6
TENT-SHAPED RETINAL DETACHMENTS IN RETINOPATHY OF PREMATURITY MICHAEL J. SHAPIRO, MD, JOSHUA ALPERT, MD, RAHUL T. PANDIT, MD Background: Most retinal detachments associated with retinopathy of prematurity (ROP) are radial, segmental, or circumferential with cicatricial extraretinal fibrovascular proliferation (EFP) at the apex of the detachment. This report describes peculiar tent- shaped retinal detachments that developed among eyes with ROP. Methods: An observational case series consisting of 9 patients and 13 eyes with tent-shaped retinal detachments. Their morphology, clinical baseline, surgical course, and final retinal status were extracted from medical records. Results: Eight had simple tent-shaped retinal detachments, three had a double tent- shaped retinal detachment, one had a chevron-based retinal detachment, and one had a star-shaped retinal detachment. Each case had a disk based stalk that extended to the apex of the traction retinal detachment and continued anteriorly; 12 stalks inserted in the retrolental space and 1 terminated in the mid vitreous. Six eyes were stage 4A, two eyes were stage 4B, and five eyes were stage 5. Vitrectomy surgery was performed on 11 eyes. Surgery resulted in retinal attachment in nine eyes, and two retinas remained detached. Conclusion: Tent-shaped retinal detachments are seen in patients with ROP. A stalk should be sought in evaluation of these eyes. Vitreous surgery focused on relieving this traction is often successful. RETINA 26:S32–S37, 2006 I n his description of the pathogenesis of late stages of retinopathy of prematurity (ROP), Machemer pre- sented observations about the development of retinal detachments in ROP. 1 He proposed a mechanical schema that showed that contraction of stage 3 ROP could account for the progression to segmental, cir- cumferential, and radial retinal detachments. He con- cluded with a general statement that “All retinal changes seen in ROP can be explained by proliferation of tissue originating in the shunt area.” Extending these observations, the international classification of ROP further described the retinal detachment config- urations. 2 The presence of a stalk originating on the optic nerve in retinal detachments associated with ROP has been frequently noted. 3–6 The dissection of the stalk was a well-documented step in membranectomy for ROP. Eller et al. studied the association of the persis- tent hyaloid artery with ROP and suggested that it was a likely part of the disturbance of vasculogenesis and angiogenesis that describe the advance stages of ROP. 7 They also speculated that the hyaloid system could provide a scaffold for fibrovascular prolifera- tion. The stalk also had practical implications since when using the open sky technique this stalk may increase the intraoperative hemorrhage, 4 and using closed vitrectomy, aggressive surgery on the stalk may lead to posterior retinal breaks. 6 Recently, the disk stalk was identified as one of the many vectors of traction involved in development of retinal detach- ment in ROP. 8 Although frequently observed, the identification of a disk stalk as a sufficient cause of retinal detachment separate from the extraretinal fi- brovascular proliferation (EFP) was often impossible From the University of Illinois at Chicago, UIC Eye Center. The authors have no proprietary interests in the material de- scribed. Reprint requests: Michael J. Shapiro, MD, UIC Eye Center, 1905 W. Taylor, Chicago, IL 60612; e-mail: [email protected] S32

TENT-SHAPED RETINAL DETACHMENTS IN ...classroster.lvpei.org/cr/images/ARCHEIVE/2019/Dr Srikanta...detachments, the area of retinal detachment was lim-ited by the disk and the laser

  • Upload
    others

  • View
    0

  • Download
    0

Embed Size (px)

Citation preview

Page 1: TENT-SHAPED RETINAL DETACHMENTS IN ...classroster.lvpei.org/cr/images/ARCHEIVE/2019/Dr Srikanta...detachments, the area of retinal detachment was lim-ited by the disk and the laser

TENT-SHAPED RETINAL DETACHMENTSIN RETINOPATHY OF PREMATURITYMICHAEL J. SHAPIRO, MD, JOSHUA ALPERT, MD, RAHUL T. PANDIT, MD

Background: Most retinal detachments associated with retinopathy of prematurity(ROP) are radial, segmental, or circumferential with cicatricial extraretinal fibrovascularproliferation (EFP) at the apex of the detachment. This report describes peculiar tent-shaped retinal detachments that developed among eyes with ROP.

