Excavated defects of the optic disc have been associated with retinaldetachment in the macula and beyond.1 Suchoptic disc anomalies encompass a large clinical spectrum ranging from single,usually temporally located, optic pits, to multiple pits, to the so-calledmorning glory syndrome. These anomalies have been attributed to abnormal closureof the fetal fissure.1 In most of the associatedretinal detachments, a retinal break has not been found, and various hypotheseshave been brought forward as to the origin of the subretinal fluid, includingthe vitreous and fluid from the subarachnoid space.2 Inmost reported cases, retinal detachment is confined to the macular area; however,extensive and even complete retinal detachment may occur. We report a casewith delayed sequential occurrence of perfluorodecalin (PFD) and then siliconeoil in the subretinal space following retinal detachment repair in a patientwith long-standing retinal detachment and proliferative vitreoretinopathyin the presence of unilateral optic pits.
A 20-year-old woman had complex retinal detachment and proliferativevitreoretinopathy in her right eye. She noted a deterioration in the visionof her right eye 1 day prior to hospital admission. Her ocular history wasremarkable for a progressive impairment in the peripheral visual field duringthe past 18 months, but she had had no dilated eye examinations in the meantime.The function of the right eye had always been worse compared with the lefteye, and it was thought to be amblyopic. On examination, visual acuity washand movements OD and 20/20 OS. Slitlamp examination was unremarkable exceptfor anterior vitreous opacities in the right eye. Funduscopy disclosed a subtotalretinal detachment with macular involvement in the right eye. There were subretinalfibrous strands in the inferior 2 quadrants and epiretinal membrane formationwith star folds peripheral to the major vascular arcades consistent with along-standing detachment and advanced proliferative vitreoretinopathy. Whilethe optic disc in the left eye appeared healthy, there were 2 pits noted inthe disc of the right eye. A small, rather subtle depression was located atthe 8-o'clock position. A larger, well-developed, obvious declivity, the bottomof which was not visible, was at the nasal disc margin (Figure 1). No diaphanous tissue was identified overlying any ofthe pits.
Optic disc of the right eye ina 20-year-old patient. A small subtle depression is located at the 8-o'clockposition (arrowheads), and a larger defect of the optic nerve head is presentnasally (arrows).
Subsequently, pars plana vitrectomy, removal of subretinal and epiretinalmembranes, inferior retinotomy, peripheral retinectomy, endolaserphotocoagulation,and silicone oil tamponade were performed. Perfluorodecalin was used intraoperatively,and all visible PFD was removed prior to air-fluid exchange and silicone oilfilling. Postoperatively, the retina was completely attached and visual acuityimproved to 20/60 OD. Two days postoperatively, several small droplets ofPFD were noted in the subretinal space inferior to the optic disc margin.The number of droplets increased during the next 2 days, and some were advancingtoward the macula (Figure 2). Therefore,another surgical procedure was performed 7 days after the initial surgeryto remove all visible subretinal PFD. This was accomplished in the presenceof the silicone oil tamponade by gently advancing the subretinal PFD dropletstoward the inferior retinotomy margin with a blunt retina spatula (DORC International,Zuidland, the Netherlands). Once released from the subretinal space, the dropletswere removed via a fluid needle and concurrent pressure increase generatedby a conventional oil pump (Geuder, Heidelberg, Germany). At the end of thisprocedure, no more PFD droplets were visible in the subneurosensory space.However, 1 day later, new subretinal PFD droplets appeared again at the inferiordisc margin. The amount of droplets increased during the subsequent days,however, at a slower rate than before. Thirteen days later, in addition tothe PFD droplets, silicone oil appeared in the same subretinal area with progressiveenlargement of the retinal detachment toward the fovea.
Two days after surgery for retinaldetachment, which included an inferior retinotomy, peripheral retinectomy,endolaser coagulation, and silicone oil tamponade, subretinal perfluorodecalindroplets appeared in the subretinal space at the inferior margin of the opticdisc. The whitish spots represent diode laser spots.
