Pneumatic Retinopexy Failures

Pneumatic Retinopexy Failures Cause, Prevention, Timing, and Management W. Sanderson Grizzard, MD, l George F. Hilton, MD, 2 Mark E. Hammer, MD, l Douglas Taren, PhD, 3 Daniel A. Brinton, MD2 Background: Pneumatic retinopexy is a procedure for reattaching the retina by injecting an expanding gas bubble and using either laser or cryopexy. The procedure is controversial because there may be a lower initial success rate, and intraocular gas may increase the risk of proliferative vitreoretinopathy. Methods: The authors performed a retrospective review of 107 unpublished consecutive cases of pneumatic retinopexy together with a literature review of 25 statistical series with primary attention to failures. Univariate and multivariate analyses were carried out on the data set, and adjusted odds ratios for risk factors associated with failure were calculated using logistic regression. Results: Initially, 74 (69%) of 107 patients had successful results, and with reoperations the success rate increased to 98%. Failure of the procedure to achieve retinal reattachment occurred soon after the initial procedure, with 86% of recorded failures occurring within the first month. The initial cause of failure was new or missed breaks in 14.9%, reopened initial breaks in 11.2%, and breaks never closed in 4.6%. Risk factors that showed a correlation with failure were patients being male (adjusted odds ratio = 2.65), eyes with preoperative visual acuity worse than 20/50 (adjusted odds ratio = 1.21), eyes with four quadrants of retinal detachment or total detachment (adjusted odds ratio = 2.03), aphakic or pseudophakic eyeS (adjusted odds ratio = 1.91), and eyes with additional pathologic findings (adjusted odds ratio = 3.14). Poor visual outcome was associated with initial visual acuity less than 20/50 (adjusted odds ratio = 15.7) and eyes with four quadrants of retinal detachment or total detachment (adjusted odds ratio = 5.01). Conclusions: Failures of pneumatic retinopexy occur early in the postoperative course. Factors known to be associated with failure of retinal reattachment using scleral buckling also were associated with failure in pneumatic retinopexy. A higher success rate in females was noted, suggesting that educational efforts may need to be greater in males. Poorer visual results occurred in patients with poor initial vision and in eyes with four quadrants of retinal detachment or total detachments Ophthalmology 1995; 102:929-936

Originally received: August 3, 1993. Revision accepted: January 30, 1995. I Department of Ophthalmology, University of South Florida College of Medicine, Tampa. 2 Department of Ophthalmology, University of California, San Francisco. 3 Department of Family and Community Medicine Health Services Section, Arizona Health Sciences Center, Tucson. Presented at Club Jules Gonin, Vienna, Austria, September 1992. Reprint requests to W. Sanderson Grizzard, MD, South Tampa Medical Center, 508 South Habana, Suite 120, Tampa, FL 33609-4138.

Retinal reattachment by injecting an expanding gas bubble with some form of retinopexy was reported independently by Dominguez! and by Hilton and Grizzard. 2 The procedure, called pneumatic retinopexy, has created some controversy because success rates from the initial procedure may be lower, and intraocular gas bubbles are thought by some to cause inflammation and increase the risk for proliferative vitreoretinopathy (PVR). 3 Results of a consecutive series of 107 patients with retinal detachment treated with pneumatic retinopexy are

929

Ophthalmology

Volume 102, Number 6, June 1995

presented. The success rate, timing of failures, direct cause of failures, relative location of new/missed breaks, and ultimate anatomic and visual results of surgery are evaluated. Additional data from the randomized, prospective, controlled trial of pneumatic retinopexy concerning timing of failures, the effect of sex on outcome, and types of subsequent intervention also are presented. 4 ,5

