8.1. Introduction
In this chapter we consider the clinical and cost effectiveness of treatments for COAG. We examine various pharmacological treatments (as in the previous chapter) as well as laser treatments and surgical procedures.
Pharmacological treatment
Eye drops are the most commonly used treatment for COAG. There are five main classes of drug available as eye drops to lower intraocular pressure (IOP); prostaglandin analogues, beta-blockers (beta receptor antagonists), carbonic anhydrase inhibitors, sympathomimetics (alpha receptor agonists), and miotics (cholinergic agonists).
Tablets of the oral carbonic anhydrase inhibitor acetazolamide are only rarely used to treat COAG. For more information on specific classes of pharmacological treatment see the introduction of chapter 7.
Laser treatment
The laser treatments under consideration in this guideline are argon laser trabeculoplasty (ALT) and selective laser trabeculoplasty (SLT).
Argon laser trabeculoplasty is an outpatient procedure. A contact lens is placed on the eye to focus an aiming beam onto the trabecular meshwork (TM) and half of the TM is treated (180 degrees) at one sitting. ALT is thought to work by activating cells called trabeculocytes and thus improving TM function. It may take up to six weeks for treatment to have the full effect and after this, if further IOP lowering is needed, the second 180 degrees of the TM is treated. Re-treatments in the same area can cause scarring of the TM and raised IOP.
Selective laser trabeculoplasty is similar to ALT but uses a different laser with a discharge of a very short duration. The spot size of the laser beam is much larger than that used for ALT so accurate identification of the TM is not as critical and the procedure is technically simpler. The mechanism of action is thought to be the same as ALT but re-treatments are said to be less likely to cause raised IOP because there is less photocoagulative damage to adjacent tissue.
Surgical treatment
The surgical treatments are classified as penetrating and non-penetrating surgery. In this guideline the penetrating surgical procedure under consideration is trabeculectomy, and the non-penetrating surgical procedures are deep sclerectomy and viscocanalostomy.
During trabeculectomy a flap of conjunctiva is dissected under the upper eyelid and a partial thickness flap of sclera is raised. A block of tissue is excised from the inner sclera exposing the iris beneath and a portion of iris is removed with the scleral flap and conjunctiva sutured back in place. Fluid from within the eye cavity filters around the edges of the scleral flap forming a fluid lake or ‘bleb’ under the conjunctiva below the upper eye lid from where it is absorbed by blood vessels of the sclera and conjunctiva into the bloodstream.
Deep sclerectomy is a variant of trabeculectomy. Instead of removing a piece of the iris and inner sclera, only a thin strip of inner sclera overlying Schlemm’s canal is removed. Fluid from the exposed canal filters slowly around the loosely applied scleral flap and a bleb is not formed.
Viscocanalostomy is a variant of deep sclerectomy. After Schlemm’s canal is deroofed it is cannulated and viscoelastic solution injected to break open the inner wall to allow easier egress of fluid from the TM into Schlemm’s canal over a larger circumference than just the area beneath the scleral flap.
8.2. Matrix of treatments considered in our clinical questions
We searched for RCT evidence comparing the effectiveness of different interventions (pharmacological, laser or surgical) for the treatment of COAG with a minimum follow up of 6 months. Below is a matrix showing where evidence was identified. A box filled with Yes represents where evidence was found and is reviewed in this chapter. A box filled with No represents where no evidence was found. In this case, no section on this comparison is included in the chapter. A box crossed out represents where the comparison was not considered for review.
Most studies relating to pharmacological treatment included patients with OHT and COAG. It was not possible to separate out the effect sizes for these populations. Therefore, we used the same evidence to assess the IOP lowering effects of pharmacological treatment relating to patients with COAG as we used for patients with OHT (Chapter 7).
Data is also presented on adverse events related to topical medications at the end of the section on pharmacological treatments (see section 8.4)
8.3. Pharmacological treatment for COAG
8.3.1. Beta-blockers versus no treatment
See Evidence Table 4, Appendix D, Forest Plots in to , Appendix E and Economic Model in Appendix F – 1.3
8.3.1.2. Economic evidence
No studies were identified. We conducted original modelling to compare various strategies for the first-choice treatment of COAG patients, including beta-blockers and no treatment. This was based on clinical evidence (see 8.3.1.1). See Appendix F – 1.3 for methods and results.
Table 8-81Beta-blockers vs. no treatment - Economic study characteristics
View in own window
Table 8-82Beta-blockers vs. no treatment- Economic summary of findings
View in own window
Study | Incremental cost (£) | Incremental effects | ICER (£/QALY) | Uncertainty |
---|
NCC-AC model | cost saving | 0.079 QALY | cost saving (a) | 95% CI: cost saving – £9,461/QALY Not sensitive to the cost of preservative-free preparations. Not sensitive to the stage of COAG. |
- a
Prostaglandin analogues are more cost-effective for this group (see ). This comparison refers to those patients for whom Prostaglandin analogues are contraindicated.
8.3.1.3. Patient views evidence
No studies were identified.
8.3.1.4. Evidence statements on beta-blockers vs. no treatment
View in own window
Clinical | There is no statistically significant difference in the number of patients with visual field progression at 2 to 6 years follow up. (LOW QUALITY) Beta-blockers are more effective than no treatment in reducing IOP from baseline at 2 to 6 years follow up. However, there is significant unexplained statistical heterogeneity within the results. This evidence relates to patients with ocular hypertension. (VERY LOW QUALITY) There is no statistically significant difference in the number of patients with an uncontrolled intraocular pressure of over 30mmHg at 2 to 10 years follow up. This evidence relates to patients with ocular hypertension. (LOW QUALITY) There were no studies which reported the number of patients with an acceptable IOP. There is no statistically significant difference in the number of patients experiencing a respiratory or cardiovascular adverse event at 5 years follow up. (LOW QUALITY) |
Economic | Beta-blockers are more cost-effective than no treatment for any stage of COAG. This evidence has minor limitations and direct applicability. |
8.3.2. Timolol at 0.5% concentration versus timolol at 0.25% concentration
See Evidence Tables 5 and 24, Appendix D and Forest Plot in , Appendix E
8.3.2.1. Clinical evidence
Table 8-83Timolol 0.5% vs. timolol 0.25% - Clinical study characteristics
View in own window
- a
- b
Not clear who was masked to treatment.
- c
There were too few patients in the study to show a clear estimate of effect.
8.3.2.2. Economic evidence
We found a cost-effectiveness study comparing two different concentrations of Timolol and sympathomimetics. We report the results of the comparison between Timolol 0.5% and Timolol 0.25% in this section, while the comparison between sympathomimetics and beta-blockers is reported in another section (8.3.9.2). See economic evidence table in Appendix D for details.
Table 8-85Timolol 0.5% vs. timolol 0.25% - Economic study characteristics
View in own window
Study | Limitations | Applicability | Other Comments |
---|
Cottle198827 | Serious (a,b) | Directly applicable | In order for the study to be applicable, Canadian costs were modified using figures from the BNF54. |
- a
- b
The same eye could be included in more than one group when the treatment was change.
Table 8-86Timolol 0.5% vs. timolol 0.25% - Economic summary of findings
View in own window
Study | Incremental cost (£) | Incremental effects | ICER | Uncertainty |
---|
Cottle198827 | Cost saving | More effective in terms of IOP control (a,b) and fewer severe adverse events (a) | Timolol 0.5% is dominant | Not reported |
- a
Not statistically significant.
- b
8.3.2.3. Patient views evidence
No studies were identified.
