- Email: [email protected]
Which Is Better? One or Two?
Which Is Better? One or Two? A Randomized Clinical Trial of Single, plate versus Double,plate Molteno Implantation for Glaucomas in Aphakia and Pseudophakia Dale K. Heuer, MD, Mary Ann Lloyd, MD, Donald A. Abrams, MD, George Baerveldt, MD, Don S. Minckler, MD, Martha B. Lee, PhD, James F. Martone, MD, MPH Purpose: Previous studies have suggested that primary double-plate Molteno implantation may be beneficial. Therefore, the authors performed a randomized clinical trial to evaluate the relative effectiveness and safety of single- versus double-plate Molteno implantation. Methods: From March 1988 to February 1990, 132 patients who underwent Molteno implantation for medically uncontrollable non-neovascular glaucomas in aphakia or pseudophakia were randomly assigned to receive either single- or double-plate implants. Results: The 1- and 2-year life-table success rates (success [survival] defined as 6 mmHg ::::; final intraocular pressure [lOP] ::::; 21 mmHg without additional glaucoma surgery or devastating complication) were 55% and 46% with single-plate implantation and 86% and 71% with double-plate implantation, respectively. The final postoperative visual acuities were within one line of the preoperative visual acuities or had improved in 73% and 80% of patients, respectively. Choroidal hemorrhages and/or effusions, corneal decompensation, flat anterior chambers, and phthisis bulbi were more common in the patients who had undergone double-plate Molteno implantation; however, transient elevations of lOP during the first few postoperative months were more common in the patients who had undergone single-plate Molteno implantation. Conclusions: Double-plate Molteno implantation more frequently affords lOP control than single-plate Molteno implantation; however, double plates are associated with greater risks of choroidal hemorrhages and/or effusions, corneal decompensation, flat anterior chambers, and phthisis bulbi. Ophthalmology 1992;99:1512-1519
Originally received: October 16, 1991. Revision accepted: May 13, 1992. From the Department of Ophthalmology, University of Southern California School of Medicine and the Doheny Eye Institute, Los Angeles. Dr. Lloyd also is currently affiliated with the Department of Veterans Affairs Outpatient Clinic, Los Angeles. Dr. Abrams is currently affiliated with the Department of Ophthalmology, Sinai Hospital, Baltimore. Dr. Martone is currently affiliated with Project Orbis, New York. Presented in part at the American Academy of Ophthalmology Annual Meeting, Atlanta, Oct/Nov 1990. Supported in part by the U. S. Department of Health and Human Services, Public Health Service Core Facilities for Clinical Vision Research
1512
Development grant EY -0340-10, National Eye Institute/NIH, Bethesda, Maryland; the Foundation for Glaucoma Research, San Francisco, California; National Glaucoma Research, a program of the American Health Assistance Foundation, Beltsville, Maryland; and Research to Prevent Blindness, Inc, New York, New York. The authors have no financial interest in Molteno implants. Dr. Baerveldt has a financial interest in an aqueous humor shunting device manufactured by another company. The views expressed herein are those of the authors and do not reflect the official policy or position of the U. S. Government. Reprint requests to Dale K. Heuer, MD, Doheny Eye Institute, 1450 San Pablo St, Los Angeles, CA 90033-4666.
Heuer et al . Single-plate versus Double-plate Molteno Implantation Table
Aqueous humor shunting devices, such as Molteno implants, anterior chamber tube shunts to encircling bands, and long Krupin-Denver valves to 180 0 bands, are being used more often in the management of glaucomas with poor surgical prognoses. With single-plate Molteno implantation for non-neovascular glaucomas in aphakia or pseudophakia, we achieved final intraocular pressures (lOPs) under 22 mmHg in 26 (63%) ofthe 41 patients on whom we were able to obtain at least 6-month follow-up; however, only 12 (29%) patients achieved final lOPs of less than 16 mmHg, and only 5 (12%) achieved final lOPs of less than 22 mmHg without antiglaucoma medications. 1 Furthermore, virtually all of the patients who were ultimately successfully controlled exhibited postoperative lOP courses similar to those seen with encapsulated filTable
1. Randomization Exclusion Criteria
Concurrent retinal detachment First stage of staged Molteno implantation performed during nonglaucoma surgery in eye with marginally functioning filtering bleb Patient unable to cooperate for unsedated lOP measurement Previous cyclodestructive procedure(s) Previous Molteno implantation(s) in eye undergoing surgery Previous scleral buckling procedure· Recent corneoscleral or corneal wound lOP = intraocular pressure. • Not absolute exclusion criteria: if at initiation of surgery, surgeon felt double-plate Molteno implantation was technically feasible despite conjunctival scarring and scleral buckling hardware present, patient was randomized.
2. Randomization, Demographic, and Preoperative Data Single-plates
Randomization: One-stage installations Surgical tube ligature release Spontaneous tube ligature release Two-stage installations· Age (years; mean ± SD)t Race:t Asian-Pacific Islander Black White: Hispanic non-Hispanic Type of Cataract Surgery:§ ECCE ICCE Pars plana lensectomy Uncertain Aphakia/Pseudophakia: ~ Aphakia Pseudophakia: Anterior chamber Posterior chamber Glaucoma diagnosis: II Open-angle glaucomas Angle-closure glaucomas Glaucomas associated with uveitis Congenital/developmental glaucomas Glaucomas associated with ocular trauma Uncertain Preoperative lOp··
37 12 25 13 19
(74%) (24%) (50%) (26%) to 84 (61.1 ± 16.2)
Double-plates 38 (75%) 13 (25%) 25 (49%) 13 (25%) 7 to 89 (62.1 ± 20.8)
2(4%) 10 (20%)
4(8%) 5 (10%)
6 (12%) 32 (64%)
4(8%) 38 (75%)
30 (60%) 16 (32%) 2(4%) 2(4%)
29 (57%) 12 (24%) 2(4%) 8 (16%)
14 (28%)
24 (47%)
9(18%) 27 (54%)
4(8%) 23 (45%)
15 (30%) 21 (42%) 5 (10%) 4(8%) 4(8%) 1 (2%) 18 to 73 (34.9 ± 12.3)
18 (35%) 17 (33%) 8(16%) 4(8%) 3(6%) 1 (4%) 15 to 60 (34.7 ± 11.3)
SD = standard deviation; ECCE = extracapsular cataract extraction; ICCE = intracapsular cataract extraction. • Thirty-one additional patients were randomly assigned to either single or double plates (16 patients and 15 patients, respectively) during the first stage of two-stage installations that were performed in combination with conventional glaucoma filtering procedures (10 patients and 13 patients, respectively) or other ocular surgery; however, they did not undergo the second stage of their two-stage installations during the study period because either their lOPs were adequately controlled or their visual potentials were subsequently judged to be inadequate to justify further intraocular surgical procedures.
t t
p = 0.78, t test. P
=
0.356, chi-square.