Data Source From January 1989 until June 1992, 107 pneumatic retinopexy procedures were performed by two of us (WSG and MEH) in the office on patients with primary rhegmatogenous retinal detachment. The indications for most patients were those published in 1985 2 (i.e., superior breaks less than 1 clock hour, good view of the entire retina, and absence of PVR), We also included patients with breaks in more than 1 clock hour of the retina, grade C-I PVR,6 and vitreous hemorrhage that occluded a view of the inferior retina. We grouped these patients under the heading "additional pathology." Specific data were abstracted from the charts and entered into a computer database. Visual acuity data were taken from standard office procedures and analyzed both as a continuous variable and dichotomized for risk analysis. Patient characteristics used in this analysis included age in years, sex, and right or left eye. Other preoperative variables analyzed were vision, number of quadrants detached, meridian of the breaks, phakia, pseudophakia, aphakia, and the presence or absence of mild PVR. The surgical variables included the type of retinopexy and the type of gas used (sulfur hexafluoride or perfluoropropane). The duration of symptoms in days before surgery also was controlled during the multivariate analyses with 61 + days being the longest duration recorded. Eyes originally were classified as phakic, aphakic, pseudophakic with anterior chamber intraocular lens, pseudophakic with a posterior chamber intraocular lens with no capsulotomy, or pseudophakic posterior chamber intraocular lens with a capsulotomy, and then finally categorized as either phakic or other because no significant differences were noted in smaller subgroups. The location of the breaks was coded I if they were located between II and 1 o'clock and 0 if the breaks were located elsewhere.

Data Analysis The data were first analyzed to determine the characteristics of the patients, eyes, and surgical procedures, The unadjusted and adjusted odds ratios were calculated for each of the variables before and during surgery in relation to initial reattachment and final vision. Adjusted odds ratios were calculated using logistic regression. Both preoperative and postoperative vision were highly skewed. Univariate analyses using the Wilcoxon twosample test, a nonparametric procedure for abnormally distributed data, was conducted when visual acuity was

930

the outcome. Visual acuity then was categorized as a dichotomous variable with 20/50 or better being one group and visual acuity less than 20/50 being the second group. This allowed us to analyze the association between factors before, during, and after surgery with final visual acuity using logistic regression. During multivariate analysis, nine cases were not included in the calculations. In six of the cases, both cryotherapy and laser therapy were used and in three of the cases, laser therapy was planned but was not performed because the hole did not close. The age was unknown for two patients. In addition, three eyes were deleted from the analysis of final vision because two never reattached and the other one had a postoperative macular hole. These latter patients were excluded from analysis because they had a specific anatomic cause for poor vision and were statistical outliers. We also analyzed our data and previously unpublished data from the randomized prospective trial4 ,5 to evaluate the timing of failure, reason for failure, and the type of intervention which was used subsequent to initial failure. Because of the unexpected finding that sex was an important variable, data from the randomized trial were evaluated to determine if sex was a risk factor for failure in that group of patients.

Results Follow-up (mean ± standard error of the mean) of all patients was 19.0 ± 1.6 months. Follow-up ranged from 1 to 44 months. The success rate with initial pneumatic surgery was 74 (69.2%) of 107 eyes. Two eyes were reattached with an additional pneumatic surgery for an initial office success rate of 76 (71 %) of 107 eyes. With re-operation, we had an ultimate success rate of 105 (98.0%) of 107 eyes. In 107 eyes, the immediate cause of failure was new or missed breaks in 16 (14.9%), re-opened initial breaks in 12 (11.2%), and breaks never closed in 5 (4.6%). In the 16 eyes in which new breaks were noted, 7 (44%) were within 2 clock hours of the original breaks and in 6 (37.5%) the subsequent breaks were within 2 to 4 clock hours of the original breaks. In only two eyes (12.5%), the break was in the opposite part of the retina. In one patient, a postoperative macular hole developed. Failure of the initial pneumatic surgery was noted soon after the procedure. Of our 33 primary failures, 16 (48.5%) never reattached and 9 redetached within 1 month (25 of 33, 76%). Only two retinas redetached after 6 months (Fig I).

