PDF
Citation
Print
INTRODUCTION
Systemic sclerosis (SSc) is a rare systemic connective tissue disease that may be associated with internal organ involvement and several systemic manifestations. The hallmarks of the disease are vasculopathy, autoimmunity, and tissue fibrosis [1]. Vital organs may be affected with dermatological, respiratory, ocular, cardiac, and renal complications [2]. SSc vasculopathy is characterized by the proliferation of intima, thickening of the basement membrane, loss of vascular endothelial cadherin, rarefaction of capillaries, platelet aggregation, and formation of microthrombi. In advanced stages, massive perivascular fibrosis, luminal occlusion, and tissue fibrosis develop [3].
Vascular damage is one of the pathogenetic factors for SSc; abnormal capillary morphology may appear before the onset of clinical symptoms. Capillary damage may reflect internal organ involvement in patients with established disease and help determine the disease stage [4]. Although large arteries may be involved, vasculopathy predominantly affects the microcirculation. Microvascular involvement may lead to mononuclear cell infiltration of the vessel wall, endothelium rupture, and obliterative lesions. These structural vascular changes and protracted vasospasm lead to insufficient blood flow, tissue injury, atrophy, and even capillary infarction [5].
The ocular microvascular changes in SSc patients are not different from peripheral vascular abnormalities, including fibroproliferative vasculopathy, concentric narrowing, atrophy, and fibrosis of small retinal vessels [6]. The retina and choroid are suitable for evaluating the microcirculation. The choroid is affected through impaired choroidal autoregulation, circulating mediators, and prolonged vasospasm, with decreased blood flow in the choroid, resulting in tissue damage. Thinning of the choroid layer with a rich vascular network and reduced retinal microvascular density have been reported [7].
Recent studies demonstrated that early retinal microvascular changes in SSc patients, particularly vessel density (VD) reduction, occur with no clinical evidence of retinopathy, suggesting a widespread vascular injury [8,9]. Thus, detecting early changes in the ocular microcirculation may facilitate early diagnosis. Evaluation of the retinal and choroidal vasculature would be optimal for identifying arteriole and capillary pathology in SSc, given that choroidal vessels have extensive blood flow and retinal tissue has the highest oxygen extraction per blood volume [10]. The disease can affect the anterior and posterior segments of the eye. Changes in the anterior segment include dry eye syndrome, eyelid skin remodeling, and conjunctival abnormalities. The disease’s impact on the posterior segment of the eye primarily affects the retinal microcirculation and the optic nerve [2]. Ocular manifestations in SSc patients are increasingly recognized, suggesting possible correlations between systemic and ocular microvascular changes [11].
Early diagnosis and adequate treatment can improve patient survival. Nail fold capillaroscopy (NFC) and optical coherence tomography angiography (OCTA) can help early detect vascular injury in patients with SSc [12]. Capillaroscopy is a safe, non-invasive, reliable diagnostic technique designed for evaluating the small vessels of the microcirculation [13,14]. OCTA is a technique used in daily clinical practice, which provides motion contrast imaging for the capillary network with high-resolution assessment of volumetric blood flow [13,14]. In addition, it is a non-invasive procedure and assesses retinal perfusion [15,16].
Recent studies shed light on retinal-choroidal abnormalities by OCTA as a new non-invasive technology for assessing vascular architecture. In the future, OCTA could play an essential role in managing these patients, acting as a potential tool for early diagnosis of SSc, discriminating between disease stages, and evaluating the effects of systemic therapy on microvascular changes. The direct link between the type of ocular affection and the evolution of systemic diseases has not been determined. Factors influencing the incidence and severity of ocular abnormalities include the degree of skin fibrosis, autoantibody profile, organ affection, microcirculatory changes severity, and the inflammatory process [17]. Few studies with controversial results have assessed the correlations between the parameters of NFC and the retinal changes by OCTA [18-20]. Our study is one of the few studies aimed at detecting the correlation between retinal microvascular changes by OCTA and nailfold capillaroscopic findings in patients with SSc.
MATERIALS AND METHODS
Study population
Forty SSc patients attending the rheumatology and rehabilitation department at Sohag University hospital and thirty healthy controls with matching age and sex were included in the study between May 2023 and July 2024 after giving written informed consent. This study was carried out in accordance with the ethical standards laid down in the Helsinki Declaration of 1975 and its later amendments in 2000, and approved by The Medical Research Ethics Committee, Faculty of Medicine - Sohag University under IRB registration number: Soh-Med-23-04-05PD. All patients included in this study gave written informed consent to participate in this research.
