Treatment outcomes in patients of livedoid vasculopathy following antithrombotic therapy: A retrospective observational study

Introduction

Livedoid vasculopathy is a rare cutaneous disorder of hypercoagulability.¹ The primary pathology in livedoid vasculopathy is the obstruction of blood circulation in small skin vessels. Initially, it was mistakenly considered a type of vasculitis. Its vaso-occlusive nature is well-established and evidenced by a typical clinical presentation, association with thrombophilia, the presence of increased levels of fibrinopeptide A, and response to fibrinolytic and antiplatelet therapies.² The exact etiopathogenesis is unclear; however, the thrombotic pathway plays a major role. It is known to be associated with various inherited and acquired disorders of thrombophilia, such as gene mutations involving factor V (Leiden), methylenetetrahydrofolate reductase (MTHFR) or prothrombin (G20210A); antithrombin and proteins C and/or S deficiency, antiphospholipid antibody (APLA) syndrome, hyperhomocysteinemia, cryoglobulinemia, cryofibrinogenemia, and increased levels of lipoprotein (a) and plasminogen activator inhibitor-1 (PAI-1). In addition, its association with venous insufficiency and connective tissue disorders, such as systemic lupus erythematosus, scleroderma, rheumatoid arthritis, and mixed connective tissue disorder, is well documented.²

Livedoid vasculopathy manifests clinically as chronic, recurrent, painful ulcers in the lower legs, ankles, and feet. These ulcers heal with atrophic, porcelain white scars known as atrophie blanche.¹ The associated pain and the chronic and recurrent nature of livedoid vasculopathy result in significant morbidity and greatly hamper the quality of life.¹ It can be mistaken for other causes of leg ulcers, and a biopsy is often necessary to reach a diagnosis. The hallmark histopathological features are the presence of intraluminal thrombosis, endothelial proliferation, and intimal thickening without inflammatory infiltration or leucocytoclasia.^2-4

Livedoid vasculopathy management is difficult due to the chronic and recurrent nature of the disease. The most effective treatment options are anticoagulants such as warfarin, heparin, low-molecular-weight heparin (LMWH), rivaroxaban, and dabigatran; and antiplatelet drugs such as aspirin, pentoxifylline, and dipyramidole. Intravenous immunoglobulin, through its anticoagulation properties, is useful in refractory cases.^5-7The data pertaining to the efficacy of various available treatment options are limited, especially in India. In addition, the clinicodemographic characteristics and associated laboratory abnormalities have not been studied adequately in Indian patients. Hence, this retrospective study attempted to assess these aspects as well.

Methods

This retrospective observational study was conducted in a tertiary care centre in India. It collected data from January 2019 to December 2022 after approval from the institute’s ethics committee (T/IM-NF/Dermat/23/91). Patients who fulfilled the following for livedoid vasculopathy were included for analysis: clinical criteria for (recurrent painful ulcers on both lower limbs with a round, oval, angulated, or stellate morphology and the absence of any other cause of leg ulcers); pathological criteria (thrombotic occlusion of cutaneous small vessels without any evidence of vasculitis); those who had received systemic anticoagulants and were followed up for a minimum of 6 months. Patients with missing clinical details in the records, incomplete or irregular follow-up, those simultaneously on other drugs that might affect the disease, or analysis of therapeutic outcomes (such as long-term NSAIDs/analgesics and immunosuppressive drugs), those who were already on antithrombotic or antiplatelet drugs prior to presenting to the institution, and those with associated connective tissue disorders, were excluded from the analysis. The following clinicodemographic details were by P.G. from the clinical notes and pathology forms: age, sex, age of onset, disease duration, history of chronic venous insufficiency (CVI), family history of livedoid vasculopathy, personal or family history of thrombophilia or antiphospholipid antibody syndrome, and previous treatment. The clinical details were retrieved from the departmental archives. Clinical notes were screened for the following features: location of the ulcers, the shape of the ulcers, the skin surrounding the ulcers, the presence of purpuric macules and papules, nodules, livedo racemosa, and atrophie blanche. The associated haematological, serological and radiological parameters such as complete hemogram, liver and renal function tests, bleeding time, clotting time, prothrombin time (PT), international standardised ratio (INR), activated partial thromboplastin time (aPTT), D-dimer, fibrinogen, protein C, protein S, lupus anticoagulant, anticardiolipin antibody, anti–β2-glycoprotein 1 antibody, homocysteine, mutations in factor V Leiden, G20210A prothrombin, antithrombin III, and MTHFR, serologies for hepatitis B, hepatitis C and HIV, and colour Doppler abnormalities were extracted from the clinical notes and entered in a predesigned proforma. The primary objective of the study was to evaluate the treatment outcomes in patients with livedoid vasculopathy at 3 months following antithrombotic therapy. The secondary objectives were to assess the following: treatment outcome at 6 months, side effects of treatment, improvement in the dermatological life quality index (DLQI), clinicodemographic features, and laboratory abnormalities associated with livedoid vasculopathy. The primary endpoint was the proportion of patients who achieved a pain visual analogue (VAS) score of zero and complete healing of livedoid vasculopathy ulcers, defined as healing of the existing ulcer(s) without development of new ulcers at 3 months after initiation of antithrombotic therapy. The secondary endpoint was the proportion of patients who achieved a pain VAS score of zero and complete healing of ulcers at 6 months.

