and without calcinosis. DIP, distal interphalangeal joint; MCP, metacarpophalangeal joint; PIP, proximal interphalangeal joint

Patients with calcinosis had a significantly higher number of lesions compared with those without calcinosis (1.48 ± 1.0 vs 1.1 ± 0.5 P = 0.05) but no significant differences were identified between calcinosis DUs and non‐calcinosis DUs with regard to dimensions (25.0 ± 32.1 mm² vs 30.2 ± 71.0 mm² P = n.s.). The exudate was more frequent in patients with calcinosis (5 26.3%) compared with non‐calcinosis DUs (2, 2.6%) P = 0.009, while no differences were identified for the appearance of the borders, perilesional skin, oedema and depth of the lesion. In our cohort, the prevalent aspect of the SC was stone in 19/24 DUs (79.2%), while the plate and the mousse aspects were identified in 3 (12.5%) and 2 (8.3%) DUs, respectively.

The group subdivision of DUs, according to Giuggioli et al, allowed us to identify 48 DUs (48.5%) in Group 1, 27 in Group 2 (27.3%), and 24 in Group 3 (24.2%). There were no significant differences between the three groups in terms of CV risk factors, a subset of the disease, autoantibodies positivity, NVC, DU characteristics, and VAS for pain.

3.3. Outcome measures and wound bed score

A similar rate of infections has been identified between calcinosis and non‐calcinosis (5, 20.8% vs 14 18.9% P = n.s.) and also the pain VAS was similar between the two groups at the first evaluation (6.0 ± 3.2 vs 5.3 ± 2.5 P = n.s., respectively) and after 2 weeks of local wound treatment (3.8 ± 2.8 vs 3.2 ± 2.7 P = n.s.). At the last evaluation, the time of complete DU healing, the pain VAS remained slightly elevated in patients with calcinosis compared to non‐calcinosis (0.9 ± 2.4 vs 0.2 ± 0.6 P = 0.03).

In our cohort of patients, all the DUs healed but patients with calcinosis DUs healed significantly more slowly (10.4 ± 7.9 weeks) compared with non‐calcinosis DUs (7.0 ± 5.7 weeks) (Figure 2A). Analysing the subdivision in groups (Figure 2B), the time to healing was higher in patients in Group 3 (10.4 ± 7.9 weeks) compared with Group 2 (5.8 ± 2.9 weeks) (P = 0.03), while no significant differences were found in Group 1 (7.8 ± 7.1 weeks).

FIGURE 2.

Different distribution of the healing time in patients with and/without calcinosis (A) and grouped according to Giuggioli et al Asterisks P < 0.05 (B)

In our cohort, the values of WBS at the first evaluation was similar between calcinosis DUs (8.96 ± 0.46) compared with non‐calcinosis (9.43 ± 0.33). A significant negative correlation was identified between the WBS and the weeks to complete healing (r = −0.237, P = 0.023). Analysing the differences between the two groups, the correlation was maintained only in the non‐calcinosis group (r = −0.46, P = 0.033), but not in the calcinosis DUs, although a trend may be identified (Figure 3).

FIGURE 3.

Correlations between wound bed score (WBS) and time to healing of the digital ulcer (expressed in weeks) in patients with calcinosis (A) and in patients without calcinosis (B)

Wound management was the same in both groups although in DUs with calcinosis, medication using silver‐containing dressing was preferred in 22 patients (91.7%) against 54 DUs without calcinosis (73%) P = 0.046.

4. DISCUSSION

In our work, we confirmed that DUs represent one of the most severe problems in patients with SSc since they require several hospital treatments, which are severely painful and hard to heal. The presence of SC is a common cause of DUs and is more frequent in patients with longer disease duration and in those with ACA positivity. The most common localization of calcinosis DUs was the fingertip and these ulcers had higher levels of exudate compared to non‐calcinosis. The most frequent aspect of SC in DUs is represented by “stone”. In our cohort, all the DUs evaluated healed, but the presence of SC represents a negative factor, as associated with significantly longer healing time. In addition, for the first time, we propose the WBS at baseline as a prognostic factor for wound healing in SSc, although the data are more consistent in patients without calcinosis, and it needs to be further validated in patients with SC.

