- Email: [email protected]
Injection of Perfluoropropane Gas to Prevent Hypotony in Eyes Undergoing Tube Implant Surgery
Injection of Perfluoropropane Gas to Prevent Hypotony in Eyes Undergoing Tube Implant Surgery WENDY A. FRANKS, FRCS, ROGER A. HITCHINGS, FRCS
Abstract: Hypotony after fistulizing surgery is common, and most cases resolve without complications. Occasionally, a flat anterior chamber in phakic or pseudophakic eyes may lead to corneal decompensation or cataract formation. In aphakic eyes that have undergone previous vitreous surgery, flat anterior chambers will not develop, but large choroidal detachments and delayed suprachoroidal hemorrhage may occur while the eye is soft. To maintain intraocular pressure (lOP) in the early postoperative period, the authors used a technique to seal the drainage tube with a Vicryl tie, combined with injection of sodium hyaluronate, sulfur hexafluoride (SF 6), or perfluoropropane (C3Fs), perioperatively in 28 eyes undergoing glaucoma tube implant surgery. Eight eyes were treated with injection of sodium hyaluronate, 8 with SF6 , and 12 eyes with C3FS. Hypotony was significantly less frequent in eyes treated with C3FS compared with sodium hyaluronate (P < 0.05). Mean lOP was significantly higher for eyes treated with C 3 Fs injection compared with sodium hyaluronate for the first 4 days after surgery (P < 0.05). Ophthalmology 1990; 97:899-903
Hypotony after tube implant surgery is common, and most cases resolve without complications. In phakic or pseudophakic eyes, a flat anterior chamber is the most frequent accompaniment of hypotony and carries the risk of corneal decompensation and cataract formation. In aphakic eyes that have undergone vitrectomy, flat anterior chambers will not develop, but large choroidal detachments and delayed suprachoroidal hemorrhage may occur. 1,2
Originally received: December 5, 1989. Revision accepted: February 20, 1990. From the Glaucoma Unit, Moorfields Eye Hospital, London. Presented as a poster at the American Academy of Ophthalmology Annual Meeting, New Orleans, Oct/Nov 1989. Supported by "The Friends of Moorfields Eye Hospital." Correspondences to Wendy A. Franks, FRCS, The Glaucoma Unit, Moorfields Eye Hospital, City Rd, London ECIV 2PD, England.
In a survey of 437 eyes undergoing fistulizing surgery (254 trabeculectomies and 178 tube implants) at this center, 12 (2%) had delayed suprachoroidal hemorrhage. Aphakia and vitrectomy were identified as the only significant risk factors. All cases of delayed suprachoroidal hemorrhage occurred with hypotony, and most cases occurred within the first week of surgery.3 To date, methods to reduce the incidence of hypotony after tube implant surgery include valved tubes, a twostage insertion procedure, and the use of sutures to occlude the lumen of the tube in the first weeks after surgery.4,5 In our experience, these methods are unsatisfactory in preventing postoperative hypotony in those eyes at greatest risk of delayed suprachoroidal hemorrhage. We report an alternative approach, applicable to aphakic and vitrectomized eyes, using injection of perfluoropropane (C 3Fg) gas to reform the globe at the end of surgery. Our results with this method show that C3Fg can prevent postoperative hypotony and therefore should reduce the risk of delayed suprachoroidal hemorrhage.
899
OPHTHALMOLOGY
m
•
JULY 1990
•
VOLUME 97
•
NUMBER 7
* *
45
J:
*
* ** $
E 35 E
0..
*
Q 25
*
15
* ;Ii
5
* 0
'"
* *
** ~* * ** ;* '" 2 3 ;Ii
:j<'
1
*
*
;Ii
** *
'" 4
* * * * ** * * 7
**
**
*
*
** **
14
28
**
Fig 1. Group 1 (sodium hyaluronate) lOP changes after tube implant surgery.