The second procedure was scleral buckling in 18 (55 %) of33 eyes; 6 required subsequent vitrectomy and all except I were successfully reattached. Vitrectomy with scleral buckling was done in II (33%) of 33 eyes, and 10 were reattached successfully. In one patient, cryopexy alone was tried but failed and a scleral buckling succeeded. In three patients, a repeat pneumatic retinopexy was attempted. In two patients, this was successful, and in one

Grizzard et al . Pneumatic Retinopexy Failures 30 24

..

18

'0 i .a

12

25 Eyes

tJ)

G> tJ)

U

E

16

:::I

Z

6

o

NR

1w

2w

3w

4w

2m

3m

4m

5m

2

2

6m

>6m

Time Before Failure

Figure 1. NR = never reattached; lw = 1 week; 2m = 2 months. Twentyfive (76%) failures occurred during the first month after surgery. Detachment after 6 months occurred in only two eyes.

patient a subsequent vitrectomy with scleral buckling was successful. The complication of subretinal gas occurred in two patients and was thought to be due to epiretinal or vitreoretinal traction on the hole. Both patients required subsequent vitrectomy, but their retinas were reattached successfully. In one patient, anteriorly trapped gas developed but the retina was attached successfully with a single pneumatic procedure. In five eyes, PVR requiring removal of the intraocular lens or use of silicone oil occurred. The timing of failures and the direct cause of failures was not published in the original report of the randomized trial of pneumatic retinopexy,4 but is presented now. The type of subsequent surgery leading to reattachment also was not published. In the pneumatic trial, there were 28 (27.1 %) initial failures in 103 eyes. Two (7%) of the 28 eyes initially never reattached, and 19 (68%) failed within 1 month (Fig 2). In 103 eyes, the immediate cause of failure in the pneumatic trial was new breaks in 20 (19.4%), re-opened initial breaks in 5 (4.8%), and breaks that never closed in I (0.9%). Two eyes (1.9%) failed because of PVR. Nine of the failed eyes could be repaired with laser or cryopexy (32%), and six additional eyes needed only another pneumatic procedure (21 %). Eleven eyes (37%) required scleral buckling, and only two eyes (8%) required vitrectomy. One of those eyes was a failure. In our series of 107 patients, results of univariate chisquare analysis and multivariate logistic regression analysis comparing preoperative and intraoperative factors with initial success are shown in Table I. Sex was the most surprising preoperative factor, with males having a failure rate of 43.6% and females a failure rate of 13.3% (P::;; 0.001, Yates correction). Significant correlation (P ::;; 0.10) with failure was found for patients who were male, eyes with preoperative visual acuity worse than 20/50, eyes with four quadrants of retinal detachment or total detachment, aphakic or pseudophakic eyes, and eyes with "additional pathology." The average age of patients and duration of symptoms was not different between successful surgery and initial failure.

Risk analysis is seen in Table I with P values derived from chi-square analysis; adjusted odds ratios and 95% confidence limits were derived from logistic regression analysis. In 73.9% of the eyes, final visual acuity was better than or equal to 20/30. In eyes with a greater proportion of retinas that were detached after initial pneumatic retinopexy, visual acuity was 20/400 or less (27.3%) compared with eyes with the retina attached (9.45%). This was significant with the Kriskal-Wallis test (P ::;; 0.05). Age was correlated negatively with visual acuity (r = -0.37; P ::;; 0.001). The association between preoperative and postoperative visual acuity was significant (r = 0.50; P ::;; 0.001). Chi-square analysis also was used to assess visual outcome for the same risk factors as above plus initial success (Table 2). Significant correlates (P ::;; 0.10) with poor visual outcome were initial poor vision, eyes with four quadrants of detachment or total detachments, and failure of the initial pneumatic procedure. The use of perfiuoropropane gas was marginally suggestive of poor visual outcome but was more likely to be used in eyes with multiple tears or in eyes in which vitreous had access to the anterior chamber. Multivariate analysis using logistic regression then was used to determine adjusted odds ratios and 95% confidence intervals. Only poor preoperative vision and eyes with four quadrants of detachment or total detachments significantly correlated with poor postoperative vision in the multivariate analysis. The patient's sex did not affect visual outcome. Analysis of our data demonstrated a higher success rate among females 39/45 (86.7%) than males 35/62 (56.5%) (P::;; 0.00 I). Data from the randomized trial showed similar results, with a higher success rate among females (34 of 39) (87%) than among males (41 of 64) (64%) (chisquare = 3.95; P::;; 0.05).

