- Email: [email protected]
Silicone oil tamponade to seal macular holes without position restrictions
Silicone Oil Tamponade to Seal Macular Holes without Position Restrictions Michael H. Goldbaum, MD,1,2 Brooks W. McCuen, 2nd, MD,3 Anne M. Hanneken, MD,4 Stuart K. Burgess, MD,1 Howard H. Chen, MD1 Objective: The authors performed a study to determine the effectiveness and safety of silicone oil as a substitute for gas to fill the vitreous cavity to treat macular holes. Design: Multicenter, nonrandomized, interventional trial. Participants: Thirty-seven consecutive patients chose vitrectomy with silicone tamponade instead of gas to treat 40 eyes with stage-2 to stage-4 idiopathic age-related macular holes. Stage-2 holes constituted 40% of the holes, and stage-3 and stage-4 holes made up 60%. Intervention: All eyes were treated with vitrectomy, manual detachment of the posterior vitreous face (not done for stage-4 holes), autologous serum instillation, and silicone fill of the vitreous cavity. After insertion of the oil, the patients resumed normal activity with no restriction of head or eye position except to avoid faceup position. The oil was removed after approximately 6 weeks. Main Outcome Measures: The authors considered the seal of the macular hole and the preoperative and postoperative logarithm of the minimum angle of resolution (logMAR) visions the most significant measures for comparison to other studies. Results: Eighty percent of all holes and 86% of holes not treated previously were sealed with a single silicone tamponade of the vitreous cavity. The logMAR value of visual acuity improved an average of 0.26 (2.6 lines) to 0.61 (20/81) for all eyes and 0.34 (3.4 lines) to 0.52 (20/66) when the macular hole sealed. Completeness of fill of the vitreous cavity with silicone affected seal of the macular hole. Three of eight eyes in which open holes developed after oil removal had less than 90% fill of the vitreous cavity by silicone. Sixty-nine percent of lenses increased opacity one grade or were removed after silicone tamponade. There were no significant adverse effects arising from silicone tamponade. Conclusions: Silicone oil tamponade of macular holes is effective and safe. Silicone may be optimal for the treatment of macular holes in persons who must travel, who cannot maintain facedown positioning, or who have monocular vision. The most important factor in the successful closure of the macular hole was the completeness of fill of the vitreous cavity with silicone oil. Ophthalmology 1998;105:2140 –2148 In 1991, Kelly and Wendel1 reported that treatment could improve visual function in eyes with full-thickness macular holes. Common to this report and others2– 4 have been surgical detachment of the posterior vitreous in the majority of cases when the vitreous remains attached to the optic nerve and macular region, filling of the vitreous cavity with a long-lasting gas bubble, and strict facedown positioning. In the original report, the retina surrounding the hole was
Originally received: October 26, 1997. Revision accepted: June 1, 1998. Manuscript no. 97544. 1 Department of Ophthalmology, Shiley Eye Center, University of California, San Diego, California. 2 Ophthalmology Section, Veterans Administration Medical Center, San Diego, California. 3 Department of Ophthalmology, Duke University, Durham, North Carolina. 4 San Diego Vitreoretinal Associates, San Diego, California. Presented at the American Academy of Ophthalmology annual meeting, San Francisco, California, October 1997. Address correspondence to Michael H. Goldbaum, MD, Department of Ophthalmology, Shiley Eye Center, University of California, San Diego, La Jolla, CA 92093-0946.
2140
reattached with 58% success.1 Since then, large series have established 69% to 73% closure of the macular holes.2,3 Efforts to improve the success rate have included removal of obvious epiretinal membranes near the hole; delamination of the internal-limiting lamina surrounding the hole; probing for occult epiretinal membranes with sharp instruments adjacent to the hole; application of autologous serum, transforming growth factor-beta, or autologous platelets over the hole; and photocoagulation.1,4 –7 Empiric observation led most surgeons to conclude that facedown positioning, such as 90% for 2 weeks followed by 50% for 2 weeks,8 improved the success rate. Many patients are unwilling to maintain a facedown position for 2 to 4 weeks or are unable to do so because of physical or medical reasons. The gas fill of the vitreous cavity precludes air travel in depressurized cabins or a gain in altitude of several thousand feet in ground transportation, because altitude-induced elevation of pressure in the eye can close retinal and choroidal circulation and infarct the retina. Longer duration of intraocular gas tamponade correlates with a higher rate of closure of the macular hole and with greater improvement in visual acuity.8 The longer-acting gases mean an extended period of restricted travel.
Goldbaum et al 䡠 Silicone Tamponade of Macular Holes without Position Restrictions We initially used silicone oil without prone positioning to fill the vitreous cavity as a substitute for gas to treat a macular hole in three eyes of two patients. One patient had to travel by air or by land over a 5000-foot pass. The other patient had recent cardiothoracic surgery that prevented facedown positioning. The silicone fill of the vitreous cavity successfully sealed the macular holes in these three eyes, and the patients experienced a benign postoperative period. Consequently, we sought to answer two questions. Is silicone tamponade of macular holes safe, and is it effective?
Materials and Methods The inclusion criteria were a centric stage-2 through stage-4 idiopathic age-related macular hole with visual acuity of 20/50 or worse. The size of the hole was estimated by comparison to a parapapillary major retinal vein (considered to be 125 m in diameter)3 in a 20° to 35° photograph or during examination. Stage-2 retinal holes were less than 400 m in diameter,9 stage-3 holes were 400 m or larger, and stage-4 holes coexisted with posterior vitreous detachment or previous vitrectomy. All patients had a cuff of subretinal fluid around the hole and saw a break in a slit of light projected across the hole (Watzke–Allen sign). The exclusion criteria were uncertainty of diagnosis of macular hole, retinal detachment greater than 1500 m diameter, macular disease that might affect visual acuity, aphakia from intracapsular cataract extraction, and a defect in the capsule or zonules sufficiently large that it might allow silicone to enter and be trapped in the anterior chamber. Consecutive patients were given the option of a single intervention of traditional gas fill with strict facedown position (90% of time for 2 weeks followed by 50% for 2 weeks)8 or silicone fill without facedown positioning but with the need for a second procedure to remove the oil. All patients chose silicone, because they considered the facedown positioning more onerous than a second procedure. Surgery was performed with monitored anesthesia care and the eye under retrobulbar anesthesia. In phakic eyes, the crystalline lens was not removed during the initial surgery or during the removal of silicone oil. The vitreous was trimmed to approximately 2 mm anterior to the optic nerve. The posterior cortical vitreous was disengaged from the optic nerve using suction from the side port of a vitrectomy instrument or a silicone-tipped aspiration cannula. The posterior vitreous face was slowly detached across the macula and posterior retina, and a complete vitrectomy with removal of peripheral vitreous was performed. The absence of the fish-strike sign with a silicone-tipped cannula confirmed the removal of the posterior vitreous face. Peripheral retinal breaks discovered on indirect ophthalmoscopy were surrounded by laser photocoagulation. We removed epiretinal membranes only if they were visible and caused obvious distortion of the retina. An air– fluid exchange was followed by removal of remnant fluid 5 to 15 minutes later. Autologous serum 0.5 ml was layered across the hole and removed after 10 minutes. Silicone oil (5000 centistoke [cs] in 38 eyes and 1000 cs in 2 eyes) warmed to body temperature was injected into the vitreous cavity through a 19-gauge 3⁄8-inch needle. Any large air bubbles found by indirect ophthalmoscopy were removed to ensure complete fill of the vitreous cavity. To facilitate adjustment of silicone fill, a reservoir of silicone was allowed to enter several centimeters into the infusion cannula. Before the infusion cannula was removed, the vitreous fill was adjusted with injection or aspiration of silicone by a syringe attached to the infusion cannula to produce a scale reading be-
tween 10 and 15 on a Schiøtz tonometer with a 5.5-g weight. This step was designed to prevent underfill or overfill of the vitreous cavity by silicone. Removal of the infusion cannula and closure of the sclerotomy generally increased the intraocular pressure to a Schiøtz scale reading between 5 and 10 (7–17 mmHg). The night of surgery, the patient was instructed to sleep facedown. Afterward, the only restriction imposed on a patient was to avoid faceup positioning to prevent remnant fluid from reaching the macular hole. One day after the oil insertion, the vitreous cavity was inspected for a visible meniscus to estimate the percentage of fill. The hole was examined through the oil for closure (Fig 1). Patients were usually able to return to work and normal activity 5 days after oil insertion. The oil was removed at the patient’s convenience after 4 weeks minimum (average, 6.5; maximum; 15 weeks). Controlled removal of the oil was performed with 500-mmHg suction from the vitrectomy console or passive efflux. When the oil was removed with suction, a 19-gauge, 3⁄8-inch beveled cannula connected to the vitrectomy console with 4-mm internal-diameter 50-cm long tubing reduced resistance to flow of the viscous oil. This tubing is commonly available for oxygen administration through a nasal cannula. When irrigation fluid began to enter the cannula and tubing, the suction was reduced quickly to avoid turbulence of the fluid in the vitreous cavity. To remove the last of the silicone, the patient’s head and eye were turned to place the sclerotomy through which the silicone was being drained at the most superior point. The last bubble of oil was removed through the cannula inserted 1.5 mm in the superiorly positioned sclerotomy. Remnant oil found on indirect ophthalmoscopy was removed with a siliconetipped needle or larger bore cannula if necessary. Silicone oil adhering to a silicone intraocular lens was removed with suction through a bent cannula with a silicone tip beveled to allow the cut end of the silicone tubing to fit flush on the back surface of the lens. Best-corrected visual acuity was assessed by an independent observer with an Early Treatment Diabetic Retinopathy Study (ETDRS) chart or with a projected Snellen chart standardized to the ETDRS chart. Visual acuity on the last examination after oil removal (or the best postoperative visual acuity if there was progressive cataract) was compared to the preoperative visual acuity. After the silicone was removed, the hole was examined with at least 16 magnification with a slit lamp through a 90-diopter or 78-diopter El Bayadi lens. To provide maximum assessment of depth, the slit was oriented horizontally and angled 15° from below. The hole was considered sealed after surgery if the hole edge was not visible or if the hole edge was visible but appeared to be in contact with the retinal pigment epithelium with no cuff of subretinal fluid (Fig 1). The lens clarity was qualitatively graded with the slit lamp from 0 (clear) to 4 (20/200 or worse view of the retina). Comparison of group pairs of discrete or continuous data for significant difference was accomplished with the t test. Group pairs of binomial data were treated similarly with comparison of two proportions. Ninety-five percent confidence intervals were generated from 1.96 times the standard error of the mean. The two planned comparisons of (1) the rate of closure of macular holes stratified by the stage of the macular holes and (2) the logarithm of the minimum angle of resolution (logMAR) vision stratified by success of closure of the macular hole were more strictly compared at ␣⫽0.025 after Bonferroni adjustment for multiple comparisons. We treated eyes as independents. We also tested individual patients as independents by averaging the variables in the two eyes when both eyes of a patient were involved.
