Chronic venous disease,platelet and haemostatic abnormalities contribute to the pathogenesis of pigmented purpuric dermatoses

Abstract

Background

To investigate the aetiology of pigmented purpuric dermatoses (PPD).

Methods

63 patients with a provisional diagnosis of PPD were assessed. Skin biopsies were performed to confirm the clinical diagnosis. Haemostasis was assessed using platelet function analyser-100 (PFA-100), light transmission aggregometry (LTA), impedance aggregometry (Multiplate) and measurement of clotting times and clotting factors. Chronic venous disease (CVD) was assessed by duplex ultrasound. When not contraindicated, patients were advised to discontinue haemostatic-modifying drugs or supplements for 4 weeks after which the laboratory measurements were repeated and the clinical resolution of PPD was assessed. Subsequently, a cohort of patients identified with CVD underwent endovenous interventions and further resolution of PPD was assessed.

Results

CVD was found in 48 patients (76.2%) while haemostatic abnormalities were found in 36 (57.1%). 30 patients (47.6%) had concurrent CVD and haemostatic abnormalities. Modifiable risk factors such as the intake of platelet inhibitors or other drugs and supplements such as fish oil were identified in 53 patients (84.1%). These could be ceased in 35 patients of whom 28 (80.0%) achieved either complete or partial resolution of PPD. Treatment of the underlying CVD was performed in 18 patients resulting in complete or partial resolution in 17 (94.4%). In seven patients (11.1%), no CVD or haemostatic abnormalities were identified, and the risk factors included dietary factors such as excessive caffeine or soft drink consumption.

Conclusion

Haemostatic abnormalities and CVD contribute to the pathogenesis of PPD. Resolution of PPD in the vast majority of patients may be achieved by cessation of modifiable risk factors and in particular platelet-modifying drugs or supplements and treatment of the underlying venous disease.

Keywords

Pigmented purpuric dermatoses capillaritis haemostatic abnormalities platelet function coagulation factors omega 3 fish oil

Introduction

Pigmented purpuric dermatoses (PPD) are a class of skin eruptions that manifest as non-blanching cayenne pepper speckled macules, patches or plaques. PPD has historically been referred to as ‘capillaritis’ although given the absence of a true vasculitis, this is a misnomer. Eight clinical variants of PPD have been described as distinct conditions such as Schamberg’s disease, Majocchi’s purpura annularis et telangiectodes and lichen aureus (Table 1). Over time, it became increasingly evident that these conditions were not distinct and in fact shared similar histopathological features, allowing them to be classed together as PPD.

Table 1.

Eight described variants of pigmented purpuric dermatoses (PPD).

1. Schamberg’s disease – progressive pigmented purpuric dermatosis

2. Majocchi’s disease – purpura annularis telangiectodes

3. Itching purpura – disseminated pruriginous angiodermatitis

4. Lichen aureus

5. Linear PPD

6. Pigmented purpuric lichenoid dermatosis (of Gougerot & Blum)

7. Eczematoid-like purpura (of Doucas and Kapetanakis)

8. Granulomatous PPD (of Kossard)

The common histological findings include extravasation of erythrocytes, marked haemosiderin deposition and a perivascular lymphocytic infiltrate. Purpura results from extravasation of erythrocytes, and the cayenne-pepper colouring of chronic lesions is due to haemosiderin-laden macrophages.¹ Given the presence of the lymphocytic infiltrate, PPD is distinct from simple petechial eruptions and is distinct from vasculitis based on the absence of vessel wall damage and other features of a true vasculitis.

The aetiology of PPD is unknown. Several risk factors have been implicated in the pathogenesis of this condition including venous hypertension,¹ capillary fragility,² and gravitational dependency.¹ Other reported triggers have included contact allergy to clothing dyes,³ food and food additives,⁴ infections,⁵ medications,^1,5 and chronic liver disease.⁶ Cell-mediated immunity and immune complexes have been implicated in the pathogenesis of PPD.^2,7–11

Although venous hypertension has been previously associated with PPD, not all patients with venous disease present with a purpuric dermatosis. In this study, we postulated that chronic venous disease (CVD) and haemostatic abnormalities contribute to the pathogenesis of this group of conditions. The aim of this study was to assess the incidence of haemostatic abnormalities and CVD in a cohort of patients with PPD. We also investigated the effect of cessation of modifiable or treatable risk factors on the clinical resolution of this condition.

Methods

Ethical approval

Ethical approval was obtained from St Vincent’s Human Research Ethics Committee (Approval 15/064).

Materials

Sodium citrate blood collection tubes (0.109 M, 3.2%, Becton Dickinson, New Jersey, USA). Hirudin blood collection tubes (Dynabyte Medical, Munich, Germany). Platelet function analyser (PFA100, Siemens Marburg, Germany), Multiplate^® Impedance Aggregometer and agonists (Dynabyte Medical, Munich, Germany), Light Transmission Aggregometer (LTA) (AggRAM, Helena Laboratories, Texas USA), and Haematology Analyser (Coulter LH750 or 780, Beckman Coulter, California, USA). Coagulation Assays were performed using a STAR Evolution Expert Analyser (Diagnostica Stago, Paris, France), with the Activated Partial Thromboplastin Time (APTT) reagent (Actin-FS, Siemens Marburg, Germany) and Prothrombin Time (PT) reagent (Neoplastin, Diagnostica Stago).

Duplex ultrasound studies were carried out using a TOSHIBA Aplio ultrasound system (TOSHIBA, Japan) and Continuous Wave (CW) Doppler (Hadeco, Smartdop, Japan).