Exclusion criteria
Patients below 18 years old, patients not capable of giving written informed consent, pregnancy, patients with other systemic diseases such as (cardiovascular diseases, cancer, severe infection, renal and hepatic diseases), active smokers, and clinical conditions that could pose an obstacle for small vascular network assessment at OCTA and NFC, patients with any other autoimmune disease rather than SSc and patients with ocular symptoms or coexisting ocular diseases (including errors of refraction as myopia, cataracts, glaucoma or a previously diagnosed retinal disease), and corneal opacities that could render the quality of the OCTA scans.
Rheumatological assessment
Patients with SSc were classified using the 2013 American College of Rheumatology/European League Against Rheumatism classification criteria [21]. All SSc patients underwent complete history, clinical, rheumatological, and laboratory evaluation. We assessed the skin thickness using the modified Rodnan skin score (mRSS), with a range of 0 to 51, using a standard 0, 1, 2, 3 scale at 17 body sites [22]. The patient was adequately positioned to allow relaxation of the area to be examined, and all recommendations and precautions were applied to ensure proper evaluation. The skin thickness for all patients was evaluated, and the mRSS was calculated by a single experienced rheumatologist who was adequately trained and well-qualified for mRSS assessment. According to the magnitude of the skin involvement, SSc patients were categorized as limited cutaneous (Lc) or diffuse cutaneous (Dc). Internal organ involvement was assessed, and a peripheral small vascular network evaluation was performed using NFC. An ophthalmological examination was done for all participants using OCTA.
Nailfold capillaroscopy
The same qualified examiner performed NFC in patients and controls in four fingers of both hands (2nd to 5th fingers). Distinctive for a scleroderma pattern in nailfold capillaroscopy, including dimensions of the capillaries, abnormal capillary patterns, and hemorrhages, were assessed. The mean capillary density per millimeter was calculated to analyze it and to correlate it with the OCTA data. The progressive obliterative character of SSc was evaluated according to Cutolo et al. [23] and classified into early, active, and late. NFC parameters were assessed for all participants on the same day of OCTA.
Ophthalmological Assessment
All cases were subjected to detailed ophthalmological examination, including refraction, uncorrected distant visual acuity (UCDVA) and best-corrected distant visual acuity (BCDVA), assessment of the anterior segment slit-lamp, tonometry, with dilated fundus examination with a +78 D lens. UCDVA and BCDVA were measured using LogMAR charts. OCT and OCTA were performed by RTVue (Optovue Inc., Fremont, CA, USA). The OCTA scans a region of 4mm×4mm. Optic nerve head (ONH) evaluation was performed using a 4.5×4.5 mm scan centered on ONH, and VD was measured in four quadrants. OCTA of the macula scan covered 6×6 mm and was centered on the fovea to evaluate vascularity at four levels. The full-thickness vascular density extending from the internal limiting membrane layer (ILM) to the retinal pigment epithelium layer (RPE) and the superficial VD (ILM-inner plexiform layer [IPL]) were measured. The foveal avascular zone in the deep and the superficial plexus was measured using the software system. The foveal VD occupied a 1-mm diameter of the small circle, while the parafoveal VD was the area of the outer circle. These circles were automatically divided into four quadrants: superior, inferior, nasal, and temporal.
Two scans were captured for each eye, and the best scan regarding quality was considered for the analysis. Color coding was presented in density map images in which the hot colors (red and orange) represented a more flow with functional perfusion areas, while cold color (blue) represented low perfusion. Other parameters were collected as central macular thickness (CMT), para and perifoveal sectors thickness, and total macular volume (TMV) using a 5×5 mm scan centered on the fovea [24].
Statistical analysis
We used the statistical package (IBM-SPSS), version 24.0 (IBM Co., Armonk, NY, USA), to analyze data. We represented the data as mean, standard deviation (SD), percentages, and numbers. We used the independent samples t-test for quantitative data to compare the means between the two groups. We used the chi-square test to compare percentages of qualitative variables. We used the Spearman’s Rho test for correlations.