Results

A total of 60 livedoid vasculopathy patients were screened for analysis, of whom 26 patients who satisfied the inclusion criteria were recruited. The rest of the patients were excluded because of incomplete data (17), incomplete/irregular follow-up (10), or being on other concomitant medications such as NSAIDs and aspirin (7). Of the 26 included patients, 20 (76.9%) were males [Figure 1]. The mean age of the cohort was 43.5 ± 12.9 years. The age of onset of the disease was 38.13 ± 12.57 years, and the median duration of illness was 51 months at the time of presentation.

Close

Figure 1:

The flowchart shows the process of patient recruitment for the study.

Export to PPT

Clinicodemographic features

A history of venous, including varicose vein, and lower leg pigmentation, was present in 6/26 (23.1%) patients. None of the patients had connective tissue disease or a personal or family history of thrombophilia prior to the onset of the disease. The detailed clinical characteristics have been tabulated in Table 1, and Figures 2a and 2b show the characteristic pathological features of LV.

Table 1: Baseline clinicodemographic and laboratory features and treatment outcomes of all the patients

Characteristics	Number
Mean age (in years)	43.5 ± 12.9
Sex
Male	20 (76.9%)
Female	6 (23.0%)
Median duration of illness (months)	51
Sites of involvement
Bilateral involvement	13 (50.0%)
Feet	26 (100%)
Ankle	26 (100%)
Lower legs	20 (76.9%)
Clinical morphology
I. Atrophie blanche	25 (96.1%)
II. Painful ulcers	26 (100%)
Reticulate	21 (80.7%)
Regular and circular	5 (19.2%)
Both	2 (7.6%)
III. Ankle edema	19 (73.0%)
IV. Purpuric macules	9 (34.6%)
V. Nodular lesions*	2 (7.6%)
Symptoms
Pain	26 (100%)
Investigations
• Hyperhomocysteinemia	19 (73.0%)
• MTHFR mutation	12 (46.2%)
• CVI	12 (46.2%)
• Fibrinogen mutation	4 (15.4%)
• Factor VIII mutation	2 (7.6%)
At baseline
Aspirin	8 (30.7%)
Rivaroxaban	18 (69.2%)
Treatment at 3 months:
Aspirin	7 (26.9%)
Rivaroxaban	11 (42.3%)
Aspirin and rivaroxaban	8 (30.7%)
Adverse effects	0 (0.0)

* Both patients had painful nodules with a maximum size of 1cm. CVI: Chronic venous insufficiency, MTHFR: Methylene tetrahydrofolate reductase, VAS: Visual analog score. Age was normally distributed (Shapiro-Wilk test). The mean (SD) was calculated using SPSS accordingly.

Close

Figure 2a:

Histopathology shows mild acanthotic epidermis and occlusion of the papillary dermal (blue arrow) without any significant perivascular inflammatory infiltrate (Haematoxylin & eosin, 100x).

Export to PPT

Close

Figure 2b:

Thrombus completely occluding the vessel (blue arrow) without any perivascular infiltration (Haematoxylin & eosin, 400x).

Export to PPT

The number of ulcers varied from 1 to 10, and the shape of the ulcers was irregular and reticulated in 21 (80.7%) and regular and circular in five (19.2%) patients. None of the patients had surrounding livedo around the ulcers. The arrangement of ulcers was non-linear in the majority (23, 88.5%) of the patients and was linear in three (11.5%) patients. Before presenting to our institution, the patients were diagnosed as follows: venous ulcers, polyarteritis nodosa, pyoderma gangrenosum, arterial ulcers, and leg ulcers, and had received systemic steroids, azathioprine, methotrexate, and cyclosporine.