~
TIME IN DAYS METHODS Eyes with intractable glaucoma that had undergone cataract surgery and anterior or core vitrectomy and required tube implant surgery were selected for treatment. Aphakic and vitrectomized eyes were treated sequentially with reformation of the globe with sodium hyaluronate (group 1), with 50% sulfur hexafluoride (SF6 ) (group 2), or with 20% C 3Fg (group 3), in an attempt to leave the eye firm at the end of surgery and to maintain intraocular pressure (lOP) in the early postoperative period. The etiology of glaucoma was similar in the three groups and included underlying trauma in six cases, previous complicated retinal detachment surgery with vitrectomy and silicone oil injection in six, complicated congenital cataract surgery in four, and glaucoma secondary to inflammatory eye disease in five. Other causes of glaucoma were neovascularization, iridocorneal endothelial syndrome, buphthalmos, phacolytic glaucoma associated with Marfan's syndrome, and aniridia. All eyes had undergone previous surgery, and the etiology of the glaucoma in most cases was complex; the exact determinants resulting in uncontrollable lOP in many of these eyes were multifactorial. The average preoperative lOP was 35 mmHg for group 1,37 mmHg for group 2, and 35 mmHg for group 3 (range, 22-50 mmHg). The average age of patients in group 1 was 44 years; in group 2,36 years; and in group 3, 36 years (range, 11-66 years). In group 1, four eyes previously had a core vitrectomy, two had anterior vitrectomy, and two underwent anterior vitrectomy at the time of tube implant surgery. In group 2, three eyes had previous core vitrectomy, one eye underwent core vitrectomy at the time of tube implant surgery, and four had anterior vitrectomy. In group 3, 900
seven eyes had undergone core vitrectomy, one had anterior vitrectomy before tube implant surgery, and four underwent anterior vitrectomy at the time of tube implant surgery. Single-plate Molteno tubes were used in 5 eyes, Joseph tubes in 10 eyes, and double-plate Molteno tubes in 13 eyes sequentially, reflecting the availability of tube implants at our center. In all cases, a Vicryl tie (Ethicon Ltd, Edinburgh, Scotland) occluding suture was placed around the tube, to prevent drainage through it, in the first 2 to 3 weeks after surgery. A general anesthetic was used, and a limbal conjunctival peritomy was done. The plate or gutter was sutured to the sclera behind the equator using 8-0 nylon. A 6-0 Vicryl tie was secured with a triple throw to occlude the lumen of the drainage tube in all cases. Where anterior vitreous threatened to block the drainage tube, an anterior vitrectomy was done with an ocutome through a corneal incision and an infusion line passing through a separate corneal paracentesis. A thick limbal-based scleral flap was fashioned, a stab incision made into the anterior chamber at the base of the flap, and the tube trimmed and passed into the anterior chamber. The scleral flap was sutured with 8-0 nylon. Ocular volume was restored with either sodium hyaluronate or gas injection via a' corneal paracentesis. Nitrous oxide anesthesia was discontinued at least 10 minutes before gas injection to prevent expansion of the bubble from diffusion of nitrous oxide from the blood stream. The conjunctiva was closed with 10-0 nylon, and subconjunctival injections of betamethasone (2 mg) and cefuroxime (125 mg) were given. After surgery, the patients were postured to avoid gas shallowing the anterior chamber and blocking the pupil in the supine position. The lOP was measured at the end of surgery and 4 hours postoperatively, using a Perkins applanation to-
FRANKS AND HITCHINGS
45
Fig 2. Group 2 (SF6) lOP changes after tube implant surgery.
* $
E E 25
*
0. 0
-
GAS INJECTION AFTER GLAUCOMA TUBE IMPLANTS
*
35
0) ~
•
*
5 0
* ** * *"* * * ** * ** \* * * 1
~
~
t
*
15
*
* *
* *
* ** * * * * * * ** ** ** ** ** 2
*
$ **
*
3 4 7 TIME IN DAYS
*
* *
**
$
*
t
14
* 28
nometer (Keeler UK Ltd, Windsor, Berkshire, England). Thereafter, lOP was measured by Goldmann applanation tonometry. Hypotony was taken to be an lOP of5 mmHg or less. Eyes were reinjected with gas ifIOP remained less than 5 mmHg on 2 consecutive days and iflarge choroidal detachments were present. The change in volume ofC 3Fg with time was measured by A-scan ultrasonography.7 Data were analyzed statistically by the Wilcoxon rank-sum method and Fisher exact test.
tachment had a top-up injection of gas on postoperative day 2, raising lOP from 0 to 12 mmHg. No other intervention was required, and the choroidal effusion resolved in 24 hours. A delayed suprachoroidal hemorrhage occurred in one eye on postoperative day 3 after a bout of coughing. The patient had refused top-up injection of intraocular gas. The eye had been hypotonous since surgery, and the Vicryl tie was seen to be broken on slit-lamp examination.
