Autonomic nervous system dysfunction correlates with microvascular damage in systemic sclerosis patients

Abstract

Objectives:

Cardiac autonomic neuropathy is among the known cardiovascular complications of systemic sclerosis and may affect the whole prognosis of the disease. The aim of our study was to assess cardiac autonomic neuropathy prevalence in our cohort of systemic sclerosis patients and compare its main features with clinical and epidemiological data, particularly with the severity of microvascular damage, as detected by nailfold videocapillaroscopy.

Methods:

Twenty-six patients with definite systemic sclerosis were consecutively enrolled at our outpatient rheumatology clinic. All patients underwent physical examination, nailfold videocapillaroscopy, and autonomic neuropathy diagnostic tests (orthostatic hypotension test, deep breathing test, lying-to-standing, and Valsalva maneuvers).

Results:

Cardiac autonomic neuropathy prevalence was 50% (13 cases). On univariate analysis, cardiac autonomic neuropathy was shown to be significantly associated with an active pattern on nailfold videocapillaroscopy (odds ratio 5.86, 95% confidence interval 1.59–9.24; p = 0.032), whereas anti-Scl-70 positivity (odds ratio, 0.24; 95% confidence interval, 0.03–2.12; p = 0.049) and C-reactive protein (odds ratio, 19.32; 95% confidence interval, 1.79–56.71; p = 0.036) reached only a borderline statistical association. The time-dependent Cox multivariate regression model showed cardiac autonomic neuropathy development to be independently associated with an active pattern on nailfold videocapillaroscopy (odds ratio, 7.19; 95% confidence interval, 1.87–8.96; p = 0.042) and anti-Scl-70 positivity (odds ratio, 5.92; 95% confidence interval, 1.06–18.43; p = 0.048).

Conclusions:

Severe microvascular damage, as detected by nailfold videocapillaroscopy, may suggest the coexistence of autonomic dysfunction and should be considered as a red flag for the identification of patients particularly at risk of cardiac morbidity and mortality.

Introduction

Systemic sclerosis (SSc) is a rare, multisystemic disorder of connective tissue, characterized by widespread inflammation, vascular abnormalities, and both skin and visceral organ fibrosis. ¹ Vascular dysfunction has long been recognized as a key factor in SSc, but its pathologic mechanisms still remain poorly understood. Some disease features (e.g. microcirculatory impairment, abnormal esophageal motility, and gastrointestinal dysfunction) may be partially explained by autonomic dysfunction.^2–4

Cardiac autonomic neuropathy (CAN) is a well-recognized complication of SSc and may affect the whole prognosis of the disease. ⁵ A number of cardiovascular complications may occur during the course of SSc, including dysrhythmias, pulmonary hypertension, and autonomic nervous system dysfunction. ⁶ Increasing evidence strongly suggests that cardiac involvement is related to recurrent focal ischemic injury causing irreversible myocardial fibrosis. ⁷ The underlying mechanism seems to be microcirculatory impairment with abnormal vasoreactivity, the so-called myocardial Raynaud’s phenomenon, possibly due to dysfunctional autonomic control.^8,9

Autonomic dysfunction is extremely common in SSc. It starts early in the disease course and may predate fibrosis development. ¹⁰ However, due to discrepancies between clinical manifestations and actual cardiac involvement, the diagnosis of impaired cardiac function may sometimes be late or difficult. For this reason, early detection is valuable for optimal treatment and implementation of preventive measures during the early stages of the disease. The current challenge is to detect early preclinical cardiac function alterations and to identify SSc patients at risk of arrhythmic complications using simple noninvasive diagnostic procedures. Conventional cardiovascular autonomic reflex tests (orthostatic hypotension test, deep breathing test, lying-to-standing, and Valsalva maneuvers) are easy to perform and well reproducible, if used under standardized conditions.^11,12

Despite uncertainties concerning central and peripheral nervous system involvement in SSc, autonomic system alterations and neurological involvement in scleroderma have been well recognized and better characterized throughout the last two decades.^13,14

Raynaud’s phenomenon is the most common and earliest manifestation of SSc and can be considered the main expression of vascular dysfunction. Raynaud’s phenomenon has been hypothesized to result from dysregulated control mechanisms and endothelial injury, yielding an imbalance between vasoconstriction and vasodilatation (in favor of vasoconstriction). ¹⁵ Since nailfold videocapillaroscopy (NVC) is a simple technique able to detect early microvascular changes in SSc patients, the aim of our study was to determine CAN prevalence in our cohort of SSc patients and to assess whether microvascular damage, as detected by NVC, as well as other clinical and epidemiological factors may help predict CAN coexistence.