§P
=
0.243, chi-square.
11 P
=
0.088, chi-square.
II P
=
•• P
=
0.911, chi-square. 0.93, t test.
1513
Ophthalmology Volume 99, Number 10, October 1992 tering blebs after conventional filtering surgery; specifically, after initiation of aqueous flow through the Molteno drainage system, patients had lOP courses characterized by initial reductions ofIOPs, subsequent rebounds in lOPs typically peaking above 21 mmHg (the upper 95% confidence limit was approximately 30 mmHg) 2 to 7 weeks after surgery, and gradually declining lOPs over the ensuing weeks to months. The qualitatively poor lOP control
with single-plate Molteno implantation, as well as final lOPs under 22 mmHg in 3 of the 5 patients who underwent installation of second single plates when the initial single plates had not controlled their lOPs, suggested that primary double-plate Molteno implantation might be beneficial. Consequently, we began a randomized clinical trial to evaluate the relative effectiveness and safety of single- versus double-plate Molteno implantation.
Table 3. Postoperative Data Single-plates Life-table (survival) analysis (Fig 1):* 6-month success rate 12-month success rate 18-month success rate 24-month success rate Conventional outcome analysis by final lOPs (Figs 2A and 2B): Complete successest Qualified successes'!' Qualified failures§ Complete failures~ Insufficient follow-up II Follow-up (mos; mean ± SD)** Percentage postoperative lOP reduction (mean ± SD): All patientstt'H Successes§§ Postoperative antiglaucoma medications (mean ± SD): All patientstt'n Successes 1111 Visual acuity outcome (Fig 3):*** Betterttt Same'!'H Worse§§§ SD
=
80% 55%
46%
Double-plates 96% 86% 82%
46%
71%
5 (10%) 20 (40%) 2 (4%)
6 to 29 (14.9 ± 68.9)
6 (12%) 32 (63%) 4 (8%) 8 (16%) 1 (2%) 7 to 30 (16.4 ± 6.8)
-136 to 74 (25 ± 43) 5 to 74 (49 ± 17)
-67 to 83 (46 ± 33) 18 to 83 (56 ± 19)
oto 4 (1.6 ± 0.9) oto 3 (1.2 ± 0.9)
o to 4 (1.2 ± 0.9) o to 3 (1.2 ± 0.8)
13 (27%) 23 (47%) 13 (27%)
12 (24%) 27 (55%) 10 (20%)
22 (44%) 1 (2%)
standard deviation .
• Success (survival) defined as final lOPs that were greater than 5 mmHg and less than 22 mmHg without additional glaucoma surgery or devastating complication; P = 0.0035 log rank test (analysis stratified by one-stage installations with surgical tube ligature release, one-stage installations with spontaneous tube ligature release, and two-stage installations).
t Patients on whom no additional glaucoma procedures have been performed (or recommended) and whose final lOPs were greater than 5 mmHg and less than 22 mmHg without antiglaucoma medications. t Patients on whom no additional glaucoma procedures have been performed (or recommended) and whose final lOPs were greater than 5 mmHg and less than 22 mmHg with antiglaucoma medications. § Patients on whom no additional glaucoma procedures have been performed (or recommended) but whose final lOPs were greater than 21 mmHg. If Patients who underwent additional glaucoma procedures (or for whom they were recommended), who lost light perception attributed to glaucoma, whose final lOPs were less than 6 mmHg, or who experienced devastating complications. II Patients with less than 6-month follow-up on whom no additional glaucoma procedures have been performed (or recommended); one is single-plate patient with 4-month follow-up, who would otherwise be categorized as qualified success; other is double-plate patient with 5-month follow-up, who would otherwise be categorized as complete success . •• Excluding patients classified as qualified failures, complete failures, and insufficient follow-up. tt Excluding patients classified as insufficient follow-up; also excluding four patients (one with single-plate and three with double-plate implants) who experienced devastating complications. H P = 0.01, t test. §§ P = 0.16, t test. Iflf P = 0.02, Wilcoxon rank sum. IIII P = 0.88, Wilcoxon rank sum. ••• Excluding patients classified as insufficient follow-up; also excluding one severely mentally handicapped patient with double-plate implant in whom visual acuity measurements not possible; P = 0.47, Wilcoxon rank sum (for statistical analysis, each acuity increment on Figure 3 assigned integer value [1 = 20/20 through 16 = no light perception]). ttt Postoperative visual acuity at last two lines better than preoperative visual acuity. tH Postoperative visual acuity within one line of preoperative visual acuity. §§§ Postoperative visual acuity at least two lines worse than preoperative visual acuity.
1514
Heuer et al . Single-plate versus Double-plate Molteno Implantation
w
!;c
II:
III III W
o o
:J III
w
-I
40%
••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
20%
••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
m
I-;" ~
::::i
o
3
6
9
12
15
18
21
24
POSTOPERATIVE FOLLOW·UP (MONTHS)
Figure 1. Life-table success rate versus postoperative follow-up (solid diamonds represent patients with single-plate Molteno implants; open circles represent patients with double-plate Molteno implants; success defined as 6 mmHg :$; final lOP:$; 21 mmHg without additional glaucoma surgery or devastating complication; P = 0.0035 log rank test [analysis stratified by one-stage installations with surgical tube ligature release, one-stage installations with spontaneous tube ligature release, and two-stage installations]).