Endovenous laser systems used included 810 nm (Diomed, UK), 980 and 1470 nm, (Biolitec, Germany) and 1500 nm (Intermedic, Spain). Class II (23–32 mmHg) graduated compression stockings (GCS; SIGVARIS, St Gallen, Switzerland); sodium tetradecyl sulphate (STS; FIBRO-VEIN, STD Pharmaceuticals Ltd, Hereford, UK); polidocanol (POL; Aethoxysklerol^®, Chemische Fabrik Kreussler & Co, Wiesbaden, Germany).

Patient inclusion

All patients were referred to the chief investigator by referring dermatologists or other vascular specialists.

Diagnosis of pigmented purpuric dermatoses

Pigmented purpuric dermatoses were diagnosed clinically as determined by at least two dermatologists (referring dermatologist and chief investigator) and confirmed by a skin biopsy when a differential diagnosis such as vasculitis or mycosis fungoides (MF) had to be excluded.

Initial assessment

All patients were consulted by the chief investigator. A complete medical history was obtained that included history of bleeding, bruising, menorrhagia, haematological disorders, venous hypertension, venous thromboembolism, surgical, gynaecological, psychiatric and family history. A complete medication history was obtained and drugs and supplements that would potentially interfere with haemostatic function were identified. These included aspirin,¹² non-steroidal anti-inflammatory drugs (NSAIDs),¹³ anticoagulants and supplements such as omega-3 polyunsaturated fatty acids (fish oil, omega-3 PUFA)^14,15 and vitamin E.¹⁶ Social history included a detailed history of nutrition, alcohol consumption, smoking and occupation.

Physical examination to determine the sub-type and the anatomical distribution of PPD was performed. Other physical signs of haemostatic disorders and concurrent venous disease were looked for and documented. A CW Doppler examination was performed to screen for venous reflux in the vicinity of the lesions. Photographs of the affected areas and limbs were taken. Patients were classified depending on the severity of the presenting venous disease using the CEAP (Clinical, Etiological, Anatomical and Pathological) classification.¹⁷

One to three punch biopsies (4 mm) of the affected skin were collected in formalin and processed following standard protocols by a routine pathology laboratory. Slides were stained with haematoxylin and eosin (H&E) and Pearl’s Prussian blue for evidence of haemosiderin deposition.

Assessment of platelet function and aggregation

Blood samples were collected using standard techniques for platelet count, function and aggregation. All samples were taken at the same time (approximately 8 a.m.) from non-fasting patients. Sodium citrate tubes were used to collect samples for platelet analyses, and Multiplate^® impedance aggregometry samples were collected in hirudin blood collection tubes.

Platelet counting was performed using a Coulter haematology analyser. Screening for abnormal platelet function was assessed by PFA-100, including testing with collagen/epinephrine (reference range [RR]: 85–165 s) and/or collagen/adenosine diphosphate (ADP; RR: 71–118 s).¹⁸ Platelet aggregation studies were performed by LTA.¹⁹ Agonists used included Adrenalin (300 μM), ADP (2.5–10 μM), Collagen (5 μg/mL), Ristocetin (0.5 and 1.2 mg/mL) and Arachidonic Acid (0.5 mg/mL).

Further testing using Multiplate^® impedance aggregometry was performed on 35 patients. Agonists used included thrombin receptor activating peptide (TRAPtest, 32 μM), collagen (COLtest, 3.2 μg/mL), arachidonic acid (AA, ASPItest, 0.5 mM), ADP (ADPtest, 6.4 μM) and ADP and prostaglandin E1 (ADP-HS; 6.4 μM ADP +0.3 nM PGE1). Aggregation was assessed over 6 min. Results were interpreted as the area under the aggregation response curve.

Coagulation testing

Conventional clotting tests performed included PT (RR: 11–15 s) and APTT (RR: 25–35 s). Clotting factor assays included fibrinogen (RR: 2–4 g/L), factor V (RR: 70–120%), factor VII (RR: 70–120%), factor VIII (RR: 70–150%), factor IX (RR: 60–150%) and von Willebrand Factor (vWF).

Other blood tests

Full blood count (FBC) was performed to detect thrombocytopenia. Renal function (EUC; electrolytes, urea and creatinine) was assessed given the association of chronic kidney disease and platelet function abnormalities. Liver function tests were performed given the association of liver disease and coagulation factor deficiencies. Investigations for paraproteins included electrophoresis (EPG) and immune electrophoresis (IEPG) to screen for multiple myeloma and its associated bleeding diathesis.

Duplex ultrasound

Vascular sonographers performed venous incompetence duplex mapping of the limbs affected by PPD, scanning from the groin to ankle in longitudinal plane using B-mode to localise and map the superficial and deep venous systems and the associated perforating veins. Colour and spectral Doppler analysis was performed to assess for reflux defined as retrograde flow lasting greater than 0.5 s for superficial veins and greater than 1 s for deep veins. An incompetence map of the venous system was developed for each patient to define the pathway of reflux.

Cessation of haemostatic modifying drugs and supplements

Patients were advised to stop medications, supplements and food or beverages suspected to interfere with the haemostatic system for 4 weeks unless medically contraindicated.

Post-cessation assessment

Patients were reviewed at 4 weeks, and site and extent of PPD were re-assessed. Serial photographs were taken. Blood tests were repeated for those whose initial investigations revealed an abnormality.

Intervention for venous disease

Patients found to have clinically relevant venous incompetence were treated with interventional techniques when medically indicated. Clinical relevance was defined as refluxing venous pathway leading to sites affected by PPD. Therefore, refluxing veins immediately below the sites of PPD were determined as relevant.