Silicone oil was then removed, and the subretinal PFD and silicone oilwas released via the inferior retinotomy (Figure 3). The vitreous cavity was now filled with air and 16% C2F6 gas. A subsequent redetachment associated with proliferativevitreoretinopathy, including epiretinal membrane formation and traction, necessitatedanother surgical procedure. This time, no PFD was used intraoperatively andprior to silicone oil tamponade, diode laser endophotocoagulation was appliedaround the inferior of the optic disc to generate chorioretinal scars thatwould prevent silicone oil from passing into the subretinal space. Three monthslater, the retina remained attached, no new subretinal PFD or silicone oiloccurred, and best-corrected visual acuity was 20/100 OD.
Intraoperative photograph showingsubretinal silicone oil inferior and temporal to the optic disc.
Congenital pits of the optic nerve head represent a rare abnormality,occurring in approximately 1 in 11 000 patients.2 Theirpathogenesis is not completely understood but is thought to be related toanomalies of closure of the fetal fissure. Pits of the disc are usually unilateralbut may occur in both eyes. Most often, they are present on the temporal edgeof the disc, but they may be multiple and have been observed in other meridians.Clinically, the pits may vary considerably in size, ranging from subtle, barelyvisible depressions to well-developed, obvious declivities.
Serous macular detachment develops in 25% to 75% of eyes with opticnerve pits.3 The pathophysiologic explanationof this complication is controversial. Several theories have been broughtforward including leakage from the vessels at the base of the pit,4 leakage from the parapapillary choroidal circulation,leakage from the cerebrospinal fluid,5,6 andleakage from the vitreous cavity.3,7
The case reported here clearly demonstrates that there may be a communicationbetween the vitreous cavity and the subretinal space via the optic pit andthat this may be the cause for retinal detachment. The most likely explanationfor the appearance of subretinal PFD and silicone oil bubbles several dayspostoperatively is a reservoir posterior to the optic disc plane not visibleby biomicroscopy and a communication between the optic disc pit and the subretinalspace. Various possibilities may be considered for such a reservoir includingthe subarachnoid space around the optic nerve or posterior herniation of abnormaltissue composing the true "floor" of the pit. While instant leakage of PFDunder the retina was noted intraoperatively in the presence of a morning glorydisc anomaly by Bartz-Schmidt and Heimann,8 thisis, to the best of our knowledge, the first description of a delayed occurrenceof PFD and silicone oil in the presence of optic pits.
It has been demonstrated in an animal study7 thatindia ink placed in the vitreous cavity flowed into the pit and subretinalspace, supporting the hypothesis that the subretinal fluid originates fromthe vitreous cavity. Bonnet3 as well asPostel et al9 reported eyes with optic pitsand associated retinal detachment managed with intravitreal gas injection,in which gas bubbles migrated through a small break in tissue overlying theoptic pits into the subretinal space postoperatively. Coll et al10 reporteda communication between the subretinal space and the vitreous cavity in themorning glory syndrome.
Our case suggests that PFD should be avoided in retinal detachment surgeryin eyes with optic pits. Furthermore, intraocular air/gas tamponade may bepreferable to silicone oil tamponade if feasible to prevent the complicationspreviously mentioned. The lower interfacial tension of silicone oil was notadequate to prevent bubbles of silicone oil from passing through the smallretinal opening within the optic disc. Endolaser photocoagulation for formationof a barrier to fluid movement was effective here to prevent silicone oilfrom passing into the subretinal space. Such laser treatment appears prudentin similar cases when air/gas tamponade is insufficient to secure retinalattachment and when silicone oil must be used. The volume of both subretinalPFD and silicone oil increased during a short period of time. It is currentlyunknown which unidirectional forces are operative to efficiently move fluidfrom the vitreous cavity through the optic pit under the neurosensory retina.A pressure gradient must be assumed whereby gravity forces are unlikely becausePFD and silicone oil owing to their specific weights would be expected tobehave differently. It may be speculated that the pressure gradient is thedifference between the intraocular pressure at the moment of migration andthe pressure within the reservoir (ie, either the extraocular tissue pressureor the pressure within the subarachnoid space).
Corresponding author: Stefan Dithmar, MD, Department of Ophthalmology,University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany(e-mail: firstname.lastname@example.org)
Dithmar S, Schuett F, Voelcker HE, Holz FG. Delayed Sequential Occurrence of Perfluorodecalin and Silicone Oilin the Subretinal Space Following Retinal Detachment Surgery in the Presenceof an Optic Disc Pit. Arch Ophthalmol. 2004;122(3):409-411. doi:10.1001/archopht.122.3.409