Use of nailfold capillaroscopy for the early diagnosis of systemic sclerosis in patients with primary biliary cholangitis

Introduction

Primary biliary cholangitis (PBC) is a chronic autoimmune cholestatic liver disease characterized by the destruction of small and medium size bile ducts. Persistent cholestasis results in elevated alkaline phosphatase serum levels, while anti-mitochondrial antibodies (AMA) serve as a distinctive marker of disease. The clinical course is variable, with potential progression to cirrhosis and liver failure if the disease is not adequately managed [1].

Notably, a significant proportion of PBC patients (approximately 53%) also present with other autoimmune diseases [2]. Among these, systemic sclerosis (SSc) is frequently associated, with 1-22% of PBC patients receiving a concurrent SSc diagnosis, depending on the criteria used for the diagnosis [3]. However, the literature on this overlap syndrome is sparse, primarily consisting of case reports and limited case series.

SSc is characterized by sclerosis affecting the skin, mucosa and muscles, with variable involvement of internal organs, such as gastrointestinal tract, lungs, kidneys and heart. The pathophysiological mechanisms of SSc are small-vessel vasculopathy, autoantibody production, fibroblast dysfunction and excessive extracellular matrix deposition [4]. Severe SSc with extensive organ involvement can preclude liver transplantation (LT) eligibility in PBC patients.

In clinical practice, symptoms and signs of SSc in PBC patients are often under-recognized or inadequately investigated, particularly by gastroenterologists, because of the predominance of subclinical forms. Moreover, rheumatologic manifestations may be confused with PBC’s extra-hepatic symptoms, complicating the diagnosis. While some researchers have proposed screening algorithms for SSc in PBC patients, formal guidelines remain unestablished. Nailfold capillaroscopy (NFC) is pivotal for early SSc diagnosis, as well as for assessing disease activity and therapeutic response [5]. This observational cross-sectional study aimed to evaluate the effectiveness of systematic clinical assessments and NFC in the early diagnosis of SSc in PBC patients.

Patients and methods

Classification and diagnostic criteria for SSc

The 2013 ACR/EULAR criteria [7] and VEDOSS criteria [6] were used for the classification and diagnosis of SSc. The 2013 ACR/EULAR criteria are a classification currently used in clinical practice; the patients can be classified as having definite SSc. The VEDOSS criteria are diagnostic criteria used for the diagnosis of very early SSc.

According to the 2013 ACR/EULAR criteria, patients with a total score ≥9 are classified as having definite SSc (Table 1). The total score is calculated by summing the weighted values assigned to each of these features: skin thickening of the fingers, fingertip lesions, telangiectasia, abnormal nailfold capillaries at NFC, pulmonary involvement, Raynaud’s phenomenon, SSc-related autoantibodies. Skin thickening of the fingers of both hands, extending proximal to the metacarpophalangeal joints, is a sufficient criterion to classify the patient as having definite SSc [7].

Table 1 2013 ACR/EULAR criteria

According to the VEDOSS criteria, Raynaud’s phenomenon, puffy fingers and detection of ANA are considered as red flags. Coexistence of these clinical features must be considered suspicious for very early SSc. Diagnosis of very early SSc can be made in patients presenting a scleroderma pattern on NFC and/or SSc-related antibodies (ACA, anti-Scl-70, anti-RNA polymerase III) in addition to the abovementioned red flags [6].

Results

Study population characteristics

During the study period 145 patients were screened for eligibility. Three were excluded because of a previous SSs diagnosis, 52 refused the NFC, and 34 refused to sign the informed consent. Finally, a total of 56 patients who met the inclusion criteria were analyzed (Fig. 1). Characteristics of the study population are summarized in Table 2: 48 (86%) were females and 8 were males, and the mean age was 62 years (range 37-84), with a median of 63 years. Mean time from diagnosis of PBC was 9 years (range 6 months to 50 years). Twelve patients had overlap of PBC with autoimmune hepatitis.