Methods: An observational case series consisting of 9 patients and 13 eyes withtent-shaped retinal detachments. Their morphology, clinical baseline, surgical course, andfinal retinal status were extracted from medical records.

Results: Eight had simple tent-shaped retinal detachments, three had a double tent-shaped retinal detachment, one had a chevron-based retinal detachment, and one had astar-shaped retinal detachment. Each case had a disk based stalk that extended to theapex of the traction retinal detachment and continued anteriorly; 12 stalks inserted in theretrolental space and 1 terminated in the mid vitreous. Six eyes were stage 4A, two eyeswere stage 4B, and five eyes were stage 5. Vitrectomy surgery was performed on 11 eyes.Surgery resulted in retinal attachment in nine eyes, and two retinas remained detached.

Conclusion: Tent-shaped retinal detachments are seen in patients with ROP. A stalkshould be sought in evaluation of these eyes. Vitreous surgery focused on relieving thistraction is often successful.

RETINA 26:S32–S37, 2006

In his description of the pathogenesis of late stages ofretinopathy of prematurity (ROP), Machemer pre-

sented observations about the development of retinaldetachments in ROP.1 He proposed a mechanicalschema that showed that contraction of stage 3 ROPcould account for the progression to segmental, cir-cumferential, and radial retinal detachments. He con-cluded with a general statement that “All retinalchanges seen in ROP can be explained by proliferationof tissue originating in the shunt area.” Extendingthese observations, the international classification ofROP further described the retinal detachment config-urations.2

The presence of a stalk originating on the opticnerve in retinal detachments associated with ROP has

been frequently noted.3–6 The dissection of the stalkwas a well-documented step in membranectomy forROP. Eller et al. studied the association of the persis-tent hyaloid artery with ROP and suggested that it wasa likely part of the disturbance of vasculogenesis andangiogenesis that describe the advance stages ofROP.7 They also speculated that the hyaloid systemcould provide a scaffold for fibrovascular prolifera-tion. The stalk also had practical implications sincewhen using the open sky technique this stalk mayincrease the intraoperative hemorrhage,4 and usingclosed vitrectomy, aggressive surgery on the stalkmay lead to posterior retinal breaks.6 Recently, thedisk stalk was identified as one of the many vectors oftraction involved in development of retinal detach-ment in ROP.8 Although frequently observed, theidentification of a disk stalk as a sufficient cause ofretinal detachment separate from the extraretinal fi-brovascular proliferation (EFP) was often impossible

From the University of Illinois at Chicago, UIC Eye Center.The authors have no proprietary interests in the material de-

scribed.Reprint requests: Michael J. Shapiro, MD, UIC Eye Center,

1905 W. Taylor, Chicago, IL 60612; e-mail: [email protected]

S32

Page 2: TENT-SHAPED RETINAL DETACHMENTS IN ...classroster.lvpei.org/cr/images/ARCHEIVE/2019/Dr Srikanta...detachments, the area of retinal detachment was lim-ited by the disk and the laser

because the vitreoretinal relationships were dominatedby the EFP.

In this article, we describe infants with ROP thatpresented with tent-shaped retinal detachments ratherthan radial, segmental, or circumferential folds. Thistent-shaped morphology is associated with persistentfetal vasculature (PFV),9 and very different from theretinal detachments generally seen in ROP. Carefulinspection revealed that these retinal detachmentswere not a result of the contraction of the extraretinalfibrovascular proliferation, but rather the contractionof a hyperplastic disk stalk probably related to thehyaloid artery. Identification of these unexpected ret-inal detachment configurations is helpful for diagnosisby ophthalmologists who examine or treat ROP.

Materials and Methods

Tent-shaped detachments were defined as a tractionretinal detachment with five criteria: one, a well de-fined small apex and relatively broad base; two, retinaldetachment apex well within the vascularized retina;three, tubular fibrous stalk connected to the retinalapex; four, stalk emanating from the optic nerve disk;and five, traction from the stalk directed into centralvitreous. Records of patients with traction retinal de-tachments from ROP seen in a pediatric retina practicewere reviewed. Cases with a documented tent-shapedretinal detachment were selected. Eyes with a radial orcircumferential fold or detachment as the predominantmorphology were excluded. All eyes that had an opticnerve disk stalk discovered during surgical dissectionwere also excluded.

