- Email: [email protected]
Use of an Intraocular Gas Tamponade to Find Retinal Breaks
USE O F AN INTRAOCULAR GAS TAMPONADE T O F I N D R E T I N A L BREAKS HARVEY L I N C O F F ,
M.D.
New York, New York INGRID KREISSIG,
M.D.
Tubingen, West Germany AND D. JACKSON C O L E M A N , M . D . ,
AND S T A N L E Y C H A N G ,
M.D.
New York, New York
We used the receding meniscus of an intraocular bubble being absorbed or the advancing meniscus of an expanding bubble to find the levels of retinal breaks in five patients with retinal detachments in which the breaks could not be found by the usual examination tech niques. When we used a receding meniscus, the subretinal fluid was drained and the volume replaced by a perfluorocarbon gas calculated to fill the eye below the probable level of the retinal break. The bubble closed the break and maintained reattachment until the gas was absorbed. The level of the meniscus when redetachment first occurred indicated the level of the break. When we used an advancing meniscus, small amounts of perfluorocarbons were injected without drainage of subretinal fluid. As the gas bubble expanded it reattached the retina from above downward. When the bubble covered the retinal break, it tamponaded it and the remaining fluid absorbed within hours. The level of the meniscus when the retina became reattached marked the level of the retinal break. We used an intraocular bubble of per fluorocarbon gas to find the level of the retinal break in an extensive detachment in which a break could not be found by the usual examination techniques. The
Accepted for publication July 5, 1983. From the Department of Ophthalmology, New York Hospital-Cornell Medical Center, New York, New York (Drs. Lincoff, Coleman, and Chang); and the Department of Ophthalmology, UniversitatsAugenklinik, Tubingen, West Germany (Dr. Kreissig). This study was supported by grants from Daphne and Albert Roe and Prince Abdullah Faisal. Presented at the meeting of the Retina Society, Cancun, Mexico, Feb. 26, 1983. Reprint requests to Harvey Lincoff, M.D., De partment of Ophthalmology, New York Hospital, 525 E. 68th St., New York, NY 10021. 510
method is more sensible and less hazard ous than the multiple encircling opera tions that have been used previously. M A T E R I A L AND M E T H O D S
We used one or more of the straightchain perfluorocarbon gases, perfluoromethane, perfluoroethane, perfluoropropane, and perfluoro-n-butane, in every case. The characteristics of the gases have been described previously. 1,2 Their exceptional insolubility causes an intraocular bubble of the gases to expand because of the diffusion of the relative ly soluble blood gases into the bubble. The expansion varies directly with the length of the carbon chain; thus, perfluoromethane expands 1.9 times, perflu-
©AMERICAN JOURNAL OF OPHTHALMOLOGY 96:510-516, 1983
VOL. 96, NO. 4
LOCATING RETINAL BREAKS
oroethane expands 3.3 times, perfluoropropane expands four times, and perfluoro-n-butane expands five times, The expanding bubble displaces the subretinal fluid and the aqueous portion of the vitreous during a one- to four-day period, slowly enough for an eye with a
511
normal outflow to compensate without an excessive increase in intraocular pressure. The five patients in this report were selected for the diagnostic gas technique because they had extensive retinal detachments but no detectable retinal breaks (except for a possible macular
Fig. 1 (LincofFand associates). Case 1. Top left, An inferior retinal detachment extending from the 2 to 10 o'clock positions. A functioning retinal break could not be found. The break on a radial buckle at the 10 o'clock position appeared to have been treated adequately. The two areas of thin retina of the 9:30 and 8:30 o'clock positions corresponded with areas of previous coagulation. Elsewhere temporally there was spotty pigmentary atrophy from coagulation on and anterior to an encircling buckle. Top right, At the operation subretinal fluid was drained and 1.5 ml of perfluoromethane was injected. A vitreous membrane (curve) was brought into sharp relief by the gas bubble. Bottom left, By the first postoperative day the bubble had expanded to a predicted volume of 2.85 ml. The retina continued to be attached. Bottom right, On the ninth day, as the meniscus of the bubble retreated across the radial buckle, the retina became redetached, indicating a functioning break at the 9:45 o'clock position.