Subjects and Methods From March 1988 to February 1990, 132 patients from the Doheny Eye Institute who underwent Molteno implantation for medically uncontrollable non-neovascular glaucomas in aphakia or pseudophakia were randomly assigned to either single- or double-plate implants (only patients who were specifically excluded by the criteria listed in Table 1 were not randomized). All patients gave informed consent; the study protocol and informed consent form regarding Molteno implantation for glaucomas with poor surgical prognoses had been approved by the
Los Angeles County/University of Southern California Medical Center Institutional Review Board (research protocol 03133). The randomization lists were generated from a random numbers table, with randomization being stratified for one-stage and two-stage installations. The surgeons first confirmed that double-plate implantation was technically feasible and then requested randomization assignments. Each assignment was read from the appropriate list for either one-stage or two-stage installation by operating room personnel; the surgeons were masked to the assignment lists. Our basic Molteno implantation techniques have been described elsewhere. 1 Glycerin-preserved donor scleral grafts were used to cover the tubes for several millimeters posterior to the limbal or pars plana penetration sites in all cases. During double-plate Molteno implantation, one of the plates was passed under the superior rectus muscle, after which both plates were secured to the sclera approximately 10 mm from the limbus with two 5-0 polyester fiber sutures on each plate. During the one-stage, single-plate implantations or during the second stage of two-stage, single-plate implantations, a second first-stage single plate was usually implanted in the adjacent superior quadrant. Separate subconjunctival injections of 12 mg dexamethasone phosphate (24 mg/ml) and 20 mg gentamicin sulfate (40 mg/ml) were administered at the conclusion of most procedures. Our postoperative medical regimen included: topical corticosteroids (usually 1% prednisolone acetate or phosphate) for 2 to 4 months in tapering doses based on intraocular and bleb capsule inflammation; topical cycloplegics (usually 1% atropine sulfate) for 4 to 6 weeks; and topical antibiotics (usually 0.3% gentamicin or 0.3% tobramycin) for 1 to 4 weeks. The adjunctive systemic antifibrosis regimen that has been recommended by Molteno et af was not administered; however, a few patients with pre-existing uveitis did receive short, tapering courses of systemic corticosteroids. Antiglaucoma medications were
Table 4. Success Rates with Alternative Outcome Criteria* Single-plates
Double-plates
Final lOP:$; mmHg
25 (51%)
38 (76%)
Final lOP:$; 21 mmHg and ~ 30% postoperative lOP reduction Final lOP:$; 18 mmHg
23 (47%) 19 (39%)
33 (66%)
Final rop :$; 15 mmHg
11 (22%)
22 (44%)
Final lOP:$; 21 mmHg without antiglaucoma medication or :$; 15 mmHg with only one topical antiglaucoma medication
9(18%)
20 (40%)
Final lOP:$; 21 mmHg without antiglaucoma medication
5 (10%)
6 (12%)
29 (58%)
* Categorizing as failures all patients who underwent additional glaucoma procedures (or for whom they were
recommended), whose eyes had final rops less than 6 mmHg, or who experienced devastating complications; excluding 2 patients with less than 6-month follow-up on whom no additional glaucoma procedures have been performed (one is single-plate patient with 4-month follow-up whose final lOP was 17 mmHg on topical timolol maleate; the other is a double-plate patient with 5-month follow-up whose final rop was 15 mmHg without antiglaucoma medication).
1515
Ophthalmology Volume 99, Number 10, October 1992
•
~o
Ci
•
•
50
::I:
E
§.
•
40
D-
Q w
i=
•
•
30
>
• • • ••• f.. .............................
c(
a:
w
20
D-
0
U)
0
10
D-
o
10
• •
•
0
40
30
20
0
••
:. • .,.J
• •
l-
•
50
~60
PREOPERATIVE lOP (mm Hg)
A
Results Seventy-five patients underwent one-stage Molteno implantation and 26 patients underwent both stages of twostage Molteno implantation during the 2-year study enrollment period. The remaining 31 patients only underwent the first stage of two-stage installations; they did not undergo the second stage during the study period because either their lOPs were adequately controlled after the conventional glaucoma filtering procedures (or other
~60
Ci
::I:
restarted when clinically judged necessary to control (or at least moderate) lOPs in the one· stage implantations until the Molteno tube ligatures either spontaneously dis· solved or were surgically released and were restarted to supplement lOP reduction as clinically indicated in all functioning implants. Further glaucoma procedures were performed only ifIOP control was clinically judged to be inadequate to prevent progressive glaucomatous optic neuropathy . A successful surgical outcome was prospectively defined as: 6 mmHg :=:; final lOP:=:; 21 mmHg, no additional glaucoma surgery (excluding surgical tube ligature release), and no devastating complication. Because of unequal lengths and frequencies of follow-up among patients (both within and between the two groups), life-table (survival) analysis was performed to facilitate comparison of the outcome of the two groups. Follow-up data were obtained from local ophthalmologists on any patients who were unable to return to the Doheny Eye Institute.
50
E
§. D-
40
O
w > i=
30
•
c(
a:
w
D-
20
.~
>
o
.a"
•
...•••
:...
0
• ················Q.···e····q..·······················
10
D-
NLP
• •
• ..·······..··0·······················
••
~ U)
0
• •
o
~
••
O~-r~--r-~~~--r-~~~--~
o
B
10
20
30
40
50
~o
PREOPERATIVE lOP (mm Hg)
Figure 2. A, preoperative lOPs versus postoperative lOPs of patients with single-plate Molteno implants (open diamonds represent postoperative lOPs of patients on no supplemental antiglaucoma medication; solid diamonds represent postoperative lOPs of patients on antiglaucoma medication[s); diagonal line at preoperative lOP = postoperative lOP; horizontal dashed lines at postoperative lOPs of 6 and 21 mmHg). B, preoperative lOPs versus postoperative lOPs of patients with doubleplate Molteno implants (open circles represent postoperative lOPs of patients on no supplemental antiglaucoma medication; solid circles represent postoperative lOPs of patients on antiglaucoma medication[s); diagonal line at preoperative lOP = postoperative lOP; horizontal dashed lines at postoperative lOPs of 6 and 21 mmHg).
1516
!::
::J
0
c(
•
HIlI 5/200
oJ
c(
::J
U)
:;
• • 0
20/80
0
20/40
0
•
20/60
•• •
0 0
0
• •
0
I-
U)
0
D-
20/25 20/15
•
0
0
20/200
w > i= c( a: w D-
0 0
0
10
....
10
0
0
0
0
C\I
0
C\I
0
0
0
--
~
~
C\I
~
C\I
~
C\I
.. •
• •e 0 0
0
0 0
C\I
0
C\I
0
•
• • 0
• 0 0
-C\I
~
LO
D...J
Z
PREOPERATIVE VISUAL ACUITY Figure 3. Preoperative visual acuities versus postoperative visual acuities (solid diamonds represent patients with single-plate Molteno implants; open circles represent patients with double-plate Molteno implants; diagonal line at preoperative visual acuity = postoperative visual acuity).
Heuer et al . Single-plate versus Double-plate Molteno Implantation ocular surgical procedures) with which the first stage installation had been performed or their visual potentials we(e subsequently judged to be inadequate to justify further intraocular surgical procedures. Although the sample size may have been insufficient to demonstrate statistically significant differences with respect to some of the patients' preoperative demographic and ocular characteristics, overall, the patients in the single- and double-plate groups appear to be relatively comparable (Table 2). The 1- and 2-year life-table success rates were 55% and 46% with single-plate implantation and 86% and 71 % with double-plate implantation, respectively (Table 3) (Fig I). The overall success rates were 50% and 75% for the singleand double-plate groups, respectively; however, only 10% and 12%, respectively, were controlled without medications (these and other alternatively defined success rates are listed in Table 4). The length of follow-up for the patients categorized as successes averaged 14.9 months and 16.4 months, respectively. The specific preoperative and postoperative lOPs are presented in Figures 2A and 2B, respectively. The postoperative visual acuities were within 1 line of the preoperative visual acuities or had improved in 73% and 80% of patients, respectively (Table 3). The specific
preoperative and postoperative visual acuities are presented in Figure 3. Sequelae of early postoperative hypotony, specifically flat anterior chambers, and choroidal hemorrhages and/ or effusions, were more common in patients who had undergone double-plate Molteno implantation. Corneal decompensation or graft rejection and phthisis bulbi also may have been more frequent in the double-plate group; however, other postoperative complications have been infrequent in both groups (Table 5). The transient postoperative hypertensive phase, through which even many of the patients who were ultimately successfully controlled passed, was less marked in the double-plate group (Fig 4B; upper 95% confidence limit was approximately 20 mmHg) than in the singleplate group (Fig 4A; upper 95% confidence limit was approximately 27 mmHg). Only patients with at least 6 months of follow-up who were categorized as successes and who had undergone surgical ligature release or a twostage installation were included in that analysis (so the date of initiation of aqueous flow could be known precisely); consequently, the number of patients for this comparison afforded insufficient statistical power to establish a significant difference.