The baseline characteristics were recorded, includ-ing the estimated gestational age and birthweight. Thehistory was reviewed using the postmenstrual age as

the standard. Preoperative evaluation focused specialattention on the retinal examination. The form of theretinal detachment was studied. The EFP was consid-ered for extent, location, vascularity, and its positionrelative to both the retina and the lens. Postoperativefollow-up ranged from 6 months to 6 years and in-cluded patient history, examination with indirect oph-thalmoscope, and slit lamp examination (when ageappropriate).

Results

Tent-shaped retinal detachments were found in 13eyes of 9 patients with ROP (Table 1). All patientswere born prematurely and also had severe ROP in thecontralateral eye. One was first examined as an adult;of the remainder, five infants were seen during theactive phase and three infants were referred during thecicatricial phase of ROP. The birthweights rangedfrom 550 g to 1800 g. Among the pediatric cases(excluding Patient 9) the average birthweight was611 g. The infants were born at estimated gestationalages that ranged from 23 to 26 weeks with a mean of24.5 weeks. Three of the infants were inborn. Fivewere referred for treatment. The one adult, ex-prema-ture neonate, was referred for long-term observation.The retinal detachments evolved under supervisionamong the inborn patients. All but two eyes among theinfants reached threshold before development of reti-nal detachment. Eight were in zone 1 and four were inzone 2. All threshold eyes were treated with laserphotocoagulation.

There were four configurations of retinal detach-ments. Eight eyes showed simple tent-shaped retinaldetachment (Figure 1). In these eight eyes, fibrouselements originated on the optic disk and connected

Table 1. Case Summary

Eye PatientBirthweight

(g)

EstimatedGestational

Age (wk) Location Eye Zone TreatmentDetachmentMorphology Stage Vitrectomy

StructuralOutcome

1 1 692 25 Inborn L 1 Laser Tent 4B LSV Attachment2 2 550 24 Inborn R 1 Laser Tent 4A LSV Attachment3 Inborn L 1 Laser Tent 4A LSV Attachment4 3 560 24 Inborn L 2 Laser Tent 4B LV Attachment5 4 646 26 Outborn R 2 Laser Tent 4A None Attachment6 L 2 Laser Tent 5 LV Attachment7 5 620 23 Outborn L 2 None Star shaped 5 LV Attachment8 6 1220 25 Outborn R 1 Laser Tent 5 LV Detachment9 L 1 Laser Double tent 4A LSV Attachment

10 7 553 23 Outborn R 1 Laser Double tent 5 LSV Attachment11 L 1 Laser Double tent 5 LSV Phthisis12 8 655 26 Outborn R 1 Laser Tent 4A LSV Attachment13 9 1800 NA Outborn L NA None Chevron 4A None Unchanged

LSV � lens sparing vitrectomy; LV � lensectomy vitrectomy.

S33TENT-SHAPED RETINAL DETACHMENTS IN ROP ● SHAPIRO ET AL

Page 3: TENT-SHAPED RETINAL DETACHMENTS IN ...classroster.lvpei.org/cr/images/ARCHEIVE/2019/Dr Srikanta...detachments, the area of retinal detachment was lim-ited by the disk and the laser

along the retina surface to the apex of the retinaldetachment within the vascularized retina, and in-serted anteriorly in the central vitreous, anterior vitre-ous, or posterior lens (Figure 2). In eight eyes therewere accompanying peripheral segmental folds due tocontraction of persistent stage 3 ROP. In Cases 1, 7, 8,

and 9 the EFP involuted and exerted no traction on theretina. Since the media were clear, ultrasounds wereonly occasionally performed. In one eye the ultra-sound showed the stalk extending from adjacent to thenerve to the retrolental space and in the second eye itextended from the disk to the central vitreous (Figure3). Connection to the condensed anterior vitreous wasnoted in five eyes, and in one a connection to vitreousthat bridged across a circumferential fold was re-corded. Among all the eyes with tent-shaped retinaldetachments, the area of retinal detachment was lim-ited by the disk and the laser induced chorioretinalatrophy. Some vascular and retinal dragging was al-ways present among cases with tent-shaped retinaldetachment.