512
AMERICAN JOURNAL OF OPHTHALMOLOGY
OCTOBER, 1983
Fig. 2 (Lincoff and associates). Case 2. Top left, The detachment is total except for a rim of attached retina anterior to the buckle superiorly. The patient had 6 diopters of myopia and a posterior staphyloma. There was a high (20%) encircling buckle with lattice degeneration on it at the 10 and 3 o'clock positions. Top right, Injection of 0.5 ml of perfluoropropane. Bottom left, By day 3 the gas bubble had expanded to a predicted volume of 2 ml and tamponaded the superior third of the retina, but the retina below the bubble continued to be detached.
hole) despite repeated examinations by two examiners who used indirect ophthalmoscopy and slit lamp biomicroscopy. In two patients reattachment was not achieved despite several procedures, in cluding encircling buckles. In three pa tients, local buckles in suspected areas had failed. In the first patient the plan was to drain subretinal fluid and restore ocular vol ume with an injection of a perfluorocarbon gas calculated to tamponade the reti na below the suspected level of the break when the patient was in an upright posi
tion. The gas bubble would temporarily close the break and maintain reattach ment until the bubble receded above the break. The level of the meniscus at the time detachment first reappeared would indicate the horizontal meridian of the retinal break. We later decided that if a total retinal detachment was present, in which the probability was 93% that the break was in the superior retina, 3 it would be better to inject a small amount (0.3 to 0.5 ml) of expanding gas without draining subreti nal fluid. As the gas bubble expanded, it
VOL. 96, NO. 4
LOCATING RETINAL BREAKS
513
Fig. 3 (Lincoff and associates). Case 2. Top left, On day 7, with the bubble retreating and no evidence of retinal flattening, 0.3 ml of perfluoro-n-butane was injected. Top right, By day 10 the bubble had expanded to an estimated 3 ml without any effect on the inferior detachment. An additional 0.3 ml of perfluoro-n-butane was injected. Bottom left, By day 13 the bubble had expanded to 4.5 ml and the inferior retina had become reattached. There was a crescent-shaped tear at the nasal edge of the optic disk.
Day 13
would displace subretinal fluid and reattach the retina from the top of the eye downward. When the bubble crossed the retinal break, it would tamponade it and the remaining subretinal fluid would be absorbed within hours. The level of the meniscus when the inferior retina reat tached would indicate the level of the retinal break. Gas injections combined with drainage of subretinal fluid were done with the patients under retrobulbar anesthesia in the operating room. Injections without drainage were done with the patients
under topical anesthesia in the outpatient treatment room. CASE REPORTS Case 1—A 52-year-old patient with 2 diopters of hyperopia had a recurring detachment after three buckling operations done over a period of six weeks. An open retinal break could not be found. The suspected area was between the 8:30 and 10 o'clock positions, an area that had been coagulated and buckled in the three previous operations (Fig. 1, top left). We estimated that approximately 3 ml of gas would be sufficient to tamponade the retina below the 8:30 o'clock position. The subretinal fluid was drained and 1.5 ml of perfluoromethane was injected (Fig. 1, top right). One day later the bubble had expanded to the predicted volume of 2.85 ml and
514
AMERICAN JOURNAL OF OPHTHALMOLOGY
covered the suspected area (Fig. 1, bottom left). The retina became redetached on the ninth day when the receding meniscus crossed the radial buckle at the 9:45 o'clock position (Fig. 1, bottom right). We con cluded that a vitreous membrane, made evident by the gas bubble, was holding open a break at the 9:45 o'clock position that appeared to have been ade quately treated. The membrane was severed by a vitrectomy limited to the area of the radial buckle and the retina became reattached. The patient's visual acuity after one year was 20/50. Case 2—A 42-year-old patient with 6 diopters of myopia had had a recurring, nearly total detachment for three months (Fig. 2, top left). He had had three previous buckle operations and 360 degrees of cryocoagulation. Without draining subretinal fluid, we injected 0.5 ml of perfluoropropane (Fig. 2, top right). By the third day it