Table 5. Complications
Corneal decompensation or graft failure Flat anterior chamber" Localized choroidal hemorrhage or partially hemorrhagic choroidal effusion Hyphema Tube or plate erosion Massive choroidal hemorrhage Phthisis bulbi Pupillary block Retinal detachment Serous choroidal effusion for which surgical drainage performed Conjunctival wound leak (without tube or plate erosion) Endophthalmitis Interplate tube block Intraocular lens implant decentration during Molteno implantation Intraoperative scleral perforation Iris-tube block Tube-cornea touch Uveitis Vitreous-tube block NA
=
Singleplates
Doubleplates
P (Chi-square Test)
3 (6%) 2 (4%)
5 (10%) 5 (10%)
0.479 0.251
0(0%) 1 (2%) 1 (2%) 1 (2%) 0(0%) 1 (2%) 1 (2%)
4(8%) 2(4%) 2(4%) 1 (2%) 2 (4%)t 1 (2%) 1 (2%)
0.043 0.570 0.570 0.989 0.157 0.989 0.989
0(0%) 1 (2%) 0(0%)
0.157 0.310 0.320
NA
2(4%) 0(0%) 1 (2%) 1 (2%)
1 (2%) 1 (2%) 0(0%) 1 (2%) 0(0%) 0(0%)
0(0%) 0(0%) 1 (2%) 0(0%) 1 (2%) 1 (2%)
0.310 0.310 0.320 0.310 0.320 0.320
NA
not applicable .
• Frank or impending anterior lens capsule to corneal endothelium touch, for which anterior c hamber reformation (and/ or other surgical procedure) performed.
t Phthisis bulbi preceded in one patient by retinal detachment, repair of which complicated by streptococcus pneumonia endophthalmitis; phthisis bulbi preceded in second patient by penetrating keratoplasty and intraocular lens implant removal, which were complicated by retinal detachment and cyclitic membrane.
1517
Ophthalmology Volume 99, Number 10, October 1992 30 25 20
ax E
§. 15 Q.
Q
10 5 0 0
M
W
00
1m
100
100
DAYS AFTER AQUEOUS FLOW INITIATED
A 30 25
c; 20 X
E
§. 15 Q.
Q
10 5
o
B
M
W
00
1m
100
180
DAYS AFTER AQUEOUS FLOW INITIATED
Figure 4. A, composite lOP profile (thick line = mean interpolated lOP; thin lines = 95% confidence limits) of nine successful single-plate patients who had undergone surgical ligature releases after one-stage installations or who had undergone both stages of two-stage installations (only those patients included so date of initiation of aqueous flow could be known precisely). B, composite lOP profile (thick line = mean interpolated lOP; thin lines = 95% confidence limits) of 20 successful double-plate patients who had undergone surgical ligature releases after one-stage installations or who had undergone both stages of two-stage installations (only those patients included so date of initiation of aqueous flow could be known precisely).
Discussion Molten0 3 performed single-, double-, or quadruple-plate implantation in a series of20 patients (the initial patients were randomly assigned; however, once "significant differences were established," subsequent patients all underwent double-plate implantation). He concluded that (except for eyes with limited inflow, such as those with glaucomas associated with uveitis) double-plate implants
1518
provided the best compromise between freedom from early postoperative hypotony and final lOP control without antiglaucoma medications. 3 Furthermore, flow rates through the bleb capsules surrounding a full-sized Molteno plate and a half-sized Molteno plate have been correlated directly to the bleb surface areas in an experimental model of glaucoma. 4 On the contrary, Hitchings et al 5 did not report any apparent difference in lOP control between 180 0 and 360 0 explants with valved tubes; however, their failure to demonstrate any difference suggests that there may be an upper limit, beyond which further increases in explant surface may not contribute substantially to lOP control (even their 180 0 explant had a planar surface area of 340 mm 2 [assuming an equatorial globe diameter of 24 mm 6 ], which is 26% greater than the approximately 270 mm 2 planar surface area of the double-plate Molteno implant). The results of our study have shown that double-plate Molteno implantation affords better lOP control (both during the early postoperative hypertensive phase and at final follow-up) than single-plate Molteno implantation. The additional surgical trauma related to the implantations of reserve first-stage single plates may have unfavorably influenced the outcome in our single-plate installations. However, our clinical practice before the initiation of this clinical trial had already evolved to nearly routine implantations of the reserve first-stage single plates, because lOP reductions had so often been inadequate after single-plate Molteno implantations. Furthermore, it remains our impression that the additional risks related to the implantations of those second single plates are negligible compared with the benefit of being able to proceed with the second stages of two-stage installations (rather than having to perform one-stage installations) in the many patients for whom lOP control has been inadequate after the initial single-plate implantation. The presence of reserve first-stage single-plate Molteno implants may have contributed to the higher rate of additional glaucoma procedures for patients in the single-plate group by lowering the threshold for the decision to perform additional glaucoma procedures. However, the relatively comparable lOPs (on comparable numbers of antiglaucoma medications) for which patients in the two groups underwent further glaucoma surgery indicate that there was not a significantly lower threshold for the decision to perform additional glaucoma procedures in the single-plate group (final lOPs before further glaucoma surgery = 17 to 80 [33.7 ± 14.9] mmHg versus 27 to 43 [34.2 ± 7.1] mmHg, respectively [P = 0.40, t test], on 1 to 4 [2.0 ± 0.7] versus 0 to 4 [1.6 ± 1.5] antiglaucoma medications, respectively [P = 0.23, Wilcoxon rank sum]). The undesirable sequelae of early postoperative hypotony have been more frequent in the double-plate group. However, because single-plate patients underwent additional surgery to achieve lOP control more frequently, they may have been more likely overall to have surgical complications. Furthermore, even among patients who did not undergo further glaucoma procedures, single-plate patients had more exaggerated early postoperative hypertensive phases than did double-plate patients. Conse-
Heuer et al . Single-plate versus Double-plate Molteno Implantation quently, considerably longer follow-up will be necessary t6 determine whether single-plate implantations (followed by installation of second single plates when necessary for further lOP reduction) or primary double-plate implantations afford the best compromise between expeditious lOP control and overall minimization of complications.
References 1. Minckler DS, Heuer DK, Hasty B, et al. Clinical experience with the single-plate Molteno implant in complicated glaucomas. Ophthalmology 1988;95:1181-8.
