Friday, September 24, 2010

Superficial and Deep Perivascular Reaction Pattern


Sometimes when you look at a slide the first thing you notice is that there is an infiltrate around vessels which can be superficial or deep or a combination of both.   Usually the cells that you will see around the vessels are lymphocytes and the condition that is most associated with a superficial perivascular lymphocytic infiltrate is an annular erythema.  Note that this is going to be one of the red non-scaly diseases because the epidermis in general isn't involved and so you won't get any scale. The classic is erythema annulare centrifugum but of course you already realise that this can present sometimes with epidermal involvement, especially if there is a little bit of spongiosis histologically. If you just have the deep perivascular infiltrating form then that will present as a red non-scaly annular rash.  Generally in EAC the perivascular infiltrate is quite tight around the superficial dermal plexus vessels and that is also a good marker for this condition.

If the vessel involvement is both superficial and deep then you go looking for areas of damage elsewhere.  The first thing you do is look at the epidermis and see if there is any involvement.  As I said a small degree of spongiosis can be a feature of EAC but it can also be a feature of pityriasis rosea.  In the latter, you would look for evidence of some red cell extravasation, particularly in the papillary dermis, you would look for some psoriasiform hyperplasia of the epidermis and a bit of parakeratosis as well corresponding with the scale.

In erythema multiforme you would expect to see damage at the basement membrane along the basal layer.  Is there any sign of vacuolisation due to a lymphocytic infiltrate along the basal layer causing damage to some of the basal layer cells.  In very mild forms of these conditions the perivascular infiltrate is what may catch your eye, but these other features should be present to allow you to make this diagnosis.

If you have a marked superficial and deep perivascular infiltrate then the other condition to think about is lupus erythematosis, though again look to the basement membrane, look to the area around hair follicles and see if there is vacuolar damage there.

Drug eruptions may also present as a perivascular infiltrate and these eruptions can be urticarial, eczematous or true morbilliform eruptions.  In these circumstances you will usually see some eosinophils and you will then look to see if there are any changes elsewhere. In urticarial drug reactions there will usually be some papillary dermal oedema and in eczematous ones there will obviously be some epidermal hyperplasia and spongiosis. In those giving a morbilliform eruption the infiltrate is mainly going to be lymphocytic with much reduced numbers of eosinophils and neutrophils and there may be some basal layer vacuolisation.

Arthropod bite reactions can give quite a marked perivascular infiltrate, both superficial and deep, but the eosinophils are the prominent feature in arthropod bite reactions.  Some of the other changes that you may see are the result of scratching the skin and that can include a bit of ulceration.  Epidermal necrosis can also occur at the actual punctum site where the insect has bitten through the epidermis.

Rarely secondary syphilis can present as a predominantly perivascular infiltrate but you should see some evidence of plasma cells in this condition.  Syphilis is one of the great imitators so again if there is any scale associated with the lesion then there is likely to be some form of either psoriasiform hyperplasia or spongiosis.

The last condition to consider with perivascular infiltrates predominating is pigmented purpuric eruption.  Here to get the hemosiderin pigmentation you have to have a degree of capillaritis with leakage of red blood cells.  It is the breakdown of the red blood cells that leaves iron pigment behind that gives both the clinical picture and the histological picture.  If you look carefully at the blood vessels in pigmented purpuric eruption, there may be a bit of endothelial cell swelling and some perivascular thickening. 

Stasis dermatitis is similar to pigmented purpuric eruption but here the other prominent feature is vascular ectasia and proliferation of vessels.  Again there may be marked red cell extravasation but also there can be an increasing fibrosis of the dermis and this is reflected clinically in the firm skin that is often seen in patients with stasis dermatitis, sometimes progressing to frank lipodermatosclerosis.

Lesions that only have a perivascular infiltrate and no epidermal involvement should not have any scale so they are going to present as red non-scaly eruptions.  The other thing to remember is that the histology depends on how early a lesion is biopsied, because as Bernard Ackerman famously said "lesions have lives" and what you see histologically depends on how early or how late the lesion actually is.  For instance, in PLEVA (pityriasis lichenoides et varioliformis acuta)  In the very early stage, you are going to see primarily a perivascular infiltrate, but if you look carefully later on as some epidermal involvement occurs then you are going to see basal vacuolar change, you are going to see a degree of parakeratosis and epidermal hyperplasia.  You may see some epidermal cell necrosis as well.  Occasionally in severe cases you will get some red cell extravasation, so it is the mixture of pathologies that allows you to come to the diagnosis of pityriasis lichenoides.

Urticaria pigmentosa can be a difficult diagnosis to make histologically.  You may predominantly see a perivascular infiltrate but when you look closely the cells aren't lymphocytes! They are in fact mast cells.  Generally these mast cells are oval in shape, sometimes they can look spindle –like.  So if you see these either oval or spindle shaped cells in a perivascular distribution especially in the upper capillary dermis then consider mastocytosis and do mast cell stains to try and delineate the mast cells better.