RESULTS
MEAN PRESSURE CURVES
There were no significant differences in the age, type of glaucoma, or preoperative lOP of patients selected for treatment in the three groups. In group 1 (the sodium hyaluronate group consisting of 8 eyes), six eyes became hypotonous within 48 hours of surgery (Fig 1). Five eyes recovered from hypotony within 14 days of surgery. One patient, treated with a Joseph tube, had chronic hypotony with permanent reduction in visual acuity. No further surgical intervention was undertaken. In group 2 (the SF6 group, 8 eyes), five eyes became hypotonous within 48 hours of surgery (Fig 2). One eye with an lOP ofless than 5 mmHg with large choroidal detachments was treated by reinjection of gas on postoperative day 3, increasing lOP from 2 to 10 mmHg. The detachment resolved in 48 hours. A second eye was hypotonous for more than 4 days, but it was decided not to reinject with gas since the choroidal detachments were not large and the patient was a child, requiring a further general anesthetic for gas injection. In group 3 (the C 3Fg group, 12 eyes), three eyes became hypotonous within 2 days of surgery (Fig 3). One eye had an lOP of 5 mmHg on the first and second postoperative days, but this increased to 10 mmHg by day 3 with no intervention. One eye with a large choroidal de-
Mean lOP was significantly higher in the group treated with C3Fg compared with sodium hyaluronate injection from postoperative days 1 to 4 (P < 0.05, Fig 4). Mean lOP was not significantly different between the SF6 and C 3Fg groups for the first 48 hours, but for the following 2 days, mean lOP was significantly higher in the C 3Fggroup (P < 0.05, Fig 4). HYPOTONY
Hypotony was significantly less likely to occur in eyes injected with C3Fg in the first 2 weeks after surgery compared with eyes injected with sodium hyaluronate (P < 0.05). The number of eyes with hypotony was greater in the SF6 group compared with C 3Fg, but the difference was not statistically significant. There was no significant difference between the number of eyes in which hypotony occurred in the groups treated with SF6 and sodium hyaluronate. Measurement of the volume ofC 3Fggas bubbles in five patients showed that this remained constant in the first days after surgery and reduced slowly after the first week. No expansion of the bubble was recorded after postoperative day 1, and no eyes required aspiration of gas. The 901
OPHTHALMOLOGY
45
-
* *
E E 25 a.. 0 15
i
:x:
JULY 1990
•
VOLUME 97
•
NUMBER 7
*
35
t»
•
*
*
**
*
~
*
i#
** * *
*
* *
5
m
0
** ?j2
*
* * ~
:I:
* $ * ** * * * *
1
2
*
** ** *
* * **
* *
* ** $!
**
* *
4
5
7
?j2
*
*
** *
3
*
*
*
Fig 3. Group 3 (C3FS) lOP changes after tube implant surgery.
** * * ** $
* * ** i# *
"*
14
28
TIME IN DAYS presence of the gas bubble over the visual axis delayed visual recovery for up to 6 weeks. High postoperative lOPs were common, and antiglaucoma medication was continued in most cases until lysis of the Vicryl tie.
DISCUSSION Hypotony after fistulizing surgery is usually due to leakage through the fistula although other factors may contribute. IfIOP is low in the postoperative period, cho-
40
roidal detachments form and provide a posterior route for aqueous outflow, perpetuating hypotony. Postoperative inflammation may result in a decrease in aqueous production, compounding hypotony.? The use of a small stab incision and the placement of a Vicryl tie are used in tube implant surgery to reduce leakage through the fistula in the early postoperative period. In phakic eyes this is generally sufficient to prevent hypotony.s In aphakic eyes, particularly those that have undergone vitrectomy, choroidal detachments form rapidly, and there is an increased risk of delayed suprachoroidal hemorrhage.
Mean lOP higher for
Mean lOP higher
C3F8 compared to
for C3F8 compared
Na hyaluronate
to SF6 days 2-3
days '-4 (P'O.05)
(P·O.05)
-8SF6
- -i<- - C3FB
····0····
m 30 :x: E E
-
a.. 20 o
Fig 4. Mean lOP change after tube implant surgery.