2. Molteno ACB, Straughan JL, Ancker E. Control of bleb fibrosis after glaucoma surgery by anti-inflammatory agents. S Afr Med J 1976;50:881-5. 3. Molteno ACB. The optimal design of drainage implants for glaucoma. Trans Ophthalmol Soc N Z 1981;33:29-41. 4. Minckler DS, Shammas A, Wilcox M, Ogden TE. Experimental studies of aqueous filtration using the Molteno implant. Trans Am Ophthalmol Soc 1987;85:368-92. 5. Hitchings RA, Joseph NH, Sherwood MB, et al. Use of onepiece valved tube and variable surface area explant for glaucoma drainage surgery. Ophthalmology 1987;94: 1079-84. 6. Hogan MJ, Alvarado JA, Weddell JE. Histology of the Human Eye: An Atlas and Textbook. Philadelphia: WB Saunders, 1971 ;50.
1519
Originally received: October 16, 1991. Revision accepted: May 13, 1992. From the Department of Ophthalmology, University of Southern California School of Medicine and the Doheny Eye Institute, Los Angeles. Dr. Lloyd also is currently affiliated with the Department of Veterans Affairs Outpatient Clinic, Los Angeles. Dr. Abrams is currently affiliated with the Department of Ophthalmology, Sinai Hospital, Baltimore. Dr. Martone is currently affiliated with Project Orbis, New York. Presented in part at the American Academy of Ophthalmology Annual Meeting, Atlanta, Oct/Nov 1990. Supported in part by the U. S. Department of Health and Human Services, Public Health Service Core Facilities for Clinical Vision Research
1512
Development grant EY -0340-10, National Eye Institute/NIH, Bethesda, Maryland; the Foundation for Glaucoma Research, San Francisco, California; National Glaucoma Research, a program of the American Health Assistance Foundation, Beltsville, Maryland; and Research to Prevent Blindness, Inc, New York, New York. The authors have no financial interest in Molteno implants. Dr. Baerveldt has a financial interest in an aqueous humor shunting device manufactured by another company. The views expressed herein are those of the authors and do not reflect the official policy or position of the U. S. Government. Reprint requests to Dale K. Heuer, MD, Doheny Eye Institute, 1450 San Pablo St, Los Angeles, CA 90033-4666.
Heuer et al . Single-plate versus Double-plate Molteno Implantation Table
Aqueous humor shunting devices, such as Molteno implants, anterior chamber tube shunts to encircling bands, and long Krupin-Denver valves to 180 0 bands, are being used more often in the management of glaucomas with poor surgical prognoses. With single-plate Molteno implantation for non-neovascular glaucomas in aphakia or pseudophakia, we achieved final intraocular pressures (lOPs) under 22 mmHg in 26 (63%) ofthe 41 patients on whom we were able to obtain at least 6-month follow-up; however, only 12 (29%) patients achieved final lOPs of less than 16 mmHg, and only 5 (12%) achieved final lOPs of less than 22 mmHg without antiglaucoma medications. 1 Furthermore, virtually all of the patients who were ultimately successfully controlled exhibited postoperative lOP courses similar to those seen with encapsulated filTable
1. Randomization Exclusion Criteria
Concurrent retinal detachment First stage of staged Molteno implantation performed during nonglaucoma surgery in eye with marginally functioning filtering bleb Patient unable to cooperate for unsedated lOP measurement Previous cyclodestructive procedure(s) Previous Molteno implantation(s) in eye undergoing surgery Previous scleral buckling procedure· Recent corneoscleral or corneal wound lOP = intraocular pressure. • Not absolute exclusion criteria: if at initiation of surgery, surgeon felt double-plate Molteno implantation was technically feasible despite conjunctival scarring and scleral buckling hardware present, patient was randomized.
2. Randomization, Demographic, and Preoperative Data Single-plates
Randomization: One-stage installations Surgical tube ligature release Spontaneous tube ligature release Two-stage installations· Age (years; mean ± SD)t Race:t Asian-Pacific Islander Black White: Hispanic non-Hispanic Type of Cataract Surgery:§ ECCE ICCE Pars plana lensectomy Uncertain Aphakia/Pseudophakia: ~ Aphakia Pseudophakia: Anterior chamber Posterior chamber Glaucoma diagnosis: II Open-angle glaucomas Angle-closure glaucomas Glaucomas associated with uveitis Congenital/developmental glaucomas Glaucomas associated with ocular trauma Uncertain Preoperative lOp··
37 12 25 13 19
(74%) (24%) (50%) (26%) to 84 (61.1 ± 16.2)
Double-plates 38 (75%) 13 (25%) 25 (49%) 13 (25%) 7 to 89 (62.1 ± 20.8)
2(4%) 10 (20%)
4(8%) 5 (10%)
6 (12%) 32 (64%)
4(8%) 38 (75%)
30 (60%) 16 (32%) 2(4%) 2(4%)
29 (57%) 12 (24%) 2(4%) 8 (16%)
14 (28%)
24 (47%)
9(18%) 27 (54%)
4(8%) 23 (45%)
15 (30%) 21 (42%) 5 (10%) 4(8%) 4(8%) 1 (2%) 18 to 73 (34.9 ± 12.3)
18 (35%) 17 (33%) 8(16%) 4(8%) 3(6%) 1 (4%) 15 to 60 (34.7 ± 11.3)
SD = standard deviation; ECCE = extracapsular cataract extraction; ICCE = intracapsular cataract extraction. • Thirty-one additional patients were randomly assigned to either single or double plates (16 patients and 15 patients, respectively) during the first stage of two-stage installations that were performed in combination with conventional glaucoma filtering procedures (10 patients and 13 patients, respectively) or other ocular surgery; however, they did not undergo the second stage of their two-stage installations during the study period because either their lOPs were adequately controlled or their visual potentials were subsequently judged to be inadequate to justify further intraocular surgical procedures.
t t
p = 0.78, t test. P
=
0.356, chi-square.
§P
=
0.243, chi-square.
11 P
=
0.088, chi-square.
II P
=
•• P
=
0.911, chi-square. 0.93, t test.
1513
Ophthalmology Volume 99, Number 10, October 1992 tering blebs after conventional filtering surgery; specifically, after initiation of aqueous flow through the Molteno drainage system, patients had lOP courses characterized by initial reductions ofIOPs, subsequent rebounds in lOPs typically peaking above 21 mmHg (the upper 95% confidence limit was approximately 30 mmHg) 2 to 7 weeks after surgery, and gradually declining lOPs over the ensuing weeks to months. The qualitatively poor lOP control
with single-plate Molteno implantation, as well as final lOPs under 22 mmHg in 3 of the 5 patients who underwent installation of second single plates when the initial single plates had not controlled their lOPs, suggested that primary double-plate Molteno implantation might be beneficial. Consequently, we began a randomized clinical trial to evaluate the relative effectiveness and safety of single- versus double-plate Molteno implantation.