RESULTS
Comparison of demographic, laboratory, and clinical data between patients and controls
This case-control study was conducted on 40 SSc patients (75% females and 25% males), recruited from the rheumatology and rehabilitation department at our university hospital, and 30 healthy subjects (70% females and 30% males). The patients’ mean age was 33.05±6.85, and the mean age of the controls was 35.8±5.46, with no significant differences between the patients group and the control group regarding age, sex, systolic and diastolic blood pressure. 65% of the patients had limited cutaneous systemic sclerosis (LcSSc), and 35% had diffuse cutaneous systemic sclerosis (DcSSc). The mean mRSS of the patients was 28.8±5.99. The most frequently reported affection was cutaneous manifestations, including pitting scars (87.5%), followed by Raynaud’s phenomenon (85%), telangiectasia (67.5%), pigmentation (62.5%), digital ulcers (35%), calcinosis (30%), and gangrene (12.5%). Gastrointestinal tract manifestations were present in 75% of our patients, and musculoskeletal manifestations in 37.5%. Twenty five percent of the patients had interstitial lung disease, and 5% had pulmonary arterial hypertension. Cardiac manifestations were encountered in 15% of the patients. Regarding the laboratory findings, the patients evidenced significantly lower levels of hemoglobin (p=0.045) in comparison to the healthy subjects, while erythrocyte sedimentation rate and C-reactive protein were significantly increased in patients compared with controls (p<0.001), regarding autoantibody positivity; 85%, 45%, and 35% of the patients had positive antinuclear antibody, anti-centromere, and anti-SCL70, respectively (Table 1).
Therapeutic data of the patients
Regarding the therapeutic data of the patients, 52.5% of the patients were receiving methotrexate, 30% mycophenolate mofetil, 17.5% cyclophosphamide, 7.5% rituximab, 7.5% iloprost, 50% calcium channel blockers, 32.5% sildenafil, 72.5% proton pump inhibitors, and 27.5% angiotensin-converting enzyme inhibitors (Supplementary Table 1).
Comparison of the capillaroscopic findings between patients and controls
The capillary density was significantly lower in patients than in controls (p<0.001). Capillaroscopic patterns significantly differed between groups, with 47.5% of the patients showing active (p<0.001), 25% showing early (p=0.003), and 20% showing late stages (p=0.009), while all controls showed normal capillaroscopic patterns (p<0.001). Moreover, the capillary dimensions differed significantly between groups, with 27.5% of the patients having giant capillaries (p=0.002), 52.5% of the patients versus 90% of the controls having normal capillary dimensions (p=0.004). Regarding the capillary shape, 92.5% of the patients had abnormal shape versus 20% of the controls. In addition, more patients had microhemorrhages than controls (p=0.027) (Table 2).
Comparison of the optical coherence tomography angiography findings between patients and controls
Regarding the OCTA findings, the CMT significantly decreased in SSc cases compared to healthy subjects (p=0.001). The superficial full VD, the superior, inferior, and medial superficial peri-foveal VD, and the superior, inferior, and temporal superficial parafoveal VD, in addition to the temporal parafoveal full VD, were significantly lower in cases than controls (p<0.001, p<0.001, p=0.005, p=0.005, p=0.019, p=0.047, p=0.040, and p=0.035, respectively), (Table 3). Figure 1 shows OCTA findings in a SSc patient
Comparison of the clinical and laboratory data between limited and diffuse systemic sclerosis patients
A significant difference was found between cases with LcSSc and DcSSc regarding disease duration (p=0.045). Patients with LcSSc showed significantly higher rates of telangiectasias (p<0.001), skin pigmentation (p=0.010), and digital pitting scars (p=0.003), while patients with DcSSc had increased ILD (p=0.026). Moreover, the positivity of anti-centromere antibodies was higher among LcSSc patients (p=0.028), while the positivity of anti-SCL70 antibodies was higher among DcSSc patients (p=0.007) (Supplementary Table 2).
Comparison of the capillaroscopic and optical coherence tomography angiography findings between limited and diffuse systemic sclerosis
There were no significant differences between LcSSc and DcSSc regarding all the capillaroscopic findings except the capillary density, which was lower in DcSSc than in LcSSc cases (p=0.029). In addition, no statistically significant differences were detected between LcSSc and DcSSc cases as regards OCTA findings.