had expanded to 2 ml and filled the upper third of the eye, but the inferior retina remained detached (Fig. 2, bottom left). On day 7, with no improvement evident and the bubble receding, we injected 0.3 ml of perfluoro-n-butane (Fig. 3, top left). By day 10 the bubble had expanded to about 3 ml and approached the optic disk but there still was no discernible flattening of the inferior detachment. We injected an additional 0.3 ml of perfluoro-n-butane (Fig. 3, top right), and, within hours after the expanding bubble crossed the optic disk, the inferior retina became reattached (Fig. 3, bottom left). We concluded that there was a break in the posterior staphyloma at the level of the optic disk, possibly a macular hole. The macula was cystic. A vitrectomy was done, and the operating micro scope showed the macula to be intact, but a break was discovered at the nasal edge of the optic disk. The break was sealed with intraocular diathermy. The retina has remained attached. The patient's visual acuity after six months was 20/100. Case 3—A 68-year-old patient with 18 diopters of myopia had had a total detachment for three days. A temporary buckle beneath a suspected area at the 1 o'clock position had no effect (Fig. 4). We thought that the break must be elsewhere in the superior third of the retina or in the posterior staphyloma. The eye was first injected with 0.5 ml of perfluoropropane and then injected three times at four-day intervals with 0.3 ml of perfluoro-n-butane. Four days after the fourth injection (day 16) the gas bubble crossed the superior edge of the optic disk, but the inferior retina remained detached. The estimated volume of the expanded bubble was then 6 ml. Ultrasonography disclosed that the anteroposterior length of the eye was 37 mm and it was apparent that more than 0.3 ml would be necessary for the bubble to tamponade the extent of the staphyloma. At the suggestion of one of the authors (D.J.C.), 0.75 ml of dilute gas was withdrawn from the eye and exchanged for an equal volume of perfluoro-n-butane (Fig. 5, top left and top right). The meniscus of the resulting bubble crossed the inferior edge of the staphyloma by the next day, and the retina became reattached overnight (Fig. 5, bottom left). The bub ble continued to expand for three more days and measured about 9 ml on day 20 (Fig. 5, bottom
OCTOBER, 1983
right). Slit-lamp biomicroscopy confirmed a break at the inferior edge of the staphyloma, and the break was treated with a laser. Thirteen days later, as the meniscus retreated across the inferior edge of the staphyloma, the break reopened and the inferior retina became redetached. A vitrectomy and intraoc ular diathermy around the break finally sealed it. The inferior retina (but not the retina in the staphyloma) later became redetached because of an iatrogenic break in the inferior periphery. A scierai buckle reattached the retina briefly, but it became rede tached again because of preretinal proliferation in the inferior periphery that could not be relieved. Cases 4 and 5—These two patients, who were treated in Tubingen, both had myopia, one 3 diopters and the other 5 diopters. They had total detach ments, apparent macular holes, and suspicious areas in the superior periphery. The initial procedure for both patients was a radial buckle beneath the area of suspicion. In one patient the upper third of the retina became reattached. In the other patient the buckle operation had no effect. Both patients then received injections of 0.7 ml of perfluoroethane after ocular massage. Each bubble expanded to about 2.3 ml and approached the upper border of the optic disk. After five days the inferior detachments were unchanged in both patients. To enlarge the bubble, we withdrew 0.5 ml of dilute gas and substituted 0.5 ml of perfluo ropropane. Within hours after the menisci of the expanding bubbles crossed the level of the maculas, the inferior retinas became reattached. In one pa tient the break, which was slightly nasal to the fovea, was treated through the gas bubble with a krypton laser. In the other patient the hole was directly on the fovea and instead of receiving laser treatment it was tamponaded for 16 days (by bending the patient's
Fig. 4 (Lincoff and associates). Case 3. A total detachment with suspected breaks at the 1 o'clock position in the periphery and at the inferior edge of a posterior staphyloma.