Livedoid vasculopathy significantly affected the quality of life, and the mean DLQI at the time of initiation of therapy was 11.96±2.27 in our patients. The aspects of DLQI that were most affected were symptoms and feelings, followed by daily activities, work/school, treatment, personal relationships, and leisure activities. The median baseline pain VAS score was 9.

Treatment outcome

Table 2 lists the therapeutic outcomes of all the patients. Out of 26 patients, 18 (69.2%) were initially treated with rivaroxaban (10mg) once daily, while eight (30.8%) were started on aspirin (150mg) alone, once daily. In addition, vitamin B12, folic acid, and pyridoxine supplementation were provided to 12 patients with MTHFR mutation and hyperhomocysteinemia.

Table 2: Treatment outcomes of all the patients with livedoid vasculopathy

Treatment outcome	Definition			N (%)
Primary endpoint	Pain VAS score of zero and complete healing of livedoid vasculopathy ulcers at three months			17 (65.3)
Secondary endpoint	Pain VAS score of zero and complete healing of livedoid vasculopathy ulcers at six months			25 (96.1)
Change in scores
Scores	Baseline	At three months	At six months	Change in score
Pain VAS	9 (IQR: 8–10)	0 (IQR: 0–3)	0 (IQR: 0–0)	P<0.001
DLQI	11.96 ± 2.27	5 (IQR:2-8)	3 (IQR:2-5)	P<0.001

DLQI: Dermatological life quality index, IQR: Interquartile range, VAS: Visual analog score. Significance threshold is 0.05

Besides pharmacotherapy, the various measures taken in our patients for ulcer management were daily cleansing, application of dressings, including absorbent foam dressing, and oral antibiotics (oral amoxicillin-clavulanic acid in five patients) when necessary. Swab cultures were performed using swab samples when clinically indicated. Staphylococcus aureus was isolated in two cases, both of which were sensitive to amoxicillin-clavulanate. Patients with chronic venous insufficiency were advised to practise leg elevation, perform regular tip-toe exercises, use compression stockings consistently, and avoid prolonged standing.

At 3 months, 17/26 (65.3%) of patients had complete ulcer healing and achieved a VAS pain score of zero, without developing any new ulcers. Of these, 11 (42.3%) patients had received rivaroxaban [Figures 3a-d and 4a-b] and 6 (23%) had received aspirin [Figures 4c-d].

Close

Figure 3a:

Ulcer on the dorsum of the foot before the therapy.

Export to PPT

Close

Figure 3b:

Complete healing of the ulcer with rivaroxaban at 3 months.

Export to PPT

Close

Figure 3c:

Ulcer around the lateral malleolus before therapy.

Export to PPT

Close

Figure 3d:

Complete healing of the ulcer with rivaroxaban at 3 months.

Export to PPT

Close

Figure 4a:

Ulcer around the lateral malleolus before therapy.

Export to PPT

Close

Figure 4b:

Complete healing of the ulcer with rivaroxaban at 3 months.

Export to PPT

Close

Figure 4c:

Ulcer on the medial aspect of the ankle.

Export to PPT

Close

Figure 4d:

Complete healing of the ulcer at 3 months following aspirin monotherapy.

Export to PPT

At 3 months, patients with incomplete response, with either the development of new ulcers, incomplete ulcer healing, or inadequate pain relief, received in addition 150 mg of aspirin (in those who were already receiving rivaroxaban or rivaroxaban (10mg) once daily (in those who were already receiving aspirin. At 6 months, 96.1% (25) of patients had complete ulcer healing and achieved a VAS pain score of 0 without the development of new ulcers.

Overall, 11 (42.3%) patients received rivaroxaban alone, 8 (30.7%) received a combination of rivaroxaban and aspirin [Figures 5a-c], and 6 had taken aspirin alone. Only one (3.8%) patient who received aspirin and apparently achieved complete remission at 3 months did not have complete healing and pain relief at 6 months. The patient developed on and off lesion after fourth months onwards. The reductions of DLQI at 3 and 6 months were significant. The changes in DLQI and VAS scores from baseline at 3 (P<0.001) and 6 months (P<0.001) were significant.

Close

Figure 5a:

Crusted ulcer before the start of therapy. This patient was misdiagnosed with polyarteritis nodosa and received oral prednisolone, azathioprine, and methotrexate without response.

Export to PPT

Close

Figure 5b:

Partial improvement of the ulcer at 3 months with rivaroxaban monotherapy.

