humanchorioretinal biopsy under controlled systemic hypotensive anaesthesia · hypotensive...

British Joutrnial of Ophthalmology, 1980, 64, 559-564

Human chorioretinal biopsy under controlledsystemic hypotensive anaesthesiaI. J. CONSTABLE, G. H. CHESTER, R. HORNE, AND J. F. HARRIOTTFrom the University of Western Australia and Royal Perth Hospital, Perth,Western Australia

SUMMARY This paper describes a simplified technique for biopsy of the retina and choroid whichhad been used in 5 human volunteers. The biopsy was carried out in 4 immediately before enuclea-tion of an eye for malignant melanoma and in 1 patient who was undergoing trabeculectomy forpainful glaucoma associated with retinitis pigmentosa. A combination of intravenous mannitol andtransient controlled systemic hypotension, induced under general anaesthesia with intravenoussodium nitroprusside, was used in 3 cases and resulted in no vitreous loss and minimal bleeding.In the 2 cases in which hypotension was not used bleeding was a definite problem, but no vitreousloss was experienced.

Understanding of many types of chorioretinalpathology has long been retarded by the lack of apractical biopsy technique. Previous workers havedescribed a full-thickness eye wall method forbiopsy of the posterior segment in rabbits, monkeys,and 1 human.' A transvitreal approach to retinalbiopsy has also been reported in animals.4 Thesemethods have not been widely adopted because ofthe complexity of the procedure and of the universalvitreous loss reported. A simplified trans-scleralapproach to chorioretinal biopsy in dogs usinghypotensive anaesthesia resulted in minimal or novitreous loss,5 so that we were encouraged to applythe procedure to 5 human volunteers.

Patients and methods

CASE 1A 35-year-old male presented to the neurologyoutpatients with a 6-week history of blurred visionin the left eye. Vision was 6/6 right, countingfingers left, and a superior field defect was notedin the left eye. On referral to the ophthalmologydepartment a large serous detachment of the inferiorretina in the left eye was noted as well as an under-lying pigmented mass that had the clinical charac-teristics of a large melanoma of the choroid.Fluorescein angiography, 32P isotope testing, andultrasonography all suggested a melanoma of the

Correspondence to Professor 1. J. Constable, UniversityDepartment of Ophthalmology. Royal Perth Hospital,Wellington Street, Perth 6000, Western Australia.

choroid. The lesion was judged to be too large forlocal radiotherapy or photocoagulation and enu-cleation was advised. A systemic survey for metas-tases was negative. Appropriate patient consent wasobtained. A local chorioretinal biopsy in the uppertemporal quadrant in the area of attached retinawas planned immediately before enucleation of thepatient's left eye. A spindle cell B melanoma wasconfirmed.

CASE 2A 45-year-old female was referred with a 3-monthhistory of blurred vision in the right eye in August1978. Her general health had been previouslyexcellent. On examination there was a large collar-stud mass occupying the midperipheral area of theinferior retina. It was pigmented and approximately6 disc diameters in size. There was a surroundingserous detachment which extended to the fovea.Fluorescein angiography, 32P uptake, and ultrasoundwere all positive for melanoma. Vision was reducedto 6/12 right eye. The left eye had 6/5 vision and noabnormalities. The mass was considered too largefor local treatment, and enucleation of the righteye was recommended. On 8 September 1978, withappropriate patient consent, a chorioretinal biopsywas carried out before right enucleation. Histologyconfirmed an epithelioid melanoma of the choroid.

CASE 3A 45-year-old man presented to the ophthalmologydepartment in February 1979 with a 4-monthhistory of blurred vision centrally in the left eye.

559

on March 3, 2021 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.64.8.559 on 1 A

ugust 1980. Dow

nloaded from

http://bjo.bmj.com/

I. J. Constable, G. H. Chester, R. Horne, and J. F. Harriott

His general health had been previously excellent.Vision was 6/5 right, 6/60 left. There was a largeraised pigmented mass 4 disc diameters in area atthe posterior pole of the left eye. It was surroundedwith a thin layer of subretinal fluid. Fluoresceinangiography and 32P uptake were both positive formelanoma. No evidence of systemic metastases wasfound on routine surveys. Enucleation was recom-mended for the left eye because of the size, position,and surrounding subretinal fluid associated with thetumour. Permission was obtained for chorioretinalbiopsy, and this was carried out under generalanaesthesia immediately before enucleation on 14March 1979. A spindle cell B melanoma wasconfirmed.