VOL. 96, NO. 4
LOCATING RETINAL BREAKS
515
Fig. 5 (Lincotf and associates). Case 3. By 16 days after several small injections of gas the bubble had expanded to the level of the optic disk. Its estimated volume was 6 ml. The inferior retina remained detached. We withdrew 0.75 ml of the bubble (top left, arrows) and substituted 0.75 ml of perfluoro-n-butane (top right). Bottom left, One day later (day 17) the bubble had reached the inferior edge of the staphyloma and covered the suspected retinal break. The inferior retina had become reattached. Bottom right, By day 20 the bubble had expanded to an estimated volume of 9 ml. head forward when the bubble receded beyond the macula). The retinas of both patients have remained attached. The visual acuity of the laser-treated pa tient after three months was 20/300; that of the other patient after three months was 20/100. DISCUSSION
The idea of using a gas bubble as a diagnostic tool developed as a result of daily observations of patients participat
ing in a clinical trial of the perfluorocarbon gases. 4 Most of these patients have had fluid-gas exchanges during vitrectomy operations for retinal detachment. Postoperatively the patients were exam ined twice daily while sitting to deter mine the level of the meniscus of the bubble and the state of the retina. If redetachment occurred, it appeared first
516
AMERICAN JOURNAL OF OPHTHALMOLOGY
in the inferior retina when the level of the meniscus rose above the level of an inad equately secured retinal break. The loca tions of the breaks, some of which were iatrogenic, were for the most part known. By observing the inferior retina as the meniscus of the bubble approached and then crossed the breaks, we were able to localize this cause of failure. It seemed likely that an absorbing bub ble of intraocular gas could also be used to indicate the level of an undetected retinal break in a detachment. Our Case 1 demonstrated the feasibility of the meth od. A diagnostic technique that requires previous drainage of subretinal fluid, however, may be regarded as relatively hazardous. The expanding bubble tech nique without drainage, used in our sec ond and third patients, is less hazardous but can be arduous and time-consuming if many injections are required. The ex change of dilute gas for pure gas, as was done in our Cases 3, 4, and 5, made it possible to obtain a diagnostic bubble of any size and, if necessary, to displace the entire vitreous volume with two injec tions. Although the expanding bubble could, in theory, displace superior fluid and make the inferior detachment more bullous, we have not observed this to any significant extent, possibly because ex pansion is slow and subretinal fluid re turns through the break to the vitreous compartment until the break is tamponaded. Another possible problem, one which we have not yet encountered, is that delayed absorption of subretinal fluid could make it difficult to determine the level of the break. Our 20 years of experi ence with nondrainage operations,6 how ever, indicated that delayed absorption of subretinal fluid is infrequent, that old fluid is absorbed as rapidly as recent fluid
OCTOBER, 1983
(Patient 1 had a detachment for six weeks, Patient 2 for three months, and Patient 4 for more than one year), and that delayed absorption affects only a fraction of the subretinal fluid (50% or more is absorbed rapidly). Thus, it should be possible to discern a distinct flattening of the inferior detachment when the meniscus crosses the break. If there is still doubt, there is a second opportunity to confirm the level of the break when the bubble begins to be absorbed; when the meniscus recedes across the break, the residual detachment should increase abruptly. Multiple breaks should not preclude using this method. With the expanding bubble technique, the first break to be closed would be the most superior one. As a consequence the upper borders of the detachment should drop to a level corresponding to the location of the sec ond break. If ophthalmoscopy and biomicroscopy still do not detect the break, a second diagnostic bubble might be indicated. In principle this technique is similar to sequential buckling of a retinal detachment from the top down ward.6 REFERENCES 1. Lincoff, H., Mardirossian, J., Lincoff, A., Ligget, P., Iwamoto, T., and Jakobiec, F.: Intravitreal longevity of three perfluorocarbon gases. Arch. Ophthalmol. 98:1610, 1980. 2. Lincoff, A., Haft, D., Ligget, P., and Reifer, C : Intravitreal expansion of perfluorocarbon bub bles. Arch. Ophthalmol. 98:1646, 1980. 3. Lincoff, H., and Gieser, R. : Finding the retinal hole. Arch. Ophthalmol. 85:565, 1971. 4. Lincoff, H., Kreissig, I., Richard, G., Coleman, J., Chang, S., and Wilcox, L. M.: The perfluo rocarbon gases in the treatment of retinal detach ment. Ophthalmology 90:546, 1983. 5. Lincoff, H., Kreissig, I.: The treatment of retinal detachment without drainage of subretinal fluid (modifications of the Custodis procedure. Part VI). Trans. Am. Acad. Ophthalmol. Otolaryngol. 76:1221, 1972. 6. Lincoff, H., Kreissig, I., and Goldbaum, M.: Reasons for failure in non-drainage operations. Mod. Probl. Ophthalmol. 12:40, 1974.