10
o 1234 902
7 14 TIME IN DAYS
28
FRANKS AND HITCHINGS
•
GAS INJECTION AFTER GLAUCOMA TUBE IMPLANTS
In this series, sodium hyaluronate injection, used to reform the globe and leave the eye firm at the end of surgery, did not prevent the subsequent development of hypotony in seven of eight eyes. Sodium hyaluronate mixes with the aqueous compartment and is cleared from the eye by the same mechanisms of aqueous outflow. Gases leave the eye by diffusion, and the more insoluble the gas, the slower its disappearance. The volume of gas inside the eye is subject to a number of variables: the volume of gas injected, carbon dioxide concentration of the blood under anesthesia, lOP, and temperature. Because the outflow facility of eyes undergoing tube implant surgery is very low, we are reluctant to use concentrations of gas which could expand significantly, causing postoperative pressure spikes. A highly insoluble gas, C 3Fg is isovolumetric at a concentration of 16%,9 so the mixture we used would be expected to have some minor expansile properties. Although SF6 is isovolumetric at 20%,9 the 50% mixture used did not produce pressure spikes, but the bubble reduced in size rapidly and lOP was not maintained over several days. Some expansion is probably desirable, to allow for aqueous production not keeping pace with any leakage of aqueous around the tube, in the first days after surgery. Ifhypotony does occur, injection can be readily repeated. Repeated fluid-gas exchange for hypotony after vitreoretinal surgery for proliferative vitreoretinopathy has been reported with some success. IO Pupil block and anterior chamber shallowing were avoided by ensuring that the patients did not lie supine while the gas bubble was present. Apart from some cases of high postoperative lOP and delayed visual recovery, there were few complications associated with gas injection. We elected to treat those with a postoperative pressure spike medically; however, it would be possible to remove gas via the corneal paracentesis track if lOP could not be reduced to an acceptable level by any other means. There is a potential risk of a gas bubble causing retinal breaks through traction on the vitreous base. II One eye in our series had a retinal detachment; however, this eye also had secondary glaucoma after vitrectomy and silicone oil injection, and the gas could not have precipitated detachment in this case. The C 3Fg is potentially toxic to the corneal2; however, no cases of corneal decompensation were seen in this series. In aphakic vitrectomized eyes gas injected into the anterior chamber passes through into the posterior segment and is unlikely to come into prolonged contact with the endothelium. Even if C 3F g does come into contact with the cornea, in the concentration used in this study, toxicity may not be apparent clinically. 13 Different types of drainage tubes were used in this series, but since all the tubes were occluded with a Vicryl tie, this should make no difference to lOP measurements in
the first days after surgery or to subsequent bleb development after lysis of the Vicryl tie.
CONCLUSION Injection of intraocular C 3Fg gas in aphakic, vitrectomized eyes is effective in restoring lOP perioperatively in eyes undergoing tube implant surgery by the method described. In most cases 20% C 3Fg maintains lOP at 5 mmHg or greater for the first week after surgery. If the eye does become hypotonous, lOP can be raised by topup injection of gas. Although SF6 is more soluble than C3Fg and is quickly absorbed, it was less effective in maintaining lOP. In aphakic, vitrectomized eyes undergoing tube implant surgery, C 3Fg injection can prevent postoperative hypotony and therefore should reduce the risk of delayed suprachoroidal hemorrhage.
REFERENCES 1. Givens K, Shields MB. Suprachoroidal hemorrhage after glaucoma filtering surgery. Am J Ophthalmol1987; 103:689-94. 2. Frenkel REP, Shin DH. Prevention and management of delayed suprachoroidal hemorrhage after filtration surgery. Arch Ophthalmol 1986; 104:1459-63. 3. Canning CR, Lavin M, McCartney ACE, et al. Delayed suprachoroidal hemorrhage after glaucoma operations. Eye 1989; 3:327-31. 4. Molteno ACB, Polkinghome PJ, Bowbyes JA. The Vicryl tie technique for inserting a draining implant in the treatment of secondary glaucoma. Aust N Z J Ophthalmol 1986; 14:343-54. 5. Hoare Naime JEA, Sherwood D, Jacob JSH, Rich WJCC. Single stage insertion of the Molteno tube for glaucoma and modifications to reduce postoperative hypotony. Br J Ophthalmol 1988: 72;846-51. 6. Jacobs PM. Intraocular gas measurement using A-scan ultrasound. Curr Eye Research 1986: 5;575-8. 7. Pederson JE, Gaasterland DE, Maclellan HM. Experimental ciliochoroidal detachment: effect on intraocular pressure and aqueous humor flow. Arch Ophthalmol1979: 97;536-41. 8. Rose GE, Lavin MJ, Hitchings RA. Silicone tubes in glaucoma surgery: the effect of technical modifications on early postoperative intraocular pressures and complications. Eye 1989: 3;553-61. 9. Killey FP, Edelhauser HF, Aaberg TM. Intraocular sulfur hexafluoride and octofluorocyclobutane: effects on intraocular pressure and vitreous volume. Arch Ophthalmol 1978: 96;511-15. 10. Stallman JB, Meyers SM. Repeated fluid gas exchange for hypotony after vitreoretinal surgery for proliferative vitreoretinopathy. Am J Ophthalmol1988; 106:147-53. 11. Johnston PB, Maguire CJF, logan WC. Failed retinal surgery caused by intraocular gas injection. Br J Ophthalmol 1988: 72;322-25. 12. Foulks GN, deJuan E, Hatchell Dl, et al. The effect of perfluoropropane on the comea in rabbits and cats. Arch Ophthalmol 1985; 105: 256-9. 13. Franks WA. Hitchings RA. Perfluoropropane. Br J Ophthalmol 1989; 73:50.