The venous pathway of reflux was treated with interventional techniques. Endovenous laser ablation (EVLA) was introduced in 2005 and was used for the trunks of saphenous veins thereafter. A range of laser systems evolving over the years were deployed including 810, 980, 1470 and 1500 nm infrared systems. Ultrasound guided sclerotherapy (UGS) was used for both trunks and tributaries prior to 2005 and exclusively for tributaries thereafter. Trunks and larger subcutaneous vessels were treated with 1.5–3% STS foam. Sub-dermal reticular veins and telangiectasias were treated with 0.5% POL foam. All foams were generated using a modified Tessari technique (1 + 4 liquid plus air fraction, LAF).²⁰ All patients were advised to wear full length GCS (23–32 mmHg) day and night for at least a week after venous interventions.

Results

Demographics

Sixty three patients with a clinical diagnosis of PPD were included in the study (Table 2). 38 patients (60.3%) were female and 25 (39.6%) were male. The median age was 46 years (range 4–78 years).

Table 2.

Patient demographics and lesion characteristics. Demographics recorded including age, gender, variants, duration and distribution of pigmented purpuric dermatoses (PPD).

Demographic	Number of patients, N (%)
Total number	63
Median age [range]	46 [4–78] years
Gender
Male	25 (39.7)
Female	38 (60.3)
Variants
Schamberg’s disease	26 (41.3)*
Lichen aureus	25 (39.7)*
Linear (unilateral) PPD	3 (4.8)
Eczematoid-like Doucas and Kapetanakis	3 (4.8)
Majocchi’s disease	2 (3.2)
Itching purpura of Lowenthal	2 (3.2)
Gougerot and Blum	1 (1.6)
Other (unspecified)	3 (4.8)
Duration
<12 months	21 (33.3)
1–5 years	31 (49.2)
6–10 years	8 (12.7)
>10 years	3 (4.7)
Median duration	2 years
Distribution
Lower limbs only	57 (90.4)
Lower and upper limbs	2 (3.2)
Lower limbs, with extensions to lower abdomen	2 (3.2)
Generalised (lower limb and trunk)	2 (3.2)
Lower Limb involvement	63 (100)
Bilateral	49 (77.8)
Unilateral	14 (22.2)

*Two patients presented with both Schamberg’s disease and lichen aureus variants of pigmented purpuric dermatoses (PPD)

Duration

The most common duration of PPD was 1–5 years in 31 patients (49.2%). PPD was present for less than 1 year in 21 patients (33.3%) and for 6–10 years in eight patients (12.7%). Three patients (4.7%) PPD had a more chronic course of greater than 10 years (Table 2).

Clinical variants

Schamberg’s disease was the most common variant seen in 26/63 patients (41.3%) followed by lichen aureus seen in 25/63 patients (39.7%). Other reported sub-types of PPD were recruited with the exception of the granulomatous variant that did not present (Table 2). Two patients presented with both Schamberg’s disease and lichen aureus.

Histology

Skin biopsies were obtained in 47 patients (74.6%). Histology in all samples demonstrated superficial perivascular infiltrate of lymphocytes with occasional foci of extravasated red blood cells within the papillary dermis and associated with deposits of haemosiderin. In one patient with a generalised eruption involving the trunk and upper limbs, a lichenoid infiltrate was also present in addition to other features described above rendering the diagnosis of Pigmented Purpuric Lichenoid Dermatosis of Gougerot and Blum. Two patients with itchy purpura and three with the Doucas Kapetanakis variant demonstrated dermal spongiosis in skin biopsies.

Anatomical distribution

All 63 patients had lower limb involvement. In 57 patients (90.4%), PPD involved the lower limbs only. In 2 patients (3.2%), PPD extended to the lower abdomen. In 2 patients (3.2%), PPD involved the lower and upper limbs and in two patients (3.2%), PPD was generalised, involving the lower limbs and the trunk. PPD was bilateral in 49 patients (77.8%) and unilateral in 14 patients (22.2%).

Associated concurrent conditions

Venous disease

Venous incompetence was found in 48 patients (76.2%) with the majority (52.4%) falling within the C2 of the CEAP classification (Table 3). The identified vessels underlying lesions of PPD were superficial tributaries as against superficial venous trunks (saphenous veins).

Table 3.

CEAP classification. CEAP (clinical, etiologic, anatomical and pathophysiological) classification of patients with pigmented purpuric dermatoses (PPD).

Venous disease	Number of patients (%)
Co-localisation with venous reflux	48 (76.2)
C0 No visible or palpable signs of venous disease	12 (19.0)
C1 Telangiectasies or reticular veins	7 (11.1)
C2 Varicose veins	33 (52.4)
C3 Oedema	1 (1.6)
C4a Pigmentation or eczema	7 (11.1)
C4b Lipodermatosclerosis or atrophie blancher	2 (3.2)
C5 Healed venous ulcer	0 (0)
C6 Active venous ulcer	1 (1.6)

Bleeding diathesis

A history of easy bruising was found in 31 patients (49.2%) and menorrhagia in 16 of 35 female patients (43.2%) (Table 4). A history of epistaxis was found in 14 patients (22%), bleeding from other sites in 14 patients (22%), and a family history of bleeding disorders was present in 11 patients (17.4%).

Table 4.

Haemostatic profile. History of a bleeding diathesis and the associated haemostatic derangements.