8.3.2.4. Evidence statements - Timolol 0.5% vs. timolol 0.25%
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. The effectiveness of Timolol 0.5% and 0.25% at reducing IOP from baseline are similar when assessed at 12 months follow-up (results for right and left eyes inconsistent but confidence intervals overlap. There is a weak suggestion of a greater effect with the higher concentration) (LOW QUALITY). There were no studies which reported the number of patients with an acceptable IOP. There were no studies which reported adverse events. |
Economic | Timolol 0.5% is less costly than Timolol 0.25% and more effective at reducing IOP without causing adverse events although this is not significant. This evidence has direct applicability but severe limitations due to the small sample size and the cross over between interventions. |
8.3.3. Prostaglandin analogues versus no treatment
See Economic Model in Appendix F – 1.3
8.3.3.1. Clinical evidence
No studies were identified.
8.3.3.2. Economic evidence
No studies were identified. We constructed an original model to compare various strategies for the first-choice treatment of COAG patients, including prostaglandin analogues and no treatment. This was based on the clinical evidence comparing beta-blockers to no treatment (see 8.3.1.1) and prostaglandin analogues to beta-blockers (see 8.3.4.1). See Appendix F – 1.3 for methods and results.
Table 8-87Prostaglandin analogues vs. no treatment - Economic study characteristics
View in own window
Table 8-88Prostaglandin analogues vs. no treatment - Economic summary of findings
View in own window
8.3.3.3. Patient views evidence
No studies were identified.
8.3.3.4. Evidence statements - Prostaglandin analogues vs. no treatment
View in own window
Clinical | No studies meeting the inclusion criteria were identified which compared prostaglandin analogues to no treatment. |
Economic | Prostaglandin analogues are more cost-effective than no treatment for any stage of COAG. This evidence has minor limitations and direct applicability. |
8.3.4. Prostaglandin analogues versus beta-blockers
See Evidence Tables 6 and 23, Appendix D, Forest Plots in to , Appendix E and Economic Model in Appendix F – 1.3
8.3.4.1. Clinical evidence
Table 8-89Prostaglandin analogues vs. beta-blockers - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline (follow up 6 to 36 months)4,17,44,47, 62,93,95,110,116,150, 156,158 | 12 | RCT | No serious limitations | Serious inconsistency (a) | No serious indirectness | No serious imprecision |
Number of patients with an acceptable IOP (follow up 6 to 12 months)4,44,47,62, 93,110,116 | 7 | RCT | No serious limitations | Serious inconsistency (a) | No serious indirectness | No serious imprecision |
Number of patients experiencing a respiratory adverse event (follow up 6 months)4,116 | 2 | RCT | No serious limitations | No serious inconsistency | No serious indirectness | Serious imprecision (b) |
Number of patients experiencing a cardiovascular adverse event (follow up 6 to 12 months)4,17,110,116,158 | 5 | RCT | No serious limitations | No serious inconsistency | No serious indirectness | Serious imprecision (b) |
Number of patients experiencing an allergic reaction (follow up 6 months)4,158 | 2 | RCT | No serious limitations | No serious inconsistency | No serious indirectness | Serious imprecision (b) |
Number of patients with hyperaemia (follow up 6 to 12 months)17,44,47,62,93,95,110,116,156,158 | 10 | RCT | No serious limitations | No serious inconsistency | No serious indirectness | No serious imprecision |
- c
Significant heterogeneity found in overall result. No specific cause for heterogeneity identified.
- d
The confidence intervals are wide making the estimate of harm uncertain.
Table 8-90Prostaglandin analogues vs. beta-blockers - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Mean change in IOP from baseline | 1342 | 1333 | not applicable | MD −1.32 (−1.79 to −0.84) | Moderate |
Number of patients with an acceptable IOP | 546/971 (56.2%) | 376/953 (39.5%) | RR 1.54 (1.21 to 1.96) | 213 more per 1000 (from 83 more to 379 more) | Moderate |
Number of patients experiencing a respiratory adverse event | 25/330 (7.6%) | 24/233 (10.3%) | RR 0.59 (0.35 to 1) | 42 fewer per 1000 (from 67 fewer to 0 more) | Moderate |
Number of patients experiencing a cardiovascular adverse event | 99/997 (9.9%) | 90/713 (12.6%) | RR 0.87 (0.67 to 1.13) | 16 fewer per 1000 (from 42 fewer to 16 more) | Moderate |
Number of patients experiencing an allergic reaction | 7/332 (2.1%) | 3/229 (1.3%) | RR 1.25 (0.31 to 5.09) | 3 more per 1000 (from 9 fewer to 53 more) | Moderate |
Number of patients with hyperaemia | 582/1778 (32.7%) | 108/1343 (8%) | RR 3.58 (2.97 to 4.32) | 206 more per 1000 (from 158 more to 266 more) | High |
8.3.4.2. Economic evidence
We found a cost-utility analysis82 comparing prostaglandin analogues to beta-blockers in a Markov Model. See economic evidence table in Appendix D for details.
We also found six economic studies10,31,48,54,125,126 comparing beta-blockers to prostaglandin analogues in a mixed population of OHT and COAG patients. Since they had more limitations and less applicability compared to other evidence available (Le Pen et al (2005)82 and NCC-AC economic model), they were not included in the GRADE tables. However, a description is reported in the economic evidence table in Appendix D.
We constructed an original model to compare various strategies for the first-choice treatment of COAG patients, including prostaglandin analogues and beta-blockers. This was based on the clinical evidence comparing prostaglandin analogues to beta-blockers (see 8.3.4.1). See Appendix F – 1.3 for methods and results.
Table 8-91Prostaglandin analogues vs. beta-blockers - Economic study characteristics
View in own window
Study | Limitations | Applicability | Other Comments |
---|
Le Pen 2005 | Serious limitations (a, b, c) | Partially applicable (d) | |
NCC-AC model | Minor limitations | Directly applicable | |
- a
- b
Clinical outcomes were not derived from a systematic search.
- c
Possible underestimation in the utilisation of ophthalmologic resources.
- d
Patients had advanced COAG. Discount of costs was 5%
Table 8-92Prostaglandin analogues vs. beta-blockers - Economic summary of findings
View in own window
8.3.4.3. Patient views evidence
One study reporting the results of a validated questionnaire found that patient satisfaction scores for eye appearance are significantly more favourable for beta-blockers compared to prostaglandin analogues but there is no statistically significant difference in patient scores for convenience of use.
8.3.4.4. Evidence statements on prostaglandin analogues vs. beta-blockers
View in own window
Clinical | There were no studies which reported visual field progression. Prostaglandin analogues are more effective than beta-blockers in reducing IOP from baseline at 6 to 36 months follow up, but the effect size is too small to be clinically effective. (MODERATE QUALITY) Prostaglandin analogues are more effective than beta-blockers in increasing the number of patients with an acceptable IOP at 6 to 12 months follow up. (MODERATE QUALITY) Significantly more patients using beta-blockers than prostaglandin analogues experienced a respiratory adverse event at 6 months follow up. (MODERATE QUALITY) There was no statistically significant difference in patients experiencing cardiovascular adverse events or an allergic reaction at 6 to 12 months follow up. (MODERATE QUALITY) Significantly more patients using prostaglandin analogues than beta-blockers experienced hyperaemia at 6 to 12 months follow up. (HIGH QUALITY) |
Economic | Prostaglandin analogues are more cost-effective than beta-blockers for any stage of COAG. This evidence has minor limitations and direct applicability. |
8.3.5. Prostaglandin analogues versus carbonic anhydrase inhibitors
See Evidence Table 23, Appendix D
8.3.5.1. Clinical evidence
No studies were identified.
8.3.5.2. Economic evidence
No studies were identified.
8.3.5.3. Patient views evidence
One study reporting the results of a validated questionnaire found no statistically significant differences between patient satisfaction scores for eye appearance and convenience of use for prostaglandin analogues compared to carbonic anhydrase inhibitors.