M.D.
New York, New York INGRID KREISSIG,
M.D.
Tubingen, West Germany AND D. JACKSON C O L E M A N , M . D . ,
AND S T A N L E Y C H A N G ,
M.D.
New York, New York
We used the receding meniscus of an intraocular bubble being absorbed or the advancing meniscus of an expanding bubble to find the levels of retinal breaks in five patients with retinal detachments in which the breaks could not be found by the usual examination tech niques. When we used a receding meniscus, the subretinal fluid was drained and the volume replaced by a perfluorocarbon gas calculated to fill the eye below the probable level of the retinal break. The bubble closed the break and maintained reattachment until the gas was absorbed. The level of the meniscus when redetachment first occurred indicated the level of the break. When we used an advancing meniscus, small amounts of perfluorocarbons were injected without drainage of subretinal fluid. As the gas bubble expanded it reattached the retina from above downward. When the bubble covered the retinal break, it tamponaded it and the remaining fluid absorbed within hours. The level of the meniscus when the retina became reattached marked the level of the retinal break. We used an intraocular bubble of per fluorocarbon gas to find the level of the retinal break in an extensive detachment in which a break could not be found by the usual examination techniques. The
Accepted for publication July 5, 1983. From the Department of Ophthalmology, New York Hospital-Cornell Medical Center, New York, New York (Drs. Lincoff, Coleman, and Chang); and the Department of Ophthalmology, UniversitatsAugenklinik, Tubingen, West Germany (Dr. Kreissig). This study was supported by grants from Daphne and Albert Roe and Prince Abdullah Faisal. Presented at the meeting of the Retina Society, Cancun, Mexico, Feb. 26, 1983. Reprint requests to Harvey Lincoff, M.D., De partment of Ophthalmology, New York Hospital, 525 E. 68th St., New York, NY 10021. 510
method is more sensible and less hazard ous than the multiple encircling opera tions that have been used previously. M A T E R I A L AND M E T H O D S
We used one or more of the straightchain perfluorocarbon gases, perfluoromethane, perfluoroethane, perfluoropropane, and perfluoro-n-butane, in every case. The characteristics of the gases have been described previously. 1,2 Their exceptional insolubility causes an intraocular bubble of the gases to expand because of the diffusion of the relative ly soluble blood gases into the bubble. The expansion varies directly with the length of the carbon chain; thus, perfluoromethane expands 1.9 times, perflu-
©AMERICAN JOURNAL OF OPHTHALMOLOGY 96:510-516, 1983
VOL. 96, NO. 4
LOCATING RETINAL BREAKS
oroethane expands 3.3 times, perfluoropropane expands four times, and perfluoro-n-butane expands five times, The expanding bubble displaces the subretinal fluid and the aqueous portion of the vitreous during a one- to four-day period, slowly enough for an eye with a
511
normal outflow to compensate without an excessive increase in intraocular pressure. The five patients in this report were selected for the diagnostic gas technique because they had extensive retinal detachments but no detectable retinal breaks (except for a possible macular
Fig. 1 (LincofFand associates). Case 1. Top left, An inferior retinal detachment extending from the 2 to 10 o'clock positions. A functioning retinal break could not be found. The break on a radial buckle at the 10 o'clock position appeared to have been treated adequately. The two areas of thin retina of the 9:30 and 8:30 o'clock positions corresponded with areas of previous coagulation. Elsewhere temporally there was spotty pigmentary atrophy from coagulation on and anterior to an encircling buckle. Top right, At the operation subretinal fluid was drained and 1.5 ml of perfluoromethane was injected. A vitreous membrane (curve) was brought into sharp relief by the gas bubble. Bottom left, By the first postoperative day the bubble had expanded to a predicted volume of 2.85 ml. The retina continued to be attached. Bottom right, On the ninth day, as the meniscus of the bubble retreated across the radial buckle, the retina became redetached, indicating a functioning break at the 9:45 o'clock position.