In five eyes more complicated retinal detachmentconfigurations developed. One eye presented with atent-shaped detachment with a chevron-shaped base(Figures 4 and 5). This shape reflected the straighten-ing of the arcade blood vessels proximal and distal tothe apex forming a chevron as a result of the anteriortraction at the apex. Three eyes presented with adouble tent with the stalk connected to two retinaldetachment apices, one nasal and one temporal. A fiftheye showed a thick stalk with tight connection 360degrees and presented with a star shape (Figure 6).

Peripapillary fibrosis was noted in two eyes (Figure7). Vitreous hemorrhage was seen in two eyes. Usingthe international classification of ROP six eyesreached stage 4A, two eyes reached stage 4B, and fiveeyes reached stage 5. Initial scleral buckle was per-formed on three eyes. In two cases the buckle reduced

Fig. 1. A simple tent-shaped retinal detachment accompanied by amild circumferential fold from cicatricial stage 3 retinopathy of pre-maturity. The stalk connects from the disk to the apex of the retinaldetachment within the vascularized retina and then extends into thevitreous and retrolental space.

Fig. 2. Preoperative 130 degree Retcam photograph of the right eye ofPatient 8. The retina within the dotted line was detached. The largearrow shows the apex of the simple tent-shaped retinal detachment. Thesmall arrows show the path of the stalk from the optic disk to the retinaldetachment apex and anterior into the midvitreous. The superior arcadevessels show mild dragging toward the apex. The periphery showschorioretinal atrophy from laser ablation and complete involution ofstage 3 Rush disease.

Fig. 3. Left, the ultrasound of a tent-shaped detachment with a stalkseen extending from nasal to the disk to the retrolental space. Right, anultrasound corresponding to Figure 2 shows a hyperplastic hyaloidartery extending into the central vitreous.

S34 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● VOL 26 ● NO 7 ● SUPPLEMENT 2006

Page 4: TENT-SHAPED RETINAL DETACHMENTS IN ...classroster.lvpei.org/cr/images/ARCHEIVE/2019/Dr Srikanta...detachments, the area of retinal detachment was lim-ited by the disk and the laser

the peripheral component of the traction retinal de-tachment; however the central traction retinal detach-ment progressed in all three cases. Vitrectomy withmembranectomy was performed on 11 eyes. Seveneyes underwent lens sparing vitrectomies; repeat lenssparing vitrectomies were performed in two eyes, andfour eyes had combined lensectomy and vitrectomy

Fig. 5. A tent-shaped retinal detachment with a chevron base. Theorigin is seen at the disk and the stalk is connected to the vessels nearthe disk. The apex is seen along the inferior arcade blood vesselslabeled as “on end view of anterior extent of stalk.” The proximal anddistal portions of the vessels are elevated, straightened, and dragged sothat the detachment base forms a chevron. The blood vessel distal toapex forms the crest of the detachment more peripherally.

Fig. 6. A star-shaped retinal detachment with 360 degree connectionsof the retina elevated and drawn towards the stalk with draping of theretina.

Fig. 7. Fibrosis of the optic nerve head with infiltration of vitreoretinalinterface. Striae as seen inferiorly are a common manifestation and mayindicate the alterations that precede the connection of the stalk to theretina. Inferior nasal there was a disk stalk.

Fig. 4. The artist presents a horizontal perspective of a tent-shapeddetachment with a chevron shaped base also seen in Figure 5. Thestalk originated at the disk and inserted on the inferior arcade vesselswith elevation and straightening of the blood vessels both distal andproximal.

S35TENT-SHAPED RETINAL DETACHMENTS IN ROP ● SHAPIRO ET AL

Page 5: TENT-SHAPED RETINAL DETACHMENTS IN ...classroster.lvpei.org/cr/images/ARCHEIVE/2019/Dr Srikanta...detachments, the area of retinal detachment was lim-ited by the disk and the laser

procedures. Two eyes failed treatment and nine reat-tached. The two cases with surgical failure were pre-operatively classified as stage 5 ROP.