8.3.5.4. Evidence statements on prostaglandin analogues vs. carbonic anhydrase inhibitors
View in own window
Clinical | No studies meeting the inclusion criteria were identified which compared prostaglandin analogues to carbonic anhydrase inhibitors. |
Economic | No studies meeting the inclusion criteria were identified which compared prostaglandin analogues to carbonic anhydrase inhibitors. |
8.3.6. Prostaglandin analogues versus sympathomimetics
See Evidence Tables 7 and 23, Appendix D and Forest Plots in to , Appendix E
8.3.6.1. Clinical evidence
Table 8-93Prostaglandin analogues vs. sympathomimetics - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline (6 to 12 months follow up)18,70 | 2 | RCT | Serious limitations (a,b) | Serious inconsistency (c) | No serious indirectness | No serious imprecision |
Number of patients with an acceptable IOP | 0 | | | | | |
Number of patients experiencing an allergic reaction (follow up mean 6 months)70 | 1 | RCT | Serious limitations (d) | No serious inconsistency | No serious indirectness | No serious imprecision |
Number of patients with hyperaemia (follow up 6 months)70 | 1 | RCT | Serious limitations (d) | No serious inconsistency | No serious indirectness | No serious imprecision |
- a
- b
Patients were not masked to treatment although observers were.
- c
Some heterogeneity in the result with one study showing a greater than 2mmHg difference in mean change in IOP from baseline with prostaglandins and the other showing less than 2mmHg. This could be due to the different follow up periods (one study - 12 months, the other - 6 months).
- d
Table 8-94Prostaglandin analogues vs. sympathomimetics - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Mean change in IOP from baseline | 337 | 343 | not applicable | MD −2.22 (−2.91 to −1.54) | Low |
Number of patients experiencing an allergic reaction | 0/187 (0%) | 16/188 (8.5%) | RR 0.03 (0 to 0.5) | 82 fewer per 1000 (from 42 fewer to 85 fewer) | Moderate |
Number of patients with hyperaemia (follow up 6 months) | 11/187 (5.9%) | 11/188 (5.9%) | RR 1.01 (0.45 to 2.26) | 1 more per 1000 (from 32 fewer to 74 more) | Moderate |
8.3.6.2. Economic evidence
No studies were identified.
8.3.6.3. Patient views evidence
One study reporting the results of a validated questionnaire found that patient satisfaction scores for convenience of use significantly favour prostaglandin analogues compared to sympathomimetics but there is no statistically significant difference in patient scores for eye appearance.
8.3.6.4. Evidence statements on prostaglandin analogues vs. sympathomimetics
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. Prostaglandin analogues are more effective than sympathomimetics in reducing IOP from baseline at 6 to 12 months follow up. (LOW QUALITY) There were no studies which reported the number of patients with an acceptable IOP. Significantly more allergic reactions were experienced by patients using sympathomimetics compared to prostaglandin analogues at 6 months mean follow up. No patient using prostaglandin analogues experienced an allergic reaction. (MODERATE QUALITY) There was no statistically significant difference in patients with hyperaemia at 6 months (MODERATE QUALITY) |
Economic | No studies meeting the inclusion criteria were identified which compared prostaglandin analogues to sympathomimetics. |
8.3.7. Carbonic anhydrase inhibitors versus no treatment
See Evidence Table 8, Appendix D and Forest Plots in ,, , Appendix E
8.3.7.2. Economic evidence
No studies were identified.
8.3.7.3. Patient views evidence
No studies were identified.
8.3.7.4. Evidence statements on carbonic anhydrase inhibitors vs. no treatment
View in own window
Clinical | There is no statistically significant difference between carbonic anhydrase inhibitors and no treatment in the number of patients converting to COAG at 5 years follow up. (MODERATE QUALITY) There is no statistically significant difference between carbonic anhydrase inhibitors and no treatment in the number of patients with visual field progression at 5 years follow up. (MODERATE QUALITY) There were no studies which reported mean change in IOP from baseline expressed as an absolute value with standard deviation. Carbonic anhydrase inhibitors are more effective than no treatment in reducing the number of patients experiencing an IOP increase to in excess of 35mmHg at 5 years follow up. (HIGH QUALITY) There were no studies which reported the number of patients with an acceptable IOP. There were no studies which reported adverse events. |
Economic | No studies meeting the inclusion criteria were identified which compared carbonic anhydrase inhibitors to no treatment. |
8.3.8. Carbonic anhydrase inhibitors versus beta-blockers
See Evidence Tables 9 and 23, Appendix D and Forest Plot in , Appendix E
8.3.8.1. Clinical evidence
Table 8-95Carbonic anhydrase inhibitors vs. beta-blockers - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline (follow up 12–18 months)92,145 | 2 | RCT | Very serious limitations (a,b,c) | No serious inconsistency | No serious indirectness | No serious imprecision |
Number of patients with an acceptable IOP | 0 | | | | | |
Number of patients with hyperaemia (follow up 18 months)92 | 1 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | No serious imprecision |
- a
- b
Not reported whether the clinicians and observers were masked to treatment.
- c
Outcomes not reported properly. One study92 does not report the standard deviations associated with the mean reductions, nor the IOP at the end of the study.
Table 8-96Carbonic anhydrase inhibitors vs. beta-blockers - Clinical summary of findings
View in own window
- a
Not enough data provided to calculate the pooled weighted mean difference. Beta-blockers were significantly better than carbonic anhydrase inhibitors in both studies. In one92 the difference was 2mmHg (confidence intervals not available), in the other 1.3mmHg (0.38, 2.22)145.
- b
An absolute effect calculation is not possible as there are no events in the control arm of the study.
8.3.8.2. Economic evidence
No studies were identified.
8.3.8.3. Patient views evidence
One study reporting the results of a validated questionnaire found that patient satisfaction scores for eye appearance significantly favour beta-blockers compared to carbonic anhydrase inhibitors but there is no statistically significant difference in patient scores for convenience of use.
8.3.8.4. Evidence statements - Carbonic anhydrase inhibitors vs. beta-blockers
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. Carbonic anhydrase inhibitors are less effective than beta-blockers in reducing IOP from baseline at 12 to 18 months follow up, but the effect size maybe too small to be clinically significant. (LOW QUALITY) There were no studies which reported the number of patients with an acceptable IOP. There is no statistically significant difference between carbonic anhydrase inhibitors and beta-blockers in the number of patients experiencing hyperaemia at 18 months follow up. (LOW QUALITY) |
Economic | No studies meeting the inclusion criteria were identified which compared carbonic anhydrase inhibitors to beta-blockers. |
8.3.9. Sympathomimetics versus beta-blockers
See Evidence Tables 10, 23 and 24, Appendix D and Forest Plots in to , Appendix E
8.3.9.1. Clinical evidence
Table 8-97Sympathomimetics vs. beta-blockers - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression (follow up 12 months)83,133 | 2 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | Serious imprecision (b) |
Mean change in IOP from baseline (follow up 12 months)152 | 1 | RCT | Very serious limitations (c,d) | No serious inconsistency | No serious indirectness | No serious imprecision |
Number of patients with an acceptable IOP | 0 | | | | | |
Number of patients experiencing an allergic reaction (follow up 12 months)133 | 1 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | No serious imprecision |
Number of patients experiencing fatigue/ drowsiness (follow up 12 months)133 | 1 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | No serious imprecision |
- a
The reporting of the methods within the studies was poor and the studies were not placebo controlled.
- b
The wide confidence intervals make the estimate of effect imprecise
- c
- d
Neither patients nor observers were masked to treatment.