Methods

Study design

This was a cross-sectional study. Twenty-six patients with definite SSc diagnosis ¹⁶ were recruited consecutively at routine control visits at the outpatient clinic of the A.O.U. Luigi Vanvitelli, Naples (Italy), according to the following inclusion criteria: age ⩽65 years, absence of myositis, absence of heart failure on clinical examination, normal chest X-ray, and normal biventricular systolic function at color Doppler echocardiography. Patients with arterial hypertension, diabetes, or any other disease potentially capable of determining autonomic dysfunction were excluded. No patient was administered specific therapeutic regimens at the time of the study. We also enrolled an age-matched control group of 26 healthy volunteers from the health personnel of our clinical facility, which was a test for autonomic nerve function.

All patients expressed their informed consent to participate in the study. Written consent was obtained according to the 1976 Declaration of Helsinki and its later amendments. The study was also approved by our local ethics committee (University of Campania “Luigi Vanvitelli,” Naples, Italy).

Clinical features and visceral involvement

All patients underwent an extensive evaluation of the signs and symptoms of SSc. In particular, the following parameters were recorded: extent of cutaneous involvement, presence of telangiectasias, calcinosis, cutaneous ulcers, Raynaud’s phenomenon, arthritis, esophageal involvement (X-ray hypomotility with or without dysphagia), and scleroderma renal crisis (proteinuria > 300 mg/24 h). Pulmonary involvement was also carefully investigated by means of high-resolution computed tomography (HRCT), spirometry, and diffusing lung capacity for carbon monoxide (DLCO). M-mode, two-dimensional (2D), and color Doppler flow conventional echocardiography was also carried out in all patients. Disease activity index was further recorded, according to the Valentini criteria for disease activity in SSc. ¹⁷

Serological parameters

The following serological markers were evaluated using standard techniques: antinuclear antibodies (ANAs, by indirect immunofluorescence on Hep-2 cell line), antiextractable nuclear antigens (ENAs, including Scl-70, RNP, Sm, SS-A/Ro, SS-B/La, CENP, Jo-1, PCNA, RNA Pol III, and PM-Scl), and complement C3 and C4.

NVC

NVC was carried out with patients seated with their hands positioned at heart level, at a room temperature of 22°C–25°C. A drop of immersion oil was applied to the nailfold to increase the translucency of the keratin layer. All fingers were examined by an experienced operator (F.M.), using a VideoCap 25videocapillaroscope (DS Medica, Freiburg, Germany), at 200× magnitude.

The following alterations were considered: enlarged and giant capillaries, hemorrhages, loss of capillaries, disorganization of the vascular array, ramified/bushy capillaries, and blood sludging.^18–20 These alterations were searched for in all fingers. A semiquantitative score system for avascular areas was used, as follows: 0 = no evidence of avascular areas; 1 = mild loss of capillaries; 2 = moderate loss of capillaries; and 3 = severe loss of capillaries. ²¹ In depth, capillary density was measured as the number of capillaries in a 1-mm span of the distal row in each finger. ²²

Patients were diagnosed for SSc according to the following patterns’ classification ²⁰ : (1) early NVC pattern (E), which included few capillary hemorrhages and enlarged/giant capillaries, a relatively well-preserved capillary distribution, and no evident loss of capillaries; (2) active NVC pattern (A), characterized by frequent giant capillaries and capillary hemorrhages, a moderate loss of capillaries, a mildly disorganized capillary architecture, some avascular areas, and absent/mild ramified capillaries; and (3) late NVC pattern (L), typically represented by an irregular enlargement of the capillaries, few/absent giant capillaries, no evidence of hemorrhages, severe loss of capillaries and large avascular areas, a harshly disorganized normal capillary array, and frequent ramified/bushy capillaries. ²³