CASE 4A 70-year-old man was found to have a raisedpigmented mass in the inferonasal quadrant of theright eye on routine examination by his ophthal-mologist in 1974. This was followed at regularintervals until May 1979, when the patient presentedwith sudden loss of vision in the right eye associatedwith vitreous haemorrhage. The patient had beenadmitted to hospital with a myocardial infarct inJanuary 1978 but had made an excellent recoveryand has now considered to be in reasonable generalhealth. Enucleation of the right eye was advisedafter consultation with several ophthalmologists,and permission was obtained to carry out a chorio-retinal biopsy immediately before this procedure.In this case controlled hypotensive anaesthesia waswithheld because of the patient's age and historyof ischaemic heart disease. Pathology confirmed a

spindle cell B melanoma.

CASE 5A 68-year-old woman presented to the ophthal-mology outpatient department in July 1978 with a

painful blind right eye and vision reduced to count-ing fingers in the left. The patient had a history ofretinitis pigmentosa with poor night vision andconstricted visual fields by the age of 20. There was

no family history of the disease. The patient was

declared legally blind in 1968 at the age of 57 andwas admitted as a chronic patient to the Braillehospital. She had a 5 year history of mild maturityonset diabetes and was also on digoxin and diureticfor congestive cardiac failure. On examination theright eye had marked central corneal oedema, a

narrowed anterior chamber, and a closed angle.The lens showed moderate nuclear sclerosis but no

signs of intumescence. No view was obtained of herfundus. Intraocular pressure was 50 mmHg right,15 mmHg left eye. The left eye showed nuclearsclerosis of the lens and end-stage retinitis pigmen-

tosa. On gonioscopy the left angle was narrow butopen. The patient was treated for 3 months withpilocarpine drops and systemic acetazolamide. Shecontinued to have intermittent pain in the right eyeand was often distressed with vomiting. She wasadmitted to hospital on 29 September 1978 withcongestive cardiac failure. This responded to in-crease in the diuretic regimen, and the patient wasagain discharged. Treatment for the chronic glau-coma in the right eye was continued, but pain anddiscomfort did not subside. The patient was thenreadmitted on 29 November 1978, at which timethe blind right eye was still painful. There waspersistent right corneal oedema. Intraocular pres-sure was 50 mmHg right, 12 mmHg left. On 1December 1978 a right trabeculectomy was plannedfor the chronic closed-angle glaucoma. Permissionwas obtained from the patient to carry out a chorio-retinal biopsy at the same time. Again no hypoten-sive anaesthesia was used. On the third postoperativeday the patient developed intermittent paroxysmaltachycardia and signs of pulmonary oedema.Despite therapy she failed to respond and died onthe fourth postoperative day.

ANAESTHETIC TECHNIQUEIn cases 1, 2, and 3 intravenous mannitol I g/kgwas given 20 minutes before retinal biopsy to softenthe eye and decrease the vitreous volume. At thetime of biopsy transient systemic hypotension wasalso induced. Preoperative anaesthetic assessmentexcluded cardiovascular and respiratory disease.The electrocardiogram was normal in these 3 cases,as was blood count and chext x-ray. Premedicationwas confined to 25 mg promethazine HCl intra-muscularly 1 hour before induction of anaesthesia.Standard anaesthetic induction included preoxy-genation, thiopentone 300 mg intravenously, muscleparalysis with intravenous suxamethonium pre-treated with D-tubocurarine, and lignocaine sprayto the vocal cords. A cuffed Oxford tube was placedin position, and anaesthesia was maintained withnitrous oxide 3 litres/minute, oxygen 1-5 litres/minute, and 1-2% halothane on a semiclosedcircuit. The ECG was monitored continuously, andbrachial blood pressure was monitored on both armswith an oscillometer. An intravenous line forinfusion of mannitol was placed in the left arm; asecond intravenous line was placed into a dorsalvein in the left foot for infusion of sodium nitro-prusside. This latter intravenous drip was set upwith 30 mg sodium nitroprusside in 500 ml of 5%dextrose. A trial dose of sodium nitroprusside wasinfused early in the operation to test the patient'sindividual response. In case 1 it was found thatsystolic blood pressure was difficult to lower

560



j.com/

Br J O


ugust 1980. Dow

nloaded from

http://bjo.bmj.com/

Hi-ma, cuhorioretitial biopsy luder conitrolled sYstemic hypotenisiv'e alnaesthesica

,4

Fig. I Schematic drawing of chorioretinal hiopszvsite in a hl/man eve.

below 80 mmHg owing to a persistent tachycardiaof 90/minute. The patient was therefore givenpractolol 1 mg intravenously, which reduced thepulse rate to 65/minute. Immediately before thefinal steps in the chorioretinal biopsy the sodiumnitroprusside infusion was allowed to drip slowlyso that the systolic blood pressure was lowered to70 mmHg over 21 minutes. This level was maintainedwith continuous infusion over 5 to 10 minutesduring which time the biopsy was taken. A total of60 to 80 ml of infusion containing 3 6 to 4 8 mgsodium nitroprusside infused at the rate of approxi-mately 8 .tggm/kg/minute was used in each case. Atthe end of the biopsy the sodium nitroprusside dripwas turned off, and blood pressure returned tonormal levels under anaesthesia within 3 minutes.