2141
Ophthalmology Volume 105, Number 11, November 1998
Figure 1. A, a 500-m hole before intervention in a patient with nonproliferative diabetic retinopathy. B, silicone tamponade. Hole is sealed. Fluid does not have access to hole. C, 3 months after silicone was removed. Hole is sealed. Visual acuity improved from 20/320 to 20/63.
Results Patient and Eye Data There were 40 eyes in 37 patients who ranged from 42 to 81 years of age, averaging 66 years. Treating individual patients as independent data yielded the same results as treating eyes as independent data. There was no significant preponderance of gender (males, 55%; 95% confidence interval [CI95], 40%–70%) or eye (right eye, 58%; CI95, 42%–73%). Forty percent of the holes were stage 2 and 60% were stage 3 or 4. The hole duration ranged from 0.7 to 22 months, averaging 5.6 months, with a median of 3 months. The mean durations of 3.5 months for stage-2 holes and 7.1 for stage-3 or stage-4 hole differed significantly (P1 ⫽ 0.02, where P1 signifies a one-tail test and P2 denotes a two-tail test). The average follow-up was 10.3 months after removal of silicone oil.
Closure Rates The holes were sealed in 32 of the 40 eyes (80%, CI95, 68%–92%) after one silicone fill (Table 1). Eight patients had an eye with an
2142
open hole after one silicone fill. Of these, five patients consented to second silicone fill, and all these holes sealed. Including reoperations, 37 eyes were sealed (92.5%, CI95 84%–100%). There was no significant difference in the rate of closed holes among the three surgeons. The stage-2, stage-3, and stage-4 holes were sealed at the same rate (75% and 83%, respectively; P2 ⫽ 0.53). The average duration of the hole before the vitrectomy and oil insertion was 7.1 months in holes that did not seal or reopened and 6.3 months in holes that sealed, and that did not differ significantly (P1 ⫽ 0.51).
Visual Acuity A change of 0.1 in the logMAR scale is equivalent to a change of one line in the ETDRS visual acuity chart. For all patients, the visual acuity increased an average of 0.26 logMAR units (2.6 lines). The visual acuity did not change significantly in eyes with holes that opened after oil removal (loss of 0.5 line); in the 32 eyes with closed holes after one silicone fill, the visual acuity improved an average of 3.4 lines (P1 ⫽ 0.03) (Fig 2). Seventy-two percent of these eyes with sealed holes increased more than one line. Of the eight eyes with open holes after one intervention, none of the
Goldbaum et al 䡠 Silicone Tamponade of Macular Holes without Position Restrictions Table 1. Outcomes Based on Stratification Closure Rate after 1 Tamponade unless Otherwise Noted
LogMAR before Silicone Inserted (20/X equivalent) 0.87 20/148 0.78 20/121 0.93 20/170
0.61 (0.47 to 0.74) 20/81 (20/59 to 20/110) 0.49 (0.26 to 0.71) 20/61 (20/36 to 20/103) 0.69 (0.52 to 0.85) 20/97 (20/67 to 20/141)
0.88 20/152 0.87 20/148
0.93 (0.62 to 1.25) 20/170 (20/83 to 20/356) 0.52 (0.39 to 0.66) 20/66 (20/49 to 20/91)
1.12 20/264 0.83 20/135
0.76 (0.42 to 1.05) 20/115 (20/53 to 20/224) 0.58 (0.44 to 0.73) 20/76 (20/55 to 20/107)
Stratification
n
All: [mean (CI95)]
40
Stage 2
16
80% (68 to 92) {92.5% (84 to 100)}* 75% (54 to 75)
Stage 3–4
24
83% (68 to 98)
Open
8‡
P2 ⫽ 0.53† 50%§
Closed
32
100%
5
40% (0 to 83)
35
86% (74 to 97)
Previous surgery Primary hole
P1 ⫽ 0.01
LogMAR after Silicone Removed* (20/X equivalent)
LogMAR Change (post-minus pre-silicone)*
>1 line Improvement*
⫺0.26 (⫺0.14 to ⫺0.39)
62% (47 to 78)
35% (20 to 50)
⫺0.29 (⫺0.03 to ⫺0.55)
69% (45 to 92)
56% (31 to 81)
⫺0.24 (⫺0.11 to ⫺0.38)
58% (38 to 78)
21% (4 to 37)
P1 ⫽ 0.38 0.05 (0.40 to ⫺0.30)
P1 ⫽ 0.26 25% (0 to 55)
⫺0.34 (⫺0.22 to ⫺0.47)
72% (56 to 87)
P1 ⫽ 0.03 ⫺0.36 (⫺0.04 to ⫺0.65)
P1 ⫽ 0.01† 60% (5 to 100)
⫺0.25 (⫺0.11 to ⫺0.39)
63% (47 to 79)
P1 ⫽ 0.58
P1 ⫽ 0.90
20/50 or Better*
P1 ⫽ 0.01 12.5% (0 to 35) 41% (24 to 58) P1 ⫽ 0.14 0% 40% (24 to 56) P1 ⫽ 0.08
P1 ⫽ 1-tail test; P2 ⫽ 2-tail test; CI95 ⫽ 95% confidence interval. * Outcomes after one or more silicone tamponades. † Two planned comparisons that required P ⱕ 0.025 for 95% certainty of difference. ‡ Number open after one silicone tamponade. § Percentage of holes open after one silicone tamponade that were closed after a second silicone tamponade.
three eyes that did not have reoperation improved one line after one silicone tamponade, and two of the five eyes that had second silicone tamponade increased at least one line. The logMAR vision improved an average of 2.9 lines for stage-2 holes and 2.4 lines for stage-3 and stage-4 holes (P2 ⫽ 0.38). Fifty-six percent (CI95, 31%– 81%) of stage-2 holes achieved a visual acuity of 20/50 or better after removal of the silicone oil, and 21% (CI95, 4%–37%) of stage-3 or stage-4 holes reached 20/50 or better (P1 ⫽ 0.01). The mean visual acuity after treatment was 20/61 (CI95, 20/36 –
20/103) for stage-2 holes and 20/97 (CI95, 20/67–20/141) for stage-3 or stage-4 holes (Table 2).
Analysis of Macular Holes Reopening or Failing to Seal In eight eyes (20%), the hole opened after the oil was removed. In none of these holes did silicone pass through the hole or under the
Figure 2. Scattergraph of preoperative logarithm of the minimum angle of resolution (logMAR) vision and postoperative logMAR vision stratified by hole closure.
2143
Ophthalmology Volume 105, Number 11, November 1998 Table 2. Conversion of LogMAR to Snellen Visual Acuity LogMAR Value 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
X for 20/X 1002 796 632 502 399 317 252 200 159 126 100 80 63 50 40 32 25 20
retina during the period of silicone fill of the vitreous cavity. Analysis of these eyes found five holes opened after temporary seal—from 9 days to 3 months after the oil was removed. Two holes reopened after cataract removal. One of these eyes had pseudophakic cystoid macular edema develop before the hole reopened. Silicone oil filled the vitreous cavity between 90% and 100% in all the sealed holes. The silicone fill of the vitreous cavity in three eyes with open holes was between 65% and 85%; these holes were open at the first examination after the oil removal.
Previous Surgery and Repeat Surgery Five eyes had previous intervention with gas tamponade for macular hole in the same or the opposite eye. All these patients expressed strong preference for the surgery that used silicone to tamponade the hole, mainly because there was no restriction on patient activity. These patients expressed great aversion to the prolonged facedown positioning with gas. Previous surgery for macular holes affected the seal rate. Three (38%) of the eight holes that reopened after silicone tamponade were reoperations for previous unsuccessful surgery with gas tamponade for macular hole. Two (6%) of the 32 sealed holes were reoperations (P1 ⫽ 0.05). With one fill of the vitreous cavity with silicone oil, 2 (40%) of 5 previous gas failures sealed; whereas, 30 (86%) of 35 primary macular holes sealed with one fill (P1 ⫽ 0.01). The visual acuity after silicone removal was 20/50 or better in 40% of primary holes and in none of the holes that had failed to seal with previous gas tamponade (P1 ⫽ 0.08).
Total Removal of Silicone Oil The mean time was 8 minutes for total removal of silicone oil from the vitreous cavity with aspiration from the vitrectomy console, detailed under the Materials and Methods section above. In all eyes, a minute amount of oil remained in the fluid in the vitreous cavity as fine droplets visible to the patient as small floaters against the backlighting and occasionally as a spot on downgaze. On B-scan ultrasound, these droplets were discernible as a delicate snow pattern.