CAN diagnostic tests

Orthostatic hypotension was diagnosed as a reduction in systolic blood pressure (SBP) ⩾20 mmHg and/or diastolic blood pressure (DBP) ⩾10 mmHg, when passing from the clinostatic to the orthostatic position. The deep breathing test was used to determine heart rate variability, both in inspiration and exhalation, while registering an electrocardiogram (ECG) for 1 min and adjusting results for age. In addition, the lying-to-standing test (i.e. standing from the supine position) was performed and the ratio between the longest R-R interval at the 30th and the shortest RR interval at the 15th beat was measured, with results again adjusted for age. Finally, a standardized Valsalva maneuver was performed for 15 s having the patient breathing against a resistance of 40 mmHg during ECG registration extended for up to 1 min after the effort. The Valsalva ratio was calculated by measuring the longest R-R interval after the maneuver and the shortest R-R interval during the maneuver. CAN diagnosis was made if at least two of the four tests were positive.

Peripheral neuropathy tests

As previously reported, peripheral neuropathy could be present in sclerodermic subjects as a consequence of nerve fibers degeneration and microvascular damage. ²⁴ The DN4 questionnaire, validated as a screening tool for painful diabetic neuropathy, was also administered to uncover potential neuropathy. ²⁵ All patients were then subjected to the Michigan Neuropathy Screening Instrument (MNSI) neurological test. On physical examination, feet were examined for the presence of deformities, dry skin, calluses, infections, and ulcers. Feet deformities included: prominence of metatarsal heads, hallux valgus, joint subluxations, and a Charcot joint. One point (for each foot) was assigned, if at least one of the above signs was present, and another one in the case of existing ulcers. The other three signs included in the questionnaire were the following: response to vibration stimulus, heel tendon reflex assessment, and response to a tactile stimulus using a 10-g monofilament. A stimulus to vibration was obtained through a 128-Hz diapason placed at the distal interphalangeal joint of the hallux. A score of either 1 or 0.5 point was assigned depending on whether the stimulus was not perceived at all or only during the entire duration of the stimulation, respectively.

The heel tendon reflex was scored with either 1 or 0.5 point in case the reflex was absent or if hyperelicitable, respectively. Similarly, the tactile stimulus was assigned either 1 or 0.5 point in the absence of perception or if not perceived on at least one of the four points of stimulus, respectively. Neuropathy was classified as probable by MNSI if the overall score was ⩾2.5.^26,27

Finally, to identify patients with an altered threshold of vibration perception according to age, ²⁷ the vibration perception threshold (VPT) examination was carried out using a neurotensiometer, with the patient lying down and vibratory stimuli, ranging from 0 to 50 µV, performed bilaterally at the medial malleolus of the ankle and at the basis of the first toe.

Endpoints

We first compared, as for autonomic nerve function investigations, our SSc population with a control group of healthy volunteers. We then focused on SSc population, aiming at the evaluation of the prevalence of dysautonomia and, as well, potential associations between diagnostic tests for dysautonomia and clinical, laboratory, and instrumental data.

Statistical analysis

As this was a pilot study, no sample size calculation was mandatory.

Categorical data were expressed as numbers and percentages, while continuous variables were reported either as medians and interquartile ranges (IQRs) or means and standard deviations (SDs), based on their distribution. Gaussian distribution was tested for each variable with the Kolmogorov–Smirnov Goodness-of-Fit test.

For categorical variables, comparison of groups was performed by applying the Fisher’s exact test for dichotomic covariates (sample size <40) or chi-squared test in the case of “m × n” contingency tables (with n >2). In this latter case, due to the small sample size, Yates correction was also applied. Continuous variables were assessed by the nonparametric Mann–Whitney U test or the Wilcoxon test.

Potentially independent associations with the outcome of interest (CAN) were probed using a multivariate logistic regression model, with stepwise selection method.

All tests were two-tailed, and a p value <0.05 was considered statistically significant. Data were analyzed using the SPSS Software, Version 24 (IBM, Armonk, NY) and STATA 14.0 software (StataCorp. 2015; StataCorp LP, College Station, TX).