512
AMERICAN JOURNAL OF OPHTHALMOLOGY
OCTOBER, 1983
Fig. 2 (Lincoff and associates). Case 2. Top left, The detachment is total except for a rim of attached retina anterior to the buckle superiorly. The patient had 6 diopters of myopia and a posterior staphyloma. There was a high (20%) encircling buckle with lattice degeneration on it at the 10 and 3 o'clock positions. Top right, Injection of 0.5 ml of perfluoropropane. Bottom left, By day 3 the gas bubble had expanded to a predicted volume of 2 ml and tamponaded the superior third of the retina, but the retina below the bubble continued to be detached.
hole) despite repeated examinations by two examiners who used indirect ophthalmoscopy and slit lamp biomicroscopy. In two patients reattachment was not achieved despite several procedures, in cluding encircling buckles. In three pa tients, local buckles in suspected areas had failed. In the first patient the plan was to drain subretinal fluid and restore ocular vol ume with an injection of a perfluorocarbon gas calculated to tamponade the reti na below the suspected level of the break when the patient was in an upright posi
tion. The gas bubble would temporarily close the break and maintain reattach ment until the bubble receded above the break. The level of the meniscus at the time detachment first reappeared would indicate the horizontal meridian of the retinal break. We later decided that if a total retinal detachment was present, in which the probability was 93% that the break was in the superior retina, 3 it would be better to inject a small amount (0.3 to 0.5 ml) of expanding gas without draining subreti nal fluid. As the gas bubble expanded, it
VOL. 96, NO. 4
LOCATING RETINAL BREAKS
513
Fig. 3 (Lincoff and associates). Case 2. Top left, On day 7, with the bubble retreating and no evidence of retinal flattening, 0.3 ml of perfluoro-n-butane was injected. Top right, By day 10 the bubble had expanded to an estimated 3 ml without any effect on the inferior detachment. An additional 0.3 ml of perfluoro-n-butane was injected. Bottom left, By day 13 the bubble had expanded to 4.5 ml and the inferior retina had become reattached. There was a crescent-shaped tear at the nasal edge of the optic disk.
Day 13
would displace subretinal fluid and reattach the retina from the top of the eye downward. When the bubble crossed the retinal break, it would tamponade it and the remaining subretinal fluid would be absorbed within hours. The level of the meniscus when the inferior retina reat tached would indicate the level of the retinal break. Gas injections combined with drainage of subretinal fluid were done with the patients under retrobulbar anesthesia in the operating room. Injections without drainage were done with the patients
under topical anesthesia in the outpatient treatment room. CASE REPORTS Case 1—A 52-year-old patient with 2 diopters of hyperopia had a recurring detachment after three buckling operations done over a period of six weeks. An open retinal break could not be found. The suspected area was between the 8:30 and 10 o'clock positions, an area that had been coagulated and buckled in the three previous operations (Fig. 1, top left). We estimated that approximately 3 ml of gas would be sufficient to tamponade the retina below the 8:30 o'clock position. The subretinal fluid was drained and 1.5 ml of perfluoromethane was injected (Fig. 1, top right). One day later the bubble had expanded to the predicted volume of 2.85 ml and
514
AMERICAN JOURNAL OF OPHTHALMOLOGY
covered the suspected area (Fig. 1, bottom left). The retina became redetached on the ninth day when the receding meniscus crossed the radial buckle at the 9:45 o'clock position (Fig. 1, bottom right). We con cluded that a vitreous membrane, made evident by the gas bubble, was holding open a break at the 9:45 o'clock position that appeared to have been ade quately treated. The membrane was severed by a vitrectomy limited to the area of the radial buckle and the retina became reattached. The patient's visual acuity after one year was 20/50. Case 2—A 42-year-old patient with 6 diopters of myopia had had a recurring, nearly total detachment for three months (Fig. 2, top left). He had had three previous buckle operations and 360 degrees of cryocoagulation. Without draining subretinal fluid, we injected 0.5 ml of perfluoropropane (Fig. 2, top right). By the third day it