Adverse Events Of the 40 eyes, 31 were phakic at the time of the oil insertion and removal. Posterior subcapsular vacuoles generally were visible
2144
transiently for the first 2 weeks after insertion of silicone oil. Over the mean observation time of 10.3 months after silicone removal, in 69% (CI95, 52%– 85%) of the phakic eyes, the lens clarity diminished at least one grade point, or a cataract was removed. Two patients had retinal detachment develop after the oil was removed. One detachment was repaired with laser and gas–fluid exchange. The other had proliferative vitreoretinopathy develop. With silicone refill after removal of epiretinal membrane, all but the inferior 20% of the retina remains attached, with improvement in vision of one line. Macular pigment changes were found in two eyes. Both eyes had a reduction in visual acuity.
Discussion Comparison to Other Studies The high rate of seal of macular holes with silicone fill of the vitreous cavity occurred even though patients returned to full activity as soon as they were able. There was no significant difference in the seal rate in stage-2 holes (75%) and stage 3- or stage-4 holes (83%), possibly because the seal rate was high in both groups of eyes. These rates compare favorably with a 75% rate of same-size, smaller, or sealed stage-2 holes10 and 69% seal rate in a comprehensive, multicentered study of stage-3 or stage-4 holes treated with gas fill of the vitreous cavity and strict facedown positioning.3 Comparing the mean visual acuity, the percentage of eyes achieving two lines’ improvement in visual acuity, and the proportion of eyes reaching 20/50 or better, the outcomes of this study were similar to the Vitrectomy for Treatment of Macular Hole Study Group investigations of stage-2 and stage-3 to stage-4 macular holes (Table 1). In that study, the mean visual acuity 6 months after treatment of stage-2 macular holes was 20/68. We calculated the CI95 in the Vitrectomy for Treatment of Macular Hole Study Group report to be 20/41 to 20/112 from 1.96 times the standard error of the mean. Although this outcome appears similar to the mean visual acuity of 20/61 (CI95, 20/36 –20/ 103) in the silicone study, examination of the confidence intervals indicates no significant difference. The population in these two studies differs somewhat, because we used Gass’s definition of stage-2 holes less than 400 m; whereas, the Vitrectomy for Treatment of Macular Hole Study Group used 300 m as the ceiling diameter for central stage-2 holes. In stage-3 and stage-4 macular holes, the mean visual acuity after silicone tamponade in this study was 20/97 (CI95, 20/67–20/141). The mean visual acuity found by the Vitrectomy for Treatment of Macular Hole Study Group after gas tamponade was 20/115 (CI95, 20/96 –20/138). Observation of the confidence intervals also shows these outcomes to be similar. Silicone contact with the crystalline lens for more than 3 months causes cataract development in virtually all eyes.11 Vitrectomy and gas fill of the vitreous cavity to treat macular holes results in cataract formation at a rate higher than occurs without surgery. The Vitrectomy for Treatment of Macular Hole Study Group found an increase of one or
Goldbaum et al 䡠 Silicone Tamponade of Macular Holes without Position Restrictions more grade points (0 – 4 scale) in 82% of treated eyes.3 Similarly, as described under adverse effects above, silicone fill of the vitreous cavity caused an increase in lens opacity (69%, CI95, 52%– 85%). The possibility of confounding variables in this measurement between these two studies is too great to draw any conclusions about differences in the rate of lens opacification between silicone and gas fill of the vitreous cavity.
Factors Affecting Sealing of Macular Hole and Elimination of Subretinal Fluid Cuff Adhesion of Retina to the Retinal Pigment Epithelium. The normal adhesion of the retina to the retinal pigment epithelium is multifactorial.12 These factors include hydrostatic forces induced by differences in fluid flow at various layer interfaces between the vitreous and the sclera, oncotic pressure from absent or reduced protein in subretinal fluid, interdigitation of photoreceptors and pigment epithelial microvilli, cell adhesion molecules between the photoreceptors and the pigment epithelium, and active transport of fluid outward by the pigment epithelium. We make the assumption that these factors are applicable to the adhesion of the fovea and perifoveal retina to the retinal pigment epithelium. After reattachment of retina detached more than a few weeks, it takes at least 4 to 6 weeks for these factors to become re-established and for the adhesive strength to reform between the retina and pigment epithelium.12,13 Before some threshold level of adhesiveness is established, fluid reaching the hole might interfere with the reparative process, allowing some or all of the adhesive factors to fail. Compared to macular holes that seal spontaneously, macular holes sealed by gas tamponade may be more likely to have adhesion enhanced by a glial membrane that occludes the hole or pulls the hole edges together.14,15 Presumably, this cicatrization increases the adhesion of the hole edge and the perifoveal retina to the retinal pigment epithelium. A macular hole can open in the absence of vitreous traction.16 For example, one patient in this study had a full-thickness macular hole develop with a cuff of subretinal fluid 8 years after vitrectomy. The presence of a glial membrane may reduce the likelihood of a hole reopening after it is closed. It is unknown whether the silicone produces adhesion that more resembles gas-sealed holes or spontaneously sealed holes.
Surgical Factors Surgical factors potentially affecting the seal of the macular hole include duration of tamponade, lack of access of fluid to the macular hole, growth factors, pressure from the tamponade pushing the retina against the pigment epithelium, control of forces to open the hole, and positioning of the eye. The duration of the gas tamponade of the retina correlates with the rate of seal of the macular hole.8 Unless the vitreous cavity is refilled, the duration of gas tamponade is limited. As the gas volume diminishes, there is more opportunity for fluid to reach and enter the macular hole. By the time that the volume of gas diminishes below 50%, the
patient is usually no longer restricted to facedown positioning. Consequently, fluid is in constant contact with the hole when the patient is erect. With the bubble still present, head and eye movements can generate strong currents of fluid across the hole, producing shear forces that can weaken an incomplete adhesion. In contrast, a complete or nearly complete fill of the vitreous cavity with silicone prevents fluid from reaching the hole most, if not all, of the time. The percentage of fill of the vitreous cavity remains constant until the oil is removed. The duration of silicone tamponade can be extended long beyond the duration of gas tamponade. The viscosity of 5000-cs oil prevents shear forces from oil flowing across the hole. Analysis of eyes in which macular holes failed to seal indicates that completeness of fill of the vitreous cavity with silicone is a factor in the seal rate. Inflammation and local growth factors augment the reparative process and improve the rate of seal of macular holes.4 – 6,17 Autologous serum appears to improve the hole closure rate (Banker AS, et al. Invest Ophthalmol Vis Sci 1997;38[Suppl]:S736). Presumably, these growth factors are present in the fluid in the vitreous cavity. In a gas-filled eye, as the gas bubble diminishes and the fluid compartment enlarges, in theory, these factors will be diluted. Because the vitreous fluid compartment remains small in a silicone-filled eye, these factors should maintain at high concentration. However, the higher concentration of growth factors may not affect the adhesion of the macular hole if the fluid is prevented from reaching the hole by the silicone oil. Silicone oil is lighter than water. But the specific gravity of silicone is much closer to water than is the specific gravity of any gas. Hence, the flotation force generated by silicone against the retina is much less than that from gas. Yet, the hole seals at least as well with silicone as with gas. The force with which the vitreous replacement holds the retina against the pigment epithelium appears to be less important than the prevention of fluid access to the hole. From the initial publication on the surgical treatment of macular holes, reports on the treatment of macular holes have stressed facedown positioning for 1 to 4 weeks as a recommended part of the procedure. Reinforcing this concept, Holekamp (Holekamp NM, Meredith TA, Mandell BA, et al. Presented at the AAO annual meeting, San Francisco, 1997) found that the major factor affecting the seal rate for different surgeons in a group practice of retinal surgeons was how well the patients followed the facedown regimen. In contrast, Tornambe et al18 found that facedown positioning was not necessary if attention was paid to producing a complete fill of the vitreous cavity with a nonexpansile 15% mixture of perfluoropropane gas. Similar to what we found with the use of silicone, completeness of fill of the vitreous cavity by gas appeared to be important in their report.
Advantages and Disadvantages of Silicone Tamponade of Macular Holes Silicone fill of the vitreous cavity to treat macular holes offers advantages. The patients can resume a normal lifestyle during the period of tamponade. Patients who have
2145
Ophthalmology Volume 105, Number 11, November 1998 monocular vision and are dependent on the eye with the macular hole can have useful vision with silicone fill of the vitreous cavity. The patients may travel by air or ascend in altitude. The silicone volume does not diminish with time; consequently, the access of fluid to the hole does not increase with time. The duration of tamponade can be extended. There are no shear forces acting on the hole. Patients who have had macular holes treated both by gas and by oil strongly prefer the use of oil. Many of these same reasons can be used to argue that silicone tamponade without position restriction is preferable to gas tamponade without position restriction proposed by Tornambe et al. However, Tornambe’s report is unique among macular hole studies in finding that facedown positioning is not necessary with gas tamponade; it remains uncertain whether his findings can be duplicated. Silicone tamponade of macular holes also has disadvantages compared to gas tamponade. A second surgical procedure is necessary to remove the oil. Remnant droplets of oil in the fluid that fills the vitreous cavity after oil removal can be detected by patients, although these are generally not bothersome. The complications that have been reported after the insertion of silicone oil into the vitreous cavity or after its removal occur mostly because of the tendency of vitreoretinal surgeons to manage complex surgical problems with intravitreous silicone left for several months or years.11,19 In contrast, eyes with age-related macular holes are not inflamed; do not have pre-existing retinal detachment; and are not likely to stimulate fibrous, pigment epithelial, or vascular proliferation. The tamponade duration to seal macular holes is not long enough to generate silicone emulsification. We have excluded eyes with poor separation of the anterior chamber and vitreous cavity to avoid silicone being trapped in the anterior chamber. For these reasons, with the exception of cataract formation, we have not observed the complications reported by others in eyes in which silicone has been used.