Discussion

Machemer identified the key factors in the devel-opment of retinal detachments from ROP as the con-traction of the circumferential shunt with stretchingand elevation of the retina, the traction from intravit-real proliferation, and the position of the shunt alongthe anterior posterior axis of the eye.1 The resultantvitreoretinal configurations in eyes with a radiallysymmetric distribution of disease centered on the diskare circumferential folds and funnel-shaped detach-ments. In eyes with radially asymmetric distributionsof fibrovascular shunt tissue the retina developed ra-dial folds, segmental folds, incomplete funnels, eccen-trically displaced funnels, and spiral funnels.10 Rec-ognizing the additional pathologic forces responsiblefor retinal detachment and understanding the vectorcombinations that lead to the different patterns ofdetachment is especially important in determining theeffective surgical approach since the membranes oftenobscure much of the retina.8 In this report, we de-scribed eyes with ROP that developed tent-shapedretinal detachments. Recognition of this uncommonpattern of retinal detachment may help during moni-toring of the ROP patient and surgical management.

The tent-shaped retinal detachments reported weremost similar to the typical radial folds described inROP. However, they presented distinct features. First,in each case the tractional element was a stalk ema-nating from the disk rather than a flat, pleated, orbunched sheet that originated from the EFP. Second,the apex of the tent-shaped retina detachment was wellwithin the vascularized retina rather than at the junc-tion of the vascular and avascular retina. Third, thestalk extended from the apex centrally into the mid-vitreous, anterior vitreous, and posterior lens ratherthan the peripheral vitreous, ciliary body, and periph-eral lens. Some of the cases revealed retinal tractionfrom both the disk stalk and extraretinal fibrovascularproliferation. However, we excluded the cases inwhich major EFP was associated with extensiveopaque vitreous proliferation to emphasize the stalkcomponent as a sufficient cause of retinal detachment.

Sporadic persistent fetal vasculature syndrome isquite rare, but findings of PFV are common amongpremature neonates.11 In normal eyes the primaryvitreous vasculature stops growing at 10 weeks, andthe branch vessels slowly atrophy. The posteriorbranches of the hyaloid trunk, vasa hyaloidea propria,regress first and are completely atrophic by 8.5

months gestational age. The hyaloid artery itselfcloses at 8 months and regresses soon afterwards.12

The neonatal cases in this report evolved in the settingof extreme prematurity, generally followed completelaser treatment, and were either inborn or early refer-rals. In our cases both retinal vasculogenesis and hya-loidal artery involution were disturbed.

The development of the normal eye in humans andmost mammals includes a chronological overlap andproportional relationship between retinal vasculariza-tion and vascular involution of the primary vitreous.In many animal models of ROP the hyaloidal systempersists and in some the hyaloidal blood vessels pro-liferate. Bischoff et al displayed persistent hyaloidalvessels in the mouse models of ROP using scanningelectron microscopy.13 They described the hyaloidvasculature as a high resistance passive system and theretinal vasculature as a low resistance active system.They suggested that oxygen exposure constricts theretinal vasculature and raises the hydrostatic pressurein the retinal vessels, driving the blood into the func-tionally closed hyaloidal vasculature. This diversionof blood causes a persistence and distention of thetunica vasculosum lentis and vasa hyaloidea propria.Larrazabal and Penn showed the arborizing hyaloidsystem with fluorescein angiography of newborn ratsrecovering in room air after 14 days of 80% oxygenexposure.14 They demonstrated an extensive and tor-tuous hyaloid vascular system as well as tortuosity ofthe major retinal vessels. This model may be the mostsimilar to the neonatal human condition. Berkowitz etal used a noninvasive method, carbogen enhancedMRI, to study the hyaloidopathy in newborn rats.15

They found that the functional volume of hyaloidalcirculation occupies a larger volume of the vitreous inROP rats compared with controls. These findings con-firm and refine the essential finding of Patz16 andalthough not in complete agreement, build on thework of Ashton.17 Bischoff also suggested a role forthe persistent vasculature in some cases of retinaldetachment. Indeed, the cicatrization of the hyaloidvascular system was a major contributor in the beaglemodel of ROP retinal detachment.18

Finally, we reported special cases in which therelationship between the stalk and the tent-shapedretinal detachments were isolated and easy to appre-ciate. More frequently, a complex mix of forces mustbe analyzed and treated. We believe that early refer-rals for failure of laser treatment may reduce thecomplexity of the retinal detachments. If this is thecase, observation of the hyperplastic disk stalk as avector contributing to retinal detachment may achievegreater importance for diagnosis and treatment of latestages of ROP.

