The Effects of Scleral Buckling on Young Rabbit Eyes

Objective  To evaluate the effects of scleral buckling surgery on the developingeye in an animal model.

Methods  Eleven young rabbits underwent scleral buckling surgery in one eye.In 6 rabbits, a 2.5 × 0.6-mm solid silicone encircling band (240 style)with a Watzke sleeve was used, and in 5 rabbits, a 2.0-mm silicone encirclingsponge (502 style) with its ends abutting, but not connected, was used. After3 months, both eyes of each animal were enucleated. Six eyes of an additional3 rabbits served as baseline controls for determining initial globe volumes.

Results  Of the eyes buckled with a 240 band, 4 of 6 developed glaucoma comparedwith 0 of 5 eyes buckled with a 502 sponge (P = .06).Migration of the element occurred in all eyes with a 240 band (3 anterior,3 posterior), while none of the 502 sponges migrated. There was no significantdifference in the mean final corneal diameter between eyes with a 240 band(P = .94) and untreated fellow eyes or between eyeswith a 502 sponge and untreated fellow eyes (P =.25). The mean axial length of eyes with a 240 band and untreated fellow eyeswas 19.57 mm and 16.83 mm, respectively (P = .009).The mean axial length of eyes with a 502 sponge and untreated fellow eyeswas 16.67 mm and 16.50 mm, respectively (P = .67).When comparing eyes with a 240 band with eyes with a 502 sponge, a significantdifference was observed in the mean axial length (P =.006) and mean volume (P = .006) between the 2 scleralbuckle groups.

Conclusions  The use of a noninterrupted solid silicone encircling band appearedto have significant effects on the growth of young rabbit eyes. The use ofan interrupted silicone sponge produced a buckling effect similar to the solidsilicone band and did not migrate, adversely affect eye growth, or resultin glaucoma.

Clinical Relevance  Noncontinuous encircling elements may prove to be as beneficial as continuousencircling elements in the treatment of tractional and rhegmatogenous retinaldetachment in infant eyes without the potentially deleterious effects associatedwith a nondivided encircling band or the need for a second surgery to dividethe encircling element.

RETINOPATHY OF PREMATU-rity (ROP) is characterized by incomplete retinalvascularization and subsequent abnormal fibrovascular proliferation. In casesthat progress to threshold ROP, laser photocoagulation or cryotherapy to theavascular retina greatly reduces the incidence of retinal detachment.^1-8 However,despite timely treatment, some infants develop retinal detachment and requirescleral buckling procedures to reattach the retina in an attempt to restoreor salvage useful vision.^9,10 Itis currently unknown what long-term effects such scleral buckling procedureshave on the developing eye. Many surgeons believe that the scleral bucklecould prevent growth of an infant's eye and therefore advocate sectioningof the exoplant approximately 3 months after the initial surgery to allowthe eye to grow.

Since the original oral presentation of our data,¹¹ therehas been only one other study documenting the effects of scleral bucklingprocedures on developing eyes.¹² The purposeof this current study was to determine if placing an encircling silicone bandaround the eyes of young rabbits has deleterious effects on eye growth. Thesecond goal was to determine if placing a silicone sponge for 360° (butnot connected end-to-end) allows for normal eye growth and obviates the needfor a second procedure to divide the element while still maintaining a goodscleral buckle effect.

Fourteen young (aged 4-7 weeks) rabbits, weighing between 600 and 850g, were used in this study. All animals in the study were treated in accordancewith institutional guidelines and the Association for Research in Vision andOphthalmology resolution on the use of animals in research.

Eleven of the rabbits were divided into 2 groups. In one group (n =6), each rabbit underwent scleral buckling surgery in the right eye, usinga 2.5 × 0.6-mm silicone encircling band (240 style; Labtician OphthalmicsInc, Oakville, Ontario) with a Watzke sleeve (Labtician Ophthalmics Inc).The left eyes in this group served as internal controls. In the second group(n = 5), each rabbit underwent scleral buckling surgery in the right eye,using a 2.0-mm silicone encircling sponge (502 style; Labtician OphthalmicsInc) with its ends abutting but not connected. The left eyes in this groupalso served as internal controls. A third group of rabbits (n = 3) servedas an additional set of control eyes for determining baseline globe volumes.Both eyes of the animals in this group were enucleated at the outset of thestudy to obtain this baseline measurement for comparison with globe volumesin buckled and control eyes 3 months later.

