BritishJournal ofOphthalmology 1994; 78: 14-18

Exenteration for benign orbital disease

Geoffrey E Rose, John E Wright

AbstractExenteration, or removal ofthe globe with partor all of the surrounding orbital contents, isgenerally reserved for malignancy. The pro-cedure may, however, be of value in themanagement of some benign orbital diseases.The indications for exenteration in manage-ment of benin orbital disease are threefold.Firstly, patients in whom diffuse disease, suchas idiopathic inflammation, has resulted in anirretrievable situation of visual loss and clinic-ally uncontrollable pain or disfigurement; inmany such cases we consider exenteration tobe preferable to the (often severe) side effectsof prolonged and inadequate medical therapy.The second group are those patients withgrossly disfiguring orbital abnormalities, suchas teratomas, extensive varices, or massiveoptic nerve tumours. The last group comprisespatients with tumours that, while histologicallybenin, may have malignant potential or showa tendency to diffuse or persistent infiltrationoforbital soft tissues. Sixteen illustrative cases

of full or partial exenteration for benigndisease are described.(BrJ Ophthalmol 1994; 78:14-18)

Enucleation is removal of the globe alone,whereas orbital exenteration refers to enuclea-tion combined with removal of other orbital softtissues - principally lacrimal gland, extraorbitalmusculature, and orbital fat.The relative safety of surgery in experienced

hands and the good aesthetic rehabilitation afterexenteration (particularly where maximumeffort is made to preserve the eyelids) makes theprocedure suitable for the treatment of someforms of benign orbital disease. The indicationsfor, and results from, such therapy are presentedin this paper.

Patients and methodsPatients having either full or partial exenterationfor benign orbital conditions were identified

... ... . .:: .:

::.:..:::*.. X.: :.

..' :&.

*: .:

*.;. ::i:i.

Figure I Residual tissue mass at apex oforbit (arrows)preventing leakage ofcerebrospinalfluid afterfullexenteratwn.

Moorfields Eye Hospital,LondonG E RoseJ E WrightCorrespondence to:Mr G E Rose, OrbitaVAdnexalClinics, Moorfields EyeHospital, City Road, LondonECIV 2PD.

Accepted for publication28 July 1993

Figure 2 Split thickness skin grafts applied to anexenteratwn socket; these results in a rapidly healhng, but deep,cavity.

Figure 3 Patient at 3 weeks after exenteration and partialmaxillectomy; direct closure ofthe eyelid skin and orbicularismuscle produces a rapidly healing and shallow cavity.

Figure 4 Forehead pedicle rotationflap used to cover anextended orbital exenteration cavity; the donor site wascovered with split thickness skin grafts.

14

..... ...

on Novem

ber 29, 2020 by guest. Protected by copyright.

http://bjo.bmj.com

/B

r J Ophthalm

ol: first published as 10.1136/bjo.78.1.14 on 1 January 1994. Dow

nloaded from

Exenterationfor benign orbital disease

Figure 5 Exenterationsocket healing by secondaryintention (A); although theslowest method ofhealing, ittypically results in a shallow,epithelial lined cavity (B).

FigSB

;..

.. ...b . t: . t . ' 4,..j_ . " ' ...... .-.','._

.,

.'4, S:

:X:

.sFig 6B

.s2'2

Fig6D

Figure 6 Broad rimmed spectacles provide usefulcamouflagefor the edge oforbital prostheses (A). Theorosthesis may be attached to the exenteration cavity by

osseointegrated(D).

Age at Visualsurgery Sex Pathology Reasonsfor exenteration acuity Type ofexenteratin12 Years M Hemifacial vascular Uncontrolled recurrent orbital 6/36 Partial, with preservation of lids

malformation haemorrhage and margins13 Years F Massive orbital varices Uncontrolled recurrent orbital 6/36 Partial, with preservation of lids

haemorrhage and margins20 Years M Orbital arteriovenous anomaly Uncontrolled recurrent orbital CF Partial, with preservation of lids

with inflammation haemorrhage and marns2 Weeks F Orbital teratoma Cosmetic Unknown Partial, with preservation of lids

and margins

FigSA

Fig 6A

adhesives (B) or by magnetsfixed to osattachments (C) and to the prosthesis

Fig6CTable I Details offour patients undergoing exenteration for massive benign structural atnormalities

15

."/ .. -,

.11i1 ..I-,.ii.