Table 3. Postoperative Data Single-plates Life-table (survival) analysis (Fig 1):* 6-month success rate 12-month success rate 18-month success rate 24-month success rate Conventional outcome analysis by final lOPs (Figs 2A and 2B): Complete successest Qualified successes'!' Qualified failures§ Complete failures~ Insufficient follow-up II Follow-up (mos; mean ± SD)** Percentage postoperative lOP reduction (mean ± SD): All patientstt'H Successes§§ Postoperative antiglaucoma medications (mean ± SD): All patientstt'n Successes 1111 Visual acuity outcome (Fig 3):*** Betterttt Same'!'H Worse§§§ SD
=
80% 55%
46%
Double-plates 96% 86% 82%
46%
71%
5 (10%) 20 (40%) 2 (4%)
6 to 29 (14.9 ± 68.9)
6 (12%) 32 (63%) 4 (8%) 8 (16%) 1 (2%) 7 to 30 (16.4 ± 6.8)
-136 to 74 (25 ± 43) 5 to 74 (49 ± 17)
-67 to 83 (46 ± 33) 18 to 83 (56 ± 19)
oto 4 (1.6 ± 0.9) oto 3 (1.2 ± 0.9)
o to 4 (1.2 ± 0.9) o to 3 (1.2 ± 0.8)
13 (27%) 23 (47%) 13 (27%)
12 (24%) 27 (55%) 10 (20%)
22 (44%) 1 (2%)
standard deviation .
• Success (survival) defined as final lOPs that were greater than 5 mmHg and less than 22 mmHg without additional glaucoma surgery or devastating complication; P = 0.0035 log rank test (analysis stratified by one-stage installations with surgical tube ligature release, one-stage installations with spontaneous tube ligature release, and two-stage installations).
t Patients on whom no additional glaucoma procedures have been performed (or recommended) and whose final lOPs were greater than 5 mmHg and less than 22 mmHg without antiglaucoma medications. t Patients on whom no additional glaucoma procedures have been performed (or recommended) and whose final lOPs were greater than 5 mmHg and less than 22 mmHg with antiglaucoma medications. § Patients on whom no additional glaucoma procedures have been performed (or recommended) but whose final lOPs were greater than 21 mmHg. If Patients who underwent additional glaucoma procedures (or for whom they were recommended), who lost light perception attributed to glaucoma, whose final lOPs were less than 6 mmHg, or who experienced devastating complications. II Patients with less than 6-month follow-up on whom no additional glaucoma procedures have been performed (or recommended); one is single-plate patient with 4-month follow-up, who would otherwise be categorized as qualified success; other is double-plate patient with 5-month follow-up, who would otherwise be categorized as complete success . •• Excluding patients classified as qualified failures, complete failures, and insufficient follow-up. tt Excluding patients classified as insufficient follow-up; also excluding four patients (one with single-plate and three with double-plate implants) who experienced devastating complications. H P = 0.01, t test. §§ P = 0.16, t test. Iflf P = 0.02, Wilcoxon rank sum. IIII P = 0.88, Wilcoxon rank sum. ••• Excluding patients classified as insufficient follow-up; also excluding one severely mentally handicapped patient with double-plate implant in whom visual acuity measurements not possible; P = 0.47, Wilcoxon rank sum (for statistical analysis, each acuity increment on Figure 3 assigned integer value [1 = 20/20 through 16 = no light perception]). ttt Postoperative visual acuity at last two lines better than preoperative visual acuity. tH Postoperative visual acuity within one line of preoperative visual acuity. §§§ Postoperative visual acuity at least two lines worse than preoperative visual acuity.
1514
Heuer et al . Single-plate versus Double-plate Molteno Implantation
w
!;c
II:
III III W
o o
:J III
w
-I
40%
••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
20%
••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
m
I-;" ~
::::i
o
3
6
9
12
15
18
21
24
POSTOPERATIVE FOLLOW·UP (MONTHS)
Figure 1. Life-table success rate versus postoperative follow-up (solid diamonds represent patients with single-plate Molteno implants; open circles represent patients with double-plate Molteno implants; success defined as 6 mmHg :$; final lOP:$; 21 mmHg without additional glaucoma surgery or devastating complication; P = 0.0035 log rank test [analysis stratified by one-stage installations with surgical tube ligature release, one-stage installations with spontaneous tube ligature release, and two-stage installations]).
Subjects and Methods From March 1988 to February 1990, 132 patients from the Doheny Eye Institute who underwent Molteno implantation for medically uncontrollable non-neovascular glaucomas in aphakia or pseudophakia were randomly assigned to either single- or double-plate implants (only patients who were specifically excluded by the criteria listed in Table 1 were not randomized). All patients gave informed consent; the study protocol and informed consent form regarding Molteno implantation for glaucomas with poor surgical prognoses had been approved by the
Los Angeles County/University of Southern California Medical Center Institutional Review Board (research protocol 03133). The randomization lists were generated from a random numbers table, with randomization being stratified for one-stage and two-stage installations. The surgeons first confirmed that double-plate implantation was technically feasible and then requested randomization assignments. Each assignment was read from the appropriate list for either one-stage or two-stage installation by operating room personnel; the surgeons were masked to the assignment lists. Our basic Molteno implantation techniques have been described elsewhere. 1 Glycerin-preserved donor scleral grafts were used to cover the tubes for several millimeters posterior to the limbal or pars plana penetration sites in all cases. During double-plate Molteno implantation, one of the plates was passed under the superior rectus muscle, after which both plates were secured to the sclera approximately 10 mm from the limbus with two 5-0 polyester fiber sutures on each plate. During the one-stage, single-plate implantations or during the second stage of two-stage, single-plate implantations, a second first-stage single plate was usually implanted in the adjacent superior quadrant. Separate subconjunctival injections of 12 mg dexamethasone phosphate (24 mg/ml) and 20 mg gentamicin sulfate (40 mg/ml) were administered at the conclusion of most procedures. Our postoperative medical regimen included: topical corticosteroids (usually 1% prednisolone acetate or phosphate) for 2 to 4 months in tapering doses based on intraocular and bleb capsule inflammation; topical cycloplegics (usually 1% atropine sulfate) for 4 to 6 weeks; and topical antibiotics (usually 0.3% gentamicin or 0.3% tobramycin) for 1 to 4 weeks. The adjunctive systemic antifibrosis regimen that has been recommended by Molteno et af was not administered; however, a few patients with pre-existing uveitis did receive short, tapering courses of systemic corticosteroids. Antiglaucoma medications were
Table 4. Success Rates with Alternative Outcome Criteria* Single-plates
Double-plates
Final lOP:$; mmHg
25 (51%)
38 (76%)
Final lOP:$; 21 mmHg and ~ 30% postoperative lOP reduction Final lOP:$; 18 mmHg
23 (47%) 19 (39%)
33 (66%)
Final rop :$; 15 mmHg
11 (22%)
22 (44%)
Final lOP:$; 21 mmHg without antiglaucoma medication or :$; 15 mmHg with only one topical antiglaucoma medication
9(18%)
20 (40%)
Final lOP:$; 21 mmHg without antiglaucoma medication
5 (10%)
6 (12%)
29 (58%)
* Categorizing as failures all patients who underwent additional glaucoma procedures (or for whom they were
recommended), whose eyes had final rops less than 6 mmHg, or who experienced devastating complications; excluding 2 patients with less than 6-month follow-up on whom no additional glaucoma procedures have been performed (one is single-plate patient with 4-month follow-up whose final lOP was 17 mmHg on topical timolol maleate; the other is a double-plate patient with 5-month follow-up whose final rop was 15 mmHg without antiglaucoma medication).