903
Abstract: Hypotony after fistulizing surgery is common, and most cases resolve without complications. Occasionally, a flat anterior chamber in phakic or pseudophakic eyes may lead to corneal decompensation or cataract formation. In aphakic eyes that have undergone previous vitreous surgery, flat anterior chambers will not develop, but large choroidal detachments and delayed suprachoroidal hemorrhage may occur while the eye is soft. To maintain intraocular pressure (lOP) in the early postoperative period, the authors used a technique to seal the drainage tube with a Vicryl tie, combined with injection of sodium hyaluronate, sulfur hexafluoride (SF 6), or perfluoropropane (C3Fs), perioperatively in 28 eyes undergoing glaucoma tube implant surgery. Eight eyes were treated with injection of sodium hyaluronate, 8 with SF6 , and 12 eyes with C3FS. Hypotony was significantly less frequent in eyes treated with C3FS compared with sodium hyaluronate (P < 0.05). Mean lOP was significantly higher for eyes treated with C 3 Fs injection compared with sodium hyaluronate for the first 4 days after surgery (P < 0.05). Ophthalmology 1990; 97:899-903
Hypotony after tube implant surgery is common, and most cases resolve without complications. In phakic or pseudophakic eyes, a flat anterior chamber is the most frequent accompaniment of hypotony and carries the risk of corneal decompensation and cataract formation. In aphakic eyes that have undergone vitrectomy, flat anterior chambers will not develop, but large choroidal detachments and delayed suprachoroidal hemorrhage may occur. 1,2
Originally received: December 5, 1989. Revision accepted: February 20, 1990. From the Glaucoma Unit, Moorfields Eye Hospital, London. Presented as a poster at the American Academy of Ophthalmology Annual Meeting, New Orleans, Oct/Nov 1989. Supported by "The Friends of Moorfields Eye Hospital." Correspondences to Wendy A. Franks, FRCS, The Glaucoma Unit, Moorfields Eye Hospital, City Rd, London ECIV 2PD, England.
In a survey of 437 eyes undergoing fistulizing surgery (254 trabeculectomies and 178 tube implants) at this center, 12 (2%) had delayed suprachoroidal hemorrhage. Aphakia and vitrectomy were identified as the only significant risk factors. All cases of delayed suprachoroidal hemorrhage occurred with hypotony, and most cases occurred within the first week of surgery.3 To date, methods to reduce the incidence of hypotony after tube implant surgery include valved tubes, a twostage insertion procedure, and the use of sutures to occlude the lumen of the tube in the first weeks after surgery.4,5 In our experience, these methods are unsatisfactory in preventing postoperative hypotony in those eyes at greatest risk of delayed suprachoroidal hemorrhage. We report an alternative approach, applicable to aphakic and vitrectomized eyes, using injection of perfluoropropane (C 3Fg) gas to reform the globe at the end of surgery. Our results with this method show that C3Fg can prevent postoperative hypotony and therefore should reduce the risk of delayed suprachoroidal hemorrhage.
899
OPHTHALMOLOGY
m
•
JULY 1990
•
VOLUME 97
•
NUMBER 7
* *
45
J:
*
* ** $
E 35 E
0..
*
Q 25
*
15
* ;Ii
5
* 0
'"
* *
** ~* * ** ;* '" 2 3 ;Ii
:j<'
1
*
*
;Ii
** *
'" 4
* * * * ** * * 7
**
**
*
*
** **
14
28
**
Fig 1. Group 1 (sodium hyaluronate) lOP changes after tube implant surgery.