Results

The main clinical-serological data of SSc patients are summarized in Table 1. Briefly, the study population mainly consisted of female subjects (24 females; 92.9%), with a median age of 58.5 years [IQR, 49–64.3 years] and a median duration of the disease of 4 years [IQR, 2–11.5 years]. Blood pressure (BP) was basically within the reference range, with a median systolic value of 120 mmHg [IQR, 117.5–130 mmHg] and a median diastolic value of 75 mmHg [IQR, 70–80 mmHg].

Table 1.

General characteristics of the study population (n = 26).

Variable
Clinical-epidemiological data
Age (years), median [IQR]	58.5 [49–64.3]
Sex, n (%)
M/F	2 (7.7)/24 (92.3)
BMI (kg/m2), median [IQR]	23.7 [22.8–28]
Laboratory parameters
ESR, median [IQR]	20.5 [12.8–31.5]
ESR, n (%)
Normal/abnormal	18 (69.2)/8 (30.8)
CRP (g/dL), median [IQR]	0.28 [0.08–0.44]
CRP, n (%)
Normal/abnormal	23 (88.5)/3 (11.5)
Autonomic tests
Vagal cardiac function test, n (%)
Positive/negative	18 (69.2)/8 (30.8)
Deep breathing test, n (%)	8 (30.8)
Valsalva maneuver, n (%)	12 (46.2)
Lying-to-standing, n (%)	8 (30.8)
Orthostatic hypotension, n (%)	10 (38.5)
VPT, n (%)	—
CAN, n (%)	13 (50)
MNSI, n (%)
Positive (⩾2.5)	2 (7.6)
Negative (<2.5)	24 (92.4)
DN4 questionnaire, n (%)
Positive (>3)	8 (30.8)
Negative (⩽3)	18 (69.2)
Clinical and SSc parameters
DLCO, median [IQR]	73 [64.5–95.5]
DLCO, n (%)
Normal/abnormal	10 (36)/16 (64)
FVC, median [IQR]	9.5 [85.3–114.3]
FVC, n (%)
Normal/abnormal	21 (80.8)/5 (19.2)
E/A ratio, n (%)
Normal/abnormal	9 (34.6)/17 (65.4)
EF, %, mean (SD)	60.8 (5.4)
mRss, mean (SD)	5.7 (9.5)
PAPs, median [IQR]	30 [22–35]
Pulmonary hypertension (PAPs >35 mmHg), n (%)	4 (15.4)
Activity index, median [IQR]	1.50 [0.88–3]
Duration of disease (years), median [IQR]	4 [2–11.5]
Videocapillaroscopic pattern, n (%)
Normal	3 (11.5)
Early	4 (15.4)
Active	13 (50)
Late	6 (23.1)
Capillary density, median [IQR]	5 [3–7]
Cutaneous involvement, n (%)
Limited	21 (80.8)
Diffuse	5 (19.2)
Gastrointestinal dysmotility, n (%)	18 (69.2)
Scleroderma renal crisis, n (%)	—
ANA positive, n (%)	26 (100)
ENA, n (%)
No markers	8 (30.8)
ACA	11 (42.3)
Scl-70	7 (26.9)
Complement, median [IQR]
C3	117 [94–135]
C4	21.8 [17.6–25.8]
SS-A/Ro positive, n (%)	2 (7.7)
SS-B/La positive, n (%)	1 (3.9)
Complement, median [IQR]
C3	117 [94–135]
C4	21.8 [17.6–25.8]
Sicca syndrome, n (%)	10 (38.5)
Dyspnea, n (%)	9 (34.6)
Ulcers, n (%)	2 (7.7)
Pitting scars, n (%)	10 (38.5)
Telangiectasias, n (%)	19 (76)

ACA: anticentromere antibodies; ANA: antinuclear antibodies; BMI: body mass index; CAN: cardiac autonomic neuropathy; CRP: C-reactive protein; DLCO: diffusing capacity of lung for carbon monoxide; EF: ejection fraction; ENA: antiextractable nuclear antigens; ESR: erythrocyte sedimentation rate; F: females; FVC: forced vital capacity; IQR: interquartile range; M: males; MNSI: Michigan neuropathy screening instrument; PAPs: pulmonary artery pressure; SD: standard deviation; VPT: vibration perception threshold.