Figure 1 Flowchart showing patient screening and outcomes

NFC, nailfold capillaroscopy; SSc, systemic sclerosis

Table 2 Features of study population

In terms of disease staging, 10 patients had hepatic cirrhosis due to PBC. Overall, 47 (84%) patients underwent liver stiffness measurement (FibroScan) and 21 (45%) had a liver stiffness value >9.6 KPa, suggestive of a high risk of liver decompensation, liver transplantation or death [10].

Regarding treatment, 53 (95%) patients were taking ursodeoxycholic acid (UDCA) and 6 (11%) were taking obeticholic acid (OCA). Of the UDCA-treated patients, 38 (72%) showed a complete biochemical response, while 3 (50%) of those on OCA normalized their liver tests. Response to therapy was defined according to EASL guidelines [1]. Concerning other medications, 9 (16%) patients were taking bezafibrate and 7 (12%) patients needed symptomatic therapy for pruritus. Three (5%) patients had undergone liver transplantation. Laboratory findings are detailed in Table 2.

Raynaud’s phenomenon was detected in 12 (21%) patients. None showed definite signs of SSc (skin sclerosis, pitting scars, digital ulcers, skin atrophy, microstomia, microcheilia), except for 1 patient who presented skin thickening of the fingers. Acrocyanosis was reported in 2 (4%) patients.

As outlined in Table 2, overall, auto-Ab screening revealed the following positivities: anti-ENA screening in 1 patient, anti-SSa/Ro in 4, anti-PML in 1 and anti-U1RNP in 1.

Regarding ANA, all patients were already tested at baseline of the hepatologic assessment and 27 (48%) were positive.

Results of NFC

As shown in Table 2, NFC abnormalities were found in 31 (55%) patients overall: 11 (20%) presented NFC minor abnormalities, 17 (30%) showed major abnormalities and 3 (5%) presented a scleroderma pattern (Fig. 2).

Figure 2 Scleroderma pattern: (A) microhemorrhages (#); (B) megacapillary (^); (C) and (D) capillary ectasia (>) and reduction in capillary density (*). The figure shows some of the typical capillaroscopic findings of scleroderma pattern: microhemorrhages, megacapillary, capillary ectasia and reduction in capillary density. This patient was diagnosed with an “early” scleroderma pattern. These pictures were taken using a digital video capillaroscope

Based on the NFC results, 20 (36%) patients were referred to undergo SSc-related autoantibody screening. All the patients with NFC abnormalities were also advised to undergo follow-up capillaroscopy.

The mean age of patients with scleroderma pattern was 55 years; the mean ages of patients with negative NFC, minor abnormalities and major abnormalities were 62, 61.9 and 61.8 years, respectively.

All PBC patients with scleroderma pattern had associated autoimmune diseases among those defined by EASL guidelines (e.g., Hashimoto’s thyroiditis, Sjögren disease or sicca complex, celiac disease, inflammatory bowel disease) [1]. These diseases were reported in 48%, 45% and 47% of patients with negative NFC, minor abnormalities and major abnormalities, respectively.

ANA positivity was observed in all patients with scleroderma pattern, and in 40%, 55% and 47% of patients with negative NFC, minor abnormalities and major abnormalities, respectively.

No relevant differences were found in biochemistry laboratory values among the categories of patients mentioned above.

Discussion

Our study integrated clinical examination, medical history, NFC and SSc-related autoantibody assessment in PBC patients, leading to the identification of 2 new SSc diagnoses. Our analysis showed a substantial prevalence of NFC abnormalities in patients with PBC, with a subsequent diagnosis of SSc in 2 patients. Based on our results, we pointed out crucial elements for SSc screening in PBC patients, proposing a diagnostic algorithm.

SSc is classified into diffuse cutaneous SSc (dcSSc) and limited cutaneous SSc (lcSSc). The distinction is based on skin sclerosis extension: the diffuse type is characterized by generalized skin sclerosis, the limited type is characterized by the sclerosis of the face and extremities of the limbs. Limited cutaneous SSc is the most common in PBC patients [11].

Pathophysiological features of SSc include small-vessel vasculopathy, autoantibody production and dysfunction of fibroblasts, leading to excessive extracellular matrix deposition. Clinical manifestations are heterogeneous; multiple organs and systems can be affected, including skin, gastrointestinal tract, lungs, kidneys and heart [7]. Systemic involvement often occurs early in the course of the disease and usually precedes symptom onset [6].