Bleeding diathesis	Number of patients (%)
Bruising and bleeding risk
Easy bruising	31 (49.2)
Menorrhagia	16 (43.2*)
Epistaxis	14 (22.2)
Other bleeds	14 (22.2)
Family history of bleeding	11 (17.4)
Haemostatic abnormalities	36 (57.1)
Abnormal platelet function	32 (50.8)
Clotting factor deficiency/vWD	8 (12.7)
Prolonged PT/APTT	4 (6.3)
Thrombocytopenia	3 (4.8)

*Percentage of 37 females. ** percentage of 35 patients tested. APTT, activated partial thromboplastin time; PT, prothrombin time; vWD, von Willebrand disease

Haemostatic abnormalities

Haemostatic abnormalities were evident in 36 patients (57.1%) (Table 4). Decreased platelet function by at least one platelet function test was found in 32 patients (50.8%). Clotting factor or vWF deficiencies were found in eight patients (12.7%), with four patients having either a prolonged PT or APTT. Thrombocytopenia was found in three patients (4.8%).

The most common method for detecting a platelet function abnormality was using the Multiplate impedance aggregometer, which found decreased platelet function with at least one agonist in 22 out of 37 patients tested (62.7%).

Risk factors

Risk factors were divided into modifiable (drugs and medications, supplements or diet) and non-modifiable (Thrombocytopenia, von Willebrand disease, clotting factor deficiencies or Haemochromatosis) categories (Table 5).

Table 5.

Modifiable and non-modifiable risk factors. Drugs, supplements, food and beverages implicated in the pathogenesis of pigmented purpuric dermatoses (PPD).

Risk factors	Number of patients (%)
Any risk factor	58 (92.1)
Modifiable risk factors	53 (84.1)*
Drugs and Medications	31 (47.6)
NSAIDs	21 (33.3)
Aspirin	8 (12.6)
Warfarin	2 (3.2)
Prednisone	1 (1.6)
Intragam	1 (1.6)
Glatiramer acetate (Copaxone)	1 (1.6)
Desensitising injections**	1 (1.6)
Supplements	27 (42.9)
Fish oil omega-3 PUFA	21 (33.3)
Tocopherol (vitamin E)	14 (22.2)
Other herbal supplements	15 (23.8)
Diet (Food and Drinks)	17 (26.9)
Non-modifiable risk factors	13 (20.6)*
von Willebrand’s disease	3 (4.8)
FVIII deficiency	3 (4.8)
Thrombocytopenia	3 (4.8)
Haemochromatosis	2 (3.2)
Hypofibrinogenaemia	1 (1.6)
Myelodysplasia	1 (1.6)

*Eight patients (12.7%) had both modifiable and non-modifiable risk factors. ** house dust mites immunotherapy. HDM, house dust mite; NSAIDs, non-steroidal anti-inflammatory drugs; PUFA, polyunsaturated fatty acids

58 patients (92.1%) had risk factors suspected to contribute to PPD while no risk factors could be identified in five patients (7.9%). Modifiable risk factors were identified in 45 patients (71.4%) while non-modifiable risk factors (factor deficiencies and platelet function abnormalities) were detected in five patients (7.9%). Eight patients (12.7%) had a combination of both modifiable and non-modifiable risk factors.

In the five patients (7.9%) in whom no risk factors were identified, two had abnormal platelet function tests the cause of which could not be determined. Genetic testing for abnormal platelet function was not performed.

Modifiable risk factors

1. Drugs: 29 patients (46.0%) were taking medications known to affect haemostatic function (Table 5). These included NSAIDs in 21 patients (33.3%), Aspirin in eight patients (12.6%), warfarin in two patients (3.2%), prednisone in one patient (1.6%), and Intragam in one patient (1.6%). One further patient (1.6%) was receiving desensitising injections for immunotherapy to house dust mites (HDM) and one patient was taking glatiramer acetate (Copaxone) for multiple sclerosis.

2. Supplements: 27 patients (42.9%) were taking supplements including fish oil omega-3 polyunsaturated fatty acids (PUFA) in 21 patients (33.3%), tocopherol/vitamin E in 14 patients (22.2%) or other herbal supplements in 15 patients (23.8%). Patients reported taking these supplements ranging from several weeks to many years.

3. Food and Drinks: In 17 patients (26.9%), excessive intake of food and beverages suspected to potentially act as risk factors were identified (Table 5). This included excessive intake of colours and preservatives in confectionaries in three patients (4.8%), caffeine consumption in two patients (3.2%), soft drinks containing tartrazine (greater than 375 mL daily) in eight patients (12.7%) and alcohol consumption (one or more bottles of red wine a day) in four patients (6.3%).

Non-modifiable risk factors

Non-modifiable risk factors were identified in 13 patients (20.6%) who had abnormal blood results (Table 5). This included underlying haematological diseases such as von Willebrand disease in three patients (4.8%), factor VIII deficiency in three patients (4.8%), thrombocytopenia in three patients (4.8%), haemochromatosis in two patients (3.2%), hypofibrinogenaemia in one patient (1.6%) and myelodysplasia in one patient (1.6%).

Other testing

No patients were diagnosed with renal or liver disease, and occasional elevations of liver enzymes were not deemed indicative.

Cessation of modifiable risk factors

Of the 53 patients whose modifiable risk factors were identified, these were addressed in 35 (68.6%), with 18 patients unable to cease their modifiable risk factor on clinical grounds (such as aspirin or warfarin use in cardiovascular disease) or were lost to follow-up. Of these 35 patients, 28 (80.0%) achieved either complete (14 patients, 40%) or partial (14 patients, 40.0%) resolution of their PPD. Seven patients (20.0%) had no resolution of their PPD (Table 6).

Table 6.