In cases 4 and 5, because of age and history ofcardiovascular disease, intravenous mannitol andsodium nitroprusside infusion were withheld. Inboth cases the general anaesthesia was induced with150 g of thiopentone after preoxygenation, andintubation was carried out after 10 mg suxametho-nium and lignocaine spray to the vocal cords.Maintenance was with a nitrous oxide/oxygenmixture in a 3 :1 ratio. Intermittent enflurane,droperidol (total 5 mg), phenoperidine (0 5 mg),and alcuronium chloride (10 mg) were given duringthe procedure. Acetazolamide (500 mg) was given

Fig. 2 Intact vitreous Aceafter chorioretinal biopsy. Pr-inttaken from a 16 mm fil/n.

..........

561

......

-,-N



j.com/

Br J O


ugust 1980. Dow

nloaded from

http://bjo.bmj.com/

I. J. Constable, G. H. Chester, R. Horne, and J. F. Harriott

intravenously to lower intraocular pressure duringthe procedure.

SURGICAL TECHNIQUEAfter conjunctival peritomy and placement of bridlesutures a biopsy site was selected in the superiortemporal quadrant posterior to the equator, exceptin case 3, in which an upper nasal site was selected.In each case the site selected was overlying attachedretina on the opposite side of the globe to themelanoma.The exact site was chosen clear of major vortex

Fig. 3 Full-thickness scleral flap closed with interrupted8-0 Vicryl sutures after chorioretinal biopsy.

Fig. 4 Fundus photography ofbiopsy site immediatelyafter closure of the scleral flap and prior toenucleation of the eye in case 1. There is no evidenceof bleeding and no elevation of the surrounding retina.

veins after transillumination of the globe. A full-thickness, circular, 270° scleral flap of 3 mm dia-meter was then raised (Fig. 1). Two 8-0 Vicrylsutures were placed through the flap and the hostbed of sclera to facilitate closure after chorioretinalbiopsy. The site was again transilluminated toselect the area for biopsy in between larger choroidalvessels. At this stage the sodium nitroprussideinfusion was started in the first 3 cases describedand this visibly blanched the choroid. The eye wasalready soft owing to infusion of mannitol 1 g/kgintravenously over the previous 20 minutes. Whensystolic blood pressure was reported by the anaes-thetist to be lowered to a level of 70 mmHg, abiopsy was taken with blunt jeweller's forceps andsmall curved Vannas scissors. The choroid andunderlying retina were slightly tented with the graspof the forceps, and the first cut exposed the vitreousface. After inserting one blade along the vitreousface the biopsy was excised with 2 more cuts. It wasimmediately placed in glutaraldehyde and theinstruments were changed for closure of the wound.In all 5 cases the vitreous face bulged, but none waslost (Fig. 2). In cases 4 and 5, in which mannitoland hypotensive anaesthesia were withheld, therewas excessive choroidal bleeding, which meant thata fragmented biopsy was obtained. The scleral flapwas closed with the preplaced sutures (Fig. 3). Incases 1 to 4 before enucleation of the globe photo-graphs were taken of the biopsy site with a portableKowa fundus camera (Fig. 4). In case 5 the eye wasnot removed but a routine trabeculectomy wascarried out in the superior temporal quadrant undera limbal based scleral flap.

Discussion

The 3 younger patients submitted to transienthypotensive anaesthesia tolerated the procedurewell and showed no postoperative side effects. Case5 unfortunately developed acute pulmonary oedema4 days postoperatively and died. After carefulinvestigation it was concluded that the anaestheticitself and the surgery were unlikely to have been thedirect cause, but that they might have contributedto the final cardiovascular decompensation.Of the 5 biopsy specimens of the choroid and

retina obtained only 3 were entirely satisfactory forelectron microscopy. The specimens tended to rollup on the end of the forceps when placed in fixative,and it was difficult to prevent artefact detachmentof the retina from the pigment epithelium. However,in 3 cases good individual specimens were obtained(Fig. 5). High power examination of these speci-mens confirmed good fixation and good resolutionof detail at the base of the pigment epithelium and

562



j.com/

Br J O


ugust 1980. Dow

nloaded from

http://bjo.bmj.com/

Huiman chorioretinal biopsy under controlled systemic hypotensive anaesthesia

Fig. 5 Electron microscopic appearance of chorioretinal biopsy from case I showing some bending of outer segmentsand slight spreading of the receptor discs. The specimen is quite adequate for interpretation. ( x 1880.)