~
TIME IN DAYS METHODS Eyes with intractable glaucoma that had undergone cataract surgery and anterior or core vitrectomy and required tube implant surgery were selected for treatment. Aphakic and vitrectomized eyes were treated sequentially with reformation of the globe with sodium hyaluronate (group 1), with 50% sulfur hexafluoride (SF6 ) (group 2), or with 20% C 3Fg (group 3), in an attempt to leave the eye firm at the end of surgery and to maintain intraocular pressure (lOP) in the early postoperative period. The etiology of glaucoma was similar in the three groups and included underlying trauma in six cases, previous complicated retinal detachment surgery with vitrectomy and silicone oil injection in six, complicated congenital cataract surgery in four, and glaucoma secondary to inflammatory eye disease in five. Other causes of glaucoma were neovascularization, iridocorneal endothelial syndrome, buphthalmos, phacolytic glaucoma associated with Marfan's syndrome, and aniridia. All eyes had undergone previous surgery, and the etiology of the glaucoma in most cases was complex; the exact determinants resulting in uncontrollable lOP in many of these eyes were multifactorial. The average preoperative lOP was 35 mmHg for group 1,37 mmHg for group 2, and 35 mmHg for group 3 (range, 22-50 mmHg). The average age of patients in group 1 was 44 years; in group 2,36 years; and in group 3, 36 years (range, 11-66 years). In group 1, four eyes previously had a core vitrectomy, two had anterior vitrectomy, and two underwent anterior vitrectomy at the time of tube implant surgery. In group 2, three eyes had previous core vitrectomy, one eye underwent core vitrectomy at the time of tube implant surgery, and four had anterior vitrectomy. In group 3, 900
seven eyes had undergone core vitrectomy, one had anterior vitrectomy before tube implant surgery, and four underwent anterior vitrectomy at the time of tube implant surgery. Single-plate Molteno tubes were used in 5 eyes, Joseph tubes in 10 eyes, and double-plate Molteno tubes in 13 eyes sequentially, reflecting the availability of tube implants at our center. In all cases, a Vicryl tie (Ethicon Ltd, Edinburgh, Scotland) occluding suture was placed around the tube, to prevent drainage through it, in the first 2 to 3 weeks after surgery. A general anesthetic was used, and a limbal conjunctival peritomy was done. The plate or gutter was sutured to the sclera behind the equator using 8-0 nylon. A 6-0 Vicryl tie was secured with a triple throw to occlude the lumen of the drainage tube in all cases. Where anterior vitreous threatened to block the drainage tube, an anterior vitrectomy was done with an ocutome through a corneal incision and an infusion line passing through a separate corneal paracentesis. A thick limbal-based scleral flap was fashioned, a stab incision made into the anterior chamber at the base of the flap, and the tube trimmed and passed into the anterior chamber. The scleral flap was sutured with 8-0 nylon. Ocular volume was restored with either sodium hyaluronate or gas injection via a' corneal paracentesis. Nitrous oxide anesthesia was discontinued at least 10 minutes before gas injection to prevent expansion of the bubble from diffusion of nitrous oxide from the blood stream. The conjunctiva was closed with 10-0 nylon, and subconjunctival injections of betamethasone (2 mg) and cefuroxime (125 mg) were given. After surgery, the patients were postured to avoid gas shallowing the anterior chamber and blocking the pupil in the supine position. The lOP was measured at the end of surgery and 4 hours postoperatively, using a Perkins applanation to-
FRANKS AND HITCHINGS
45
Fig 2. Group 2 (SF6) lOP changes after tube implant surgery.
* $
E E 25
*
0. 0
-
GAS INJECTION AFTER GLAUCOMA TUBE IMPLANTS
*
35
0) ~
•
*
5 0
* ** * *"* * * ** * ** \* * * 1
~
~
t
*
15
*
* *
* *
* ** * * * * * * ** ** ** ** ** 2
*
$ **
*
3 4 7 TIME IN DAYS
*
* *
**
$
*
t
14
* 28
nometer (Keeler UK Ltd, Windsor, Berkshire, England). Thereafter, lOP was measured by Goldmann applanation tonometry. Hypotony was taken to be an lOP of5 mmHg or less. Eyes were reinjected with gas ifIOP remained less than 5 mmHg on 2 consecutive days and iflarge choroidal detachments were present. The change in volume ofC 3Fg with time was measured by A-scan ultrasonography.7 Data were analyzed statistically by the Wilcoxon rank-sum method and Fisher exact test.
tachment had a top-up injection of gas on postoperative day 2, raising lOP from 0 to 12 mmHg. No other intervention was required, and the choroidal effusion resolved in 24 hours. A delayed suprachoroidal hemorrhage occurred in one eye on postoperative day 3 after a bout of coughing. The patient had refused top-up injection of intraocular gas. The eye had been hypotonous since surgery, and the Vicryl tie was seen to be broken on slit-lamp examination.