had expanded to 2 ml and filled the upper third of the eye, but the inferior retina remained detached (Fig. 2, bottom left). On day 7, with no improvement evident and the bubble receding, we injected 0.3 ml of perfluoro-n-butane (Fig. 3, top left). By day 10 the bubble had expanded to about 3 ml and approached the optic disk but there still was no discernible flattening of the inferior detachment. We injected an additional 0.3 ml of perfluoro-n-butane (Fig. 3, top right), and, within hours after the expanding bubble crossed the optic disk, the inferior retina became reattached (Fig. 3, bottom left). We concluded that there was a break in the posterior staphyloma at the level of the optic disk, possibly a macular hole. The macula was cystic. A vitrectomy was done, and the operating micro scope showed the macula to be intact, but a break was discovered at the nasal edge of the optic disk. The break was sealed with intraocular diathermy. The retina has remained attached. The patient's visual acuity after six months was 20/100. Case 3—A 68-year-old patient with 18 diopters of myopia had had a total detachment for three days. A temporary buckle beneath a suspected area at the 1 o'clock position had no effect (Fig. 4). We thought that the break must be elsewhere in the superior third of the retina or in the posterior staphyloma. The eye was first injected with 0.5 ml of perfluoropropane and then injected three times at four-day intervals with 0.3 ml of perfluoro-n-butane. Four days after the fourth injection (day 16) the gas bubble crossed the superior edge of the optic disk, but the inferior retina remained detached. The estimated volume of the expanded bubble was then 6 ml. Ultrasonography disclosed that the anteroposterior length of the eye was 37 mm and it was apparent that more than 0.3 ml would be necessary for the bubble to tamponade the extent of the staphyloma. At the suggestion of one of the authors (D.J.C.), 0.75 ml of dilute gas was withdrawn from the eye and exchanged for an equal volume of perfluoro-n-butane (Fig. 5, top left and top right). The meniscus of the resulting bubble crossed the inferior edge of the staphyloma by the next day, and the retina became reattached overnight (Fig. 5, bottom left). The bub ble continued to expand for three more days and measured about 9 ml on day 20 (Fig. 5, bottom
OCTOBER, 1983
right). Slit-lamp biomicroscopy confirmed a break at the inferior edge of the staphyloma, and the break was treated with a laser. Thirteen days later, as the meniscus retreated across the inferior edge of the staphyloma, the break reopened and the inferior retina became redetached. A vitrectomy and intraoc ular diathermy around the break finally sealed it. The inferior retina (but not the retina in the staphyloma) later became redetached because of an iatrogenic break in the inferior periphery. A scierai buckle reattached the retina briefly, but it became rede tached again because of preretinal proliferation in the inferior periphery that could not be relieved. Cases 4 and 5—These two patients, who were treated in Tubingen, both had myopia, one 3 diopters and the other 5 diopters. They had total detach ments, apparent macular holes, and suspicious areas in the superior periphery. The initial procedure for both patients was a radial buckle beneath the area of suspicion. In one patient the upper third of the retina became reattached. In the other patient the buckle operation had no effect. Both patients then received injections of 0.7 ml of perfluoroethane after ocular massage. Each bubble expanded to about 2.3 ml and approached the upper border of the optic disk. After five days the inferior detachments were unchanged in both patients. To enlarge the bubble, we withdrew 0.5 ml of dilute gas and substituted 0.5 ml of perfluo ropropane. Within hours after the menisci of the expanding bubbles crossed the level of the maculas, the inferior retinas became reattached. In one pa tient the break, which was slightly nasal to the fovea, was treated through the gas bubble with a krypton laser. In the other patient the hole was directly on the fovea and instead of receiving laser treatment it was tamponaded for 16 days (by bending the patient's
Fig. 4 (Lincoff and associates). Case 3. A total detachment with suspected breaks at the 1 o'clock position in the periphery and at the inferior edge of a posterior staphyloma.