Bruch's membrane, the receptors and the internallimiting membrane of the retina respectively (Fig. 6).The poor specimens were obtained from the 2 eyesin which hypotensive anaesthesia was not employed.Because of the bleeding it was not possible neatlyto excise the biopsy specimens, and indeed theywere removed piecemeal because of the risk ofcutting into the vitreous face in the presence ofblood. However, individual cell layers could beassessed in these 2 specimens. It is to be noted thatrenal and liver biopsies are often less then ideal.The choroidal bleeding was surprising in the case

of the eye with end-stage retinitis pigmentosa. Ithad been presumed that, because of choroidal andretinal atrophy, bleeding would not be a problem.

Nevertheless even in the cases not subjected toocular and systemic hypotension it was shown thatchoroid and retina could be excised without losingvitreous. It is suggested that, if the problems ofhandling such minute specimens of choroid andretina can be overcome, this technique may havewidespread application, particularly in young

patients in whom controlled hypotensive anaes-thesia can be employed without significant risk. Inthis context it is to be noted that sodium nitro-prusside infusion is used to induce transient hypo-tension in a wide variety of surgical fields, notablyvascular, cerebral, and cardiac surgery. In theseareas the hypotension is often maintained for longperiods without ill effects. In the case of retinalbiopsy it is necessary for only 5 to 10 minutes.

It is obviously of great importance to determinewhether excision of a 1 mm piece of choroid andretina without disturbance of the vitreous face wouldlead to retinal detachment in humans in the shortterm or long term. To date, the definitive experi-ment, which would require a volunteer to submit ablind eye perhaps from optic atrophy to the pro-cedure, has not been carried out, It is noteworthy,however, that retinal detachment must be extremelyrare in patients with retinal dystrophies, as we havebeen unable to extract any reports from the litera-ture or from anecdotal sources to date. It is possiblein those patients who have widespread pigment

563



j.com/

Br J O


ugust 1980. Dow

nloaded from

http://bjo.bmj.com/

r

Fig. 6 Composite higher-power electron micrographs of

human chorioretinal biopsyspecimen. (A) Basal section of

retinal pigment epithelial (rpe)cell (x 8640). (B) Apical sectionof rpe cell (x 7808). (C) Junctionof inner and outer segments ofrods (x 10 000). (D) Innersurface of retina showinginternal limiting membrane(x 11 200).

D

epithelial disturbances that the risk of retinaldetachment is minimal because of pathologicalretinal adhesion.

If this simplified chorioretinal biopsy techniquecan be shown in a large number of cases to be freeof the dual risks of massive haemorrhage andsignificant vitreous loss, then a wide range oflaboratory techniques could be employed to eluci-date chorioretinal diseases at present poorlyunderstood.

We thank Dr D. C. Wilson for referring two of the cases,

Mr C. Barry for clinical photography and Miss H. Deadyfor secretarial help.

References

'Peyman GA, Meisels HI, Batko KA, Vlchek JK. Full-thickness eye wall biopsy. I. An experimental approach tothe tissue diagnosis and study of choroidal and retinallesions. Invest Ophthalmol 1975; 14: 484-7.2Peyman GA, Homer P, Kasbeer R, Vlchek J. Full thicknesseye wall biopsy. II. In primates. Invest Ophthalmol 1975;14: 565-7.3Peyman GA, Fishman GA, Sanders DR, Apple DJ, VlchekJK. Biopsy of human scleral-chorioretinal ti-sue. InvestOphthalmol 1975; 14: 707-10.

4Griffin JR, Straatsma BR, Kreiger AE. Transvitreal chorio-retinal biopsy in the rabbit. Am J Ophthalmol 1975; 79:25-38.5Constable IJ, Slatter DH, Horne R. Chorioretinal biopsy indogs. Inivest Ophthalmol 1979; 19: 603.

564 I. J. Constable, G. H. Chester, R. Hormie, and J. F. Harriolt



j.com/

Br J O


ugust 1980. Dow

nloaded from

http://bjo.bmj.com/

humanchorioretinal biopsy under controlled systemic hypotensive anaesthesia · hypotensive...

Documents