RESULTS
MEAN PRESSURE CURVES
There were no significant differences in the age, type of glaucoma, or preoperative lOP of patients selected for treatment in the three groups. In group 1 (the sodium hyaluronate group consisting of 8 eyes), six eyes became hypotonous within 48 hours of surgery (Fig 1). Five eyes recovered from hypotony within 14 days of surgery. One patient, treated with a Joseph tube, had chronic hypotony with permanent reduction in visual acuity. No further surgical intervention was undertaken. In group 2 (the SF6 group, 8 eyes), five eyes became hypotonous within 48 hours of surgery (Fig 2). One eye with an lOP ofless than 5 mmHg with large choroidal detachments was treated by reinjection of gas on postoperative day 3, increasing lOP from 2 to 10 mmHg. The detachment resolved in 48 hours. A second eye was hypotonous for more than 4 days, but it was decided not to reinject with gas since the choroidal detachments were not large and the patient was a child, requiring a further general anesthetic for gas injection. In group 3 (the C 3Fg group, 12 eyes), three eyes became hypotonous within 2 days of surgery (Fig 3). One eye had an lOP of 5 mmHg on the first and second postoperative days, but this increased to 10 mmHg by day 3 with no intervention. One eye with a large choroidal de-
Mean lOP was significantly higher in the group treated with C3Fg compared with sodium hyaluronate injection from postoperative days 1 to 4 (P < 0.05, Fig 4). Mean lOP was not significantly different between the SF6 and C 3Fg groups for the first 48 hours, but for the following 2 days, mean lOP was significantly higher in the C 3Fggroup (P < 0.05, Fig 4). HYPOTONY
Hypotony was significantly less likely to occur in eyes injected with C3Fg in the first 2 weeks after surgery compared with eyes injected with sodium hyaluronate (P < 0.05). The number of eyes with hypotony was greater in the SF6 group compared with C 3Fg, but the difference was not statistically significant. There was no significant difference between the number of eyes in which hypotony occurred in the groups treated with SF6 and sodium hyaluronate. Measurement of the volume ofC 3Fggas bubbles in five patients showed that this remained constant in the first days after surgery and reduced slowly after the first week. No expansion of the bubble was recorded after postoperative day 1, and no eyes required aspiration of gas. The 901
OPHTHALMOLOGY
45
-
* *
E E 25 a.. 0 15
i
:x:
JULY 1990
•
VOLUME 97
•
NUMBER 7
*
35
t»
•
*
*
**
*
~
*
i#
** * *
*
* *
5
m
0
** ?j2
*
* * ~
:I:
* $ * ** * * * *
1
2
*
** ** *
* * **
* *
* ** $!
**
* *
4
5
7
?j2
*
*
** *
3
*
*
*
Fig 3. Group 3 (C3FS) lOP changes after tube implant surgery.
** * * ** $
* * ** i# *
"*
14
28
TIME IN DAYS presence of the gas bubble over the visual axis delayed visual recovery for up to 6 weeks. High postoperative lOPs were common, and antiglaucoma medication was continued in most cases until lysis of the Vicryl tie.
DISCUSSION Hypotony after fistulizing surgery is usually due to leakage through the fistula although other factors may contribute. IfIOP is low in the postoperative period, cho-
40
roidal detachments form and provide a posterior route for aqueous outflow, perpetuating hypotony. Postoperative inflammation may result in a decrease in aqueous production, compounding hypotony.? The use of a small stab incision and the placement of a Vicryl tie are used in tube implant surgery to reduce leakage through the fistula in the early postoperative period. In phakic eyes this is generally sufficient to prevent hypotony.s In aphakic eyes, particularly those that have undergone vitrectomy, choroidal detachments form rapidly, and there is an increased risk of delayed suprachoroidal hemorrhage.
Mean lOP higher for
Mean lOP higher
C3F8 compared to
for C3F8 compared
Na hyaluronate
to SF6 days 2-3
days '-4 (P'O.05)
(P·O.05)
-8SF6
- -i<- - C3FB
····0····
m 30 :x: E E
-
a.. 20 o
Fig 4. Mean lOP change after tube implant surgery.