Discussion Between 1986 and 1989, there were 26 published series that provided information concerning a single-surgery 20

...,

19 Eyes

16

tJ)

G>

U

12

'0

... G>

.a

E

8

8

:::I

Z

4 0

NR

1w

2w

3w

4w

2m

3m

4m

5m

6m

>6m

Time Before Failure

Figure 2. NR = never reattached; 1 w = 1 week; 2 m = 2 months. Nineteen (63%) failures occurred during the first month after surgery. Detachment after 6 months occurred in only one eye.

931

Ophthalmology

Volume 102, Number 6, June 1995

Table 1. Percent of Eyes that Failed Initial Reattachment with One Pneumatic Procedure Variables Sex M F Preoperative visual acuity Worse than 20/50 20/50 or better Quads detached 4 and total 1,2,3 Location of tear 11-1 hrs Other Lens status Aphakic/ pseudophakic Phakic Additional pathology Yes No PVR Yes No Vitreous hemorrhage Yes No Retinopexy Cryotherapy Laser Gas SF 6 C 3 F8

AOR 95%CI

No. of Eyes

% Failed

62 45

43.6 13.3

0.001

2.65

1.99-3.32

77

36.4 16.7

0.Q48

1.21

0.38-2.04

30 22 85

45.5 27.1

0.096

2.03

1.26- 2.80

47 60

34.0 28.3

0.526

1.21

0.63-1.80

55

38.2

0.091

1.91

1.23-2.58

52

23.1

15 92

73.3 23.9

O.OOlt

3.14

2.33-3.96

7 100

71.4 28.0

0.028t

5.l2

0.46-56.49

7 100

71.4 28.0

O.071t

1.02

0.08-12.47

74 23

28.4 21.7

0.530

1.15

0.49-2.69

66 41

34.9 24.4

0.255

1.61

0.97-2.25

p.

AOR

AOR = adjusted odds ratio derived from logistic regression; CI = confidence interval; PVR = proliferative vitreoretinopathy; SF6 = sulfur hexafluoride; C 3FB = perfluoropropane . • Derived from univariate, chi-square analysis.

t

A two-tailed Fisher's exact test was performed f or small sample sizes.

success rate for pneumatic retinopexy (Table 3). 2,4,7- 30 Single-surgery success rates ranged from 53% to 100%. The initial success rate combining all series (including the current series) was 78.7%. Twenty-one series reported an ultimate success rate ranging from 92% to 100%, with an overall final success rate of 1287 (98%) in 1309 patients. The current report of pneumatic retinopexy in 107 patients showed an initial success rate of 69.2% (n = 74). Retinas in two additional patients were reattached successfully with a subsequent gas injection for an initial office success rate of 71 % (n = 76). If only eyes with accepted indications were counted (" additional pathology" eyes excluded), the initial success rate was 75.3% (70/93). The ultimate success rate was 98% (105/107). Our initial success rate is somewhat lower than published series because we included patients with stage C-I PVR, vitreous hemorrhage, and extended areas of retinal breaks. Three pa-