S..-t

Adie

....

....

on Novem

ber 29, 2020 by guest. Protected by copyright.

http://bjo.bmj.com

/B

r J Ophthalm

ol: first published as 10.1136/bjo.78.1.14 on 1 January 1994. Dow

nloaded from

Rose, Wright

from the records of the orbital clinic at Moor-fields Eye Hospital. Of 17 such patients withknown benign disease, complete clinical noteswere available for review in 16 cases.

Exenteration can be achieved by dissection ofthe tissue planes central to the orbital periosteum(leaving periosteum on bone, usefully termed'partial exenteration'), or in the plane betweenthe orbital periosteum and the bone. The lattertechnique, with preservation of an intact peri-

Figure 6E The prosthesismay, altematively, bespectacle mounted.

Figure 7 Recrrent orbitalhaemorrhage (A)from avascular malformation thatinvolved the orbit, maxilla,and hard palate (B). Orbitalexenteration and partialmaxillectomy was required tocontrol the condition. rig/A

F-p.

osteum around the exenteration specimen ('full'exenteration) is the favoured treatment forextensive malignancy within the orbit or eyelids.The eyelid complex, in part or in total, may be

saved during exenteration'-5 and a surgeon'sefforts should be so directed, especially withbenign diseases, to thereby hasten and improveaesthetic rehabilitation. It is neither practicable,nor advisable, to remove all tissue from theorbital apex (Fig 1) because the risk of cerebro-spinal fluid leak is higher if apical tissues areresected.

After removal of the orbital soft tissues, thecavity may be grafted with split thickness skin(Fig 2), covered by direct closure of the lid skin(Fig 3), or a rotation flap from a neighbouringarea (Fig 4), or may be left to heal by secondaryintention (Fig 5A). Although healing by second-ary intention is the slowest therapy and can takemany months, granulation leads to a shallow,saucer-like cavity (Fig 5B), rather than the deepcavity typical of skin grafting. For lid sacrificingexenteration, rarely necessary for benigndisease, we favour closure of eyelid skin or,wherever this is not possible, a healing of thecavity by secondary intention.The majority of patients with benign disease

can be treated by lid sparing exenteration, withlater fitting ofan artificial eye. Aesthetic rehabili-tation can otherwise be achieved by a suitableorbital prosthesis; the discontinuity betweena prosthesis and the periorbital tissue maybe conveniently hidden by a broad rimmedspectacle frame (Fig 6A). The prosthesis may beattached directly to the cavity, by the use ofadhesives (Fig 6B) or osseointegrated magneticattachments (Fig 6C, D), or may be carried onthe spectacle frame (Fig 6E).

ResultsPatients with benign orbital disease underwentexenteration for three principal reasons - formassive structural lesions, for disease withuncontrollable pain, or for extensive benigndisease or tumours.

MASSIVE STRUCTURAL LESIONSMost patients within this group had largevascular anomalies with recurrent haemorrhageand gross proptosis, although one patient had alarge orbital teratoma (Table 1). Lid sparingexenteration was possible in all of these cases; inone case the malformation extended throughoutthe orbit and maxilla (Fig 7) and requiredextensive midfacial surgery at another hospital.The teratoma was resected at another hospital(for neonatal anaesthetic care) and the child isable to wear a socket conformer 2 years later(Fig 8).The disease has been controlled in all four

cases, although two patients have problems withthe fitting and stability of their artificial eyes.

UNCONTROLLABLE PAINFour patients underwent exenteration forchronic idiopathic sclerosing orbital inflamma-tion and one had surgery for a massive recur-Fig 7B

16 on N

ovember 29, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.78.1.14 on 1 January 1994. D

ownloaded from

Exenterationfor benign orbital disease

Figure 8 Orbital teratomain a neonate (A); the child isable to retain a socketconformer 2 years after lid-sparing exenteration (B).

