PROGRESSIVE HEMIFACIAL ATROPHY

Progressive hemifacial atrophy often referred to as Romberg's disease (or Parry Romberg Syndrome), is a pathological process involving progressive wasting of the skin, subcutaneous fat, muscle and occasionally, bones of the face. It was first described by Parry in 1825 and by Romberg in 1846. Eulenburg described the entity as "progressive facial hemi atrophy" in 1871.

INCIDENCE

Incidence of progressive hemifacial atrophy is unknown but is relatively rare. Most plastic surgeons see less than a dozen cases in their working lifetime. It is unilateral in 95% of cases and is seen more commonly in females than males with a ratio of 1.5:1.

PRESENTATION

The onset is slow and progressive and begins usually during the first two decades of life, more often between the ages of 5 and 15. The progression of the atrophy usually lasts from 2 to 10 years, following which the process seems to enter a stable or "burn out" phrase. Clinically, subcutaneous tissue and skin are initially affected. The earliest signs include subcutaneous wasting in the malar or lateral mental regions, but may begin in the brow and paramedian forehead (coup de sabre). Pigmented, atrophic skin is another sign of the disease in its early stages. Later, as the disease progresses, facial musculature may atrophy and bone hypopasia may ensue. The facial skeleton is more likely to be affected when the disease onset is in the first decade of life. When the disease is peri-orbital, it can be associated with ophthalmic manifestations, which can include papillary disturbances, exudative neuroretinopathy and optic nerve dysfunction. There can be underlying spasm of involved muscles (esp. the masseter) and there can also be CNS manifestations including learning difficulties, multiple migraine headaches and even epilepsy.

DIFFERENT DIAGNOSIS

Linear scleroderma "en coup de sabre" (or localized linear scleroderma) is a well recognized entity by dermatologists and rheumatologists. These specialists are less likely to see Romberg's disease which tend to present to plastic or caniofacial specialists. There has been a debate (especially in the early German literature) about the connection of hemiatrophy and scleroderma for many years. This is important because most physicians believe that the course of scleroderma can be affected by systematic treatment. It is the authors' belief that the 2 conditions are separate. Clinically, linear scleroderma may present in childhood and it involves intense loss of subcutaneous fat with ensuing thinning and pigmentation of the skin. It is commonly seen in the paramedian forehead region, hence "en coup de sabre". If it overlies skull suture lines, these may be slow to close (or remain unclosed). However, one does not usually see underlying muscle or bone atrophy. The disease is more likely to start later in life and be progressive after the second decade. Patches of peripheral scleroderma occurring elsewhere on the body (usually the trunk) will clinch the diagnosis. Ocular manifestations are usually limited to atrophy of fat leading to enophthalmos but not neuro-ophthalmological problems. Underlying CNS manifestations are rare.

Historically, chronically inflammation and scarring can be seen in skin specimens in both Rombergs and scleroderma. However, the dermal sclerosis is said to be more intense with scleroderma and in Rombergs disease the elastin fibres are said to be preserved. In addition, there is apparently prolonged nerve conduction in areas affected by scleroderma which do not exist in Rombergs. Anti-nuclear anti-body titres are often raised with active linear scleroderma, but rarely so with Romberg's disease.

AETIOLOGYS

There have been a number of theories on the etiology of Romberg's disease and these include trigemial neuritis, a chronic autoimmune neurovasculitis, a chronic infestation with a neurotropic virus (eg. Herpes) and an increased sympathetic nerve activity triggering facial atrophy. A similar condition in rats has been initiated by performing a cervical sympathectomy during the new born. There are enough differences both clinically and histologically to make a strong case for two different diseases. The paediatric rheumatologists tend to treat active linear scleroderma with anti-autoimmune agents such as Steroids, D-penicillamine and Methotrexate. There is anecdotal evidence that the rapid progress of the disease is slowed and the" burn out" earlier. There seems to be no evidence that systemic treatment for the true Romberg's disease pattern of illness is of benefit.

