Purpura-associated congenital lymphedema

ARTICLE

Auteur(s) : Samantha Berti¹, Alessandro Pieri², Torello Lotti³, Alberto Duranti¹, John Panelos⁴, Maurizio De Martino¹, Silvia Moretti³

¹Department of Pediatric Medicine, Anna Meyer Children’s Hospital, Viale Pieraccini, 24, 50132 Florence, Italy
²Department of Cardiology, University of Florence, Florence, Italy
³Department of Dermatology, University of Florence, Florence, Italy
⁴Department of Human Oncology and Pathology, University of Florence, Florence, Italy

accepté le 28 Mai 2009

An 8-year-old girl was referred to our Department showing several purpuric lesions localized on the right thigh and a two-month deterioration of congenital primary lymphedema of the right lower limb. Ulceration of a purpuric lesion was witnessed a few days prior to presentation, whereas swelling was evident below the knee from birth (figure 1A). Past medical history of cellulites or other clinical or structural abnormalities, with the exception of syndactylia and brachydactyly of the toes (figure 1B) was not observed. A family history of leg swelling was not present; however the maternal grandmother did have varicose veins. The patient was affected by Milroy disease, also known as hereditary lymphedema type I (not confirmed by molecular analysis).

Laboratory studies revealed normal circulating immune complexes and the absence of autoantibodies. Color duplex investigation (CDI) of the right lower extremity showed collateral varicose veins coming from an insufficient anterior saphenous vein. Histopathological examination, performed on a purpuric lesion of the right thigh, demonstrated lichenoid dermatitis with mild spongiosis in epidermal foci. A dermal moderate mixed inflammatory infiltrate and dilated papillary vessels replete with erythrocytes were noted (figure 2). Extravasated red blood cells were seen in the papillary dermis. Direct immunofluorescence, performed on lesional skin, revealed IgG, IgA, IgM and C3 perivascular deposits and IgM and C3 junctional deposits. These findings were consistent with small vessel vasculitis. An X-ray of the right toes did not show a bone involvement, concluding that it was just a skin syndactylia.

On the basis of clinical, histopathological and instrumental findings, we diagnosed lichenoid pigmented purpuric dermatosis of Gougerot and Blum in a patient with Milroy Disease complicated by anterior saphenous vein insufficiency. We treated the patient with topical steroids and compression stockings, until phlebectomy resolved the cutaneous lesions (figure 3). The edema improved slightly after removal of the saphenous vein. Ulcer healing was delayed because of lymphorragia. We reported this case not only for the rarity of Milroy disease, but also for its novel association with lichenoid pigmented purpura and its cutaneous immunopathological findings.

Discussion

Milroy disease is clinically characterized by lower limb swelling from birth, accompanied by other signs and symptoms, such as large caliber legs veins (23%), recurrent episodes of cellulites (20%), papillomatosis (10%), upslanting toenails (10%) and, in males, hydrocele (37%) which is the most common finding after congenital edema [2, 3]. To be considered “typical” Milroy disease, the onset of lymphedema should be congenital, confined to the lower limbs (especially to the dorsum of the feet), and have a “woody” feel [4]. It is very unusual for edema to be present in any other region in a patient with Milroy disease, although pleural effusion, nuchal translucency and hydrops fetalis have been reported in utero [5]. In 2005, Brice G et al. [2] studied 71 patients from 10 families with Milroy disease. The onset of lymphedema was congenital in 97% of the patients. The edema was symmetrical in 55%, bilateral in 85% and extended below the knee in 94%. An association with urethral abnormalities was reported in 4%. According to this study, there were no major structural abnormalities or consistent dysmorphic features. Only 16/71 patients had an active treatment for lymphedema. Treatment was limited to massage and compression with hosiery or pneumatic therapy. Two female patients had undergone surgery; one had unilateral reduction surgery during childhood, while the other had papilloma removal from the dorsum of the toes in her twenties.

The clinical features of our patient were in accordance with current understanding of “typical” Milroy disease: the swelling was congenital (it appears at birth or in the first year of life), confined to the lower limbs, both legs are affected, displayed involvement of the dorsum of the feet, (it was evident below the knee) and had a “woody” feel. Edema was found associated with toe syndactylia and there was large caliber leg vein insufficiency, as documented by the Color Duplex investigation (CDI) of the right lower extremity, that showed collateral varicose veins coming from an insufficient anterior saphenous vein. There was no history of cellulites or other clinical or structural abnormalities. The patient was treated from birth with compression stockings and massages.

Traditional X-ray lymphangiography or isotope lymphangiography (lymphoscintigraphy) often demonstrates lymphatic truncal aplasia or hypoplasia with minimal or no collateral dermal formation [1, 6]. Lymphatic hypoplasia can be noted only when lymphangiography is successful. Failure of lymphangiography can be due to a lack of lymphatics in the foot. Lymphoscintigraphy can show no, or little, uptake in the inguinal nodes as a result of no, or little, uptake by peripheral lymphatics of the foot, as well [2]. This explanation might account for the unsuccessful lymphangiography results of the right lower limb seen in our patient. A lymph scan of left lower limb was normal.