Prior to surgery, all animals were anesthetized with an intramuscularinjection of 50 mg/kg of ketamine hydrochloride and 5 mg/kg of xylazine hydrochloride.The animals were prepared and draped in the usual sterile fashion. After a360° peritomy was made, dissection was performed, and the rectus muscleswere isolated with 4-0 silk sutures. An encircling 240 band or 502 spongewas placed under the extraocular muscles in the equatorial plane in the righteye of each rabbit.

In the eyes with the 240 band, 8-0 nylon sutures were used to securethe element in all 4 quadrants. In the eyes with the 502 sponge, 8-0 nylonsutures were used to anchor the element to the sclera in all 4 quadrants.Although a 502 sponge was placed for 360°, its abutting ends were notsutured end-to-end. In both groups of rabbits, 7-0 polyglactin 910 (Vicryl)sutures were used to close the conjunctiva. All eyes with an encircling bandor encircling sponge were given a similar amount of scleral indentation, estimatedintraoperatively through the use of binocular indirect ophthalmoscopy.

An anterior chamber paracentesis was performed to normalize the intraocularpressure. All animals tolerated the procedure well, and there were no complicationsas a result of anesthesia. Immediately following euthanasia, enucleation wasperformed in the standard fashion in all eyes (control and experimental) 3months after the start of the study. The globes were then analyzed for thevariables of interest. The 3 animals used as controls for initial globe volumedetermination had both eyes enucleated at the outset of the study to obtainthis baseline measurement for later comparison with experimental and fellowcontrol eye volumes.

Systat software (Systat, Cary, NC) was used to carry out the statisticalcalculations for this study. Prior to surgery, animal weight and corneal diameterdata were recorded. At the time of enucleation, the following parameters werecollected: animal weight, corneal diameter, axial length, eye volume, presenceor absence of glaucoma, and presence or absence of element migration. Forthe 3 additional control animals, globe volume was assessed in each of the6 eyes at the outset of the study.

Mean preoperative corneal diameter measurements were compared with meancorneal diameter measurements at the time of enucleation in each eye, usingthe Wilcoxon signed rank test. The mean corneal diameter measurements at thetime of enucleation in the experimental eyes and their fellow control eyeswere compared against each other, also using the Wilcoxon signed rank test.The mean corneal diameter measurements at the time of enucleation in the eyesbuckled with the 240 band and those buckled with the 502 band were comparedusing the Mann-Whitney test.

The Mann-Whitney test was used to compare mean axial length measurementsfor eyes with the 240 and 502 elements. The Wilcoxon signed rank test wasused to compare the mean axial length measurements in the operated on vs thecontrol eyes. The difference in mean axial length measurements between experimentaleyes (eyes buckled with either the 240 band or the 502 sponge) and their fellowcontrol eyes was compared using the Mann-Whitney test.

Globe volume determination was determined by measuring the displacementof water in a 10-mL graduated cylinder. For the experimental group of eyes,the Mann-Whitney test was used to compare the volume measurements betweeneyes with the 240 and 502 elements. This test was also employed to comparethe difference in mean volume measurements between these 2 groups. The volumemeasurements from the 3 additional control animals (n = 6 eyes) were comparedseparately with the volume measurements from the eyes with the 240 and 502elements.

Glaucoma was defined as an intraocular pressure greater than 21 mm Hgas measured by Tono-Pen (Mentor, Norwell, Mass) following topical instillationof tetracaine. Intraocular pressure was measured just prior to euthanasia.We used the χ² test to compare the association of glaucomawith either element (240 or 502).

Element Migration. A χ² analysiswas employed to characterize the association between the presence or absenceof postoperative element migration and element type (240 or 502).

Table 1 presents the preoperativerabbit eye characteristics. The mean weight (849 g), corneal diameter (10.4mm), axial length (13.7 mm), and globe volume (1.53 mL) characteristics ofthe control group killed at baseline were recorded for initial volume determination. Table 2 presents pertinent findings determinedat the time of postoperative enucleation.

The 11 rabbits in the study had a mean weight of 701 g (range, 594-908g) preoperatively and a mean weight of 3089 g (range, 2554-3715 g) at thetime of euthanasia. The mean weight of the 3 control animals used to determineinitial globe volume was 849 g (range, 820-886 g).