Export to PPT

Close

Figure 5c:

Complete healing of the ulcer with the combination of rivaroxaban and aspirin at 6 months.

Export to PPT

None of the patients reported any side effects, and the PT/INR tests done at months one, three and six were within normal limits during the follow-up. Data on various laboratory abnormalities in the patients who received the rivaroxaban or aspirin and their response at 3 months are provided [Supplementary Table S1]. The INR reports of all our patients at baseline, 3 months, and 6 months have been tabulated in Supplementary Table S2.

Supplementary Table S1

Download

Supplementary Table S2

Download

Discussion

This study presents the data regarding the therapeutic outcomes and clinicodemographic and laboratory data of livedoid vasculopathy patients in a tertiary care centre in India. These patients were treated with rivaroxaban, aspirin, or a combination of both.

Livedoid vasculopathy is a rare disorder characterised by lower leg ulcers, livedo racemosa, and atrophie blanche. Our study showed a male preponderance; in contrast, prior studies reported it as being common among females.^8,9 This is probably due to the small sample size, which may not represent the true prevalence in the population. In addition, the exclusion of 34 cases due to ineligibility could be another factor. The affected patients had livedoid vasculopathy for a median duration of more than 4 years (51 months), signifying the prolonged disabling effect of the disease and possibly its social and economic impact, besides the disease-associated morbidity itself.

The most common clinical findings in our cohort were painful leg ulcers and pedal oedema, besides atrophie blanche. Purpuric macules and painful nodules were rare. This may be due to the late presentation of the patients to our tertiary care centre. In addition, none of our patients had livedo racemosa. Prior studies reported livedo racemosa to be a common feature.^4,9 In a systematic review, it was seen in 85% of the patients.⁸ The absence of livedo racemosa in our patients can be explained by the late presentation and the darker skin phototype, which can hinder the appreciation of a faint network of erythema. In concordance with prior studies, there was no difference in the involvement of the right or left limb.

Treatment of livedoid vasculopathy has the following three targets: relieving pain, healing existing ulcers, and preventing the development of new ulcers. Two-thirds of our patients achieved pain control and ulcer healing using monotherapy, either rivaroxaban (10 mg/day)(11/17) or aspirin (150mg/day) (6/17), at 3 months. These patients did not report any breakthrough lesions. Previous studies have shown the response to monotherapy with rivaroxaban or aspirin.^9-13 Rivaroxaban, a direct Factor Xa inhibitor, is a commonly used anticoagulant to treat livedoid vasculopathy in a dose of 10-20mg/day. A systematic review reported it to be the most commonly (54%) used anticoagulant in patients with livedoid vasculopathy and was preferred over low-molecular-weight heparin (LMWH), heparin, warfarin sodium, and fluindione.¹⁴ It provides more consistent and predictable anticoagulation than oral warfarin. It prevents the formation of intravascular thrombi and relieves ischemia of the epidermis and dermis, hence resulting in pain reduction. Rivaroxaban was reported to be effective in complete (31%) and partial (68%) ulcer healing in 31% and 68% of patients, respectively.¹⁵ However, at 3 months, 7/18 of the patients who were on rivaroxaban and 2/8 patients who were on aspirin had inadequate pain control, ulcer healing, and/or were developing new ulcers and received combination therapy. Due to the small number of cases, we could not do a subgroup analysis to characterise the patients who responded to monotherapy at 3 months versus those who did not. At 6 months, all but one patient had achieved pain control and ulcer healing without new ulcer occurrence. The patient who did not achieve pain control or ulcer healing was on aspirin (150mg) once daily only, despite being under control at 3 months. A combination of rivaroxaban and aspirin was needed in eight patients to achieve pain control and ulcer healing. Due to the small sample size, we could not compare the characteristics of the patients who needed combination therapy versus those who did not. Prior reports suggest that monotherapy may not be effective for all patients, and greater efficacy is noted with combination treatment.^4,8,14,15 Aspirin has been used in combination with low-dose warfarin to control APLA-associated livedoid vasculopathy.¹⁶ A similar therapeutic efficacy has been noted in combining aspirin and rivaroxaban in patients with cardiovascular diseases.¹⁷ The superior efficacy of combination therapy suggests that drugs with different mechanisms of action may be better at controlling the livedoid vasculopathy episode and preventing recurrence.