Correlation of the modified Rodnan skin score with the nail fold capillary density and optical coherence tomography angiography findings in systemic sclerosis patients
The mRSS score was negatively correlated with the nail fold capillary density (r=–0.989, p<0.001), CMT (r=–0.428, p=0.006), superficial full VD (r=–0.692, p<0.001), superior (r=–0.359, p=0.023), temporal (r=–0.522, p=0.001), and medial (r=–0.402, p=0.010) superficial perifoveal VD, superior superficial parafoveal VD (r=–0.387, p=0.014) and temporal perifoveal full VD (r=–0.327, p=0.039) (Table 4).
Correlation of the nail fold capillary density with the optical coherence tomography angiography findings in systemic sclerosis patients
The nail fold capillary density was positively correlated with CMT (r=0.417, p=0.007), the superficial full VD (r=0.664, p<0.001), the superior (r=0.346, p=0.029), temporal (r=0.555, p<0.001), and medial (r=0.424, p=0.006) superficial perifoveal VD, the superior superficial parafoveal VD (r=0.404, p=0.010), and temporal perifoveal full VD (r=0.318, p=0.045) (Table 5). Correlations of the NFC density and mRSS with the central macular thickness and superficial full VD are shown in Supplementary Figure 1.
DISCUSSION
The pathophysiology of SSc suggests that changes in the microcirculation are not limited to the nailfolds but are a generalized phenomenon affecting other anatomical regions, including the ocular microvasculature, particularly the choroid, as a highly vascularized tissue [25]. Ocular manifestations in SSc patients are increasingly recognized, suggesting a possible association between ocular microvascular changes and systemic vascular abnormalities [11]. Previous studies found significant impairment of the choriocapillaris and retinal vascular plexuses and correlated these changes with skin disease and NFC involvement. Thus, early detection of ocular micro-vasculopathy may help identify SSc and discriminate between disease stages [17]. Our study is one of the few studies aimed at detecting the correlation between retinal microvascular changes by OCTA and NFC findings in patients with SSc.
When we compared the NFC results between patients and controls, our patients showed more abnormal NFC patterns, dimensions, shapes, and microhemorrhages than the controls. In addition, our patients had decreased capillary density compared to the controls. Our findings agree with Emrani et al. [26], who found that the main capillaroscopic changes in SSc patients were giant capillaries with capillary dropout and decreased capillary density. Also, Elsayed et al. [27] mentioned that 96% of SSc patients have abnormal capillaroscopic findings. Jakhar et al. [28] found abnormal capillary morphology in 97.8% of SSc patients.
Regarding the OCTA findings in our patients, our patients had significantly decreased CMT, superficial full VD, superior, inferior, and medial superficial perifoveal VD, superior, inferior, and temporal superficial parafoveal VD, and temporal para-foveal full VD compared to the controls. Our findings agree with previous studies, which found that the superficial and deep capillary plexus VD were significantly decreased in SSc patients compared to healthy controls [9,11,29,30]. Mihailovic et al. [19] found reduced VD in the choriocapillaris, ONH, and macula in SSc patients. Güven et al. [8] found that SSc patients had significantly lower foveal and macular thickness than controls. Abd El-Hameed et al. [31] found that SSc patients had low superficial VD and concluded that once SSc has been diagnosed, retinal and choroidal microvascular evaluations using OCTA should be regularly done to identify any sub-clinical lesions. A recent meta-analysis revealed significantly decreased vessel densities in both the superficial and deep capillary plexuses among SSc patients compared to controls [32].
On comparing LcSSc with DcSSc, we found that DcSSc cases had significantly lower capillary density than LcSSc cases. In agreement with our findings, Jakhar et al. [28] found more frequent capillary morphology abnormalities and decreased capillary density in patients with diffuse cutaneous disease. Moreover, according to our findings, DcSSc patients had more ILD than LcSSc. Our findings agree with Perelas et al. [33]. Moreover, patients with LcSSc have higher positivity of anticentromere antibodies, while patients with DcSSc have higher positivity of anti-SCL70 antibodies. Bobeica et al. [34] mentioned that anticentromere antibodies are more specific for limited disease, and anti-SCL70 antibodies are more specific for diffuse disease.