There was a significant difference in the preoperative and postoperativecorneal diameter measurements for both control (P =.003) and experimental (P = .003) eyes. The meanfinal corneal diameter for eyes with the 240 band was 13.01 mm compared with12.81 mm for fellow control eyes (P = .94). The meanfinal corneal diameter for eyes with the 502 sponge was 12.19 mm comparedwith 12.31 mm for fellow eyes (P = .25). No significantdifference in corneal diameter was found postoperatively between eyes withthe 240 band and eyes with the 502 sponge (P = .07).Similarly, no significant difference in corneal diameter was noted betweeneyes with either 240 or 502 elements and their fellow control eyes, respectively(P = .26).

A significant difference in axial length was found when 240 band eyesand 502 sponge eyes were compared postoperatively (P =.006). The mean axial length of eyes with the 240 band and untreated felloweyes was 19.57 mm and 16.82 mm, respectively (P =.009). The mean axial length of eyes with the 502 sponge and untreated felloweyes was 16.67 mm and 16.50 mm, respectively (P =.67).

A significant difference in the globe volume was found when we comparedthe eyes with the 240 band with those with the 502 sponge (P = .006). As Table 2 demonstrates,eyes buckled with a 240 band were an average of 0.42 mL larger than the fellowunbuckled eyes, and eyes buckled with a 502 sponge were an average of 0.11mL smaller than the fellow unbuckled eyes. These mean differences in finalglobe volume between buckled eyes and their fellow control eyes were foundto be statistically significant (P = .03).

Of the eyes buckled with a 240 band, 4 of 6 developed glaucoma comparedwith 0 of 5 eyes buckled with a 502 sponge (P = .006).

Element migration was noted in all 6 eyes with the 240 band (3 migratedanteriorly, and 3 migrated posteriorly). All of the 240 bands that migratedanteriorly also demonstrated evidence of glaucoma (Figure 1A), while only 1 of the 3 bands that migrated posteriorly(Figure 1B) had secondary glaucoma.Of the 5 eyes implanted with the 502 sponge, none showed evidence of elementmigration (Figure 2). When comparingthe element type with the incidence of migration, a significant differencewas found (P = .004). Cheese-wiring of the scleralanchoring sutures through the sclera was seen with all of the encircling bandsand with none of the 502 sponges.

Our findings support the widely held belief that the placement of anencircling band in a developing eye can have deleterious effects on oculargrowth and physiology. Anterior and posterior migration of the element, highincidence of glaucoma, and negative influence on eye growth were all seriouscomplications associated with the undivided 240 encircling band. Therefore,our original presentation of this data¹¹ anda similar subsequent report¹² support the notionthat sectioning of the 240 band is recommended once adequate retinal appositionhas been achieved to help prevent the described complications.

Although perhaps not directly applicable to the treatment of infantilepediatric retinal detachments in humans, the results of the present animalstudy demonstrate clearly that a noncontinuous encircling element is bettertolerated in a developing eye than a continuous, nondivided element. No eyesbuckled with the 502 sponge had evidence of glaucoma or element migration.There was no significant difference in axial length between control eyes andeyes buckled with a 502 sponge at the end of 3 months.

We found no significant difference in corneal diameter between eyesbuckled with a 240 band and those buckled with a 502 sponge or between eyesbuckled with a 240 band and fellow control eyes. This lack of a differencein corneal diameter despite the presence of a significant difference in axiallength and globe volume calls into question the utility of corneal diametermeasurements as a clinical indicator of ocular growth or restriction in buckledeyes. This is especially notable because when comparing the 2 buckling elements,we found a significant difference in axial length and globe volume measurementbetween eyes buckled with a 240 band and a 502 sponge.

Although this animal study was small, it stresses important points regardingthe treatment approach to pediatric retinal detachments with scleral bucklingsurgery when lens sparing vitrectomy is not appropriate.¹³ Thisstudy offers surgeons an experimental basis for selecting a noncontinuousencircling element as an alternative to placing a continuous encircling elementfor the treatment of pediatric retinal detachments. A nondivided encirclingband was not well tolerated by the developing eye in a rabbit model.

Corresponding author and reprints: Philip J. Ferrone, MD, 600 NorthernBlvd, Suite 216, Great Neck, NY 11021 (e-mail: p_ferrone@hotmail.com).

Submitted for publication January 31, 2003; final revision receivedJune 20, 2003; accepted July 23, 2003.

This study was supported by grant EY06520 from the National Eye Institute,Bethesda, Md (Dr Hartzer).

We thank Labtician Ophthalmics Inc for their generous provision of thescleral buckling elements used in this study.