Eyes for Which Silicone Tamponade Should Be Avoided Silicone bubbles in the anterior chamber are undesirable although acceptable. Silicone constantly in contact with most of the corneal endothelium is unacceptable because of the high probability of corneal complications.11 Silicone is not a good choice when the oil can enter and be trapped in the anterior chamber through defects in the capsule or zonules. Large posterior capsulotomies may still permit silicone tamponade if they are central, with the edges in apposition to the posterior lens surface. Small peripheral capsular and zonular defects could prevent silicone tamponade. Because air–fluid exchange is a step in this surgery, a reasonable test of the quality of the barrier between the vitreous cavity and the anterior chamber is how well the barrier keeps air out of the anterior chamber when the vitreous cavity is filled with air. If no air bubble enters the anterior chamber, it should be safe to use silicone tamponade.
2146
In aphakic eyes with the capsule removed, it is difficult to fill the vitreous cavity more than 90% with silicone oil without it contacting most or all of the corneal endothelium, even with an inferior iridotomy. Consequently, aphakia from intracapsular cataract extraction is a relative contraindication for silicone tamponade of a macular hole. Yet, aphakic eyes with intact posterior capsule and zonules can safely have silicone tamponade. Acknowledgment. The authors thank Fred Wright, the Biostatistics Core of the University of California San Diego Cancer Center, for advice in the statistical analysis in this study.
References 1. Kelly NE, Wendel RT. Vitreous surgery for idiopathic macular holes. Results of a pilot study. Arch Ophthalmol 1991; 109:654 –9. 2. Wendel RT, Patel AC, Kelly NE, et al. Vitreous surgery for macular holes. Ophthalmology 1993;100:1671– 6. 3. Freeman WR, Azen SP, Kim JW, et al. Vitrectomy for the treatment of full-thickness stage 3 or 4 macular holes. Results of a multicentered randomized clinical trial. The Vitrectomy for Treatment of Macular Hole Study Group. Arch Ophthalmol 1997;115:11–21. 4. Glaser BM, Michels RG, Kuppermann BD, et al. Transforming growth factor-beta 2 for the treatment of full-thickness macular holes. A prospective randomized study. Ophthalmology 1992;99:1162–72. 5. Liggett PE, Skolik SDS, Horio B, et al. Human autologous serum for the treatment of full-thickness macular holes. A preliminary study. Ophthalmology 1995;102:1071– 6. 6. Gaudric A, Massin P, Paques M, et al. Autologous platelet concentrate for the treatment of full-thickness macular holes. Graefes Arch Clin Exp Ophthalmol 1995;233:549 –54. 7. Del Priore LV, Kaplan HJ, Bonham RD. Laser photocoagulation and fluid– gas exchange for recurrent macular hole. Retina 1994;14:381–2. 8. Thompson JT, Smiddy WE, Glaser BM, et al. Intraocular tamponade duration and success of macular hole surgery. Retina 1996;16:373– 82. 9. Gass JDM. Reappraisal of biomicroscopic classification of stages of development of a macular hole. Am J Ophthalmol 1995;119:752–9. 10. Ruby AJ, Williams DF, Grand MG, et al. Pars plana vitrectomy for treatment of stage 2 macular holes. Arch Ophthalmol 1994;112:359 – 64. 11. Federman JL, Schubert HD. Complications associated with the use of silicone oil in 150 eyes after retina–vitreous surgery. Ophthalmology 1988;95:870 – 6. 12. Marmor MF. Mechanisms of normal retinal adhesion. In: Ryan SJ, ed. Retina, 2nd ed. St. Louis: Mosby, 1994; v. 3, chap. 123. 13. Yoon YH, Marmor MF. Rapid enhancement of retinal adhesion by laser photocoagulation. Ophthalmology 1988; 95:1385– 8. 14. Madreperla SA, Geiger GL, Funata M, et al. Clinicopathologic correlation of a macular hole treated by cortical vitreous peeling and tamponade. Ophthalmology 1994;101: 682– 6. 15. Funata M, Wendel RT, de la Cruz Z, Green WR. Clinicopathologic study of bilateral macular holes treated with pars plana vitrectomy and gas tamponade. Retina 1992;12:289 – 98.
Goldbaum et al 䡠 Silicone Tamponade of Macular Holes without Position Restrictions 16. Gordon LW, Glaser BM, Ie D, et al. Full-thickness macular hole formation in eyes with a pre-existing complete posterior vitreous detachment. Ophthalmology 1995;102:1702–5. 17. Nagineni CN, Kutty RK, Detrick B, Hooks JJ. Inflammatory cytokines induce intercellular adhesion molecule-1 (ICAM-1) mRNA synthesis and protein secretion by human retinal pigment epithelial cell cultures. Cytokine 1996;8: 622–30.
18. Tornambe PE, Poliner LS, Grote K. Macular hole surgery without face-down positioning. A pilot study. Retina 1997;17: 179 – 85. 19. Casswell AG, Gregor ZJ. Silicone oil removal. II. Operative and postoperative complications. Br J Ophthalmol 1987;71: 898 –902.
Discussion by Paul E. Tornambe, MD Goldbaum et al present a multicenter retrospective pilot study that asks the question, “Is silicone tamponade of macular holes safe and effective?” Surgery was performed by 3 surgeons and included 40 eyes, 8 of which had previous macular hole surgery. Forty percent were stage 2 (defined as a hole 400 m or less in diameter), 60% were stage 3 or 4. The mean duration of the holes was 3 months. Preoperative acuity was 20/50 or worse. All surgeons used the same surgical technique consisting of a pars plana vitrectomy, removal of the posterior hyaloid, removal of epiretinal membranes when visually present and resulting in significant macular distortion, autologous serum, and a silicone oil fill. Facedown positioning was done over the first postoperative night only. No facedown positioning was done thereafter and no restrictions to the patient’s lifestyle were required. Patients were free to travel by air or ascend to any elevation. The oil was removed, on average approximately 6.5 weeks after insertion. Removal was done in all eyes, took less than 10 minutes, and was not associated with any operative complications except failure to remove all the oil droplets. Follow-up after oil removal averaged 10.3 months. The 1-fill operative success for all 40 eyes is 80%, and the rate for the 35 eyes with virgin macular holes is 86%. The 1-fill operative success rate for the five holes operated on previously with gas was 40%. Eight eyes failed permanent closure, with two holes reopening after cataract surgery. Of these, five underwent repeat macular hole surgery (MHS) with silicone oil and all closed, giving an overall attachment rate of 92.5%. Acuity improved an average of 3.4 lines. However, eyes cured with one procedure attained more visual improvement than eyes that failed the first attempt. The mean postoperative visual acuity was 20/61 for stage-2 holes and 20/97 for stage-3 and stage-4 holes. Visual acuity of 20/50 or better was attained in 56% of stage-2 holes and in 21% of stage-3 or stage-4 holes. No eyes that failed the first procedure ultimately attained 20/50 or better visual acuity. The authors’ objective was to determine whether macular hole surgery with silicone tamponade without facedown positioning is safe and effective. Their data, when compared with other studies, suggest that this technique probably is effective. However, without prospective randomization, it cannot be proved better or worse than other techniques. Many variables have been shown to be important and are likely inter-related, including size and duration of the hole and preoperative visual acuity. Goldbaum’s 86% one operation closure rate for primary MHS repair compares favorably with the literature. However, the visual results appear disappointing, especially for stage-2 eyes, in which only 56% attained 20/50 or better visual acuity. Tornambe et al1 found preoperative acuity a better predictor of
From Retina Consultants, San Diego, Mericose Eye Institute, La Jolla, California.
visual outcome than hole stage. Eighty percent of eyes with 20/100 or better preoperative visual acuity attained 20/50 or better postoperative visual acuity. Stage-2 holes usually have better preoperative visual acuity, and one would therefore expect better postoperative visual acuity than Goldbaum reports. This may be explained by the limited follow-up of only 10.3 months and the progression of cataract change. Leonard et al2 recently has shown that macular function improves over the first 3 postoperative years. However, there may be other explanations for less visual return. Silicone may be toxic and limit potential photoreceptor recovery. Surgical technique also must be considered. Goldbaum’s group removed only obvious epiretinal macular membrane (ERM) tissue and only if it significantly distorted the macula. Our recent experience with the Tano diamond-dusted soft-tip cannula (Synergetics, St. Louis, MO) suggests that a fine ERM, not frequently obvious, surrounds the hole and usually is attached firmly to the edges of the hole. This membrane can be removed without dissecting the underlying glistening membrane, which probably is internal limiting membrane. We believe that this subtle membrane may be important in the pathogenesis and persistence of the macular dehiscence and believe that removing it allows the edges of the hole to reapproximate. Removing the membrane may also stimulate growth factor production responsible for the formation of glial tissue that bridges the hole. We agree with Goldbaum’s observation that there are two types of postoperative endpoints. In one, the hole is flat against the RPE but the round defect is visible. In the other, the hole shrinks in size to a pinpoint defect. The former hole persistence may be caused by incomplete removal of surface traction, which prevents the reapproximation of the macular dehiscence and limits optimal visual recovery. Goldbaum’s group’s failure to routinely attempt to remove this less-than-obvious tissue may have prevented reapproximation of the edges of the hole and thus limited visual return. The authors did not mention which holes had ERM dissection regarding the type of hole closure endpoint and the degree of visual recovery. What information can we extract from this effort and apply clinically? First, MHS using silicone oil without facedown positioning works and likely does so by isolating the hole from liquid vitreous. Silicone is much less buoyant than gas. This provides additional evidence that the buoyancy of the bubble, and therefore prone positioning, may not be as important as waterproofing the hole. It would be interesting to evaluate visual fields on these patients. Gas bubble buoyancy may play an adverse role in the development of post-MHS field defects. This study also indicates that for those patients who do not wish to assume facedown positioning or must travel by air or to higher elevations, silicone oil may be reasonable. This technique may be particularly applicable to young patients with phakia, such as children or those young adults who wish an
2147
Ophthalmology Volume 105, Number 11, November 1998 uninterrupted lifestyle and are less prone to postvitrectomy cataract change. The study also finds that single operation success is important, and best acuity results after a single attempt. One reason for failure, as Goldbaum indicates, may be subtotal silicone oil fill. The major drawback of this approach, in addition to possible silicone toxicity, is the need for a second operation and those tiny silicone oil bubbles that are always left behind. In pseudophakic eyes, in which altitude is not an issue, we would prefer to use a perfluoropropane gas fill without prone positioning. The controversy still continues whether or not to combine lens replacement surgery with MHS. Goldbaum and others have observed that most eyes will develop a significant cataract and some holes reopen
2148
shortly after cataract surgery, which suggests more reasons to combine cataract extraction with macular hole surgery.