VOL. 96, NO. 4
LOCATING RETINAL BREAKS
515
Fig. 5 (Lincotf and associates). Case 3. By 16 days after several small injections of gas the bubble had expanded to the level of the optic disk. Its estimated volume was 6 ml. The inferior retina remained detached. We withdrew 0.75 ml of the bubble (top left, arrows) and substituted 0.75 ml of perfluoro-n-butane (top right). Bottom left, One day later (day 17) the bubble had reached the inferior edge of the staphyloma and covered the suspected retinal break. The inferior retina had become reattached. Bottom right, By day 20 the bubble had expanded to an estimated volume of 9 ml. head forward when the bubble receded beyond the macula). The retinas of both patients have remained attached. The visual acuity of the laser-treated pa tient after three months was 20/300; that of the other patient after three months was 20/100. DISCUSSION
The idea of using a gas bubble as a diagnostic tool developed as a result of daily observations of patients participat
ing in a clinical trial of the perfluorocarbon gases. 4 Most of these patients have had fluid-gas exchanges during vitrectomy operations for retinal detachment. Postoperatively the patients were exam ined twice daily while sitting to deter mine the level of the meniscus of the bubble and the state of the retina. If redetachment occurred, it appeared first
516
AMERICAN JOURNAL OF OPHTHALMOLOGY
in the inferior retina when the level of the meniscus rose above the level of an inad equately secured retinal break. The loca tions of the breaks, some of which were iatrogenic, were for the most part known. By observing the inferior retina as the meniscus of the bubble approached and then crossed the breaks, we were able to localize this cause of failure. It seemed likely that an absorbing bub ble of intraocular gas could also be used to indicate the level of an undetected retinal break in a detachment. Our Case 1 demonstrated the feasibility of the meth od. A diagnostic technique that requires previous drainage of subretinal fluid, however, may be regarded as relatively hazardous. The expanding bubble tech nique without drainage, used in our sec ond and third patients, is less hazardous but can be arduous and time-consuming if many injections are required. The ex change of dilute gas for pure gas, as was done in our Cases 3, 4, and 5, made it possible to obtain a diagnostic bubble of any size and, if necessary, to displace the entire vitreous volume with two injec tions. Although the expanding bubble could, in theory, displace superior fluid and make the inferior detachment more bullous, we have not observed this to any significant extent, possibly because ex pansion is slow and subretinal fluid re turns through the break to the vitreous compartment until the break is tamponaded. Another possible problem, one which we have not yet encountered, is that delayed absorption of subretinal fluid could make it difficult to determine the level of the break. Our 20 years of experi ence with nondrainage operations,6 how ever, indicated that delayed absorption of subretinal fluid is infrequent, that old fluid is absorbed as rapidly as recent fluid
OCTOBER, 1983
(Patient 1 had a detachment for six weeks, Patient 2 for three months, and Patient 4 for more than one year), and that delayed absorption affects only a fraction of the subretinal fluid (50% or more is absorbed rapidly). Thus, it should be possible to discern a distinct flattening of the inferior detachment when the meniscus crosses the break. If there is still doubt, there is a second opportunity to confirm the level of the break when the bubble begins to be absorbed; when the meniscus recedes across the break, the residual detachment should increase abruptly. Multiple breaks should not preclude using this method. With the expanding bubble technique, the first break to be closed would be the most superior one. As a consequence the upper borders of the detachment should drop to a level corresponding to the location of the sec ond break. If ophthalmoscopy and biomicroscopy still do not detect the break, a second diagnostic bubble might be indicated. In principle this technique is similar to sequential buckling of a retinal detachment from the top down ward.6 REFERENCES 1. Lincoff, H., Mardirossian, J., Lincoff, A., Ligget, P., Iwamoto, T., and Jakobiec, F.: Intravitreal longevity of three perfluorocarbon gases. Arch. Ophthalmol. 98:1610, 1980. 2. Lincoff, A., Haft, D., Ligget, P., and Reifer, C : Intravitreal expansion of perfluorocarbon bub bles. Arch. Ophthalmol. 98:1646, 1980. 3. Lincoff, H., and Gieser, R. : Finding the retinal hole. Arch. Ophthalmol. 85:565, 1971. 4. Lincoff, H., Kreissig, I., Richard, G., Coleman, J., Chang, S., and Wilcox, L. M.: The perfluo rocarbon gases in the treatment of retinal detach ment. Ophthalmology 90:546, 1983. 5. Lincoff, H., Kreissig, I.: The treatment of retinal detachment without drainage of subretinal fluid (modifications of the Custodis procedure. Part VI). Trans. Am. Acad. Ophthalmol. Otolaryngol. 76:1221, 1972. 6. Lincoff, H., Kreissig, I., and Goldbaum, M.: Reasons for failure in non-drainage operations. Mod. Probl. Ophthalmol. 12:40, 1974.