Diffuse SSc (dcSSc) was diagnosed in 5 patients (19.2%), Raynaud’s phenomenon was present in 26 patients (100%), 19 patients (76%) had telangiectasias, 2 patients (7,7%) were suffering from digital ischemic changes, and 10 patients (38.5%) showed pitting scars. Esophageal dysmotility was found in 18 patients (69.2%), while basal pulmonary fibrosis by chest radiography was detected in 7 patients (26.9%). For what concerns scleroderma renal manifestation of SSc, no patient has shown proteinuria >300 mg/24 h. ANA was positive in all patients (100%), 11 patients (42.3%) had detectable anticentromere antibodies (ACA), while 7 patients (26.9%) tested positive for anti-Scl-70. We also measured the complement, with a median C3 of 117 [IQR 94–135] and C4 of 21.8 [IQR 17.6–25.8]. Moreover, as for antiextractable nuclear antigens (anti-ENAs), we found SS-A/Ro positive in two patients (7.7%), while only one had SS-B/La positive. All other anti-ENAs were instead absent in our study population.

NVC revealed an active pattern in 13 cases (50%), a late pattern in 6 patients (21.4%), and an early pattern in 4 patients (15.4%), with a median capillary density of 5 [IQR, 3–7]. No abnormalities were observed in three subjects (11.5%).

With regard to neuropathy assessment, VPT results were normal in all subjects. DN4 and MNSI questionnaires were positive in eight (30.8%) and two patients (7.6%), respectively. Abnormal results during the Valsalva maneuver were recorded in 12 patients (46.2%); 10 subjects (38.5%) had orthostatic hypotension. The deep breathing test and lying-to-standing maneuver were both positive in eight patients (30.8%). CAN prevalence was 50% (13 cases) in our cohort of patients. All data are shown in Table 1.

We thus compared autonomic nerve function investigations in our SSc population with an age-matched control group of healthy volunteers, showing significant differences in all autonomic tests between the two groups (Table 2). Hence, the SSc study population was divided into two subgroups according to the presence or absence of CAN (Table 3); potential differences between the two subgroups with regard to such covariates as age, duration of disease, gastrointestinal dysmotility, sicca syndrome, cutaneous involvement, and type of NVC pattern were then investigated. Differences were tested by adjusting the model for the duration of disease; therefore, a univariate logistic regression model including the duration of disease as a time covariate was performed.

Table 2.

Autonomic nerve function comparison of SSc subgroup with a healthy volunteer population (n = 52).

	SSc (n = 26)	Healthy volunteers (n = 26)	p
Clinical-epidemiological data
Age (years), median [IQR]	58.5 [49–64.3]	53.5 [32.8–69.3]	0.301
Sex, n (%)			1.000
M/F	2 (7.7)/24 (92.3)	3 (11.5)/23 (88.5)
BMI (kg/m2), median [IQR]	23.7 [22.8–28]	24.8 [21.7–32.7]	0.509
Autonomic tests
Vagal cardiac function test, n (%)			<0.001
Positive/negative	18 (69.2)/8 (30.8)	0 (–)/26 (100)
Deep breathing test, n (%)	8 (30.8)	0 (–)	0.007
Valsalva maneuver, n (%)	12 (46.2)	2 (7.7)	0.005
Lying-to-standing, n (%)	8 (30.8)	0 (–)	0.007
Orthostatic hypotension, n (%)	10 (38.5)	0 (–)	0.002
VPT, n (%)	–	2 (7.7)	0.471
CAN, n (%)	13 (50)	0 (–)	<0.001
MNSI, n (%)			0.471
Positive (⩾2.5)	2 (7.6)	0 (–)
Negative (<2.5)	24 (92.4)	26 (100)
DN4 questionnaire, n (%)			0.028
Positive (>3)	8 (30.8)	1 (3.8)
Negative (⩽3)	18 (69.2)	25 (96.2)

BMI: body mass index; CAN: cardiac autonomic neuropathy; F: females; IQR: interquartile range; M: males; MNSI: Michigan neuropathy screening instrument; SD: standard deviation; VPT: vibration perception threshold.

Table 3.

Univariate and multivariate analysis according to cardiac autonomic neuropathy (n = 26).