References 1. Tornambe PE, Poliner LS, Grote K. Macular hole surgery without face-down positioning. A pilot study. Retina 1997;17: 179 – 85. 2. Leonard RE, Smiddy WE, Flynn HW, Feuer W. Long-term visual outcomes in patients with successful macular hole surgery. Ophthalmology 1997;104:1648 –52.
Originally received: October 26, 1997. Revision accepted: June 1, 1998. Manuscript no. 97544. 1 Department of Ophthalmology, Shiley Eye Center, University of California, San Diego, California. 2 Ophthalmology Section, Veterans Administration Medical Center, San Diego, California. 3 Department of Ophthalmology, Duke University, Durham, North Carolina. 4 San Diego Vitreoretinal Associates, San Diego, California. Presented at the American Academy of Ophthalmology annual meeting, San Francisco, California, October 1997. Address correspondence to Michael H. Goldbaum, MD, Department of Ophthalmology, Shiley Eye Center, University of California, San Diego, La Jolla, CA 92093-0946.
2140
reattached with 58% success.1 Since then, large series have established 69% to 73% closure of the macular holes.2,3 Efforts to improve the success rate have included removal of obvious epiretinal membranes near the hole; delamination of the internal-limiting lamina surrounding the hole; probing for occult epiretinal membranes with sharp instruments adjacent to the hole; application of autologous serum, transforming growth factor-beta, or autologous platelets over the hole; and photocoagulation.1,4 –7 Empiric observation led most surgeons to conclude that facedown positioning, such as 90% for 2 weeks followed by 50% for 2 weeks,8 improved the success rate. Many patients are unwilling to maintain a facedown position for 2 to 4 weeks or are unable to do so because of physical or medical reasons. The gas fill of the vitreous cavity precludes air travel in depressurized cabins or a gain in altitude of several thousand feet in ground transportation, because altitude-induced elevation of pressure in the eye can close retinal and choroidal circulation and infarct the retina. Longer duration of intraocular gas tamponade correlates with a higher rate of closure of the macular hole and with greater improvement in visual acuity.8 The longer-acting gases mean an extended period of restricted travel.
Goldbaum et al 䡠 Silicone Tamponade of Macular Holes without Position Restrictions We initially used silicone oil without prone positioning to fill the vitreous cavity as a substitute for gas to treat a macular hole in three eyes of two patients. One patient had to travel by air or by land over a 5000-foot pass. The other patient had recent cardiothoracic surgery that prevented facedown positioning. The silicone fill of the vitreous cavity successfully sealed the macular holes in these three eyes, and the patients experienced a benign postoperative period. Consequently, we sought to answer two questions. Is silicone tamponade of macular holes safe, and is it effective?
Materials and Methods The inclusion criteria were a centric stage-2 through stage-4 idiopathic age-related macular hole with visual acuity of 20/50 or worse. The size of the hole was estimated by comparison to a parapapillary major retinal vein (considered to be 125 m in diameter)3 in a 20° to 35° photograph or during examination. Stage-2 retinal holes were less than 400 m in diameter,9 stage-3 holes were 400 m or larger, and stage-4 holes coexisted with posterior vitreous detachment or previous vitrectomy. All patients had a cuff of subretinal fluid around the hole and saw a break in a slit of light projected across the hole (Watzke–Allen sign). The exclusion criteria were uncertainty of diagnosis of macular hole, retinal detachment greater than 1500 m diameter, macular disease that might affect visual acuity, aphakia from intracapsular cataract extraction, and a defect in the capsule or zonules sufficiently large that it might allow silicone to enter and be trapped in the anterior chamber. Consecutive patients were given the option of a single intervention of traditional gas fill with strict facedown position (90% of time for 2 weeks followed by 50% for 2 weeks)8 or silicone fill without facedown positioning but with the need for a second procedure to remove the oil. All patients chose silicone, because they considered the facedown positioning more onerous than a second procedure. Surgery was performed with monitored anesthesia care and the eye under retrobulbar anesthesia. In phakic eyes, the crystalline lens was not removed during the initial surgery or during the removal of silicone oil. The vitreous was trimmed to approximately 2 mm anterior to the optic nerve. The posterior cortical vitreous was disengaged from the optic nerve using suction from the side port of a vitrectomy instrument or a silicone-tipped aspiration cannula. The posterior vitreous face was slowly detached across the macula and posterior retina, and a complete vitrectomy with removal of peripheral vitreous was performed. The absence of the fish-strike sign with a silicone-tipped cannula confirmed the removal of the posterior vitreous face. Peripheral retinal breaks discovered on indirect ophthalmoscopy were surrounded by laser photocoagulation. We removed epiretinal membranes only if they were visible and caused obvious distortion of the retina. An air– fluid exchange was followed by removal of remnant fluid 5 to 15 minutes later. Autologous serum 0.5 ml was layered across the hole and removed after 10 minutes. Silicone oil (5000 centistoke [cs] in 38 eyes and 1000 cs in 2 eyes) warmed to body temperature was injected into the vitreous cavity through a 19-gauge 3⁄8-inch needle. Any large air bubbles found by indirect ophthalmoscopy were removed to ensure complete fill of the vitreous cavity. To facilitate adjustment of silicone fill, a reservoir of silicone was allowed to enter several centimeters into the infusion cannula. Before the infusion cannula was removed, the vitreous fill was adjusted with injection or aspiration of silicone by a syringe attached to the infusion cannula to produce a scale reading be-
tween 10 and 15 on a Schiøtz tonometer with a 5.5-g weight. This step was designed to prevent underfill or overfill of the vitreous cavity by silicone. Removal of the infusion cannula and closure of the sclerotomy generally increased the intraocular pressure to a Schiøtz scale reading between 5 and 10 (7–17 mmHg). The night of surgery, the patient was instructed to sleep facedown. Afterward, the only restriction imposed on a patient was to avoid faceup positioning to prevent remnant fluid from reaching the macular hole. One day after the oil insertion, the vitreous cavity was inspected for a visible meniscus to estimate the percentage of fill. The hole was examined through the oil for closure (Fig 1). Patients were usually able to return to work and normal activity 5 days after oil insertion. The oil was removed at the patient’s convenience after 4 weeks minimum (average, 6.5; maximum; 15 weeks). Controlled removal of the oil was performed with 500-mmHg suction from the vitrectomy console or passive efflux. When the oil was removed with suction, a 19-gauge, 3⁄8-inch beveled cannula connected to the vitrectomy console with 4-mm internal-diameter 50-cm long tubing reduced resistance to flow of the viscous oil. This tubing is commonly available for oxygen administration through a nasal cannula. When irrigation fluid began to enter the cannula and tubing, the suction was reduced quickly to avoid turbulence of the fluid in the vitreous cavity. To remove the last of the silicone, the patient’s head and eye were turned to place the sclerotomy through which the silicone was being drained at the most superior point. The last bubble of oil was removed through the cannula inserted 1.5 mm in the superiorly positioned sclerotomy. Remnant oil found on indirect ophthalmoscopy was removed with a siliconetipped needle or larger bore cannula if necessary. Silicone oil adhering to a silicone intraocular lens was removed with suction through a bent cannula with a silicone tip beveled to allow the cut end of the silicone tubing to fit flush on the back surface of the lens. Best-corrected visual acuity was assessed by an independent observer with an Early Treatment Diabetic Retinopathy Study (ETDRS) chart or with a projected Snellen chart standardized to the ETDRS chart. Visual acuity on the last examination after oil removal (or the best postoperative visual acuity if there was progressive cataract) was compared to the preoperative visual acuity. After the silicone was removed, the hole was examined with at least 16 magnification with a slit lamp through a 90-diopter or 78-diopter El Bayadi lens. To provide maximum assessment of depth, the slit was oriented horizontally and angled 15° from below. The hole was considered sealed after surgery if the hole edge was not visible or if the hole edge was visible but appeared to be in contact with the retinal pigment epithelium with no cuff of subretinal fluid (Fig 1). The lens clarity was qualitatively graded with the slit lamp from 0 (clear) to 4 (20/200 or worse view of the retina). Comparison of group pairs of discrete or continuous data for significant difference was accomplished with the t test. Group pairs of binomial data were treated similarly with comparison of two proportions. Ninety-five percent confidence intervals were generated from 1.96 times the standard error of the mean. The two planned comparisons of (1) the rate of closure of macular holes stratified by the stage of the macular holes and (2) the logarithm of the minimum angle of resolution (logMAR) vision stratified by success of closure of the macular hole were more strictly compared at ␣⫽0.025 after Bonferroni adjustment for multiple comparisons. We treated eyes as independents. We also tested individual patients as independents by averaging the variables in the two eyes when both eyes of a patient were involved.