Fig8A Fig8B

Table 2 Details offive patients undergoing exenteration for chronic severe pain uncontrolled by conventional medical therapy

Age atsurgery Visual(years) Sex Pathology Reasonsfor exenteration acuity Type ofexenteration

11 F Chronic/fibrotic inflammation Uncontrolled pain and disease. NPL Partial, with preservation of lidsSteroid side effects and margins

75 M Chronic/fibrotic and necrotising Uncontrolled pain and disease. Bare PL Partial, with preservation of lidsinflammation - Steroid side effects and margins

30 F Chronic/fibrotic inflammation Uncontrolled pain and disease. NPL Full, with closure of lid skin overwith vasculitis Steroid side effects cavity

67 M Chronic/fibrotic inflammation Uncontrolled pain and disease 6/60 Full, with lid skin reflected intothe cavity

60 F Extradural meningioma Uncontrolled pain and disease NPL Full, with preservation of lidsand margins

NPL=no perception of light, PL=perception of light.

Figure 9 Gross idiopathicorbital inflammatory diseasein a 1-year-old girl.Prolonged systemic steroidtherapy, which failed tocontrol either the pain or thedisease, resulted in hirsutism,pyelonephritis, hypertension,and widespread cutaneousviral papillomas (arrow).

rence of an orbital meningioma. All had severepain (such as to cause two patients to contem-plate suicide) and progressive disease which was

uncontrolled by systemic steroids or radio-therapy (Table 2). Three had serious side effectsfrom prolonged steroid therapy; these includedgastric surgery for ulceration (one case), hyper-tension (two cases), Cushingoid facies andhirsutism (two cases), chronic pyelonephritis(one case), and widespread cutaneous viral papil-lomas (Fig 9).

Pain control was almost immediate afterexenteration in patients with orbital inflamma-tion; at follow up the disease appears inactive inall patients and drug therapy has ceased in three.The patient with orbital meningioma hadimproved comfort and cosmesis (being able to

wear an artificial eye), but had mild continuedpain due to an intracranial tumour.

Table 3 Details ofseven patients with widespread and uncontrolled benign orbital disease requiring exenteration

Age atsurgery Visual(years) Sex Pathology Reasonsfor exenteration acuity Type ofexenteration

42 F Optic nerve meningioma with Massive proptosis with exposure of NPL Partial, with preservation of lidsfibrosis globe and margins

43 F Optic nerve meningioma Massive proptosis with exposure of NPL Partial, with preservation of lidsglobe and margins

15 F Optic nerve glioma Massive proptosis with exposure of NPL Partial, with preservation of lidsglobe and margins

72 M Chronic/fibrotic and necrotising Massive proptosis with exposure of NPL Partial, with preservation of lidsinflammation globe and margins

27 M Recurrent lacrimal gland Infiltrating and extensive tumour 6/9 Full, with sacrifice of eyelidspleomorphic adenoma recurrence

37 M Fibrous histiocytoma, Infiltrating and extensive tumour NPL Full, with preservation of lidsrecurrence and margins

21 F Haemangiopericytoma Infiltrating/extensive tumour with 6/12 Full, with sacrifice of eyelidsmalignant potential

NPL=no perception of light.

17

......

on Novem

ber 29, 2020 by guest. Protected by copyright.

http://bjo.bmj.com

/B

r J Ophthalm

ol: first published as 10.1136/bjo.78.1.14 on 1 January 1994. Dow

nloaded from

Rose, Wright

IiEW,~~~~Q~.....

Figure 10 A massive optic nerve glioma which resulted inexposure ofthe globe and subsequent exenteration.

Figurel11 Extensivebenign,but diffutsely infiltrative,haemangiwperscytoma requiring orbital exenteration.

UNCONTROLLED BENIGN DISEASE WITHOUTr PAIN

Uncontrolled, but painless, disease was our

conmnonest single indication for partial (four) or

full (three) exenteration (Table 3), Gross propto-

sis, due to large retrobulbar masses, had resulted

in exposure of the globe in 4/7 cases (three opticnerve tumours, one orbital inflammation; Fig10). The other three patients underwent full

(periosteal removing) exenteration for widelyinfiltrative benign tumours with malignant

potential one haemangiopericytoma (Fig 11),one fibrous histiocytoma, and -one recurrent

pleomorphic adenoma.