Parameter	Univariate analysis				Multivariate analysis
	CAN
	Yes (n = 13)	No (n = 13)	OR [95% CI]*	p*	OR [95% CI]*	p*
Age (years), median [IQR]	58 [48–64.5]	60 [49–64]	1.01 [0.93–1.08]	0.970
Activity index, median [IQR]	2 [0.75–3]	1.5 [0.75–3]	1.15 [0.72–1.85]	0.560
Duration of disease (years), median [IQR]	3 [2–7.5]	7 [2–17]	0.91 [0.80–1.04]	0.154
Videocapillaroscopic pattern, n (%)
None	—	2 (15.4)	—	0.239	—	0.289
Early	3 (23.1)	2 (15.4)	6.19 [0.89–16.43]	0.076	4.16 [0.86–6.75]	0.089
Active	8 (61.5)	5 (38.5)	5.86 [1.59–9.24]	0.032	7.19 [1.87–8.96]	0.042
Late	2 (15.4)	4 (30.8)	2.98 [0.59–12.76]	0.121	3.17 [0.94–14.66]	0.392
Cutaneous involvement, n (%)
Limited	11 (84.6)	10 (76.9)
Diffuse	2 (15.4)	3 (23.1)	1.65 [0.23–11.99]	0.621
ENA, n (%)
No markers	5 (38.5)	3 (23)	—	0.412	—	0.192
ACA	6 (46.2)	5 (38.5)	0.72 [0.11–4.62]	0.729	2.56 [0.69–10.76]	0.328
Scl-70	2 (15.4)	5 (38.5)	0.24 [0.03–2.12]	0.049	5.92 [1.06–18.43]	0.048
Autonomic tests
Deep breathing test, n (%)	8 (61.5)	—	4.71 [1.59–13.68]	0.026
Valsalva maneuver, n (%)	10 (76.9)	2 (15.4)	18.33 [2.52–43.26]	0.004
Lying-to-standing, n (%)	7 (53.8)	1 (7.7)	14.00 [1.39–76.35]	0.025
Orthostatic hypotension, n (%)	7 (53.8)	3 (23)	3.89 [0.72–21.06]	0.115
DN4 questionnaire, n (%)			0.94 [0.62–1.42]	0.754
Positive (>3)	4 (30.8)	13 (100)
Negative (⩽3)	9 (69.2)	-
C-reactive protein (g/dL), median [IQR]	0.40 [0.15–0.66]	0.14 [0.06–0.33]	19.32 [1.79–56.71]	0.036

ACA: anticentromere antibodies; CAN: cardiac autonomic neuropathy; CI: confidence interval; ENA: extractable nuclear antigens; IQR: interquartile range; OR: odds ratio.

On univariate analysis, a statistically significant difference between the two subgroups was observed for the active NVC pattern (odds ratio (OR), 5.86; 95% confidence interval (CI), 1.59–9.24; p = 0.032); significance was indeed only borderline for anti-Scl-70 positivity (OR, 0.24; 95% CI, 0.03–2.12; p = 0.049) and CRP levels (OR, 19.32; 95% CI, 1.79–56.71; p = 0.036). Other variables significantly different on univariate analysis between the two patient subgroups were the deep breathing test and the Valsalva maneuver (Table 3). Next, a multivariable logistic regression model, including variables statistically significant on univariate analysis, was performed. However, due to the limited number of events, the model was fitted only on two covariates, namely, NVC pattern and ENA. A significant independent association was thus shown between CAN development and both active NVC pattern (OR, 7.19; 95% CI, 1.87–8.96; p = 0.042) and anti-Scl-70 positivity (OR, 5.92; 95% CI, 1.06–18.43; p = 0.048). Detailed data are reported in Table 3.

Discussion

Although not always routinely assessed, CAN is among the known complications of SSc. In our study population, CAN prevalence was found to be particularly high, probably due to the thorough assessment our patients went through. Specifically, a high percentage of patients were found to be affected by orthostatic hypotension, consistently with previous reports showing parasympathetic dysfunction, sympathetic overactivity, and depression of the heart rate circadian rhythm in SSc patients.^28,29 The involvement of the parasympathetic system as a cause of dysautonomia was also inferred by the abnormal results obtained during the lying-to-standing, deep breathing, and Valsalva maneuvers.^30,31 However, since CAN occurrence has been found highly prevalent in a systematic review of the literature, ³² ad hoc studies in larger patient populations may even disclose far higher rates of CAN coexistence in SSc.