2141
Ophthalmology Volume 105, Number 11, November 1998
Figure 1. A, a 500-m hole before intervention in a patient with nonproliferative diabetic retinopathy. B, silicone tamponade. Hole is sealed. Fluid does not have access to hole. C, 3 months after silicone was removed. Hole is sealed. Visual acuity improved from 20/320 to 20/63.
Results Patient and Eye Data There were 40 eyes in 37 patients who ranged from 42 to 81 years of age, averaging 66 years. Treating individual patients as independent data yielded the same results as treating eyes as independent data. There was no significant preponderance of gender (males, 55%; 95% confidence interval [CI95], 40%–70%) or eye (right eye, 58%; CI95, 42%–73%). Forty percent of the holes were stage 2 and 60% were stage 3 or 4. The hole duration ranged from 0.7 to 22 months, averaging 5.6 months, with a median of 3 months. The mean durations of 3.5 months for stage-2 holes and 7.1 for stage-3 or stage-4 hole differed significantly (P1 ⫽ 0.02, where P1 signifies a one-tail test and P2 denotes a two-tail test). The average follow-up was 10.3 months after removal of silicone oil.
Closure Rates The holes were sealed in 32 of the 40 eyes (80%, CI95, 68%–92%) after one silicone fill (Table 1). Eight patients had an eye with an
2142
open hole after one silicone fill. Of these, five patients consented to second silicone fill, and all these holes sealed. Including reoperations, 37 eyes were sealed (92.5%, CI95 84%–100%). There was no significant difference in the rate of closed holes among the three surgeons. The stage-2, stage-3, and stage-4 holes were sealed at the same rate (75% and 83%, respectively; P2 ⫽ 0.53). The average duration of the hole before the vitrectomy and oil insertion was 7.1 months in holes that did not seal or reopened and 6.3 months in holes that sealed, and that did not differ significantly (P1 ⫽ 0.51).
Visual Acuity A change of 0.1 in the logMAR scale is equivalent to a change of one line in the ETDRS visual acuity chart. For all patients, the visual acuity increased an average of 0.26 logMAR units (2.6 lines). The visual acuity did not change significantly in eyes with holes that opened after oil removal (loss of 0.5 line); in the 32 eyes with closed holes after one silicone fill, the visual acuity improved an average of 3.4 lines (P1 ⫽ 0.03) (Fig 2). Seventy-two percent of these eyes with sealed holes increased more than one line. Of the eight eyes with open holes after one intervention, none of the
Goldbaum et al 䡠 Silicone Tamponade of Macular Holes without Position Restrictions Table 1. Outcomes Based on Stratification Closure Rate after 1 Tamponade unless Otherwise Noted
LogMAR before Silicone Inserted (20/X equivalent) 0.87 20/148 0.78 20/121 0.93 20/170
0.61 (0.47 to 0.74) 20/81 (20/59 to 20/110) 0.49 (0.26 to 0.71) 20/61 (20/36 to 20/103) 0.69 (0.52 to 0.85) 20/97 (20/67 to 20/141)
0.88 20/152 0.87 20/148
0.93 (0.62 to 1.25) 20/170 (20/83 to 20/356) 0.52 (0.39 to 0.66) 20/66 (20/49 to 20/91)
1.12 20/264 0.83 20/135
0.76 (0.42 to 1.05) 20/115 (20/53 to 20/224) 0.58 (0.44 to 0.73) 20/76 (20/55 to 20/107)
Stratification
n
All: [mean (CI95)]
40
Stage 2
16
80% (68 to 92) {92.5% (84 to 100)}* 75% (54 to 75)
Stage 3–4
24
83% (68 to 98)
Open
8‡
P2 ⫽ 0.53† 50%§
Closed
32
100%
5
40% (0 to 83)
35
86% (74 to 97)
Previous surgery Primary hole
P1 ⫽ 0.01
LogMAR after Silicone Removed* (20/X equivalent)
LogMAR Change (post-minus pre-silicone)*
>1 line Improvement*
⫺0.26 (⫺0.14 to ⫺0.39)
62% (47 to 78)
35% (20 to 50)
⫺0.29 (⫺0.03 to ⫺0.55)
69% (45 to 92)
56% (31 to 81)
⫺0.24 (⫺0.11 to ⫺0.38)
58% (38 to 78)
21% (4 to 37)
P1 ⫽ 0.38 0.05 (0.40 to ⫺0.30)
P1 ⫽ 0.26 25% (0 to 55)
⫺0.34 (⫺0.22 to ⫺0.47)
72% (56 to 87)
P1 ⫽ 0.03 ⫺0.36 (⫺0.04 to ⫺0.65)
P1 ⫽ 0.01† 60% (5 to 100)
⫺0.25 (⫺0.11 to ⫺0.39)
63% (47 to 79)
P1 ⫽ 0.58
P1 ⫽ 0.90
20/50 or Better*
P1 ⫽ 0.01 12.5% (0 to 35) 41% (24 to 58) P1 ⫽ 0.14 0% 40% (24 to 56) P1 ⫽ 0.08
P1 ⫽ 1-tail test; P2 ⫽ 2-tail test; CI95 ⫽ 95% confidence interval. * Outcomes after one or more silicone tamponades. † Two planned comparisons that required P ⱕ 0.025 for 95% certainty of difference. ‡ Number open after one silicone tamponade. § Percentage of holes open after one silicone tamponade that were closed after a second silicone tamponade.
three eyes that did not have reoperation improved one line after one silicone tamponade, and two of the five eyes that had second silicone tamponade increased at least one line. The logMAR vision improved an average of 2.9 lines for stage-2 holes and 2.4 lines for stage-3 and stage-4 holes (P2 ⫽ 0.38). Fifty-six percent (CI95, 31%– 81%) of stage-2 holes achieved a visual acuity of 20/50 or better after removal of the silicone oil, and 21% (CI95, 4%–37%) of stage-3 or stage-4 holes reached 20/50 or better (P1 ⫽ 0.01). The mean visual acuity after treatment was 20/61 (CI95, 20/36 –
20/103) for stage-2 holes and 20/97 (CI95, 20/67–20/141) for stage-3 or stage-4 holes (Table 2).
Analysis of Macular Holes Reopening or Failing to Seal In eight eyes (20%), the hole opened after the oil was removed. In none of these holes did silicone pass through the hole or under the
Figure 2. Scattergraph of preoperative logarithm of the minimum angle of resolution (logMAR) vision and postoperative logMAR vision stratified by hole closure.
2143
Ophthalmology Volume 105, Number 11, November 1998 Table 2. Conversion of LogMAR to Snellen Visual Acuity LogMAR Value 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
X for 20/X 1002 796 632 502 399 317 252 200 159 126 100 80 63 50 40 32 25 20
retina during the period of silicone fill of the vitreous cavity. Analysis of these eyes found five holes opened after temporary seal—from 9 days to 3 months after the oil was removed. Two holes reopened after cataract removal. One of these eyes had pseudophakic cystoid macular edema develop before the hole reopened. Silicone oil filled the vitreous cavity between 90% and 100% in all the sealed holes. The silicone fill of the vitreous cavity in three eyes with open holes was between 65% and 85%; these holes were open at the first examination after the oil removal.
Previous Surgery and Repeat Surgery Five eyes had previous intervention with gas tamponade for macular hole in the same or the opposite eye. All these patients expressed strong preference for the surgery that used silicone to tamponade the hole, mainly because there was no restriction on patient activity. These patients expressed great aversion to the prolonged facedown positioning with gas. Previous surgery for macular holes affected the seal rate. Three (38%) of the eight holes that reopened after silicone tamponade were reoperations for previous unsuccessful surgery with gas tamponade for macular hole. Two (6%) of the 32 sealed holes were reoperations (P1 ⫽ 0.05). With one fill of the vitreous cavity with silicone oil, 2 (40%) of 5 previous gas failures sealed; whereas, 30 (86%) of 35 primary macular holes sealed with one fill (P1 ⫽ 0.01). The visual acuity after silicone removal was 20/50 or better in 40% of primary holes and in none of the holes that had failed to seal with previous gas tamponade (P1 ⫽ 0.08).
Total Removal of Silicone Oil The mean time was 8 minutes for total removal of silicone oil from the vitreous cavity with aspiration from the vitrectomy console, detailed under the Materials and Methods section above. In all eyes, a minute amount of oil remained in the fluid in the vitreous cavity as fine droplets visible to the patient as small floaters against the backlighting and occasionally as a spot on downgaze. On B-scan ultrasound, these droplets were discernible as a delicate snow pattern.
Adverse Events Of the 40 eyes, 31 were phakic at the time of the oil insertion and removal. Posterior subcapsular vacuoles generally were visible
2144
transiently for the first 2 weeks after insertion of silicone oil. Over the mean observation time of 10.3 months after silicone removal, in 69% (CI95, 52%– 85%) of the phakic eyes, the lens clarity diminished at least one grade point, or a cataract was removed. Two patients had retinal detachment develop after the oil was removed. One detachment was repaired with laser and gas–fluid exchange. The other had proliferative vitreoretinopathy develop. With silicone refill after removal of epiretinal membrane, all but the inferior 20% of the retina remains attached, with improvement in vision of one line. Macular pigment changes were found in two eyes. Both eyes had a reduction in visual acuity.