932

tients had breaks in more than I clock hour and all of them had unsuccessful reattachment after the first surgery. Five had preoperative vitreous hemorrhage and one had intraoperative vitreous hemorrhage. Reattachments of four of these patients failed. Seven had preoperative PVR, and reattachments of five of these failed. Tornambe and colleagues27 reported a series in which eyes also exceeded usual criteria for pneumatic retinopexy. They found that multiple breaks in multiple quadrants, large tears, associated vitreous hemorrhage, or stage C-l or less PVR could be managed successfully with pneumatic retinopexy. We did not have the same success with these problems. The immediate cause of initial failure was detailed in 12 of the above 26 series. There were 676 eyes published in these 12 articles, of which there were 161 failures (23.8%). The immediate cause offailure was new or missed breaks in 87 eyes (12.8%), reopened breaks in 25 (3.6%),

Grizzard et al . Pneumatic Retinopexy Failures Table 2. Percent of Eyes with a Final Visual Acuity Less Than 20/50

Variables Sex M F Preoperative visual acuity Worse than 20/50 20/50 or better Quads detached 4 and total 1,2,3 Location of tear 11-1 hrs Other Lens status Aphakic/pseudophakic Phakic Additional pathology Yes No PVR Yes No Vitreous hemorrhage Yes No Retinopexy Cryotherapy Laser Gas C 3 FS SF6 Initial success No Yes

No.

% Visual Acuity < 20/50

62 45

33.9 35.6

0.056

1.58

77

46.8 3.3

0.001

15.73

1.59-155.68

22

68.2 25.8

0.001

5.00

1.33-18.76

85 47 60

34.0 35.0

0.918

1.06

0.35-3.26

55 22

38.2 30.8

0.420

0.87

0.27-2.79

15 92

53.3 31.5

O.l00t

0.36

0.04-3.67

7 100

42.8 34.0

0.691t

1.35

0.10-18.18

7 100

57.1 33.0

0.232t

1.34

0.06-29.39

74 23

32.4 30.4

0.858

1.15

0.31-4.32

41 66

46.3 27.3

0.044

3.07

0.89-10.56

33 74

54.6 25.7

0.004

3.08

0.75-12.74

30

p.

AOR

AOR 95%CI 0.50-5.02

AOR = adjusted odds ratio derived from logistic regression; CI = confidence interval; PVR = proliferative vitreoretinopathy; C3FS = perfluoropropane; SF 6 = sulfur hexafluoride . • Derived from univariate, chi-square analysis.

t A two-tailed Fisher's exact test was performed for small sample sizes. PVR in 19 (2.8%), breaks never closed in 7 (1.0%), reason not given in 7 (1.0%), and other miscellaneous reasons in 16 (2.3%) (Table 4).

Our data also demonstrated a higher than expected number of patients whose retina failed to initially reattach. Retinas in 16 of our 33 initial failures did not reattach, suggesting poor postoperative compliance, poor case selection, missed breaks, or very prominent vitreoretinal traction (Fig 1). The immediate cause of failure in our series was new or missed breaks in 14.9%, reopened initial break in 11.2%, and breaks never closed in 4.6%. Forty-four percent of new or missed breaks occurred within 2 clock hours of the original breaks. The timing of failure was not published in the initial report of the randomized trial of pneumatic retinopexy

versus scleral buckling4 but is presented in the current series. Figure 2 shows that failure occurs early, with 19 failures (63% offailures) occurring during the first month. A similar result occurred in our series, with 25 failures (76% of failures) occurring during the first month. Chi-square analysis and multivariate analysis using logistic regression showed a markedly higher failure rate for males (43.6%) than for females (13.3%). This sex difference was confirmed by a review of the data from the randomized triaL This highly statistically significant difference may indicate that compliance is more difficult to obtain from males than from females, or it may represent some difference in ocular pathology not appreciated previously. Idiopathic macular holes are more likely to occur in females,31,32 and there may be some underlying difference