Table 8-98Sympathomimetics vs. beta-blockers - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Visual field progression | 22/357 (6.2%) | 29/294 (9.9%) | RR 0.92 (0.56 to 1.52) | 8 fewer per 1000 (from 44 fewer to 51 more) | Low |
Mean change in IOP from baseline | 22 | 22 | not applicable | MD −0.26 (−0.65, 0.13) | Low |
Number of patients experiencing an allergic reaction | 20/221 (9%) | 0/222 (0%) | RR 41.18 (2.18 to 676.76) | not estimable (a) | Moderate |
Number of patients experiencing fatigue/ drowsiness | 44/221 (19.9%) | 38/222 (17.1%) | RR 1.16 (0.79 to 1.72) | 27 more per 1000 (from 36 fewer to 123 more) | Moderate |
- a
An absolute effect calculation is not possible as there are no events in the control arm of the study.
8.3.9.2. Economic evidence
We identified a cost-effectiveness study where sympathomimetics were compared to beta-blockers. See economic evidence table in Appendix D for details.
Table 8-99Sympathomimetics vs. beta-blockers - Economic study characteristics
View in own window
Study | Limitations | Applicability | Other Comments |
---|
Cottle199827 | Serious limitations (a, b) | Directly applicable | In order for the study to be applicable, Canadian costs were modified using figures from the BNF. |
- a
- b
The same eye could be included in more than one group when the treatment was change.
Table 8-100Sympathomimetics vs. beta-blockers - Economic summary of findings
View in own window
Study | Incremental cost (£)per patient per year | Incremental effects
(a) | ICER | Uncertainty |
---|
Cottle199827 | £10 | 10% (b) | £100/patient with controlled IOP and no adverse event. | Not reported |
- a
Additional patients whose IOP is controlled with no severe adverse events
- b
Not statistically significant
8.3.9.3. Patient views evidence
One study reporting the results of a validated questionnaire found that patient satisfaction scores for convenience of use significantly favour beta-blockers compared to sympathomimetics but there is no statistically significant difference in patient scores for eye appearance.
8.3.9.4. Evidence statements - Sympathomimetics vs. beta-blockers
View in own window
Clinical | There is no statistically significant difference between sympathomimetics and beta-blockers in the number of people with visual field progression at 12 months follow up. (LOW QUALITY) There is no statistically significant difference between sympathomimetics and beta-blockers in reducing IOP from baseline at 12 months follow up. (LOW QUALITY) There were no studies which reported the number of patients with an acceptable IOP. Significantly more allergic reactions were experienced by patients using sympathomimetics than beta-blockers at 12 months follow up. No patient using beta-blockers experienced an allergic reaction. (MODERATE QUALITY) There is no statistically significant difference between sympathomimetics and beta-blockers in the number of patients experiencing fatigue or drowsiness at 12 months follow up. (MODERATE QUALITY) |
Economic | Sympathomimetics are more costly than beta-blockers but more effective at controlling IOP without causing adverse events, although this is not significant. However due to the small sample size, the cross over between interventions, and the contradiction with the clinical evidence, the findings of this study were deemed unreliable. |
8.3.10. Miotics versus beta-blockers
See Evidence Table 11, Appendix D
8.3.10.1. Clinical evidence
Table 8-101Miotics vs. beta-blockers - Clinical study characteristics
View in own window
- a
- b
The studies do not provide standard deviations for IOP change from baseline and although visual field testing results are reported they are not valid as miotics constrict the pupil..
- c
One study141 was very old.
Table 8-102Miotics vs. beta-blockers - Clinical summary of findings
View in own window
- a
Unable to provide a pooled estimate. The mean change in IOP from baseline between arms is similar suggesting no difference between miotics and beta-blockers.
8.3.10.2. Economic evidence
We found a cost-effectiveness study comparing beta-blockers, sympathomimetics and miotics. We report the results of the comparison between beta-blockers and miotics in this section, while the comparison between sympathomimetics and beta-blockers is reported in another section (8.3.9.2). See economic evidence table in Appendix D for details.
Table 8-103Miotics vs. beta-blockers - Economic study characteristics
View in own window
Study | Limitations | Applicability | Other Comments |
---|
Cottle199827 | Serious limitations (a,b) | Directly applicable | In order for the study to be applicable, Canadian costs were modified using figures from the BNF54. |
- a
- b
The same eye could be included in more than one group when the treatment was changed.
Table 8-104Miotics vs. beta-blockers - Economic summary of findings
View in own window
Study | Incremental cost (£) | Incremental effects | ICER | Uncertainty |
---|
Cottle199827 | Cost saving | More effective in terms of IOP control (a,b) but more severe adverse events (a) | Pilocarpine 1.0% is dominant | Not reported |
8.3.10.3. Patient views evidence
No studies were identified.
8.3.10.4. Evidence statements - Miotics vs. beta-blockers
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. There is no statistically significant difference between miotics and beta- blockers in reducing IOP from baseline at 17 to 24 months follow up. (LOW QUALITY) There were no studies which reported the number of patients with an acceptable IOP. There were no studies which reported adverse events. |
Economic | Miotics are less costly than beta-blockers and more effective at reducing IOP. However they could cause more adverse events although this finding is not statistically significant. Due to the small sample size and the cross over between interventions, the findings of this study were deemed unreliable. |
8.3.11. Fixed combination of carbonic anhydrase inhibitors plus beta-blockers versus prostaglandin analogues
See Evidence Table 12, Appendix D and Forest Plots in to , Appendix E
8.3.11.1. Clinical evidence
Table 8-105Fixed combination of carbonic anhydrase inhibitors + beta-blockers vs. prostaglandin analogues - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline (follow up 6 months)115 | 1 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | No serious imprecision |
Number of patients with an acceptable IOP | 0 | | | | | |
Number of patients experiencing a respiratory adverse event (follow up 6 months)115 | 1 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | Serious imprecision (c) |
Number of patients with hyperaemia (follow up 6 months)115 | 1 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | No serious imprecision |
- a
- b
Only assessors of IOP measurements were masked to treatment.
- c
The confidence intervals are broad making the effect size imprecise.
Table 8-106Fixed combination of carbonic anhydrase inhibitors + beta-blockers vs. prostaglandin analogues - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Mean change in IOP from baseline | 30 | 35 | not applicable | MD −0.30 (−1.32 to 0.72) | Moderate |
Number of patients experiencing a respiratory adverse event | 1/30 (3.3%) | 0/35 (0%) | RR 3.48 (0.15 to 82.48) | not estimable (a) | Low |
Number of patients with hyperaemia | 4/30 (13.3%) | 18/35 (51.4%) | RR 0.26 (0.1 to 0.68) | 380 fewer per 1000 (from 164 fewer to 463 fewer) | Moderate |
- a
An absolute effect calculation is not possible as there are no events in the control arm of the study.
8.3.11.2. Economic evidence
No studies were identified.
8.3.11.3. Patient views evidence
No studies were identified.