Multiorgan involvement of SSc may contraindicate LT in PBC patients with end-stage liver disease. All potential LT candidates should undergo a multidisciplinary evaluation before their inclusion on the transplant list. Cardiovascular, respiratory and renal function strongly impact the final decision about the reliability of LT [12]; therefore, patients with PBC and late-stage SSc could be excluded from the waiting list. Pulmonary arterial hypertension particularly increases the mortality after LT [12].

In PBC patients, SSc coexistence could alter the disease course, impacting transplant eligibility and prognosis. However, the literature on the influence of SSc on PBC prognosis is mixed, with some studies suggesting a better prognosis, while others report greater mortality due to SSc complications [13]. A higher incidence of a first episode of spontaneous bacterial peritonitis and septicemia was reported in patients with PBC and SSc, compared to PBC patients without SSc; an increased risk of infection due to immune abnormalities may explain this issue [13].

No disease-modifying drugs for SSc are currently available; however, early screening and organ-specific management appear to reduce disease progression and mortality. Proper investigation strategies should be adopted for the early identification of deterioration in pulmonary, gastrointestinal, renal and cardiac function. Current treatments include symptomatic therapies, immunosuppression, steroids and biological drugs. Targeted therapies are available for pulmonary arterial hypertension [14]. Patients may need hepatotoxic drugs to treat SSc complications (e.g., bosentan and methotrexate); thus, hepatic function should be carefully monitored.

Hudson et al studied 408 patients with SSc and pointed out that the mean time from the onset of RP to SSc diagnosis was 5±6.3 years, while the time from non-Raynaud’s symptoms to SSc diagnosis was 2.6±4.5 years [15]. SSc is an uncommon disease for most physicians: diagnosis and treatment are often challenging, in particular for patients who are not sent to referral centers [14].

The prevalence of SSc in PBC patients is quite substantial [16]; thus, a special diagnostic algorithm should be planned and used by gastroenterologist whenever needed. Diagnosis of SSc should be made by investigating early SSc signs before the patient complains about advanced symptoms of the disease.

Rigamonti et al [13] proposed an algorithm for SSc screening and diagnosis in PBC patients, focusing on the use of SSc-associated antibodies and NFC. However, official guidelines and indications for the diagnosis in PBC patients are lacking, and SSc identification is often delayed.

To our knowledge, this is the first study to investigate the prevalence of SSc in patients with PBC using 2013 ACR/EULAR and VEDOSS criteria to detect very early SSc, and the usefulness of NFC as a screening tool in this specific population.

RP was detected in 21.4% of our PBC patients; 12.5% of these were diagnosed with SSc. None of the patients without RP was diagnosed with SSc. This phenomenon is very common in patients with PBC, and it is a manifestation of SSc in a significant percentage of cases. SSc without RP is quite rare [17]. Based on this evidence, the indication to perform NFC can be properly given in all patients presenting RP at first evaluation and at any time during the clinical follow up.

Performing NFC at first is a useful approach, because it allows the patient to meet a rheumatologist who could properly evaluate all symptoms and signs. Moreover, SSc-related auto-antibody screening can be prescribed based on NFC results; thus, unnecessary extensive tests could be avoided.

Our study found NFC abnormalities in 55% of PBC patients, aligning with the literature on its prevalence in PBC populations and suggesting a role for endothelial dysfunction in PBC [18]. However, these findings must be interpreted cautiously, given the high prevalence of capillaroscopic abnormalities in healthy individuals. Hoerth et al [19] reported morphological changes in nailfold capillaries in 67% of healthy volunteers. Fahrig et al found morphological anomalies in 34% of the healthy population studied [20]. In another study, Fahrig et al reported capillary branching in 78% of the healthy subjects investigated, meandering in 94%, tortuous loops in 64%, hemorrhagic extravasation in 25% and apical dilatation in 19% [21]. Despite the similar prevalence of capillaroscopic abnormalities, the same findings should be considered carefully in people with PBC, given the high incidence of SSc in these patients. Proper follow up and further investigations, when needed, should be recommended when capillaroscopic abnormalities are found in patients diagnosed with PBC.