S36 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● VOL 26 ● NO 7 ● SUPPLEMENT 2006

Page 6: TENT-SHAPED RETINAL DETACHMENTS IN ...classroster.lvpei.org/cr/images/ARCHEIVE/2019/Dr Srikanta...detachments, the area of retinal detachment was lim-ited by the disk and the laser

Key words: hyaloid artery, persistent fetal vascu-lature (PFV), persistent and hyperplastic primary vit-reous (PHPV), retinal detachment, ROP, pediatric ret-inal detachment, tent-shaped retinal detachment.

References

1. Machemer R. Description and pathogenesis of late stages ofretinopathy of prematurity. Ophthalmology 1985;92:1000–1004.

2. An international classification of retinopathy of prematurity.II. The classification of retinal detachment. The InternationalCommittee for the Classification of the Late Stages of Reti-nopathy of Prematurity. Arch Ophthalmol 1987;105:906–912.

3. Trese MT, Droste PJ. Long-term postoperative results of aconsecutive series of stages 4 and 5 retinopathy of prematu-rity. Ophthalmology 1998;105:992–997.

4. Jabbour NM, Eller AE, Hirose T, et al. Stage 5 retinopathy ofprematurity. Prognostic value of morphologic findings. Oph-thalmology 1987;94:1640–1646.

5. Mintz-Hittner HA, Miyashiro MJ, Knight-Nanan DM, et al.Vitreoretinal findings similar to retinopathy of prematurity ininfants with compound heterozygous protein S deficiency.Ophthalmology 1999;106:1525–1530.

6. Chong LP, Machemer R, de Juan E. Vitrectomy for advancedstages of retinopathy of prematurity. Am J Ophthalmol 1986;102:710–716.

7. Eller AW, Jabbour NM, Hirose T, Schepens CL. Retinopathyof prematurity. The association of a persistent hyaloid artery.Ophthalmology 1987;94:444–448.

8. Recchia FM, Capone A Jr. Contemporary understanding and

management of retinopathy of prematurity. Retina 2004;24:283–292.

9. Goldberg MF. Persistent fetal vasculature (PFV): an inte-grated interpretation of signs and symptoms associated withpersistent hyperplastic primary vitreous (PHPV). LIV Ed-ward Jackson Memorial Lecture. Am J Ophthalmol 1997;124:587–626.

10. Trese MT Retinopathy of prematurity. In: Ryan SJ, ed. Ret-ina. 2nd ed. St. Louis: Mosby, 1989: v. III, chapter 154,2449–2462.

11. Renz BE, Vygantas CM. Hyaloid vascular remnants in hu-man neonates. Ann Ophthalmol 1977;9:179–184.

12. Mann I. The development of the human eye. New York:Grune and Stratton, Inc., 1950.

13. Bischoff PM, Wajer SD, Flower RW. Scanning electronmicroscopic studies of the hyaloid vascular system in new-born mice exposed to O2 and CO2. Graefes Arch Clin ExpOphthalmol 1983;220:257–263.

14. Larrazabal LI, Penn JS. Fluorescein angiography of the new-born rat: implications in oxygen-induced retinopathy. InvestOphthalmol Vis Sci 1990;31:810–818.

15. Berkowitz BA, Lukaszew RA, Mullins CM, Penn JS. Im-paired hyaloidal circulation function and uncoordinated ocu-lar growth patterns in experimental retinopathy of prematu-rity. Invest Ophthalmol Vis Sci 1998;39:391–396.

16. Patz A. The effect of oxygen on immature retinal vessels.Invest Ophthalmol 1965;4:988–999.

17. Ashton N. Donders lecture, 1967. Some aspects of the com-parative pathology of oxygen toxicity in the retina. Br JOphthalmol 1968;52:505–531.

18. Ozawa Y, Nishina S, Akiya S, Uemura Y, Azuma N. Oph-thalmoscopical and pathological features of oxygen-inducedvitreoretinopathy in the neonatal beagle. Invest OphthalmolVis Sci 1997;38:S747.

S37TENT-SHAPED RETINAL DETACHMENTS IN ROP ● SHAPIRO ET AL