Discussion Comparison to Other Studies The high rate of seal of macular holes with silicone fill of the vitreous cavity occurred even though patients returned to full activity as soon as they were able. There was no significant difference in the seal rate in stage-2 holes (75%) and stage 3- or stage-4 holes (83%), possibly because the seal rate was high in both groups of eyes. These rates compare favorably with a 75% rate of same-size, smaller, or sealed stage-2 holes10 and 69% seal rate in a comprehensive, multicentered study of stage-3 or stage-4 holes treated with gas fill of the vitreous cavity and strict facedown positioning.3 Comparing the mean visual acuity, the percentage of eyes achieving two lines’ improvement in visual acuity, and the proportion of eyes reaching 20/50 or better, the outcomes of this study were similar to the Vitrectomy for Treatment of Macular Hole Study Group investigations of stage-2 and stage-3 to stage-4 macular holes (Table 1). In that study, the mean visual acuity 6 months after treatment of stage-2 macular holes was 20/68. We calculated the CI95 in the Vitrectomy for Treatment of Macular Hole Study Group report to be 20/41 to 20/112 from 1.96 times the standard error of the mean. Although this outcome appears similar to the mean visual acuity of 20/61 (CI95, 20/36 –20/ 103) in the silicone study, examination of the confidence intervals indicates no significant difference. The population in these two studies differs somewhat, because we used Gass’s definition of stage-2 holes less than 400 m; whereas, the Vitrectomy for Treatment of Macular Hole Study Group used 300 m as the ceiling diameter for central stage-2 holes. In stage-3 and stage-4 macular holes, the mean visual acuity after silicone tamponade in this study was 20/97 (CI95, 20/67–20/141). The mean visual acuity found by the Vitrectomy for Treatment of Macular Hole Study Group after gas tamponade was 20/115 (CI95, 20/96 –20/138). Observation of the confidence intervals also shows these outcomes to be similar. Silicone contact with the crystalline lens for more than 3 months causes cataract development in virtually all eyes.11 Vitrectomy and gas fill of the vitreous cavity to treat macular holes results in cataract formation at a rate higher than occurs without surgery. The Vitrectomy for Treatment of Macular Hole Study Group found an increase of one or
Goldbaum et al 䡠 Silicone Tamponade of Macular Holes without Position Restrictions more grade points (0 – 4 scale) in 82% of treated eyes.3 Similarly, as described under adverse effects above, silicone fill of the vitreous cavity caused an increase in lens opacity (69%, CI95, 52%– 85%). The possibility of confounding variables in this measurement between these two studies is too great to draw any conclusions about differences in the rate of lens opacification between silicone and gas fill of the vitreous cavity.
Factors Affecting Sealing of Macular Hole and Elimination of Subretinal Fluid Cuff Adhesion of Retina to the Retinal Pigment Epithelium. The normal adhesion of the retina to the retinal pigment epithelium is multifactorial.12 These factors include hydrostatic forces induced by differences in fluid flow at various layer interfaces between the vitreous and the sclera, oncotic pressure from absent or reduced protein in subretinal fluid, interdigitation of photoreceptors and pigment epithelial microvilli, cell adhesion molecules between the photoreceptors and the pigment epithelium, and active transport of fluid outward by the pigment epithelium. We make the assumption that these factors are applicable to the adhesion of the fovea and perifoveal retina to the retinal pigment epithelium. After reattachment of retina detached more than a few weeks, it takes at least 4 to 6 weeks for these factors to become re-established and for the adhesive strength to reform between the retina and pigment epithelium.12,13 Before some threshold level of adhesiveness is established, fluid reaching the hole might interfere with the reparative process, allowing some or all of the adhesive factors to fail. Compared to macular holes that seal spontaneously, macular holes sealed by gas tamponade may be more likely to have adhesion enhanced by a glial membrane that occludes the hole or pulls the hole edges together.14,15 Presumably, this cicatrization increases the adhesion of the hole edge and the perifoveal retina to the retinal pigment epithelium. A macular hole can open in the absence of vitreous traction.16 For example, one patient in this study had a full-thickness macular hole develop with a cuff of subretinal fluid 8 years after vitrectomy. The presence of a glial membrane may reduce the likelihood of a hole reopening after it is closed. It is unknown whether the silicone produces adhesion that more resembles gas-sealed holes or spontaneously sealed holes.
Surgical Factors Surgical factors potentially affecting the seal of the macular hole include duration of tamponade, lack of access of fluid to the macular hole, growth factors, pressure from the tamponade pushing the retina against the pigment epithelium, control of forces to open the hole, and positioning of the eye. The duration of the gas tamponade of the retina correlates with the rate of seal of the macular hole.8 Unless the vitreous cavity is refilled, the duration of gas tamponade is limited. As the gas volume diminishes, there is more opportunity for fluid to reach and enter the macular hole. By the time that the volume of gas diminishes below 50%, the
patient is usually no longer restricted to facedown positioning. Consequently, fluid is in constant contact with the hole when the patient is erect. With the bubble still present, head and eye movements can generate strong currents of fluid across the hole, producing shear forces that can weaken an incomplete adhesion. In contrast, a complete or nearly complete fill of the vitreous cavity with silicone prevents fluid from reaching the hole most, if not all, of the time. The percentage of fill of the vitreous cavity remains constant until the oil is removed. The duration of silicone tamponade can be extended long beyond the duration of gas tamponade. The viscosity of 5000-cs oil prevents shear forces from oil flowing across the hole. Analysis of eyes in which macular holes failed to seal indicates that completeness of fill of the vitreous cavity with silicone is a factor in the seal rate. Inflammation and local growth factors augment the reparative process and improve the rate of seal of macular holes.4 – 6,17 Autologous serum appears to improve the hole closure rate (Banker AS, et al. Invest Ophthalmol Vis Sci 1997;38[Suppl]:S736). Presumably, these growth factors are present in the fluid in the vitreous cavity. In a gas-filled eye, as the gas bubble diminishes and the fluid compartment enlarges, in theory, these factors will be diluted. Because the vitreous fluid compartment remains small in a silicone-filled eye, these factors should maintain at high concentration. However, the higher concentration of growth factors may not affect the adhesion of the macular hole if the fluid is prevented from reaching the hole by the silicone oil. Silicone oil is lighter than water. But the specific gravity of silicone is much closer to water than is the specific gravity of any gas. Hence, the flotation force generated by silicone against the retina is much less than that from gas. Yet, the hole seals at least as well with silicone as with gas. The force with which the vitreous replacement holds the retina against the pigment epithelium appears to be less important than the prevention of fluid access to the hole. From the initial publication on the surgical treatment of macular holes, reports on the treatment of macular holes have stressed facedown positioning for 1 to 4 weeks as a recommended part of the procedure. Reinforcing this concept, Holekamp (Holekamp NM, Meredith TA, Mandell BA, et al. Presented at the AAO annual meeting, San Francisco, 1997) found that the major factor affecting the seal rate for different surgeons in a group practice of retinal surgeons was how well the patients followed the facedown regimen. In contrast, Tornambe et al18 found that facedown positioning was not necessary if attention was paid to producing a complete fill of the vitreous cavity with a nonexpansile 15% mixture of perfluoropropane gas. Similar to what we found with the use of silicone, completeness of fill of the vitreous cavity by gas appeared to be important in their report.
Advantages and Disadvantages of Silicone Tamponade of Macular Holes Silicone fill of the vitreous cavity to treat macular holes offers advantages. The patients can resume a normal lifestyle during the period of tamponade. Patients who have
2145
Ophthalmology Volume 105, Number 11, November 1998 monocular vision and are dependent on the eye with the macular hole can have useful vision with silicone fill of the vitreous cavity. The patients may travel by air or ascend in altitude. The silicone volume does not diminish with time; consequently, the access of fluid to the hole does not increase with time. The duration of tamponade can be extended. There are no shear forces acting on the hole. Patients who have had macular holes treated both by gas and by oil strongly prefer the use of oil. Many of these same reasons can be used to argue that silicone tamponade without position restriction is preferable to gas tamponade without position restriction proposed by Tornambe et al. However, Tornambe’s report is unique among macular hole studies in finding that facedown positioning is not necessary with gas tamponade; it remains uncertain whether his findings can be duplicated. Silicone tamponade of macular holes also has disadvantages compared to gas tamponade. A second surgical procedure is necessary to remove the oil. Remnant droplets of oil in the fluid that fills the vitreous cavity after oil removal can be detected by patients, although these are generally not bothersome. The complications that have been reported after the insertion of silicone oil into the vitreous cavity or after its removal occur mostly because of the tendency of vitreoretinal surgeons to manage complex surgical problems with intravitreous silicone left for several months or years.11,19 In contrast, eyes with age-related macular holes are not inflamed; do not have pre-existing retinal detachment; and are not likely to stimulate fibrous, pigment epithelial, or vascular proliferation. The tamponade duration to seal macular holes is not long enough to generate silicone emulsification. We have excluded eyes with poor separation of the anterior chamber and vitreous cavity to avoid silicone being trapped in the anterior chamber. For these reasons, with the exception of cataract formation, we have not observed the complications reported by others in eyes in which silicone has been used.
Eyes for Which Silicone Tamponade Should Be Avoided Silicone bubbles in the anterior chamber are undesirable although acceptable. Silicone constantly in contact with most of the corneal endothelium is unacceptable because of the high probability of corneal complications.11 Silicone is not a good choice when the oil can enter and be trapped in the anterior chamber through defects in the capsule or zonules. Large posterior capsulotomies may still permit silicone tamponade if they are central, with the edges in apposition to the posterior lens surface. Small peripheral capsular and zonular defects could prevent silicone tamponade. Because air–fluid exchange is a step in this surgery, a reasonable test of the quality of the barrier between the vitreous cavity and the anterior chamber is how well the barrier keeps air out of the anterior chamber when the vitreous cavity is filled with air. If no air bubble enters the anterior chamber, it should be safe to use silicone tamponade.
2146
In aphakic eyes with the capsule removed, it is difficult to fill the vitreous cavity more than 90% with silicone oil without it contacting most or all of the corneal endothelium, even with an inferior iridotomy. Consequently, aphakia from intracapsular cataract extraction is a relative contraindication for silicone tamponade of a macular hole. Yet, aphakic eyes with intact posterior capsule and zonules can safely have silicone tamponade. Acknowledgment. The authors thank Fred Wright, the Biostatistics Core of the University of California San Diego Cancer Center, for advice in the statistical analysis in this study.