8.3.11.4. Evidence statements - Fixed combinations of carbonic anhydrase inhibitors + beta-blockers vs. prostaglandin analogues
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. There is no statistically significant difference between a fixed combination of carbonic anhydrase inhibitors + beta-blockers and prostaglandins alone in reducing IOP from baseline at 6 months follow up. (MODERATE QUALITY) There were no studies which reported the number of patients with an acceptable IOP. There is no statistically significant difference between a fixed combination of carbonic anhydrase inhibitors + beta-blockers and prostaglandins alone in the number of patients experiencing a respiratory adverse event at 6 months follow up. (LOW QUALITY) Prostaglandins result in significantly more patients with hyperaemia than a fixed combination carbonic anhydrase inhibitor + beta-blockers at 6 month follow up. (MODERATE QUALITY) |
Economic | No studies meeting the inclusion criteria were identified which compared fixed combinations of carbonic anhydrase inhibitors + beta-blockers to prostaglandin analogues alone. |
8.3.12. Fixed combination of prostaglandin analogues plus beta-blockers versus prostaglandin analogues
See Evidence Table 12, Appendix D and Forest Plots in to , Appendix E
8.3.12.1. Clinical evidence
Table 8-107Fixed combination of prostaglandin analogues + beta-blockers vs. prostaglandin analogues - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline (follow up 6 months)61,116 | 2 | RCT | Serious limitations (a,b) | serious (c) | No serious indirectness | Serious imprecision (d) |
Number of patients with an acceptable IOP of <18mmHg (follow up 6 months)61,116 | 2 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | Serious imprecision (d) |
Number of patients experiencing a respiratory adverse event (follow up 6 months)116 | 1 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | Serious imprecision (d) |
Number of patients experiencing a cardiovascular adverse event (follow up 6 months)116 | 1 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | Serious imprecision (d) |
Number of patients with hyperaemia (follow up 6 months)116 | 1 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | Serious imprecision (d) |
- a
- b
- c
There is significant unexplained statistical heterogeneity within the results. In one study the fixed combination is statistically more effective than prostaglandin analogues in reducing IOP[HIGGINBOTHAM2002A}, in the other there is no statistical difference and the point estimate favours prostaglandin analogues116.
- d
The confidence intervals are broad making the effect size imprecise.
Table 8-108Fixed combination of prostaglandin analogues + beta-blockers vs. prostaglandin analogues - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Mean change in IOP from baseline | 278 | 287 | not applicable | MD −0.34 (−1.81 to 1.13) | Very low |
Number of patients with an acceptable IOP of <18mmHg | 93/278 (33.5%) | 90/287 (31.4%) | RR 1.07 (0.84 to 1.36) | 22 more per 1000 (from 50 fewer to 113 more) | Low |
Number of patients experiencing a respiratory adverse event | 3/140 (2.1%) | 6/147 (4.1%) | RR 0.53 (0.13 to 2.06) | 19 fewer per 1000 (from 36 fewer to 43 more) | Low |
Number of patients experiencing a cardiovascular adverse event | 5/140 (3.6%) | 1/147 (0.7%) | RR 5.25 (0.62 to 44.38) | 30 more per 1000 (from 3 fewer to 304 more) | Low |
Number of patients with hyperaemia | 4/140 (2.9%) | 2/147 (1.4%) | RR 2.10 (0.39 to 11.28) | 15 more per 1000 (from 9 fewer to 144 more) | Low |
8.3.12.2. Economic evidence
No studies were identified.
8.3.12.3. Patient views evidence
No studies were identified.
8.3.12.4. Evidence statements - Fixed combination of prostaglandin analogues + beta-blockers vs. prostaglandin analogues
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. There is no statistically significant difference between a fixed combination of prostaglandin analogues + beta-blockers and prostaglandins alone in reducing IOP from baseline at 6 months follow up. (VERY LOW QUALITY) There is no statistically significant difference between a fixed combination of prostaglandin analogues + beta-blockers and prostaglandins alone in the number of patients with an acceptable IOP of <18mmHg at 6 months follow up. (LOW QUALITY) There is no statistically significant difference between a fixed combination of prostaglandin analogues + beta-blockers and prostaglandins alone in the number of patients experiencing a respiratory adverse event at 6 months follow up. (LOW QUALITY) There is no statistically significant difference between a fixed combination of prostaglandin analogues + beta-blockers and prostaglandins alone in the number of patients experiencing a cardiovascular adverse event at 6 months follow up. (LOW QUALITY) There is no statistically significant difference between a fixed combination of prostaglandin analogues + beta-blockers and prostaglandins alone in the number of patients experiencing hyperaemia at 6 months follow up. (LOW QUALITY) |
Economic | No studies meeting the inclusion criteria were identified which compared fixed combinations of prostaglandin analogues + beta-blockers to prostaglandin analogues alone. |
8.3.13. Fixed combination of prostaglandin analogues plus beta-blockers versus beta-blockers
See Evidence Table 12, Appendix D and Forest Plots in to , Appendix E
8.3.13.1. Clinical evidence
Table 8-109Fixed combination of prostaglandin analogues + beta-blockers vs. beta-blockers - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline (follow up 6 months)61,116 | 2 | RCT | Serious limitations (a,b) | Serious inconsistency (c,d) | No serious indirectness | Serious imprecision (e) |
Number of patients with an acceptable IOP of <18mmHg (follow up 6 months)
61,116 | 2 | RCT | Serious limitations (a,b) | Serious inconsistency (c) | No serious indirectness | Serious imprecision (e) |
Number of patients experiencing a respiratory adverse event (follow up 6 months)116 | 1 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | Serious imprecision (e) |
Number of patients experiencing a cardiovascular adverse event (follow up 6 months)116 | 1 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | Serious imprecision (e) |
Number of patients with hyperaemia (follow up 6 months)116 | 1 | RCT | Serious limitations (a,b) | No serious inconsistency | No serious indirectness | Serious imprecision (e) |
- a
- b
- c
There is significant unexplained statistical heterogeneity within the results.
- d
In one study the fixed combination is statistically and clinically more effective than beta-blockers in reducing IOP61, in the other there is no statistical difference116. The confidence intervals do not overlap.
- e
The confidence intervals are broad making the effect size imprecise.
Table 8-110Fixed combination of prostaglandin analogues + beta-blockers vs. beta-blockers - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Mean change in IOP from baseline | 278 | 289 | not applicable | MD −1.75 (−4.00 to 0.51) | Very low |
Number of patients with an acceptable IOP of <18mmHg | 93/278 (33.5%) | 48/289 (16.6%) | RR 2.03 (1.50 to 2.75) | 171 more per 1000 (from 83 more to 290 more) | Very low |
Number of patients experiencing a respiratory adverse event | 3/140 (2.1%) | 7/149 (4.7%) | RR 0.46 (0.12 to 1.73) | 25 fewer per 1000 (from 41 fewer to 34 more) | Low |
Number of patients experiencing a cardiovascular adverse event | 5/140 (3.6%) | 2/149 (1.3%) | RR 2.66 (0.52 to 13.49) | 22 more per 1000 (from 6 fewer to 162 more) | Low |
Number of patients with hyperaemia | 4/140 (2.9%) | 1/149 (0.7%) | RR 4.26 (0.48 to 37.63) | 23 more per 1000 (from 4 fewer to 256 more) | Low |
8.3.13.2. Economic evidence
No studies were identified.
8.3.13.3. Patient views evidence
No studies were identified.