Our data did not reveal a specific capillaroscopic pattern for SSc in PBC patients. The high prevalence of capillaroscopic abnormalities in the healthy population makes it harder to find typical features in patients with PBC. At present, data on this issue are lacking; therefore, further studies involving a large number of patients are needed to better define the existence of a specific capillaroscopic pattern for PBC.

In our study, a scleroderma pattern on NFC was reported in 3 patients and was associated with SSc in 2 cases (66%). The third patient with scleroderma pattern did not fulfil the diagnostic criteria for SSc, and therefore a longitudinal observation was recommended. “Incomplete overlap” between PBC and SSc is also possible, and may represent an evolutive stage toward definite SSc [22].

Little is known about the clinical significance of isolated SSc features, such as capillaroscopic pathological patterns in patients affected by PBC without SSc symptoms and signs. Both PBC and SSc are characterized by elevated levels of the profibrotic cytokines transforming growth factor-β and interleukin-6, resulting in dysregulation of inflammatory processes and fibrotic tissue deposition [11]. The vessel impairment may explain the presence of isolated SSc signs in patients with PBC [18].

In our study, all the patients with a scleroderma pattern on NFC shared common features: they were female, they had an associated autoimmune disease and they had a positive ANA test. All these aspects are strongly consistent with the autoimmune nature of PBC and SSc [2].

Adding the number of newly diagnosed patients to the number of those already diagnosed with SSc (excluded from the study), the SSc prevalence among PBC patients in our center is about 11.5%. This value is consistent with literature data (5-12%) [23]. Thanks to our study, the number of SSc diagnoses in our center increased by 40%.

Based on our findings, we propose a diagnostic algorithm for SSc in PBC patients as follows (Fig. 3). RP should always be checked at baseline assessment in the PBC patient. If present, the patient should undergo NFC; otherwise, the patient should be regularly checked over the years by the gastroenterologist for ex novo occurrence of RP. If a scleroderma pattern is detected on NFC, the patient should undergo screening for SSc-related autoantibodies (ACA, anti-Scl-70, ANA and anti-ENA). In patients without a scleroderma pattern, SSc-related antibody screening and/or capillaroscopic follow up, or no other intervention, should be scheduled, according to the entity of abnormalities found on NFC. Patients with scleroderma pattern and/or with SSc-related autoantibodies should be assessed for fulfilment of the 2013 ACR-EULAR criteria [7] or VEDOSS criteria [6]. Patients diagnosed with SSc, according to these criteria, should be referred to a rheumatologist. The same diagnostic pathway should be clearly followed by patients with symptoms and signs of SSc at baseline assessment, even though they may have a later stage of the disease.

Figure 3 Diagnostic algorithm

NFC, nailfold capillaroscopy; PBC, primary biliary cholangitis; RP, Raynaud’s phenomenon; SSc, systemic sclerosis; ACA, anti-centromere antibodies; anti-Scl-70, antitopoisomerase I; ANA, antinuclear antibodies; ENA, extractable nuclear antigens

Several limitations of the present study must be acknowledged. First, as it was a cross-sectional study, patients were evaluated at a single point in time, without a specific follow up, except for capillaroscopy at 12 months from baseline. This design limits the ability to identify false negatives (patients who might develop SSc later) or observe disease progression. Without follow up, it is impossible to definitively assess whether patients classified as negative for SSc might develop the disease in the future; thus, the accuracy of the diagnostic algorithm could not be estimated. Second, 52 patients refused NFC, and 3 refused to sign informed consent, which could have introduced selection bias. The patients who agreed to participate may differ from those who declined in ways that are relevant to the study outcomes. Third, a larger sample would provide more robust conclusions.

In conclusion, NFC can help anticipate the SSc diagnosis in PBC patients, leading to the possibility of early treatment and prevention of major complications. Gastroenterologists should vigilantly monitor for clinical signs, particularly RP, and employ NFC as needed. Our proposed algorithm, focusing on early detection and appropriate follow up, could significantly impact patient outcomes by facilitating timely SSc diagnosis and management.