References 1. Kelly NE, Wendel RT. Vitreous surgery for idiopathic macular holes. Results of a pilot study. Arch Ophthalmol 1991; 109:654 –9. 2. Wendel RT, Patel AC, Kelly NE, et al. Vitreous surgery for macular holes. Ophthalmology 1993;100:1671– 6. 3. Freeman WR, Azen SP, Kim JW, et al. Vitrectomy for the treatment of full-thickness stage 3 or 4 macular holes. Results of a multicentered randomized clinical trial. The Vitrectomy for Treatment of Macular Hole Study Group. Arch Ophthalmol 1997;115:11–21. 4. Glaser BM, Michels RG, Kuppermann BD, et al. Transforming growth factor-beta 2 for the treatment of full-thickness macular holes. A prospective randomized study. Ophthalmology 1992;99:1162–72. 5. Liggett PE, Skolik SDS, Horio B, et al. Human autologous serum for the treatment of full-thickness macular holes. A preliminary study. Ophthalmology 1995;102:1071– 6. 6. Gaudric A, Massin P, Paques M, et al. Autologous platelet concentrate for the treatment of full-thickness macular holes. Graefes Arch Clin Exp Ophthalmol 1995;233:549 –54. 7. Del Priore LV, Kaplan HJ, Bonham RD. Laser photocoagulation and fluid– gas exchange for recurrent macular hole. Retina 1994;14:381–2. 8. Thompson JT, Smiddy WE, Glaser BM, et al. Intraocular tamponade duration and success of macular hole surgery. Retina 1996;16:373– 82. 9. Gass JDM. Reappraisal of biomicroscopic classification of stages of development of a macular hole. Am J Ophthalmol 1995;119:752–9. 10. Ruby AJ, Williams DF, Grand MG, et al. Pars plana vitrectomy for treatment of stage 2 macular holes. Arch Ophthalmol 1994;112:359 – 64. 11. Federman JL, Schubert HD. Complications associated with the use of silicone oil in 150 eyes after retina–vitreous surgery. Ophthalmology 1988;95:870 – 6. 12. Marmor MF. Mechanisms of normal retinal adhesion. In: Ryan SJ, ed. Retina, 2nd ed. St. Louis: Mosby, 1994; v. 3, chap. 123. 13. Yoon YH, Marmor MF. Rapid enhancement of retinal adhesion by laser photocoagulation. Ophthalmology 1988; 95:1385– 8. 14. Madreperla SA, Geiger GL, Funata M, et al. Clinicopathologic correlation of a macular hole treated by cortical vitreous peeling and tamponade. Ophthalmology 1994;101: 682– 6. 15. Funata M, Wendel RT, de la Cruz Z, Green WR. Clinicopathologic study of bilateral macular holes treated with pars plana vitrectomy and gas tamponade. Retina 1992;12:289 – 98.
Goldbaum et al 䡠 Silicone Tamponade of Macular Holes without Position Restrictions 16. Gordon LW, Glaser BM, Ie D, et al. Full-thickness macular hole formation in eyes with a pre-existing complete posterior vitreous detachment. Ophthalmology 1995;102:1702–5. 17. Nagineni CN, Kutty RK, Detrick B, Hooks JJ. Inflammatory cytokines induce intercellular adhesion molecule-1 (ICAM-1) mRNA synthesis and protein secretion by human retinal pigment epithelial cell cultures. Cytokine 1996;8: 622–30.
18. Tornambe PE, Poliner LS, Grote K. Macular hole surgery without face-down positioning. A pilot study. Retina 1997;17: 179 – 85. 19. Casswell AG, Gregor ZJ. Silicone oil removal. II. Operative and postoperative complications. Br J Ophthalmol 1987;71: 898 –902.
Discussion by Paul E. Tornambe, MD Goldbaum et al present a multicenter retrospective pilot study that asks the question, “Is silicone tamponade of macular holes safe and effective?” Surgery was performed by 3 surgeons and included 40 eyes, 8 of which had previous macular hole surgery. Forty percent were stage 2 (defined as a hole 400 m or less in diameter), 60% were stage 3 or 4. The mean duration of the holes was 3 months. Preoperative acuity was 20/50 or worse. All surgeons used the same surgical technique consisting of a pars plana vitrectomy, removal of the posterior hyaloid, removal of epiretinal membranes when visually present and resulting in significant macular distortion, autologous serum, and a silicone oil fill. Facedown positioning was done over the first postoperative night only. No facedown positioning was done thereafter and no restrictions to the patient’s lifestyle were required. Patients were free to travel by air or ascend to any elevation. The oil was removed, on average approximately 6.5 weeks after insertion. Removal was done in all eyes, took less than 10 minutes, and was not associated with any operative complications except failure to remove all the oil droplets. Follow-up after oil removal averaged 10.3 months. The 1-fill operative success for all 40 eyes is 80%, and the rate for the 35 eyes with virgin macular holes is 86%. The 1-fill operative success rate for the five holes operated on previously with gas was 40%. Eight eyes failed permanent closure, with two holes reopening after cataract surgery. Of these, five underwent repeat macular hole surgery (MHS) with silicone oil and all closed, giving an overall attachment rate of 92.5%. Acuity improved an average of 3.4 lines. However, eyes cured with one procedure attained more visual improvement than eyes that failed the first attempt. The mean postoperative visual acuity was 20/61 for stage-2 holes and 20/97 for stage-3 and stage-4 holes. Visual acuity of 20/50 or better was attained in 56% of stage-2 holes and in 21% of stage-3 or stage-4 holes. No eyes that failed the first procedure ultimately attained 20/50 or better visual acuity. The authors’ objective was to determine whether macular hole surgery with silicone tamponade without facedown positioning is safe and effective. Their data, when compared with other studies, suggest that this technique probably is effective. However, without prospective randomization, it cannot be proved better or worse than other techniques. Many variables have been shown to be important and are likely inter-related, including size and duration of the hole and preoperative visual acuity. Goldbaum’s 86% one operation closure rate for primary MHS repair compares favorably with the literature. However, the visual results appear disappointing, especially for stage-2 eyes, in which only 56% attained 20/50 or better visual acuity. Tornambe et al1 found preoperative acuity a better predictor of
From Retina Consultants, San Diego, Mericose Eye Institute, La Jolla, California.
visual outcome than hole stage. Eighty percent of eyes with 20/100 or better preoperative visual acuity attained 20/50 or better postoperative visual acuity. Stage-2 holes usually have better preoperative visual acuity, and one would therefore expect better postoperative visual acuity than Goldbaum reports. This may be explained by the limited follow-up of only 10.3 months and the progression of cataract change. Leonard et al2 recently has shown that macular function improves over the first 3 postoperative years. However, there may be other explanations for less visual return. Silicone may be toxic and limit potential photoreceptor recovery. Surgical technique also must be considered. Goldbaum’s group removed only obvious epiretinal macular membrane (ERM) tissue and only if it significantly distorted the macula. Our recent experience with the Tano diamond-dusted soft-tip cannula (Synergetics, St. Louis, MO) suggests that a fine ERM, not frequently obvious, surrounds the hole and usually is attached firmly to the edges of the hole. This membrane can be removed without dissecting the underlying glistening membrane, which probably is internal limiting membrane. We believe that this subtle membrane may be important in the pathogenesis and persistence of the macular dehiscence and believe that removing it allows the edges of the hole to reapproximate. Removing the membrane may also stimulate growth factor production responsible for the formation of glial tissue that bridges the hole. We agree with Goldbaum’s observation that there are two types of postoperative endpoints. In one, the hole is flat against the RPE but the round defect is visible. In the other, the hole shrinks in size to a pinpoint defect. The former hole persistence may be caused by incomplete removal of surface traction, which prevents the reapproximation of the macular dehiscence and limits optimal visual recovery. Goldbaum’s group’s failure to routinely attempt to remove this less-than-obvious tissue may have prevented reapproximation of the edges of the hole and thus limited visual return. The authors did not mention which holes had ERM dissection regarding the type of hole closure endpoint and the degree of visual recovery. What information can we extract from this effort and apply clinically? First, MHS using silicone oil without facedown positioning works and likely does so by isolating the hole from liquid vitreous. Silicone is much less buoyant than gas. This provides additional evidence that the buoyancy of the bubble, and therefore prone positioning, may not be as important as waterproofing the hole. It would be interesting to evaluate visual fields on these patients. Gas bubble buoyancy may play an adverse role in the development of post-MHS field defects. This study also indicates that for those patients who do not wish to assume facedown positioning or must travel by air or to higher elevations, silicone oil may be reasonable. This technique may be particularly applicable to young patients with phakia, such as children or those young adults who wish an
2147
Ophthalmology Volume 105, Number 11, November 1998 uninterrupted lifestyle and are less prone to postvitrectomy cataract change. The study also finds that single operation success is important, and best acuity results after a single attempt. One reason for failure, as Goldbaum indicates, may be subtotal silicone oil fill. The major drawback of this approach, in addition to possible silicone toxicity, is the need for a second operation and those tiny silicone oil bubbles that are always left behind. In pseudophakic eyes, in which altitude is not an issue, we would prefer to use a perfluoropropane gas fill without prone positioning. The controversy still continues whether or not to combine lens replacement surgery with MHS. Goldbaum and others have observed that most eyes will develop a significant cataract and some holes reopen
2148
shortly after cataract surgery, which suggests more reasons to combine cataract extraction with macular hole surgery.
References 1. Tornambe PE, Poliner LS, Grote K. Macular hole surgery without face-down positioning. A pilot study. Retina 1997;17: 179 – 85. 2. Leonard RE, Smiddy WE, Flynn HW, Feuer W. Long-term visual outcomes in patients with successful macular hole surgery. Ophthalmology 1997;104:1648 –52.