We found a significant negative correlation between the mRSS score and nailfold capillary density. Our findings agree with Santos et al. [35], who found that decreased capillary density was associated with more skin progression in SSc. In contrast, Haque et al. [36] did not find any association between skin thickness score and both NFC variables or NFC patterns. We also found negative correlations between the mRSS and CMT, superficial full VD, superior, temporal, and medial superficial perifoveal VD, superior superficial parafoveal VD, and temporal perifoveal full VD, which was consistent with Mihailovic et al. [19], who reported a significant negative correlation between the mRSS and the VD of the superficial capillary plexus. Ataş et al. [37] found a statistically significant negative correlation between mRSS and choroidal vascularity index. Also, Foti et al. [11] found a negative correlation between mRSS and the superficial VD, but it was insignificant. On the contrary, Romanowska-Prochnicka et al. [38] found a positive correlation between the parafoveal mean superficial VD and the mRSS. Shenavandeh et al. [39] found that the scleroderma type, disease duration, and skin score were not associated with retinal findings. Different disease stages, severity, sample size, and population may explain the variability of findings.
According to our findings, the nail fold capillary density was positively correlated with the CMT, superficial full VD, superior, temporal, and medial superficial perifoveal VD, superior superficial parfoveal VD, and temporal perifoveal full VD. Our findings agree with Jakhar et al. [28], who mentioned that patients with SSc usually have retinal microvascular abnormalities consistent with the characteristic vascular changes of SSc, and the severity of these retinal lesions correlates with changes in NFC. Thus, NFC could be a valuable marker for retinal affection in those patients. Mihailovic et al. [19] and Hysa et al. [40] found positive correlations between the capillary density by NFC and the VD of the choriocapillaris. Gambini et al. [29] found that the mean capillary density at capillaroscopy was positively correlated with the VD at the level of the superficial and deep capillary plexus. Hysa et al. [40] and Paczwa et al. [41] found that peripheral microvascular damage was correlated with impaired ocular microcirculation in SSc patients. They suggested that the ocular microvascular alterations mirror the peripheral microvascular involvement, even without overt clinical symptoms, and highlighted the role of OCTA in assessing microvascular changes in SSc. However, Shenavandeh et al. [39] found no correlation between the NFC abnormalities and retinal vascular changes except hemorrhages and retinal tortuosity, and they suggested that the mechanism or the type of changes in the nailfold vasculature and the retinal vessels may be different.
The negative correlations between the mRSS and the OCTA findings and the positive correlations between the nail fold capillary density and OCTA findings highlight the importance of both mRSS and NFC in predicting microvascular changes in the eye. However, the mRSS usually reflects the degree of skin thickening resulting from fibrosis rather than the microvasculopathy. In comparison, NFC detects peripheral microvascular abnormalities that mimic the retinal and choroidal microvascular changes. Thus, the NFC may be more accurate and allow for the early prediction of retinal microvascular abnormalities. Our findings are supported by previous studies reporting that retinal and choroidal vessel abnormalities detected by OCTA mirror the peripheral microvascular changes detected by NFC in SSc patients [28,40]. Thus, any SSc patient with severe NFC findings or decreased nailfold capillary density should be referred for OCTA examination, even without ocular manifestations; this may help predict retinal microvascular affection and allow early therapeutic interventions.
The strength of this study is that it involves a group of SSc patients without ocular manifestations, which allows for the early prediction of ocular microvascular changes. We assessed the peripheral microvascular changes by NFC and the ocular microcirculation by OCTA. Both are non-invasive, reliable diagnostic techniques. We correlated the nailfold capillary density with the OCTA findings to shed light on the relation between the peripheral and ocular microvascular changes in SSc patients.
Our study has some limitations; first, it is a small sample size, which may affect the data generability. Nevertheless, considering the rarity of the disease, the number of SSc patients in this study is acceptable and accounts for a homogeneous study population. Second, it is a cross-sectional study; therefore, we cannot comment on the value of NFC and OCTA for the follow-up of disease progression. Third, performing the mRSS by a single examiner carries the disadvantage of being an individual assessor dependent. However, performing it by a single trained, experienced rheumatologist decreases the inter-individual variability. Daungkum et al. [42] considered performing the mRSS by a single examiner as a point of strength, reducing the inter-individual variability in their study. Khanna et al. [43] recommended that the same assessor examine the patient during their trial.
CONCLUSION
The nailfold capillary density measured by NFC positively correlates with the retinal VD measured by OCTA, suggesting that NFC could be a valuable marker for retinal vessel involvement in SSc patients. In addition, our results highlight the importance of combining OCTA with NFC for diagnosing and monitoring microvascular changes in SSc patients.
SUPPLEMENTARY DATA
Supplementary data can be found with this article online at https://doi.org/10.4078/jrd.2024.0124
XML Download