8.3.13.4. Evidence statements on fixed combination of prostaglandin analogues + beta-blockers vs. beta-blockers
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. There is no statistically significant difference between a fixed combination of prostaglandin analogues + beta-blockers and beta-blockers alone in reducing IOP from baseline at 6 months follow up. (VERY LOW QUALITY) A fixed combination of prostaglandin analogues + beta-blockers is significantly more effective than beta-blockers alone in increasing the number of patients with an acceptable IOP of <18mmHg at 6 months follow up. (VERY LOW QUALITY) There is no statistically significant difference between a fixed combination of prostaglandin analogues + beta-blockers and beta-blockers alone in the number of patients experiencing a respiratory adverse event at 6 months follow up. (LOW QUALITY) There is no statistically significant difference between a fixed combination of prostaglandin analogues + beta-blockers and beta-blockers alone in the number of patients experiencing a cardiovascular adverse event at 6 months follow up. (LOW QUALITY) There is no statistically significant difference between a fixed combination of prostaglandin analogues + beta-blockers and beta-blockers alone in the number of patients experiencing hyperaemia at 6 months follow up. (LOW QUALITY) |
Economic | No studies meeting the inclusion criteria were identified which compared fixed combinations of prostaglandin analogues + beta-blockers to beta- blockers alone. |
8.3.14. Fixed combination of sympathomimetics plus beta-blockers versus beta-blockers
See Evidence Table 12, Appendix D and Forest Plots in to , Appendix E
8.3.14.1. Clinical evidence
Table 8-111Fixed combination of sympathomimetics + beta-blockers vs. beta-blockers - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline | 0 | | | | | |
Number of patients with an acceptable IOP of <17.5mmHg (mean follow up across all visits)135 | 1 | RCT | No serious limitations | No serious inconsistency | No serious indirectness | (a) |
Number of patients experiencing a respiratory adverse event (follow up 12 months)135 | 1 | RCT | No serious limitations | No serious inconsistency | No serious indirectness | None |
Number of patients experiencing a cardiovascular adverse event (follow up 12 months)135 | 1 | RCT | No serious limitations | No serious inconsistency | No serious indirectness | None |
- a
Outcomes are not reported properly. Mean diurnal IOP pressures are not reported. Standard deviations for each mean are not reported.
Table 8-112Fixed combination of sympathomimetics + beta-blockers vs. beta-blockers - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Number of patients with an acceptable IOP of <17.5mHg | 202/385 (52.5%) | 127/392 (32.4%) | RR 1.62 (1.36 to 1.92) | 201 more per 1000 (from 117 more to 298 more) | High |
Number of patients experiencing an allergic reaction | 100/385 (26%) | 47/392 (12%) | RR 2.17 (1.58 to 2.97) | 140 more per 1000 (from 70 more to 236 more) | High |
Number of patients with hyperaemia | 56/385 (14.5%) | 29/392 (7.4%) | RR 1.97 (1.28 to 3.01) | 72 more per 1000 (from 21 more to 149 more) | High |
8.3.14.2. Economic evidence
No studies were identified.
8.3.14.3. Patient views evidence
No studies were identified.
8.3.14.4. Evidence statements on fixed combination of sympathomimetics + beta-blockers vs. beta-blockers
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. There were no studies which reported mean change in IOP from baseline expressed as an absolute value with standard deviation. A fixed combination of sympathomimetics + beta-blockers is more effective than beta-blockers alone in increasing the number of patients with an acceptable IOP of <17.5mmHg at a mean follow up across all visits. (HIGH QUALITY) A fixed combination of sympathomimetics + beta-blockers resulted in significantly more people experiencing an allergic reaction than beta- blockers alone at 12 months follow up. (HIGH QUALITY) A fixed combination of sympathomimetics + beta-blockers resulted in significantly more patients experiencing hyperaemia than beta-blockers alone at 12 months follow up. (HIGH QUALITY) |
Economic | No studies meeting the inclusion criteria were identified which compared fixed combination of sympathomimetics + beta-blockers to beta-blockers alone. |
8.3.15. Separate combination of carbonic anhydrase inhibitors plus beta-blockers versus prostaglandin analogues
See Evidence Table 13, Appendix D and Forest Plots in to , Appendix E
8.3.15.1. Clinical evidence
Table 8-113Separate combination of carbonic anhydrase inhibitors + beta-blockers vs. prostaglandin analogues - Clinical study characteristics
View in own window
- a
- b
- c
Masked outcome assessment was not mentioned in one study117
- d
Serious statistical heterogeneity was observed between studies which may have been due to different dosages of CAI applied. One study121 applied CAI at a dosage of 3/day rather than the recommended 2/day for use alongside a beta-blocker.
- e
The confidence intervals are broad making the effect size imprecise.
Table 8-114Separate combination of carbonic anhydrase inhibitors + beta-blockers vs. prostaglandin analogues - Clinical summary of findings
View in own window
8.3.15.2. Economic evidence
No studies were identified.
8.3.15.3. Patient views evidence
No studies were identified.
8.3.15.4. Evidence statements - Separate combination of carbonic anhydrase inhibitors + beta-blockers vs. prostaglandin analogues
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. There is no statistically significant difference between a separate combination of carbonic anhydrase inhibitors + beta-blockers and prostaglandin analogues alone in reducing IOP from baseline at 6 months follow up. (LOW QUALITY) A separate combination of carbonic anhydrase inhibitors + beta-blockers is less effective than prostaglandin analogues alone in increasing the number of patients with an acceptable IOP of <21mmHg at 24 months follow up. (VERY LOW QUALITY) There were no studies which reported adverse events. |
Economic | No studies meeting the inclusion criteria were identified which compared separate combinations of carbonic anhydrase inhibitors plus beta-blockers to prostaglandin analogues alone. |
8.3.16. Separate combination of prostaglandin analogues plus beta-blockers versus prostaglandin analogues
See Evidence Tables 13 and 24, Appendix D and Forest Plots in to , Appendix E
8.3.16.1. Clinical evidence
Table 8-115Separate combination of prostaglandin analogues plus beta-blockers versus prostaglandin analogues - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline (follow up 6 months)13,91 | 2 | RCT | Very serious limitations (a,b,c) | No serious inconsistency | No serious indirectness | Serious imprecision (d) |
Number of patients with an acceptable IOP of approx <18mmHg (follow up 6 months)13 | 1 | RCT | Very serious limitations (b,c,e) | No serious inconsistency | No serious indirectness | None |
Number of patients experiencing a respiratory adverse event (follow up 6 months)13 | 1 | RCT | Very serious limitations (b,c,e) | No serious inconsistency | No serious indirectness | Serious imprecision (d) |
Number of patients with hyperaemia (follow up 6 months)13,91 | 2 | RCT | Very serious limitations (a,b,c) | No serious inconsistency | No serious indirectness | Serious imprecision (d) |
- a
Only one study reports the method of randomisation. This study has a 90% weighting on the estimate of effect.
- b
- c
Only observers were masked to treatment.
- d
The confidence intervals are broad making the effect size imprecise.
- e
Table 8-116Separate combination of prostaglandin analogues plus beta-blockers versus prostaglandin analogues - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Mean change in IOP from baseline | 79 | 81 | not applicable | MD −0.66 (−1.44 to 0.13) | Very low |
Number of patients with an acceptable IOP of approx <18mmHg | 30/45 (66.7%) | 32/46 (69.6%) | RR 0.96 (0.72 to 1.27) | 28 fewer per 1000 (from 195 fewer to 188 more) | Low |
Number of patients experiencing a respiratory adverse event | 1/49 (2%) | 0/50 (0%) | RR 3.06 (0.13 to 73.34) | not estimable (a) | Very low |
Number of patients with hyperaemia | 27/79 (34.2%) | 18/81 (22.2%) | RR 1.54 (0.98 to 2.44) | 120 more per 1000 (from 4 fewer to 320 more) | Very low |
- a
An absolute effect calculation is not possible as there are no events in the control arm of the study.
8.3.16.2. Economic evidence
We found a cost-effectiveness analysis based on a retrospective cohort study143. Patients who failed treatment with beta-blockers were either treated with a prostaglandin analogue in monotherapy or this was added to the beta-blocker already prescribed. Two studies based on the same cohort study reported the cost-effectiveness analysis after one year125 and two year126 follow-up of patients treated with either beta-blockers, prostaglandin analogues or an unfixed combination of a prostaglandin analogue plus beta-blocker. The comparison of beta-blockers with the fixed combination is reported in 8.3.17.2. See economic evidence table in Appendix D for details of the studies.