The sympathetic nervous system plays a major role in thermoregulation; therefore, the autonomic nervous system may play a role in the pathogenesis of Raynaud’s phenomenon, via both central and peripheral (local) mechanisms, ³³ resulting in microvascular dysfunction. Parasympathetic and sympathetic dysfunctions in SSc have been reported in the literature^{5,30,31,34,35}; however, in most cases, conventional autonomic testing, ³⁶ which lacks standardization and reproducibility,^37,38 has been used and there is often no accurate information about patient selection. Moreover, large differences in the incidence of autonomic dysfunction in SSc patients have been reported, ranging from 0% ³⁹ to 100%, ⁵ with the CREST variant claimed to be specifically correlated with autonomic impairment by Hermosillo et al. ⁴⁰ In our study, however, we did not find a substantial difference between the limited and diffuse variant of SSc with regard to CAN prevalence.

The 26 SSc patients of our study were strictly selected and results clearly showed a correlation between NVC abnormalities and autonomic dysfunction, consistently with recent reports suggesting the role of neural control of vascular tone in the pathogenesis of Raynaud’s phenomenon.^41,42 Other than Raynaud’s phenomenon, autonomic dysfunction may also explain such typical features of scleroderma, as esophageal dysmotility.^43,44 The current hypothesis on the pathogenesis of RP suggests the presence of a state of sensory nervous system failure in SSc, leading to unopposed endothelial and platelet control of vascular tone. Endothelial injury and platelet activation in SSc may induce a shift in vascular wall function, from a vasodilatory function to a provasospastic function, unopposed by a vasodilatory sensory input. Thus, enhanced vasospasm is generated. ⁴⁵

NVC is a simple, noninvasive, and reproducible technique able to detect microvascular abnormalities and disease progression over time. ²⁰ Unlike Di Franco et al., ⁴¹ a significant association between the active NVC pattern and CAN was found in our patients. Since the association between the active NVC pattern and presence of CAN was found in patients with definite SSc diagnosis but not in those with pre-SSc, a concomitant involvement of skin and sympathetic and parasympathetic cardiac fibers may represent prerequisites for CAN development.

The median disease duration in our study was 4 years, similarly to what reported by Othman et al. ⁴⁵ This relatively short disease duration resulted in a decreased prevalence of patients with overt visceral involvement in our study cohort, which could suggest an earlier involvement of nerve fibers respect to the generalized organ damage. However, these data are still controversial and need further insights.

In the present study, we also found an association between CAN occurrence and anti-Scl-70 positivity. Interestingly, Ferri et al. ²⁸ found a correlation between anti-Scl-70 antibodies, sinus node activity, and relative risk of death in older SSc patients. Indeed, anti-Scl-70 positivity is known to be correlated with a severe outcome and to be associated with alterations in the cardiac rhythm in SSc patients. ⁴⁶ However, we cannot rule out that the relationship between the presence of anti-Scl-70 and cardiac arrhythmias may be due to the high prevalence of both these factors in SSc, without a clear pathophysiological link between them. Therefore, all these associations may as well be an expression of a more active and severe disease, thus representing negative prognostic factors overall.

The main limitation of this study is the small sample size, which does not allow us to generalize our findings at the moment. However, we must underline that SSc is a rare disease, which implies per se small populations. Of course, this would need a cautious interpretation of our findings, even though the evidence from the literature about an association with the microvascular damage of the nerve fibers and mostly with the involved pathogenetic mechanism is really poor. Hence, future prospective, larger cohort study would be helpful to strengthen our conclusions. Meanwhile, in agreement with Ferri et al., ⁹ autonomic cardiac neuropathy may represent an important prognostic feature, which should be looked at for a more complete evaluation of all SSc patients. An abnormal autonomic nervous control of the heart may impact the mortality rate for cardiac disease, particularly in the case of subclinical cardiac involvement. As shown in this study, the coexistence of autonomic dysfunction with severe microvascular damage, as detected by NVC, may thus facilitate identification of those patients particularly at the risk of cardiac death and implementation of consequent preventive treatment strategies, other than allowing for a thorough assessment of SSc complications.