TREATMENT

The treatment of true Romberg's disease is symptomatic. The timing of any surgical intervention is generally agreed to be the best following exhaustion of the disease course and completion of facial growth. Photographic documentation and confirmation that the disease is actually burnt out is actually important and most authors concur on a waiting period of one or two years before proceeding with reconstruction. There is good anecdotal evidence of melting away of any reconstructive efforts if they are performed during the active period of the disease. Other authors (Wolfe) have suggested that bringing in a good blood supply from another area can interrupt the progression of the atrophy. They therefore suggest that it is not best to wait until the condition is quiescent but to move earlier and to rely on any non-vascularised grafts. Again, the evidence for this rationale is purely anecdotal. Facial contour restoration employing a myriad of reconstructive techniques, have been described including fat grafts, dermis fat grafts, lipo injections, pedicle flaps, bone and cartilage grafts, microvascular free tissue transfer, orthognathic corrective maxillary surgery and alloplastic implants. Probably all of these have a place if properly utilized in the appropriate individual case. For muscle spasm associated with Rombergs disease, Botox has been used with success and there have been anecdotal reports of improvement if free vascularised tissues are bought into the adjacent area. CNS manifestations are treated symptomatically but said to be somewhatrefractory.

FAT GRAFTS AND DERMIS GRAFTS

Fat grafts have been described for use in contour reconstruction of the face since van der Muelen's account in 1889. Multiple reports over the past century- Neuber 1893, Lexer 1910, Peer 1956, Ellenbogen 1986, Billings and May 1989, have documented excellent aesthetic results. Fat auto grafts are said to vascularise at day 4 and after a process of inflammation and fibrosis, resorption occurs until roughly 40 to 70% of the graft remains in patients followed from 1 to 3 years. Dermis fat grafts have also been employed with excellent contour restoration and survival. Fat and dermis fat grafts would appear to be appropriate for mild to moderate contour deformities of the hemiface in Romberg's disease.

LIPOINJECTION OR LIPOFILLING

Microlipoaspiration and lipoinjection is a process by which fat is harvested from one area of the body (usually the periumbilical region, thigh or buttock) through a thin cannula and injected into the region contour deformity. There are variations on this technique which include direct harvest and injection, centrifugation of the harvest fat with the decanting of the supernated blood and serum and removal of the excess oil before mono or multiplanar subcutaneous injections. Sometimes, various hormones or cytokines have been added to the fat (eg. insulin or insulin like growth factor) in the hope that this will give a better take. Fat grafts transferred by lipoinjection are subject to the same local resorbable factors as those described by standard fat grafts and a resorption rate of up to 50% has been noted. Certainly there seems to be more success with the Coleman technique where the fat is carefully harvested, centrifuged and injected without pressure into thin (1.5mm) tunnels of well vascularised recipient tissue. A limitation of these techniques is that a small to moderate volume of fat only can be injected into any given spot, as large volumes will jeopardize revascularisation. The benefits however, are that the process can be repeated multiple times and there is minimal donor site morbidity. Again, like fat and dermis fat grafting, small to moderate contour defects can be treated with this technique and have found it useful in linear scleroderma as well.

FREE TISSUE TRANSFER

Facial contour restoration with free transfer was first described by Harashina in 1977. Following his description of free groin transfer others have used omentum, parascapular, rectus abdominus, latissimus dorsi flaps. Unlike fat grafting before them, the free vascularised flaps transferred for facial contour reconstruction provided large volumes of reliable tissue. Disadvantages of the free tissue reconstruction include donor site scarring and morbidity, recipient site scars, longer operative times and hospital stays, plus a definite incidence of flap loss.

(In the above passage Dr Holmes quotes many surgeons over the years that have performed the above surgery.)

The more commonly used flaps are the groin flap. These have the advantage of including relatively large volumes of subcutaneous tissue and fascia whilst leaving relatively acceptable donor site scars. The Para scapular flap in particular provides the benefit of a long, reliable pedicle and a large subcutaneous facial territory that can be used as an extension from the flap skin island. The skin paddle can be de-epithelialised and used in location of the greatest contour deformity. The facial flaps with time has led to better techniques of fixating the soft tissues to bone and using dissected septae when using more fluid flaps such as omentum.

ORTHOGNATHIC SURGERY

In some cases, constriction of growth by the atrophy process in younger patients lead to considerable deformity of the jaws in the long term. Just as for hemifacial microsomia, orthognathic surgery can be of great benefit in correcting the underlying skeletal abnormality before soft tissue flaps are added.

ALLOGRAFTS

Injectable allografts such as silicone have been tried and are now officially rejected by reputable plastic surgery bodies. Hydroxyapatite has also been used but it is difficult to contour and is expensive, the new soluble and moldable Hydroxypatite (bone source) has been remarkable in craniofacial surgery.