In contrast with the literature, ¹⁴ , ¹⁵ , ¹⁶ patients with SC and DUs were comparable in terms of a subset of disease (lc‐SSc and dc‐SSc), although we identified a higher prevalence of ACA positivity in patients with SC. ¹⁶ , ¹⁷ , ¹⁸ Our data confirm that patients with SC were older than those without SC ¹⁴ and, although the mean disease duration was not significantly different between the two groups, the presence of calcinosis is extremely rare in patients with early disease.

Fingertip localization, consistent with the literature, ¹⁰ , ¹¹ may be explained as being the area most frequently exposed to trauma, ¹⁹ while we did not notice any different distribution in the two hands or a higher prevalence in one finger compared to the others. ²⁰ In contrast with the literature, ¹⁸ , ²¹ no difference was identified in the NVC pattern between patients with or without calcinosis, probably because in our cohort the prevalence of a “late” pattern was the most common in both groups, as it has been identified as being associated with DUs by itself. ²²

In partial contrast with the literature, ¹⁰ , ¹¹ we did not confirm a higher prevalence of infections in patients with SC. It is noteworthy to add that in patients with calcinosis, we preferred silver‐containing dressings that have a higher antimicrobial effect compared with other dressings, as these DUs are at a very high risk of infection.

Although different systemic or topical treatments have been proposed for the treatment of calcinosis, ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ in our experience the use of a standard protocol of mechanical debridement with a scalpel, aimed at removing the calcium deposits was able to treat all the DUs with SC, although we observed a longer healing time in this group of patients. ¹² This longer time may be linked to the difficult removal of the calcium deposit, which is painful for the patients, even if topical anaesthetics are applied. Our therapeutic approach may be not considered a surgical intervention, ²⁸ as we did not perform a radical calcification removal. However, we only remove the calcinosis visible under the DUs, ²⁹ which is usually responsible for the DU pathogenesis, and after this removal, the DUs usually heal quickly. The mechanical and/or autolytic debridement in SSc DUs, although performed only by a small number of rheumatology centres, is strongly encouraged as may reduce ulcer‐related pain and improve tissue healing. ³⁰ , ³¹

The WBS, which we propose for the first time in patients with SSc, may increase the possibility of a prognostic stratification of patients with DUs according to the clinical characteristics of the lesions at baseline. Although this scoring system is mainly designed for the evaluation of ulcers at the lower limbs, ⁷ , ³² it is an easy‐to‐use scoring system and showed a good correlation with healing time also in SSc DUs; however, the data need to be evaluated in larger cohorts. The main limitation that we observe in the application of this score in SSc DUs is the evaluation of the wound bed, as the presence of calcium deposits at this level is not included in the WBS. While instrumental methods such as laser Doppler speckled imaging ³³ may predict the healing time in patients with SSc and DUs, the WBS could be a fast and easy method that may be commonly applied in clinical practice.

The main strength of our study was that we included in the analysis a homogeneous cohort of consecutive patients with SSc, followed in a highly specialised centre, treated with a standardised treatment for the wounds. The therapeutic approach that we propose is feasible and applicable in different outpatient clinics. Moreover, for the first time, we identified in the WBS an easy‐to‐use and reliable prognostic factor for predicting the healing of DUs in SSc. One of the main limitations of our work is that we identified SC only by the clinical aspect and not by plain radiogram of the hands; however, in our opinion, when calcinosis in the hands is associated with DUs, it is easy to recognise because the calcium deposits are directly visible or identifiable with the tip of the scalpel. Moreover, although we did not use a validated scoring system, ³⁴ we preferred a clinical classification of the SC, which is more useful in the clinical practice.

5. CONCLUSION

DUs in patients with SSc are common, difficult to heal and are associated with severe pain and impact on hand functionality, thus significantly reducing the patient's quality of life. The management of DUs requires specific competences and the availability of a team with trained nurses and dedicated physicians; in particular, the removal of calcium deposits should be carried out only in dedicated centres where the wound healing rate can be higher. Data regarding topical treatment, local management, and the most effective wound dressing for patients with SSc‐related DUs (and in particular for those with SC) are lacking and further data from larger studies are required.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

ETHICS STATEMENT

The results of the procedures reported in this study are in accordance with the ethical standards and the study was conducted in accordance with the Declaration of Helsinki. The study was approved by the local ethics committee and all subjects signed the informed consent for their participation in this study.

Barsotti S, Venturini V, Di Battista M, et al. The impact of skin calcinosis on digital ulcers in patients with SSc: clinical and prognostic stratification using the “wound bed score”. Int Wound J. 2020;17:1783–1790. 10.1111/iwj.13467