933

Ophthalmology

Volume 102, Number 6, June 1995

Table 3. Retinal Reattachment with Pneumatic Retinopexy, 1986-1990

References AIgvere et af Ambler et al8 Bovey et al9 Brinton and Hilton 10 Chan and Wessels ll Chen et al 12 Dominguez et al13 Dominguez et aP4 Friberg and Eller l5 Gnad et aP6 Hilton and Grizzard2 Hilton et aP7 Kellyl8 Liggett l9 Lowe et afo McAllister et afl McAllister et aIl2 Menchini et al23 Mortensen and Sj0lie24 Packo 25 Poliner and Grand26 Tornambe et aIl7 T ornambe et al4 van Effenterre et al2s Vygantas 29 Zakka30 Current study Total (%)

No. of Eyes 35 101 27 70 38 51 31 43 14 27 20 100 268 30 55 4 56 36 12 33 39 34 103 60 12 20 107 1426

Reattached Surgery" No. 29 78 18 57 30 32 29

40 12 24 18 84 217 16 45 4

Reattached Reoperationst (%)

No.

(83)

33 101 27 68 38

(94) (100) (100) (97) (100)

30 42 14 27 20 98 265

(97) (98) (100) (100) (100) (98) (99) (98) (100) (100) (92) (100) (95) (97) (99) (98) (92) (95) (98) (98)

(77)

(67) (82) (79) (63) (94) (93) (86) (89) (90) (84) (81) (53) (82) (100)

40

(71)

54 4 56

31 10 27 24 28 83 51 8 16 74 1125

(86) (83) (82) (62) (82) (81) (85) (67) (80) (69) (79)

11 33 37 33 102 59 11 19 105 1287

(%)

• Some series included supplemental postoperative laser photocoagulation or cryotherapy. t Only 1309 eyes had published data regarding re-operations.

in vitreoretinal relations that could lead to this observation. A similar sex difference in success rate has not been noted in scleral buckling series. 33 Risk factors for failure in scleral buckling (i.e., extent of the detachment, poor preoperative vision, aphakia/ pseudophakia, vitreous hemorrhage, and early PVR) are also risk factors for failure of pneumatic retinopexy.33,34 As expected, visual outcome correlated with initial vision (r = 0.60; P < 0.001). This was observed in the scleral buckling procedure as well. 35 Failure of the initial pneumatic procedure led to poorer visual outcome if only univariate analysis was considered. Multivariate analysis, however, had a 95% confidence interval of 0.75 to 12.74, which failed to confirm significance if other factors were considered. Ambler and associates 8 also noted preservation of good visual results even in eyes that failed the initial pneumatic procedure.

934

The initial success rate from pneumatic retinopexy is equal to or slightly less than that obtained by initial scleral buckling. 4,5 O'Malley and Swearingen36 in a retrospective study of 100 "pneumatic retinopexy eligible eyes," had a higher initial success rate (96%) but used historical controls of pneumatic retinopexy to compare results. No controlled trial was done by these authors. Good results can be maximized by careful preoperative examination, avoiding eyes with early signs ofPVR, such as star folds or rolled posterior edges, and carefully educating the patient and assessing his or her ability to comply with postoperative instructions. A high number of patients in whom the retina fails to reattach or a higher failure rate in males may indicate poor compliance with instructions. In eyes with initial failures of pneumatic retinopexy, the retina usually could be reattached with a subsequent

Grizzard et al . Pneumatic Retinopexy Failures Table 4. Failure of the First Pneumatic Retinopexy Procedure: Twelve Published Series, 1986-1993 (676 eyes)

References

No. of Failures

New/Missed Breaks

Ambler et alB Chan and Wessels 1J Chen et al 12

23 8 20

11 6 7

Dominguez et alB Hilton and Grizzard2 Lowe et aIlo McAllister et al21 T ornambe et al27

2 2 10 16 6

1 6 9 3

T ornambe and Hilton4 van Effenterre et aIlB Zakka30 Current study Total failures % of 676 eyes

28 9

PVR

=

Break Reopened 5

PVR

Break Never Closed

Reason Not Given

1 1 5

6

Other 6, 1, 1, 1,

SRF not reabsorbed anterior gas enlarged break SRF shift to flat break

1 1 1

1 2 1 3

1, subretinal gas 5, SFR not reabsorbed 1

4 33 161 23.8

19 7 2 16 87 12.8

5

12 25 3.6

3 1 1 19 2.8

1, bubble too small 1 5 7 1.0

7 1.0

16 2.3

proliferative vitreoretinopathy; SRF = subretinal fluid.

procedure and an initial attempt with pneumatic retinopexy does not reduce the chance of ultimate anatomic success. 4 .5 Failures of pneumatic retinopexy usually occur early, and late redetachment 6 months after pneumatic retinopexy occurs rarely.