Lid sparing exenteration was performed in

five patients (with fitting of artificial eyes) and

the two others were fitted with adhesive mounted

Table 4 Comparison ofother series

Proportion forReference benign disease Benign orbital diseases requiring exenteration

Levin and Dutton 16/99 (16%) Mucormycosis (5) Haemangiopericytoma (2)(1991)6 Orbital cellulitis (1) Fibrous histiocytoma (1)

Lymphangioma (1) Schwannoma (1)Meningioma (3) Benign lymphoid disease (1)Lacrimal gland pleomorphic adenoma (1)

Bartley et al (1989)' 3/102 (3%) Mucormycosis (1)Wegener's granuloma (1)Severe facial burn (1)

Rathbun et al (1971)' 3/48 (6%) Recurrent pleomorphic adenoma (2)*Progressive granuloma (1)

Naquin (1954)' 8/48 (17%) Meningioma (2)Pleomorphic adenoma (2)*Progressive granuloma (3)Tuberculous periostitis (1)

*Possibly malignant mixed tumours of lacrimal gland.

(one) or spectacle mounted (one) orbital pros-theses.

DiscussionExenteration for benign orbital disease wouldnot appear to have been specifically addressed inthe ophthalmic literature, although cases havebeen included within several reviews of orbitalexenteration.' The proportion ofbenign diseasevaries between the series, but the spectrum ofbenign disease overlaps with our experiencedescribed in this paper (Table 4); the presentpaper intentionally excludes several patientswith major periorbital fungal infections (mucoror aspergillosis), such cases being referred tocolleagues for consideration of extensive cranio-facial surgery.

Bartley and associates suggest that orbitalexenteration may be useful in non-malignantconditions, such as neurofibromatosis with ablind eye, in advanced mucormycosis, or withthe severely contracted or painful socket.7Similarly, exenteration has been described in thetreatment of mucormycosis,) recurrent menin-gioma,5 severely contracted sockets,'0 orbitalneurofibromatosis," and in a case of deep orbitalmyiasis. 12

In our experience, full or partial exenterationwith sparing of the complete eyelid complex,wherever possible, has proved valuable in themanagement of advanced or poorly controlledbenign orbital disease, particularly where visionhas been severely damaged and chronic peri-orbital pain can be controlled only with systemicsteroids or other immunosuppressive drugs.Cure by exenteration should be considered pre-ferable to side effects from such long termmedical therapy. Preservation of the completeeyelids should be attempted in all cases (possiblein 12/16 of this series) because it speeds both thehealing and aesthetic rehabilitation of suchpatients.We thank Mr Alan Mushin and Mr Richard Collin for theopportunity to examine patients under their care at Moorfields EyeHospital (Fig 8A [AM] and 2, 6B-D, 8B [JROC]).

1 Coston TO, Small RG. Orbital exenteration - simplified. TransAm Ophthalmol Soc 1981; 79: 136-52.

2 Baylis H, Shorr N, McCord CD Jr, Tanenbaum M. Eviscera-tion, enucleation and exenteration. In: McCord CD,Tanenbaum M, eds. Oculoplastic surgery. New York:Raven Press, 1987: 419-24.

3 Rootman J. Orbital surgery. In: Rootman J, eds. Diseases oftheorbit. A multidisciplinary approach. Philadelphia: Lippincott,1988: 603-4.

4 Shields JA. Diagnosis and management of orbital tumours.Philadelphia: Saunders, 1989: 63-4.

5 Shields JA, Shields CL, Suvarnamani C, Tantisira M, Shah P.Orbital exenteration with eyelid sparing: indications, tech-nique and results. Ophthalmic Surg 1991; 22: 292-7.

6 Levin PS, Dutton JJ. A 20-year series of orbital exenteration.AmJ Ophthalmol 1991; 112: 496-501.

7 Bartley GB, Garrity JA, Waller RR, Henderson JW, IlstrupDM. Orbital exenteration at the Mayo Clinic, 1%7-1986.Ophthalmology 1989; %: 468-74.

8 Rathbun JE, Beard C, Quickert MH. Evaluation of 48 cases oforbital exenteration. AmJ Ophthalmol 1971; 72: 191-9.

9 Naquin HA. Exenteration of the orbit. Arch Ophthalmol 1954;51: 850-62.

10 Small RG, Lafuente H. Exenteration of the orbit in selectedcases of severe orbital contracture. Ophthalmology 1983; 90:236-8.

11 Jackson IT, Laws ER Jr, Martin RD. The surgical manage-ment of orbital neurofibromatosis. Plast Reconstr Surg 1983;71: 751-8.

12 Kersten RC, Shoukrey NM, Tabbara KF. Orbital myiasis.Ophthalmology 1986; 93: 1228-32.

18 on N

ovember 29, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.78.1.14 on 1 January 1994. D

ownloaded from