Milroy disease typically segregates in an autosomal dominant manner and linkage to chromosomal locus 5q35.3 was reported in 1998 and 1999 [1, 7, 8]. Subsequently, the vascular endothelial growth factor receptor 3 (VEGFR-3) gene located in this region was found to be the causative gene [1, 9]. This receptor, activated by VEGF-C and VEGF-D, is required for lymphatic development during embryogenesis [2]. Several mutations of this gene have been identified: most of them are missense mutations within the tyrosine kinase domain I or II, leading to VEGFR-3 gene malfunction and developmental failure of most peripheral lymphatic vessels in the lower limbs [1, 2, 10]. Despite the germline mutation, lymphedema appears only to develop in the lower limbs of humans (or the hindlimbs of transgenice mice), but the reason for this is unclear. The onset of edema at birth, before walking begins, suggests that orthostatic factors are not responsible for the limitation of edema to the lower limbs [2]. Theoretically, gene therapy could be a treatment option, as successful reversal of the phenotype has been achieved in transgenic mice.

A family history of Milroy disease must always be investigated and is considered a strong indicator of correct diagnosis. Although new mutations in a dominant inherited disorder are unexpected, Carver C et al. recently presented three cases with congenital onset of lymphedema (confirmed by the identification of VEGFR3 mutations), but no familial component, concluding that it is not necessary for diagnosis [4]. Likewise, in our patient, the typical features of Milroy disease were present in the absence of a family history of leg swelling. Our patient’s findings were probably the result of a novel gene mutation involved in Milroy disease (Explanation to follow). The diagnosis can now be made on clinical grounds and eventually confirmed by molecular analysis, even if not all individuals with Milroy disease have identifiable mutations in VEGFR-3 [2]. We concur with the study of Brice G et al. [2], in that the term Milroy disease should been reserved for those cases of primary lymphedema fulfilling the criteria proposed by Milroy: familial, congenital, located to lower limbs, and non progressive. However VEGFR-3 gene studies would be helpful when new mutations occur or when the parent may be non-penetrant.

Our patient showed a two-month history of several purpuric lesions localized on the right thigh coupled with congenital primary lymphedema that progressively worsened during this period. A few days before the department visit, one of the purpuric lesions showed ulceration. On the basis of clinical and histopathological findings, we diagnosed a lichenoid pigmented purpura of Gougerot and Blum, probably due to an insufficient anterior saphenous vein, as documented by the resolution of the cutaneous lesion after phlebectomy. This unique association between lichenoid pigmented purpura of Gougerot and Blum and Milroy disease has never before been reported in literature.

Furthermore our patient showed atypical cutaneous immunopathological findings that were more consistent for a small vessel vasculitis rather than for purpura. In fact direct immunofluorescence revealed perivascular deposits of IgG, IgA, IgM and C3 and junctional deposits of IgM and C3. Vasculitis related to the immune-mediated inflammation generated by deep vein insufficiency would have been the mostly likely evolution of this patient had early surgical intervention not been performed.

Acknowledgements

References

2 Brice G, Child AH, Evans A, et al. Milroy disease and the VEGFR-3 mutation phonotype. J Med Genet 2005; 42: 98-102.

3 Brice G, Mansour S, Bell R, et al. Analysis of the phenotypic abnormalities in lymphoedema distichiasis syndrome in 74 patients with FOXC2 mutations or linkage to 16q24. J Med Genet 2000; 39: 478-83.

4 Carver C, Brice G, Mansour S, Ostergaard P, Mortimer P, Jeffery S. Three children with Milroy disease and de novo mutations in VEGFR3. Clin Genet 2007; 71: 187-9.

5 Ghalamkarpour A, Morlot S, Raas-Rothschild A, et al. Hereditary lymphedema type I associated with VEGFR3 mutation: the first de novo case and atypical presentations. Clin Genet 2006; 70: 330-5.

6 Greenlee R, Hoyme HE, Witte MH, Crowe P, Witte CL. Developmental disorders of the lymphatic system. Lymphology 1993; 26: 156-68.

7 Ferrell RE, Levinson KL, Esman JH, et al. Hereditary lymphedema: evidence for linkage and genetic heterogeneity. Hum Mol Genet 1998; 7: 2073-8.

8 Evans AL, Brice G, Sotirova V, et al. Mapping of primary congenital lymphedema to the 5q35.3 region. Am J Hum Genet 1999; 64: 547-55.

9 Irrthum A, Karkkainen MJ, Devriendt K, Alitalo K, Vikkuka M. Congenital hereditary lymphedema caused by a mutation that inactives VEGFR3 tyrosinase kinase. Am J Hum Genet 2000; 67: 295-301.

10 Karkkainen MJ, Ferrell RE, Lawrence EC, et al. Missense mutations interfere with VEGFR-3 signalling in primary limphoedema. Nat Genet 2000; 25: 153-9.