References 1. Dominguez A. Cirugia precoz y ambulatoria del desprendimento de retina. Arch Soc Esp Oftalmol 1985;48:47-54. 2. Hilton GF, Grizzard WS. Pneumatic retinopexy. A twostep outpatient operation without conjunctival incision. Ophthalmology 1986;93:626-41. 3. Griffiths PG, Richardson J. Causes of proliferative retinopathy following pneumatic retinopexy [letter]. Arch Ophthalmol 1990; 108: 1515. 4. Tornambe PE, Hilton GF. The Retinal Detachment Study Group. Pneumatic retinopexy. A multicenter randomized controlled clinical trial comparing pneumatic retinopexy with scleral buckling. Ophthalmology 1989;96:772-84. 5. Tornambe PE, Hilton GF, Brinton DA, et al. Pneumatic Retinopexy. A two-year follow-up study of the multicenter clinical trial comparing pneumatic retinopexy with scleral buckling. Ophthalmology 1991;98:1115-23. 6. The Retina Society Terminology Committee. The classification of retinal detachment with proliferative vitreoretinopathy. Ophthalmology 1983;90:121-5. 7. Algvere P, Hallniis K, Palmqvist BM. Success and complications of pneumatic retinopexy. Am J Ophthalmol 1988; 106:400-4. 8. Ambler JS, Meyers SM, Zegarra H, Paranandi L. Reoperations and visual results after failed pneumatic retinopexy. Ophthalmology 1990;97:786-90.

9. Bovey EH, Bucher PJM,Gonvers M, Gailloud C. Le traitement du decollement de retine par cryocoagulation et injection intravitreenne de gaz. Premiers resultats. Klin Monatsbl Augenheilkd 1988; 192:468-70. 10. Brinton DA, Hilton GF. Clinical experience: Oakland, California. In: Tornambe PE, Grizzard WS, eds. Pneumatic Retinopexy: A Clinical Symposium, 1988. Chicago: Greenwood, 1989; 130-4. II. Chan CK, Wessels IF. Delayed subretinal fluid absorption after pneumatic retinopexy. Ophthalmology 1989;96: 16911700. 12. Chen JC, Robertson JE, Coonan P, et al. Results and complications of pneumatic retinopexy. Ophthalmology 1988;95:601-5. 13. Dominguez A, Boyd BF, Gordon S. Repeated insufflation of expansive gas. Highlights Ophthalmol Lett 1986;14(5): 1-14. 14. Dominguez A, Fonseca A, Gomez-Montana J. Gas tamponade for ambulatory treatment of retinal detachment. In: Acta XXV Concilium Ophthalmolgicum (Rome), 1986. Vol. 2;2038-45. 15. Friberg TR, Eller AW. Pneumatic repair of primary and secondary retinal detachments using a binocular indirect ophthalmoscope laser delivery system. Ophthalmology 1988;95:187-93. 16. Gnad HD, Skorpik Ch, Paroussis P, et al. C3Fs-gas als innere Tamponade in der Netzhautchirugie-eine "pneumatische" Ablatio-Operation ohne Eroffnung der Bindehaut. Fortschr Ophthalmol 1987;84:462-4. 17. Hilton GF, Kelly NE, Salzano TC, et al. Pneumatic retinopexy: A collaborative report of the first 100 cases. Ophthalmology 1987;94:307-14. 18. Kelly NE. Clinical experience: Sacramento, California. In: Tornambe PE, Grizzard WS, eds. Pneumatic Retinopexy:

935

Ophthalmology

19.