Pain was a significant problem for all patients, with a median baseline VAS score of 9; this is in line with a previous study.⁹ Either with monotherapy or combination therapy, there was significant improvement in pain and DLQI over time. None of our patients had received any additional painkillers to achieve pain control. A prior study assessing the impairment in quality of life by DLQI and the Short Form 36 questionnaire showed greater impairment in symptoms and feelings, activities of daily living, and work/school. The mean DLQI score was 11.37 in that study, which increased to 17.83 when the disease was active. In addition, the impact on QOL was greater in patients with active disease than inactive disease.¹

Our patients did not show features of an abnormal coagulation profile (PT/INR) during the treatment period, and there were no adverse effects from monotherapy or combination therapy. Previous studies have reported hemarthrosis, nose bleeding, gingival bleeding, bleeding during dental procedures, bleeding from minor wounds, and dysmenorrhea and dyspnoea with rivaroxaban in patients withlivedoid vasculopathy.^9,14,18 These were not seen in our patients, perhaps due to the lower dose of rivaroxaban (10mg) received by them. The target INR for anticoagulation in livedoid vasculopathy is relatively lower than for APLA syndrome (where the target INR is 2-3), which suggests the former can be controlled without any risk of serious bleeding.¹⁹ Besides, the dose of rivaroxaban to prevent thromboembolism in APLA syndrome is higher, i.e., 20mg, leading to increased chances of side effects. All our patients were monitored for PT/INR during the therapy, and these were within the normal range, demonstrating that ulcer healing and prevention of recurrence in livedoid vasculopathy can be achieved without increased risk of bleeding, as reported previously. A prior case report describes three patients of livedoid vasculopathy who were able to achieve remission with a target INR of 1-1.5 using low-dose warfarin of 1-2mg.²⁰ Rivaroxaban exhibits a safer profile and does not need regular monitoring, unlike vitamin K antagonists such as warfarin, heparin, and LMWH. Furthermore, rivaroxaban has a significantly lower incidence of cerebral bleeding even at a dose of 20mg, as compared to warfarin.²¹

One of the important steps in the management of livedoid vasculopathy is to identify the underlying etiology, associations, or risk factors. In our cohort, hyperhomocysteinemia was detected in 73.1% (19) of patients, which is consistent with previously reported rates ranging from 62% to 83%.^13,22,23 Additionally, a heterozygous mutation in the MTHFR gene was identified in 46.2% (12) of our patients, which is somewhat lower than the 75% prevalence reported in earlier literature.^22,24 Venous stasis in livedoid vasculopathy has been reported in up to 38% of patients, which is comparable to our findings of 46%.^24-26 Although the role of hyperhomocysteinemia or chronic venous insufficiency in etiopathogenesis cannot be ascertained, hyperhomocysteinemia could be an important factor leading to hypercoagulability in the vessels of the lower limb. In addition, patients with hyperhomocysteinemia need concomitant vitamin B12, folic acid, and pyridoxine supplementation to restore the methyl tetrahydrofolate cycle. The effect of venous stasis in the pathogenesis of livedoid vasculopathy also cannot be ignored, as evidenced by the fact that it is seldom seen on the upper limbs. Livedoid vasculopathy is still misdiagnosed, especially by the primary physician, possibly due to a lack of awareness about the disease and/or a lack of diagnostic resources. It is frequently mistaken for other causes of leg ulcers, such as venous ulcers, polyarteritis nodosa, atherosclerosis, and others, as evidenced by our study. Many of these patients had received prior systemic steroids for the ulcers, which can be detrimental and should not be used except for patients with associated connective tissue disorders such as systemic lupus erythematosus, rheumatoid arthritis, Sjögren syndrome, and mixed connective tissue disorder.

Limitations

Limitations of this study include its retrospective nature, small sample size, and the relatively short duration of follow-up. Data regarding the time to complete ulcer healing and the time taken to achieve a VAS score of 0 were not available. We were not able toanalyse factors associated with poor response to therapy in the monotherapy groups, such as age, sex, body mass index, duration of the disease, and associated laboratory abnormalities. In our cohort, a few patients received oral antibiotics and absorbent dressings, which may have had a short-term effect on ulcer healing; however, this could not be addressed in our analysis.

Conclusion

Livedoid vasculopathy poses a diagnostic and therapeutic problem with frequent misdiagnoses and wrong treatment. This is the first study from India to evaluate treatment outcomes in livedoid vasculopathy using rivaroxaban, aspirin, or both. Monotherapy with rivaroxaban or aspirin can effectively manage livedoid vasculopathy, successfully achieve pain control, and improve the quality of life. However, a fraction of patients may need a combination of the two to achieve therapeutic goals. Both monotherapy and combination therapy are safe and not associated with significant side effects.