Table 8-117Separate combination of prostaglandin analogues plus beta-blockers versus prostaglandin analogues - Economic study characteristics
View in own window
Study | Limitations | Applicability | Other Comments |
---|
Stewart2002143 | Serious limitations (a, b, c) | Partially applicable (d, e) | |
Rouland2003125 | Serious limitations (a, b) | Partially applicable (d, f) | |
Rouland2005126 | Serious limitations (a, b) | Partially applicable (d, f) | Same study as above but different outcomes reported. |
- a
- b
- c
- d
- e
Patients were previously prescribed a topical beta-blocker as monotherapy.
- f
Table 8-118Separate combination of prostaglandin analogues plus beta-blockers versus prostaglandin analogues - Economic summary of findings
View in own window
- a
Not statistically significant.
- b
Significance not reported.
8.3.16.3. Patient views evidence
No studies were identified
8.3.16.4. Evidence statements - Separate combination of prostaglandin analogues plus beta-blockers versus prostaglandin analogues
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. There is no statistically significant difference between a separate combination of prostaglandin analogues + beta-blockers and prostaglandin analogues alone in reducing IOP from baseline at 6 months follow up. (VERY LOW QUALITY) There is no statistically significant difference between a separate combination of prostaglandin analogues + beta-blockers and prostaglandin analogues alone in increasing the number of patients with an IOP of approx <18 mmHg at 6 months follow up. (LOW QUALITY) There is no statistically significant difference between a separate combination of prostaglandin analogues + beta-blockers and prostaglandin analogues alone in the number of patients experiencing a respiratory adverse event at 6 months follow up. (VERY LOW QUALITY) There is no statistically significant difference between a separate combination of prostaglandin analogues + beta-blockers and prostaglandin analogues alone in the number of patients experiencing hyperaemia at 6 months follow up. (VERY LOW QUALITY) |
Economic | Separate combinations of prostaglandin analogues plus beta-blockers are more effective (not statistically significant) but more costly than prostaglandin analogues alone. This evidence has serious limitations and partial applicability. |
8.3.17. Separate combination of prostaglandin analogues plus beta-blockers versus beta-blockers
See Evidence Table 13, Appendix D and Forest Plots in to , Appendix E
8.3.17.1. Clinical evidence
Table 8-119Separate combination of prostaglandin analogues + beta-blockers vs. beta-blockers - Clinical study characteristics
View in own window
- a
Outcomes not reported properly. Mean diurnal IOP pressures are not reported. Standard deviations for each mean are not reported.
- b
Only 77% of those randomised were included in the analysis.
Table 8-120Separate combination of prostaglandin analogues + beta-blockers vs. beta-blockers - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Number of patients with an acceptable IOP of approx <17mmHg | 55/114 (48.2%) | 11/112 (9.8%) | RR 4.91 (2.72 to 8.88) | 383 more per 1000 (from 169 more to 772 more) | High |
Number of patients with hyperaemia | 52/145 (35.9%) | 13/145 (9%) | RR 4.00 (2.28 to 7.02) | 270 more per 1000 (from 115 more to 542 more) | Moderate |
8.3.17.2. Economic evidence
We found two studies based on the same cohort study reporting the cost-effectiveness analysis after one year125 and two year126 follow-up of patients treated with either beta-blockers, prostaglandin analogues or an unfixed combination of a prostaglandin analogue plus beta-blocker. The comparison of prostaglandin analogues with the fixed combination is reported in 8.3.16.2. See economic evidence table in Appendix D for details of the studies.
Table 8-121Separate combination of prostaglandin analogues + beta-blockers vs. beta-blockers - Economic study characteristics
View in own window
Study | Limitations | Applicability | Other Comments |
---|
Rouland2003125 | Serious limitations (a, b) | Partially applicable (c, d) | |
Rouland2005126 | Serious limitations (a, b) | Partially applicable (c, d) | Same study as above but different outcomes reported. |
Table 8-122Separate combination of prostaglandin analogues + beta-blockers vs. beta-blockers - Economic summary of findings
View in own window
- a
Significance not reported.
8.3.17.3. Patient views evidence
No studies were identified.
8.3.17.4. Evidence statements - Separate combination of prostaglandin analogues + beta-blockers vs. beta-blockers
View in own window
Clinical | There were no studies which reported the number of patients with visual field progression. There were no studies which reported mean change in IOP from baseline expressed as an absolute value with standard deviation. A separate combination of prostaglandin analogues + beta-blockers is more effective than beta-blockers alone in increasing the number of patients who reach an IOP of approx <17mmHg at 6 months follow up. (HIGH QUALITY) Significantly more patients using a fixed combination of prostaglandin analogues + beta-blockers compared to beta-blockers alone experienced hyperaemia at 6 months follow up. (MODERATE QUALITY) |
Economic | Separate combinations of prostaglandin analogues plus beta-blockers are more effective (significance not reported) but more costly than beta- blockers alone. This evidence has serious limitations and partial applicability. |
8.4. Adverse events associated with pharmacological treatments
Some important adverse events were not well reported in the randomised controlled trials. This is particularly the case for beta-blockers which have been associated, or an association has been suggested, with serious respiratory or cardiovascular adverse events109, a change in respiratory or cardiovascular function35,139, depression137 or falls and syncope46,103. Further evidence is reviewed here from comparative observational studies where patients had been using medications for a minimum of six months, the same time period used for the RCT reviews. A summary of the evidence identified from both RCTs and observational studies are included below.
See Evidence Table 14, Appendix D
Table 8-123Summary of adverse events evidence associated with topical medications
View in own window
Adverse event | Evidence from reviewed RCTs | Evidence from observational studies |
---|
Respiratory adverse events | Some evidence in studies of beta- blockers reviewed earlier in this chapter but these are mostly too small to show an effect. | Large observational study shows evidence of increased harm with beta-blockers |
Cardiovascular adverse events | Some evidence in studies to beta- blockers but these are mostly too small to show an effect. | No studies |
Change in respiratory or cardiovascular function | No studies | No studies |
Depression | No studies | Large observation study shows no difference between beta-blockers & other medications |
Syncope and falls | No studies | No studies |
8.4.1.1. Clinical evidence
Table 8-124Adverse events associated with topical medications - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
New prescription for reversible airways obstruction (follow up 6 months)74,75 | 1 | Observational study | No serious limitations | No serious inconsistency | No serious indirectness | None |
New prescription for reversible airways obstruction (follow up 12 months)74,75 | 1 | Observational study | No serious limitations | No serious inconsistency | No serious indirectness | None |
New prescription for reversible airways obstruction AND a new Read code for asthma or COPD (follow up 6 months)74,75 | 1 | Observational study | No serious limitations | No serious inconsistency | No serious indirectness | None |
New prescription for reversible airways obstruction AND a new Read code for asthma or COPD (follow up 12 months)74,75 | 1 | Observational study | No serious limitations | No serious inconsistency | No serious indirectness | None |
Number of patients taking at least 4 prescriptions of anti- depressants | 1 | Observational study | No serious limitations | No serious inconsistency | No serious indirectness | None |
Table 8-125Adverse events associated with topical medications - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
New prescription for reversible airways obstruction (follow up 6 months) | 49/2645 (1.9%) | 55/9094 (0.6%) | HR 2.79 (1.88 to 4.15) (a) | 11 more per 1000 (from 5 more to 19 more) | Low |
New prescription for reversible airways obstruction (follow up 12 months) | 81/2645 (3.1%) | 112/9094 (1.2%) | HR 2.29 (1.71 to 3.07) (a) | 15 more per 1000 (from 8 more to 24 more) | Low |
New prescription for reversible airways obstruction AND a new Read code for asthma or COPD (follow up 6 months) | 115/2645 (4.3%) | 172/9094 (1.9%) | HR 2.18 (1.71 to 2.79) (a) | 22 more per 1000 (from 13 more to 33 more) | Low |
New prescription for reversible airways obstruction AND a new Read code for asthma or COPD (follow up 12 months) | 191/2645 (7.2%) | 354/9094 (3.9%) | HR 1.77 (1.48 to 2.12) (a) | 29 more per 1000 (from 18 more to 42 more) | Low |
Number of patients taking at least 4 prescriptions of antidepressants | 715/5846 (12.2%) | 95/752 (12.6%) | OR 0.96 (0.77 to 1.21) | 5 fewer per 1000 (from 27 fewer to 23 more) | Low |
- a
Adjusted analysis used a proportional hazards model, corrected for age, sex, use of systemic beta-blockers, use of non-steroidal anti-inflammatory drugs, use of nitrates, smoking, season of presentation, and number of visits to general practitioners.