20. 21. 22. 23.

24. 25.

26.

27.

936

Volume 102, Number 6, June 1995

A Clinical Symposium, 1988. Chicago: Greenwood, 1989;151-4. Liggett PE. Clinical experience: Doheny Eye Institute, Los Angeles, California. In: Tornambe PE, Grizzard WS, eds. Pneumatic Retinopexy: A Clinical Symposium, 1988. Chicago: Greenwood, 1989; 155-6. Lowe MA, McDonald HR, Campo RV, et al. Pneumatic retinopexy. Surgical results. Arch Ophthalmol 1988;106: 1672-6. McAllister IL, Zegarra H, Meyers SM, Gutman FA. Treatment of retinal detachments with multiple breaks by pneumatic retinopexy. Arch Ophthalmol 1987;105:913-16. McAllister IL, Meyers SM, Zegarra H, et al. Comparison of pneumatic retinopexy with alternative surgical techniques. Ophthalmology 1988;95:877-83. Menchini U, Scialdone A, Davi G, Introini V. Trattamento di casi selezionati di distacco retinico regmatogeno mediante pneumoretinopessia. Atti Del 67 Congresso Societa Oftalmologica Italiana 1987. Mortensen KK, Sj0lie AK. Retinal detachment treated by pneumatic retinopexy. Acta Ophthalmol 1988;66:1879. Packo K. Clinical experience: Chicago, Illinois. In: Tornambe PE, Grizzard WS, eds. Pneumatic Retinopexy: A Clinical Symposium, 1988. Chicago: Greenwood, 1989;1659. Poliner LS, Grand MG. Clinical experience: St. Louis, Missouri. In: Tornambe PE, Grizzard WS, eds. Pneumatic Retinopexy: A Clinical Symposium, 1988. Chicago: Greenwood,1989;127-9. Tornambe PE, Hilton GF, Kelly NF, et al. Expanded in-

28.

29.

30.

31. 32. 33. 34.

35. 36.

dications for pneumatic retinopexy. Ophthalmology 1988;95:597-600. van Effenterre G, Haut J, Larricart P, et al. Gas tamponade as a single technique in the treatment of retinal detachment: is vitrectomy needed? A comparative study of 120 cases. Graefes Arch Clin Exp Ophthalmol 1987;225:254-8. Vygantas CM. Clinical experience: pneumatic retinopexy with C4FS (octafiuorocyclobutane). In: Tornambe PE, Grizzard WS, eds. Pneumatic Retinopexy: A Clinical Symposium" 1988. Chicago: Greenwood, 1989;135-41. Zakka K. Clinical experience: utilizing 15% C3Fs Los Angeles, California. In: Tornambe PE, Grizzard WS, eds. Pneumatic Retinopexy: A Clinical Symposium, 1988. Chicago: Greenwood, 1989;161-4. Aaberg TM. Macular holes: a review. Surv Ophthalmol 1970; 15: 139-62. Aaberg TM, Blair CJ, Gass JDM. Macular holes. Am J Ophthalmol 1970;69:555-62. Cowley M, Conway BP, Campochiaro PA, et al. Clinical risk factors for proliferative vitreoretinopathy. Arch OphthalmoI107:1147-51. Tornquist R, Bodin L, Tornquist P. Retinal detachment. A study of a population-based patient material in Sweden 1971-1981. IV. Prediction of surgical outcome. Acta Ophthalmo1 1988;66:637-42. Burton TC. Recovery of visual acuity after retinal detachment involving the macula. Trans Am Ophthalmol Soc 1982;80:475-97. O'Malley P, Swearingen K. Scleral buckle with diathermy for simple retinal detachments. 100 pneumatic retinopexy eligible eyes. Ophthalmology 1992;99:269-77.