Resolution of PPD. Resolution of pigmented purpuric dermatoses (PPD) was assessed following cessation of modifiable risk factors followed by treatment of chronic venous disease (CVD) when indicated. Resolution was assessed as complete, partial or no resolution.

Clinical resolution	Number of patients (%)
Cessation of modifiable risk factors	35
Resolution	28 (80.0)
Complete	14 (40.0)
Partial	14 (40.0)
No Resolution	7 (20.0)
Treatment of CVD	18
Resolution	17 (94.4)
Complete	6 (33.3)
Partial	11 (61.1)
No Resolution	1 (5.6)

CVD, chronic venous disease.

Treatment of venous disease

Patients with modifiable risk factors who achieved only partial or no resolution of their PPD, as well as those with non-modifiable risk factors, were offered treatment of their venous disease if clinically indicated. Treatments were performed in a total of 18 patients. Of these patients, 17 patients (89.5%) achieved complete (n = 6, 33.3%) or partial (n = 11, 61.1%) resolution of their PPD. Only one patient (5.6%) had no resolution to their PPD following venous intervention (Table 6).

Discussion

Pigmented purpuric dermatoses are a group of conditions with an unknown aetiology presenting with similar clinical and histopathological features. This study has demonstrated important and novel associations of PPD with CVD (76.2%) and haemostatic abnormalities (57.1%). We also demonstrated resolution of PPD in 80% of patients following cessation of modifiable risk factors (such as the intake of fish oil) and in further 94.4% following intervention for superficial venous disease. We propose that these two factors, CVD and haemostatic abnormalities, play a significant role in the aetiology of PPD (Figure 1 and 2).

Figure 1.

(a) Pigmented Purpuric Dermatosis (PPD) localised to incompetent tributaries of the great saphenous vein. (b) Near complete resolution following cessation of fish oil and treatment of CVD.

Figure 2.

(a) Pigmented Purpuric Dermatosis (PPD) and stasis dermatitis localised to distal segment of an incompetent great saphenous vein and Cokett perforators in a patient with von Willebrand disease. (b) Partial resolution following treatment of CVD.

Venous hypertension and ‘gravitational dependence’ have been previously associated with the onset of PPD.^7,9 In this study, we determined venous hypertension to be a common risk factor for most variants of PPD. However, PPD is an uncommon condition and is not a typical finding in most patients presenting with venous hypertension. Our results emphasise the importance of the co-presence of other associated risk factors such as haemostatic abnormalities in the pathogenesis of this condition. Haemostatic abnormalities trigger the eruption by resulting in capillary leakage and red cell extravasation. Meanwhile, venous hypertension results in haemosiderin deposition and localised perivascular lymphocytic cuffing of dermal venules. The combined effect is the clinical presentation of petechial-type eruption intermixed with haemosiderin pigmentation localised to sites of venous hypertension and the histological findings of red cell extravasation, hemosiderin deposition and a perivascular lymphocytic infiltrate typical of PPD. In this model, venous hypertension acts as a ‘prerequisite’ and haemostatic abnormalities act as triggers.

Platelets with abnormal function predispose to abnormal haemostasis. Furthermore, platelets influence lymphocyte function via direct cell–cell contact and/or soluble mediators and in particular enhance adhesion and cell migration of T, NK and B cells. Perivascular lymphocytic cuffing is a well-known finding in patients with chronic venous hypertension and is considered a secondary phenomenon.²¹ We propose that the lymphocytic infiltrate in PPD is most likely a secondary rather than a primary phenomenon. Interaction of lymphocytes with platelets plays a major role in the pathogenesis of a range of vascular conditions and its role in PPD should be explored in future studies.

In this study, 57.1% of patients demonstrated some form of haemostatic abnormality, predominantly due to the use of drugs and supplements that affect platelet function. We used PFA-100 as a screening method followed by light transmission and impedance aggregometry. Consistently, a recent case report described PPD in a child with a platelet delta storage pool disorder, resulting in decreased platelet aggregation detectable using light transmission aggregometry.²²

The decreased platelet function observed in 50.8% of our patients is most likely an underestimation given that we obtained the impedance aggregometry (Multiplate) system later in this study and hence this test was only performed in just over half of the patient population. Impedance aggregometry is a quantitative measure of platelet function and hence more specific than light transmission method which for our institution is reported as a qualitative assessment only. In a previous study, 801 patients underwent both tests where light transmission missed the diagnosis in 34 patients detected by Multiplate.²³ Future studies should incorporate impedance aggregometry and flow cytometry as more specific methods of detecting platelet function abnormalities.

We detected coagulation factor abnormalities including deficiencies in factor VIII and vWF. Factor VIII and vWF are essential in platelet adhesion in high shear environments and their deficiency has been associated with a bleeding diathesis. Interestingly, two patients presenting with linear PPD presented with factor VIII deficiency. The association between Factor VIII deficiencies and linear PPD needs to be explored in future studies and has not been reported before.

An important finding in this study was that a number of reversible factors were identified, the cessation of which resulted in resolution of PPD. 84.1% of patients were taking drugs, medications, supplements such as fish oil or other modifiable triggers. Cessation of these modifiable risk factors resulted in the complete resolution of the eruption in 80% of patients. These included antiplatelet agents such as aspirin and platelet function modifiers such as fish oil, NSAIDs and Vitamin E. Fish oil supplementation is very popular in Australia with 12% of the population consuming this supplement.²⁴ Patients with PPD should be specifically asked for history of intake of fish oil and other supplements such as krill oil and flaxseed oil.