1515
Ophthalmology Volume 99, Number 10, October 1992
•
~o
Ci
•
•
50
::I:
E
§.
•
40
D-
Q w
i=
•
•
30
>
• • • ••• f.. .............................
c(
a:
w
20
D-
0
U)
0
10
D-
o
10
• •
•
0
40
30
20
0
••
:. • .,.J
• •
l-
•
50
~60
PREOPERATIVE lOP (mm Hg)
A
Results Seventy-five patients underwent one-stage Molteno implantation and 26 patients underwent both stages of twostage Molteno implantation during the 2-year study enrollment period. The remaining 31 patients only underwent the first stage of two-stage installations; they did not undergo the second stage during the study period because either their lOPs were adequately controlled after the conventional glaucoma filtering procedures (or other
~60
Ci
::I:
restarted when clinically judged necessary to control (or at least moderate) lOPs in the one· stage implantations until the Molteno tube ligatures either spontaneously dis· solved or were surgically released and were restarted to supplement lOP reduction as clinically indicated in all functioning implants. Further glaucoma procedures were performed only ifIOP control was clinically judged to be inadequate to prevent progressive glaucomatous optic neuropathy . A successful surgical outcome was prospectively defined as: 6 mmHg :=:; final lOP:=:; 21 mmHg, no additional glaucoma surgery (excluding surgical tube ligature release), and no devastating complication. Because of unequal lengths and frequencies of follow-up among patients (both within and between the two groups), life-table (survival) analysis was performed to facilitate comparison of the outcome of the two groups. Follow-up data were obtained from local ophthalmologists on any patients who were unable to return to the Doheny Eye Institute.
50
E
§. D-
40
O
w > i=
30
•
c(
a:
w
D-
20
.~
>
o
.a"
•
...•••
:...
0
• ················Q.···e····q..·······················
10
D-
NLP
• •
• ..·······..··0·······················
••
~ U)
0
• •
o
~
••
O~-r~--r-~~~--r-~~~--~
o
B
10
20
30
40
50
~o
PREOPERATIVE lOP (mm Hg)
Figure 2. A, preoperative lOPs versus postoperative lOPs of patients with single-plate Molteno implants (open diamonds represent postoperative lOPs of patients on no supplemental antiglaucoma medication; solid diamonds represent postoperative lOPs of patients on antiglaucoma medication[s); diagonal line at preoperative lOP = postoperative lOP; horizontal dashed lines at postoperative lOPs of 6 and 21 mmHg). B, preoperative lOPs versus postoperative lOPs of patients with doubleplate Molteno implants (open circles represent postoperative lOPs of patients on no supplemental antiglaucoma medication; solid circles represent postoperative lOPs of patients on antiglaucoma medication[s); diagonal line at preoperative lOP = postoperative lOP; horizontal dashed lines at postoperative lOPs of 6 and 21 mmHg).
1516
!::
::J
0
c(
•
HIlI 5/200
oJ
c(
::J
U)
:;
• • 0
20/80
0
20/40
0
•
20/60
•• •
0 0
0
• •
0
I-
U)
0
D-
20/25 20/15
•
0
0
20/200
w > i= c( a: w D-
0 0
0
10
....
10
0
0
0
0
C\I
0
C\I
0
0
0
--
~
~
C\I
~
C\I
~
C\I
.. •
• •e 0 0
0
0 0
C\I
0
C\I
0
•
• • 0
• 0 0
-C\I
~
LO
D...J
Z
PREOPERATIVE VISUAL ACUITY Figure 3. Preoperative visual acuities versus postoperative visual acuities (solid diamonds represent patients with single-plate Molteno implants; open circles represent patients with double-plate Molteno implants; diagonal line at preoperative visual acuity = postoperative visual acuity).
Heuer et al . Single-plate versus Double-plate Molteno Implantation ocular surgical procedures) with which the first stage installation had been performed or their visual potentials we(e subsequently judged to be inadequate to justify further intraocular surgical procedures. Although the sample size may have been insufficient to demonstrate statistically significant differences with respect to some of the patients' preoperative demographic and ocular characteristics, overall, the patients in the single- and double-plate groups appear to be relatively comparable (Table 2). The 1- and 2-year life-table success rates were 55% and 46% with single-plate implantation and 86% and 71 % with double-plate implantation, respectively (Table 3) (Fig I). The overall success rates were 50% and 75% for the singleand double-plate groups, respectively; however, only 10% and 12%, respectively, were controlled without medications (these and other alternatively defined success rates are listed in Table 4). The length of follow-up for the patients categorized as successes averaged 14.9 months and 16.4 months, respectively. The specific preoperative and postoperative lOPs are presented in Figures 2A and 2B, respectively. The postoperative visual acuities were within 1 line of the preoperative visual acuities or had improved in 73% and 80% of patients, respectively (Table 3). The specific
preoperative and postoperative visual acuities are presented in Figure 3. Sequelae of early postoperative hypotony, specifically flat anterior chambers, and choroidal hemorrhages and/ or effusions, were more common in patients who had undergone double-plate Molteno implantation. Corneal decompensation or graft rejection and phthisis bulbi also may have been more frequent in the double-plate group; however, other postoperative complications have been infrequent in both groups (Table 5). The transient postoperative hypertensive phase, through which even many of the patients who were ultimately successfully controlled passed, was less marked in the double-plate group (Fig 4B; upper 95% confidence limit was approximately 20 mmHg) than in the singleplate group (Fig 4A; upper 95% confidence limit was approximately 27 mmHg). Only patients with at least 6 months of follow-up who were categorized as successes and who had undergone surgical ligature release or a twostage installation were included in that analysis (so the date of initiation of aqueous flow could be known precisely); consequently, the number of patients for this comparison afforded insufficient statistical power to establish a significant difference.