10
o 1234 902
7 14 TIME IN DAYS
28
FRANKS AND HITCHINGS
•
GAS INJECTION AFTER GLAUCOMA TUBE IMPLANTS
In this series, sodium hyaluronate injection, used to reform the globe and leave the eye firm at the end of surgery, did not prevent the subsequent development of hypotony in seven of eight eyes. Sodium hyaluronate mixes with the aqueous compartment and is cleared from the eye by the same mechanisms of aqueous outflow. Gases leave the eye by diffusion, and the more insoluble the gas, the slower its disappearance. The volume of gas inside the eye is subject to a number of variables: the volume of gas injected, carbon dioxide concentration of the blood under anesthesia, lOP, and temperature. Because the outflow facility of eyes undergoing tube implant surgery is very low, we are reluctant to use concentrations of gas which could expand significantly, causing postoperative pressure spikes. A highly insoluble gas, C 3Fg is isovolumetric at a concentration of 16%,9 so the mixture we used would be expected to have some minor expansile properties. Although SF6 is isovolumetric at 20%,9 the 50% mixture used did not produce pressure spikes, but the bubble reduced in size rapidly and lOP was not maintained over several days. Some expansion is probably desirable, to allow for aqueous production not keeping pace with any leakage of aqueous around the tube, in the first days after surgery. Ifhypotony does occur, injection can be readily repeated. Repeated fluid-gas exchange for hypotony after vitreoretinal surgery for proliferative vitreoretinopathy has been reported with some success. IO Pupil block and anterior chamber shallowing were avoided by ensuring that the patients did not lie supine while the gas bubble was present. Apart from some cases of high postoperative lOP and delayed visual recovery, there were few complications associated with gas injection. We elected to treat those with a postoperative pressure spike medically; however, it would be possible to remove gas via the corneal paracentesis track if lOP could not be reduced to an acceptable level by any other means. There is a potential risk of a gas bubble causing retinal breaks through traction on the vitreous base. II One eye in our series had a retinal detachment; however, this eye also had secondary glaucoma after vitrectomy and silicone oil injection, and the gas could not have precipitated detachment in this case. The C 3Fg is potentially toxic to the corneal2; however, no cases of corneal decompensation were seen in this series. In aphakic vitrectomized eyes gas injected into the anterior chamber passes through into the posterior segment and is unlikely to come into prolonged contact with the endothelium. Even if C 3F g does come into contact with the cornea, in the concentration used in this study, toxicity may not be apparent clinically. 13 Different types of drainage tubes were used in this series, but since all the tubes were occluded with a Vicryl tie, this should make no difference to lOP measurements in
the first days after surgery or to subsequent bleb development after lysis of the Vicryl tie.
CONCLUSION Injection of intraocular C 3Fg gas in aphakic, vitrectomized eyes is effective in restoring lOP perioperatively in eyes undergoing tube implant surgery by the method described. In most cases 20% C 3Fg maintains lOP at 5 mmHg or greater for the first week after surgery. If the eye does become hypotonous, lOP can be raised by topup injection of gas. Although SF6 is more soluble than C3Fg and is quickly absorbed, it was less effective in maintaining lOP. In aphakic, vitrectomized eyes undergoing tube implant surgery, C 3Fg injection can prevent postoperative hypotony and therefore should reduce the risk of delayed suprachoroidal hemorrhage.
REFERENCES 1. Givens K, Shields MB. Suprachoroidal hemorrhage after glaucoma filtering surgery. Am J Ophthalmol1987; 103:689-94. 2. Frenkel REP, Shin DH. Prevention and management of delayed suprachoroidal hemorrhage after filtration surgery. Arch Ophthalmol 1986; 104:1459-63. 3. Canning CR, Lavin M, McCartney ACE, et al. Delayed suprachoroidal hemorrhage after glaucoma operations. Eye 1989; 3:327-31. 4. Molteno ACB, Polkinghome PJ, Bowbyes JA. The Vicryl tie technique for inserting a draining implant in the treatment of secondary glaucoma. Aust N Z J Ophthalmol 1986; 14:343-54. 5. Hoare Naime JEA, Sherwood D, Jacob JSH, Rich WJCC. Single stage insertion of the Molteno tube for glaucoma and modifications to reduce postoperative hypotony. Br J Ophthalmol 1988: 72;846-51. 6. Jacobs PM. Intraocular gas measurement using A-scan ultrasound. Curr Eye Research 1986: 5;575-8. 7. Pederson JE, Gaasterland DE, Maclellan HM. Experimental ciliochoroidal detachment: effect on intraocular pressure and aqueous humor flow. Arch Ophthalmol1979: 97;536-41. 8. Rose GE, Lavin MJ, Hitchings RA. Silicone tubes in glaucoma surgery: the effect of technical modifications on early postoperative intraocular pressures and complications. Eye 1989: 3;553-61. 9. Killey FP, Edelhauser HF, Aaberg TM. Intraocular sulfur hexafluoride and octofluorocyclobutane: effects on intraocular pressure and vitreous volume. Arch Ophthalmol 1978: 96;511-15. 10. Stallman JB, Meyers SM. Repeated fluid gas exchange for hypotony after vitreoretinal surgery for proliferative vitreoretinopathy. Am J Ophthalmol1988; 106:147-53. 11. Johnston PB, Maguire CJF, logan WC. Failed retinal surgery caused by intraocular gas injection. Br J Ophthalmol 1988: 72;322-25. 12. Foulks GN, deJuan E, Hatchell Dl, et al. The effect of perfluoropropane on the comea in rabbits and cats. Arch Ophthalmol 1985; 105: 256-9. 13. Franks WA. Hitchings RA. Perfluoropropane. Br J Ophthalmol 1989; 73:50.
903