In this study, fish oil inhibited collagen-induced platelet aggregation as detected by Multiplate platelet aggregometry analysis. Whilst some studies showed ADP-induced inhibitory effects on platelet aggregation,²⁵ our findings were consistent with those that showed fish oil inhibited collagen-induced platelet aggregation.^14,15,26 Fish oil, comprising omega-3 fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), is thought to have antithrombotic effects through various mechanisms, including augmenting the susceptibility of plasma lipoproteins to oxidation and improvement of endothelial dysfunction.^27,28 In porcine studies, fish oil causes relaxation of the coronary arteries²⁹ and microvessels³⁰ due to the effects of EPA, which is the major omega 3-fatty acid of fish oil. There was also enhanced relaxation of the endothelium of arteries³⁰ and femoral veins,³¹ which may be due to the effects of nitric oxide on platelet aggregation and adhesion, leukocyte adhesion and smooth muscle cell proliferation.³¹ While these effects are beneficiary in those settings, in the context of PPD, fish oil is a contributory factor. The effect may be dose dependent. In this study, our patients consumed 3250 mg of fish oil on average. Future studies could explore the dose-response relationship between the intake of fish oil and onset of PPD.

Other than drugs and supplements, dietary factors seem to play an important role in some patients presenting with PPD. In this study, one patient presenting with itchy purpura consumed several bottles per day of caramel-containing soft drinks such as Coke. Tartrazine (E102), a common food additive found in many soft drinks and confectioneries, has been associated with PPD. Tartrazine has also been associated with altered platelet function.³² This dietary association with soft drinks has been described previously in one patient.⁴ In this patient, clinical disease resolved upon the cessation of soft drinks and recurred with re-challenge several times. The underlying mechanism may be due to the effect of tartrazine on collagen-induced platelet aggregation.^32–35

Three patients in this study were found to consume at least one bottle of red wine per day. For one of these patients, their PPD improved (partially) upon the cessation of red wine consumption and recurred with re-challenge. This patient also had venous disease of lower limbs and his PPD improved significantly following cessation of red wine consumption and treatment of venous disease. Intake of flavonoid-rich foods such as wine, blueberries, black tea and dark chocolate has been associated with abnormal platelet function.³⁶ Red wine in particular has been associated with decreased platelet function, unlike white wine which has been conversely associated with increased platelet function.³⁷

Other than NSAIDS, fish oil and aspirin, other drugs such as calcium channel blockers, beta-blockers, statins, angiotensin-converting enzyme (ACE) inhibitors, antihistamines, antidepressants and analgesics have been associated with the onset of PPD.³⁸ This association is most strongly correlated with the Schamberg’s variant.^5,39,40 The pathogenic mechanism in these cases remains unknown; however, a number of these agents are associated with decreased platelet function. Antidepressants, in particular selective serotonin reuptake inhibitors (SSRI),^41,42 calcium channel blockers⁴³ and ACE inhibitors,^44,45 are all well known to decrease platelet function. Specifically, platelet aggregation may be decreased by beta-blockers,⁴⁶ statins^47,48 and antihistamines.^49,50

Despite venous hypertension appearing to be a common underlying risk factor for most variants of PPD, some variants do not seem to be associated with venous disease. This is particularly more relevant in generalised eruptions involving the trunk. One such patient presented with the Gourgerot-Blum lichenoid eruption. This particular patient was noted to have no platelet or coagulation abnormalities. One patient with itchy purpura secondary to dietary factors also presented with a generalised eruption. Hence, a generalised eruption involving the trunks should trigger the search for other underlying risk factors beyond venous disease even though venous disease may still be present and contribute to a more intense eruption on the lower limbs.

Pigmentation of PPD is distinct from pigmentation seen in venous disease and pigmentation associated with petechial eruptions. Fresh lesions of PPD present with pin-point non-blanchable ‘petechial’ macules and patches on a background of cayenne pepper pigmentation. Older patches may lack the fresh petechial lesions. It is possible that PPD is more prevalent in the community but misdiagnosed as a petechial eruption or venous pigmentation. Skin biopsy is essential to confirm the diagnosis. In addition, mycosis fungoides (MF) have been reported to mimic PPD and must be excluded. In this study, no patients were diagnosed with MF. Furthermore, having observed this cohort of patients in a 12-year period, none of the patients progressed to MF. In our opinion, it is more likely that MF can mimic PPD rather than PPD progressing to MF.

We recommend a panel of investigations for patients presenting with PPD (Table 7). All patients should have at least one skin biopsy to confirm the diagnosis and exclude possible MF. Patients with a lower limb eruption should have a detailed venous incompetence mapping that includes all subcutaneous and in particular sub-dermal tributaries. A simple assessment of venous trunks will not suffice, and the assessment should incorporate a high-frequency probe to interrogate the superficial tributaries. CW-Doppler is more sensitive than pulsed-wave duplex and may need to be utilised to localise the sites of incompetence underlying the patches of PPD. Trans-illumination may be helpful to visualise the vessels (Figure 3). Blood tests should include an assessment of platelet function. We recommend PFA-100 as an initial screening test followed by platelet aggregation studies and in particular impedance aggregometry (Multiplate) if available. A simple platelet count is inadequate and an assessment of platelet function should be undertaken. Although we screened all patients for paraproteins, none of the patients showed an abnormality and hence EPG and IEPG do not need to be routinely performed unless indicated.

Table 7.

Recommended investigations to assess pigmented purpuric dermatoses (PPD).