Table 5. Complications
Corneal decompensation or graft failure Flat anterior chamber" Localized choroidal hemorrhage or partially hemorrhagic choroidal effusion Hyphema Tube or plate erosion Massive choroidal hemorrhage Phthisis bulbi Pupillary block Retinal detachment Serous choroidal effusion for which surgical drainage performed Conjunctival wound leak (without tube or plate erosion) Endophthalmitis Interplate tube block Intraocular lens implant decentration during Molteno implantation Intraoperative scleral perforation Iris-tube block Tube-cornea touch Uveitis Vitreous-tube block NA
=
Singleplates
Doubleplates
P (Chi-square Test)
3 (6%) 2 (4%)
5 (10%) 5 (10%)
0.479 0.251
0(0%) 1 (2%) 1 (2%) 1 (2%) 0(0%) 1 (2%) 1 (2%)
4(8%) 2(4%) 2(4%) 1 (2%) 2 (4%)t 1 (2%) 1 (2%)
0.043 0.570 0.570 0.989 0.157 0.989 0.989
0(0%) 1 (2%) 0(0%)
0.157 0.310 0.320
NA
2(4%) 0(0%) 1 (2%) 1 (2%)
1 (2%) 1 (2%) 0(0%) 1 (2%) 0(0%) 0(0%)
0(0%) 0(0%) 1 (2%) 0(0%) 1 (2%) 1 (2%)
0.310 0.310 0.320 0.310 0.320 0.320
NA
not applicable .
• Frank or impending anterior lens capsule to corneal endothelium touch, for which anterior c hamber reformation (and/ or other surgical procedure) performed.
t Phthisis bulbi preceded in one patient by retinal detachment, repair of which complicated by streptococcus pneumonia endophthalmitis; phthisis bulbi preceded in second patient by penetrating keratoplasty and intraocular lens implant removal, which were complicated by retinal detachment and cyclitic membrane.
1517
Ophthalmology Volume 99, Number 10, October 1992 30 25 20
ax E
§. 15 Q.
Q
10 5 0 0
M
W
00
1m
100
100
DAYS AFTER AQUEOUS FLOW INITIATED
A 30 25
c; 20 X
E
§. 15 Q.
Q
10 5
o
B
M
W
00
1m
100
180
DAYS AFTER AQUEOUS FLOW INITIATED
Figure 4. A, composite lOP profile (thick line = mean interpolated lOP; thin lines = 95% confidence limits) of nine successful single-plate patients who had undergone surgical ligature releases after one-stage installations or who had undergone both stages of two-stage installations (only those patients included so date of initiation of aqueous flow could be known precisely). B, composite lOP profile (thick line = mean interpolated lOP; thin lines = 95% confidence limits) of 20 successful double-plate patients who had undergone surgical ligature releases after one-stage installations or who had undergone both stages of two-stage installations (only those patients included so date of initiation of aqueous flow could be known precisely).
Discussion Molten0 3 performed single-, double-, or quadruple-plate implantation in a series of20 patients (the initial patients were randomly assigned; however, once "significant differences were established," subsequent patients all underwent double-plate implantation). He concluded that (except for eyes with limited inflow, such as those with glaucomas associated with uveitis) double-plate implants
1518
provided the best compromise between freedom from early postoperative hypotony and final lOP control without antiglaucoma medications. 3 Furthermore, flow rates through the bleb capsules surrounding a full-sized Molteno plate and a half-sized Molteno plate have been correlated directly to the bleb surface areas in an experimental model of glaucoma. 4 On the contrary, Hitchings et al 5 did not report any apparent difference in lOP control between 180 0 and 360 0 explants with valved tubes; however, their failure to demonstrate any difference suggests that there may be an upper limit, beyond which further increases in explant surface may not contribute substantially to lOP control (even their 180 0 explant had a planar surface area of 340 mm 2 [assuming an equatorial globe diameter of 24 mm 6 ], which is 26% greater than the approximately 270 mm 2 planar surface area of the double-plate Molteno implant). The results of our study have shown that double-plate Molteno implantation affords better lOP control (both during the early postoperative hypertensive phase and at final follow-up) than single-plate Molteno implantation. The additional surgical trauma related to the implantations of reserve first-stage single plates may have unfavorably influenced the outcome in our single-plate installations. However, our clinical practice before the initiation of this clinical trial had already evolved to nearly routine implantations of the reserve first-stage single plates, because lOP reductions had so often been inadequate after single-plate Molteno implantations. Furthermore, it remains our impression that the additional risks related to the implantations of those second single plates are negligible compared with the benefit of being able to proceed with the second stages of two-stage installations (rather than having to perform one-stage installations) in the many patients for whom lOP control has been inadequate after the initial single-plate implantation. The presence of reserve first-stage single-plate Molteno implants may have contributed to the higher rate of additional glaucoma procedures for patients in the single-plate group by lowering the threshold for the decision to perform additional glaucoma procedures. However, the relatively comparable lOPs (on comparable numbers of antiglaucoma medications) for which patients in the two groups underwent further glaucoma surgery indicate that there was not a significantly lower threshold for the decision to perform additional glaucoma procedures in the single-plate group (final lOPs before further glaucoma surgery = 17 to 80 [33.7 ± 14.9] mmHg versus 27 to 43 [34.2 ± 7.1] mmHg, respectively [P = 0.40, t test], on 1 to 4 [2.0 ± 0.7] versus 0 to 4 [1.6 ± 1.5] antiglaucoma medications, respectively [P = 0.23, Wilcoxon rank sum]). The undesirable sequelae of early postoperative hypotony have been more frequent in the double-plate group. However, because single-plate patients underwent additional surgery to achieve lOP control more frequently, they may have been more likely overall to have surgical complications. Furthermore, even among patients who did not undergo further glaucoma procedures, single-plate patients had more exaggerated early postoperative hypertensive phases than did double-plate patients. Conse-
Heuer et al . Single-plate versus Double-plate Molteno Implantation quently, considerably longer follow-up will be necessary t6 determine whether single-plate implantations (followed by installation of second single plates when necessary for further lOP reduction) or primary double-plate implantations afford the best compromise between expeditious lOP control and overall minimization of complications.
References 1. Minckler DS, Heuer DK, Hasty B, et al. Clinical experience with the single-plate Molteno implant in complicated glaucomas. Ophthalmology 1988;95:1181-8.
2. Molteno ACB, Straughan JL, Ancker E. Control of bleb fibrosis after glaucoma surgery by anti-inflammatory agents. S Afr Med J 1976;50:881-5. 3. Molteno ACB. The optimal design of drainage implants for glaucoma. Trans Ophthalmol Soc N Z 1981;33:29-41. 4. Minckler DS, Shammas A, Wilcox M, Ogden TE. Experimental studies of aqueous filtration using the Molteno implant. Trans Am Ophthalmol Soc 1987;85:368-92. 5. Hitchings RA, Joseph NH, Sherwood MB, et al. Use of onepiece valved tube and variable surface area explant for glaucoma drainage surgery. Ophthalmology 1987;94: 1079-84. 6. Hogan MJ, Alvarado JA, Weddell JE. Histology of the Human Eye: An Atlas and Textbook. Philadelphia: WB Saunders, 1971 ;50.
1519