Skin biopsy	1–2 x 4 mm punch biopsies from representative lesions
Venous incompetence mapping	To include a high-frequency probe and to interrogate in erect position all superficial tributaries
Blood tests	Full blood count
	Renal function
	Liver function
	Coagulation screen and clotting factor assays
	Prothrombin time (PT)
	Activated partial thromboplastin time (APTT)
	Factor VIII
	Factor IX von Willebrand factor (vWF)
	Platelet function Testing
	Platelet function analyser (PFA-100^TM)
	Light Transmission aggregometry
	Multiplate impedance aggregometry

Figure 3.

Trans-illumination showing the colocalisation of a subdermal vein with the clinical eruption of Pigmented Purpuric Dermatosis (PPD).

In conclusion, we have identified the co-existence of venous hypertension and haemostatic abnormalities as the key contributory factors to the pathogenesis of PPD (Figure 4). We recommend investigation of platelet function as well as clotting factors in this patient population. Supplements affecting platelet function such as fish oil should be stopped. To achieve a more comprehensive and long-term resolution of PPD, venous hypertension, especially in the region of the affected limb, should be treated.

Figure 4.

Study design flowchart.

Footnotes

Acknowledgements

We would like to thank our research assistants Dr Anne Pilotelle and Arash Bargriz and medical student Lina Salem for assistance with laboratory testing, data collection and entry.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

Ethical approval was obtained from St Vincent’s Human Research Ethics Committee (Approval 15/064).

Guarantor

Contributorship

KP designed the study, collected data, performed all clinical consultations and interventions, analysed the data and drafted the article. BK collected and analysed the data and drafted the article. AO’C collected and analysed the data and drafted the article. PK collected and analysed the data and drafted the article. DC collected and analysed the data and drafted the article.

ORCID iDs

Kurosh Parsi

David Connor

References

Newton

Raimer

. Pigmented purpuric eruptions. Dermatol Clin 1985; 3: 165–169.

Tristani-Firouzi

Meadows

Vanderhooft

. Pigmented purpuric eruptions of childhood: a series of cases and review of literature. Pediatr Dermatol 2001; 18: 299–304.

Wong

Ratnam

. A report of two cases of pigmented purpuric dermatoses treated with PUVA therapy. Acta Derm Venereol 1991; 71: 68–70.

Weiss

Karrer

Landthaler

, et al. Diet-induced pigmented purpuric dermatosis. Australas J Dermatol 2014; 55: e51–53.

Taketuchi

Chinen

Ichikawa

, et al. Two cases of unilateral pigmented purpuric dermatosis. J Dermatol 2001; 28: 493–498.

Georgala

Katoulis

Symeonidou

, et al. Persistent pigmented purpuric eruption associated with mycosis fungoides: a case report and review of the literature. J Eur Acad Dermatol Venereol 2001; 15: 62–64.

Sardana

Sarkar

Sehgal

. Pigmented purpuric dermatoses: an overview. Int J Dermatol 2004; 43: 482–488.

Iwatsuki

Aoshima

Tagami

, et al. Immunofluorescence study in purpura pigmentosa chronica. Acta Derm Venereol 1980; 60: 341–345.

Ratnam

Peters

. Purpura simplex (inflammatory purpura without vasculitis): a clinicopathologic study of 174 cases. J Am Acad Dermatol 1991; 25: 642–647.

10.

Klug

Haustein

. Ultrastructure of macrophage-lymphocyte interaction in purpura pigmentosa progressiva. Dermatologica 1976; 153: 209–217.

11.

Simon

Heese

Götz

. Immunopathological investigations in purpura pigmentosa chronica. Acta Derm Venereol 1989; 69: 101–104.

12.

Buchanan

Hirsh

. Effect of aspirin on hemostasis and thrombosis. N Engl Reg Allergy Proc 1986; 7: 26–31.

13.

Schafer

. Effects of nonsteroidal antiinflammatory drugs on platelet function and systemic hemostasis. J Clin Pharmacol 1995; 35: 209–219.

14.

McEwen

Morel-Kopp

M-C

Tofler

, et al. The effect of omega-3 polyunsaturated fatty acids on fibrin and thrombin generation in healthy subjects and subjects with cardiovascular disease. Semin Thromb Hemost 2015; 41: 315–322.

15.

McEwen

Morel-Kopp

M-C

Chen

, et al. Effects of omega-3 polyunsaturated fatty acids on platelet function in healthy subjects and subjects with cardiovascular disease. Semin Thromb Hemost 2013; 39: 25–32.

16.

Liu

Wallmon

Olsson-Mortlock

, et al. Mixed tocopherols inhibit platelet aggregation in humans: potential mechanisms. Am J Clin Nutr 2003; 77: 700–706.

17.

Eklöf

Rutherford

Bergan

, et al. Revision of the CEAP classification for chronic venous disorders: consensus statement. J Vasc Surg 2004; 40: 1248–1252.

18.

Mammen

Comp

Gosselin

, et al. PFA-100 system: a new method for assessment of platelet dysfunction. Semin Thromb Hemost 1998; 24: 195–202.

19.

Born

. Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature 1962; 194: 927–929.

20.

Tessari

Cavezzi

Frullini

. Preliminary experience with a new sclerosing foam in the treatment of varicose veins. Dermatol Surg Off Publ Am Soc Dermatol Surg Al 2001; 27: 58–60.

21.

Smith

. The causes of skin damage and leg ulceration in chronic venous disease. Int J Low Extrem Wounds 2006; 5: 160–168.

22.

Gkalea

Tang

Favier

, et al. Progressive pigmented purpuric dermatosis and platelet delta storage pool deficiency in a child. Pediatr Blood Cancer 2019; 66: e27748.

23.

Paniccia

Antonucci

Maggini

, et al. Comparison of methods for monitoring residual platelet reactivity after clopidogrel by point-of-care tests on whole blood in high-risk patients. Thromb Haemost 2010; 104: 587–292.

24.

Australian Bureau of Statistics

. Australian Health Survey: Nutrition First Results - Foods and Nutrients 2011–2012, https://www.abs.gov.au/statistics/health/health-conditions-and-risks/australian-health-survey-nutrition-first-results-foods-and-nutrients/latest-release (2014, accessed 5 September 2014).

25.

Chen

Jokela

, et al. Effect of stable fish oil on arterial thrombogenesis, platelet aggregation, and superoxide dismutase activity. J Cardiovasc Pharmacol 2000; 35: 502–505.

26.

Agren

Väisänen

Hänninen

, et al. Hemostatic factors and platelet aggregation after a fish-enriched diet or fish oil or docosahexaenoic acid supplementation. Prostaglandins Leukot Essent Fatty Acids 1997; 57: 419–421.

27.

Mozaffarian

JHY

. Omega-3 fatty acids and cardiovascular disease: effects on risk factors, molecular pathways, and clinical events. J Am Coll Cardiol 2011; 58: 2047–2067.

28.

De Caterina

. n–3 fatty acids in cardiovascular disease. N Engl J Med 2011; 364: 2439–2450.

29.

Shimokawa

Lam

Chesebro

, et al. Effects of dietary supplementation with cod-liver oil on endothelium-dependent responses in porcine coronary arteries. Circulation 1987; 76: 898–905.

30.

Shimokawa

Vanhoutte

. Dietary omega 3 fatty acids and endothelium-dependent relaxations in porcine coronary arteries. Am J Physiol 1989; 256: H968–H973.

31.

Tagawa

Hirooka

Shimokawa

, et al. Long-term treatment with eicosapentaenoic acid improves exercise-induced vasodilation in patients with coronary artery disease. Hypertens Res Off J Jpn Soc Hypertens 2002; 25: 823–829.

32.

Gallagher

Splansky

Bernstein

. Inhibition of platelet aggregation by tartrazine and a pyrazolone analogue in normal and allergic individuals. Clin Allergy 1980; 10: 683–690.

33.

Gerber

Payne

Oelz

, et al. Tartrazine and the prostaglandin system. J Allergy Clin Immunol 1979; 63: 289–294.

34.

Williams

Pawlowicz

Davies

. Aspirin-like effects of selected food additives and industrial sensitizing agents. Clin Exp Allergy J Br Soc Allergy Clin Immunol 1989; 19: 533–537.

35.

Kalinke

Wüthrich

. [Purpura pigmentosa progressiva in type III cryoglobulinemia and tartrazine intolerance. A follow-up over 20 years]. Hautarzt Z Dermatol Venerol Verwandte Geb 1999; 50: 47–51.

36.

Hooper

Kroon

Rimm

, et al. Flavonoids, flavonoid-rich foods, and cardiovascular risk: a meta-analysis of randomized controlled trials. Am J Clin Nutr 2008; 88: 38–50.

37.

Seigneur

Bonnet

Dorian

, et al. Effect of the consumption of alcohol, white wine, and red wine on platelet function and serum lipids. J Appl Cardiol 1990; 5: 215–222.

38.

Crowson

Magro

Zahorchak

. Atypical pigmentary purpura: a clinical, histopathologic, and genotypic study. Hum Pathol 1999; 30: 1004–1012.

39.

Kwon

Lee

. Figurate purpuric eruptions on the trunk: acetaminophen-induced rashes. J Dermatol 1998; 25: 756–758.

40.

Adams

Gadenne

. Glipizide-induced pigmented purpuric dermatosis. J Am Acad Dermatol 1999; 41: 827–829.

41.

McCloskey

Postolache

Vittone

, et al. Selective serotonin reuptake inhibitors: measurement of effect on platelet function. Transl Res J Lab Clin Med 2008; 151: 168–172.

42.

Reikvam

A-G

Hustad

Reikvam

, et al. The effects of selective serotonin reuptake inhibitors on platelet function in whole blood and platelet concentrates. Platelets 2012; 23: 299–308.

43.

Mehta

. Influence of calcium-channel blockers on platelet function and arachidonic acid metabolism. Am J Cardiol 1985; 55: 158B–164B.

44.

Bauriedel

Skowasch

Schneider

, et al. Antiplatelet effects of angiotensin-converting enzyme inhibitors compared with aspirin and clopidogrel: a pilot study with whole-blood aggregometry. Am Heart J 2003; 145: 343–348.

45.

Moser

Callahan

Cheung

, et al. ACE inhibitor effects on platelet function in stages I-II hypertension. J Cardiovasc Pharmacol 1997; 30: 461–467.

46.

Bonten

Plaizier

CEI

Snoep

J-JD

, et al. Effect of β-blockers on platelet aggregation: a systematic review and meta-analysis. Br J Clin Pharmacol 2014; 78: 940–949.

47.

Rosenson

Tangney

. Antiatherothrombotic properties of statins: implications for cardiovascular event reduction. JAMA 1998; 279: 1643–1650.

48.

Kearney

Fitzgerald

. The anti-thrombotic effects of statins. J Am Coll Cardiol 1999; 33: 1305–1307.

49.

Petríková

Jancinová

Nosál

, et al. H (1)- antihistamines and activated blood platelets. Inflamm Res Off J Eur Histamine Res Soc Al 2006; 55(Suppl 1): S51–S52.

50.

Nosál

Drábiková

Jancinová

, et al. Antiplatelet and antiphagocyte activity of H1-antihistamines. Inflamm Res Off J Eur Histamine Res Soc Al 2005; 54(Suppl 1): S19–S20.