8.4.1.2. Economic evidence
No economic studies were identified which compared the cost implications of adverse events with different treatment. The cost of asthma was included in the NCC-AC model on treatment. It was estimated as £147 per year11. See Appendix F – 1.3 for details.
8.4.1.3. Evidence Statements – adverse events
View in own window
Clinical | Significantly more patients using beta-blockers compared to those not using beta-blockers required a new prescription for reversible airways obstruction and/or a new Read code for asthma or COPD. (LOW QUALITY) There is no statistically significant difference between beta-blockers and other medications in the number of patients who are prescribed anti-depressants. (LOW QUALITY) |
Economic | No economic studies were identified which compared the cost implications of adverse events with different treatment. The annual cost of asthma was estimated and used in the NCC-AC model on treatment (Appendix F – 1.3). |
8.5. Laser treatment for COAG
8.5.1. Selective laser trabeculoplasty versus argon laser trabeculoplasty
See Evidence Table 15, Appendix D and Forest Plots in to
8.5.1.1. Clinical evidence
Table 8-126Selective laser trabeculoplasty vs. argon laser trabeculoplasty - Clinical study characteristics
View in own window
- a
Studies are supplemented by data from the Cochrane systematic reviews Rolim 2007124.
- b
- c
- d
All patients were maintained on current IOP lowering medications throughout study and some patients previously received ALT treatment.
Table 8-127Selective laser trabeculoplasty vs. argon laser trabeculoplasty - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Mean change in IOP from baseline | 89 | 87 | not applicable | MD 0.18 (−1.45 to 1.81) | Moderate |
Number of patients with an unacceptable IOP | 35/89 (39.3%) | 27/87 (31%) | 1.27 (0.84 to 1.90) | 84 more per 1000 (from 50 fewer to 249 more) | Low |
Complications: PAS formation | 1/89 (1.1%) | 1/87 (1.1%) | 0.98 (0.06 to 15.38 | 0 fewer per 1000 (from 10 fewer to 158 more) | Low |
8.5.1.2. Economic evidence
No studies were identified.
8.5.1.3. Patient views evidence
No studies were identified.
8.5.1.4. Evidence statements - Selective laser trabeculoplasty vs. argon laser trabeculoplasty
8.5.2. Laser trabeculoplasty versus pharmacological treatment
See Evidence Table 15, Appendix D and Forest Plot in
8.5.2.1. Clinical evidence
Table 8-128Laser trabeculoplasty vs. pharmacological treatment - Clinical study characteristics
View in own window
- a
- b
- c
One study104 included 51% OHT patients.
- d
- e
Although there was no statistical heterogeneity observed other differences between studies were noted in length of follow up, IOP failure criteria, laser modality, laser degrees of treatment, class of medications, mean baseline IOP and COAG population (previously untreated or treated). One study104 tested different in laser degrees of treatment against prostaglandin analogues. For the purposes of comparison the 360 degree was selected.
Table 8-129Laser trabeculoplasty vs. pharmacological treatment - Clinical summary of findings
View in own window
8.5.2.2. Economic evidence
No studies were identified.
8.5.2.3. Patient views evidence
No studies were identified.
8.5.2.4. Evidence statements - Laser trabeculoplasty vs. pharmacological treatment
View in own window
Clinical | There were no studies which reported number of patients with visual field progression. There were no studies which reported mean change in IOP from baseline expressed as an absolute value with standard deviation. There is no statistically significant difference between laser trabeculoplasty and pharmacological treatment in terms of number of patients with an unacceptable IOP at 2 to 48 months follow up. (VERY LOW QUALITY) There were no studies which reported complications lasting longer than 1 week. |
Economic | No studies meeting the inclusion criteria were identified which compared laser trabeculoplasty to pharmacological treatment. |
8.5.3. Laser trabeculoplasty plus pharmacological treatment versus pharmacological treatment
See Evidence Table 15, Appendix D and Forest Plot in
8.5.3.1. Clinical evidence
Table 8-130Laser trabeculoplasty + pharmacological treatment vs. pharmacological treatment- Clinical study characteristics
View in own window
- a
Studies are supplemented by data from the Cochrane systematic reviews Rolim 2007124.
- b
- c
I-squared value of 81% indicates high statistical heterogeneity which may have been due to the studies being from very different populations. One study102 is exclusively in Afro-Caribbean patients. Variations between studies are also noted in laser degrees of treatment and mean baseline IOP.
- d
Table 8-131Laser trabeculoplasty + pharmacological treatment vs. pharmacological treatment - Clinical summary of findings
View in own window
8.5.3.2. Economic evidence
No studies were identified.
8.5.3.3. Patient views evidence
No studies were identified.
8.5.3.4. Evidence statements - Laser trabeculoplasty + pharmacological treatment vs. pharmacological treatment
View in own window
Clinical | There were no studies which reported number of patients with visual field progression. There were no studies which reported mean change in IOP from baseline expressed as an absolute value with standard deviation. There is no statistically significant difference between laser trabeculoplasty + pharmacological treatment and pharmacological treatment alone in terms of number of patients with an unacceptable IOP at 12 months follow up. (VERY LOW QUALITY) There were no studies which reported complications lasting longer than 1 week. |
Economic | No studies meeting the inclusion criteria were identified which compared laser trabeculoplasty + pharmacological treatment to pharmacological treatment. |
8.5.4. Laser trabeculoplasty versus trabeculectomy
See Evidence Table 15, Appendix D and Forest Plot in
8.5.4.1. Clinical evidence
Table 8-132Laser trabeculoplasty vs. trabeculectomy - Clinical study characteristics
View in own window
- a
Studies are supplemented by data from the Cochrane systematic reviews Rolim 2007124.
- b
One study98 does not report masking of outcome assessment.
- c
Although there is no statistical heterogeneity observed at 0 – 6 months follow up, the I-squared value is high (51%) for 3 – 24 months follow up.
- d
Differences between studies are noted in IOP failure criteria, laser degrees of treatment and mean baseline IOP.
Table 8-133Laser trabeculoplasty vs. trabeculectomy - Clinical summary of findings
View in own window
8.5.4.2. Economic evidence
No studies were identified.
8.5.4.3. Patient views evidence
No studies were identified
8.5.4.4. Evidence statements - Laser trabeculoplasty vs. trabeculectomy
8.6. Surgical treatment for COAG
8.6.1. Trabeculectomy versus pharmacological treatment
Evidence Table 16, Appendix D, Forest Plots in to and Economic Model in Appendix F - 1.3
8.6.1.1. Clinical evidence
Table 8-134Trabeculectomy vs. pharmacological treatment- Clinical study characteristics
View in own window
- a
- b
- c
- d
- e
Other differences in study populations are noted in baseline IOP, severity of COAG and race.
Table 8-135Trabeculectomy vs. pharmacological treatment - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Visual field progression | 47/98 (48%) | 52/97 (53.6%) | 0.81 (0.38 to 1.73) | 102 fewer per 1000 (from 332 fewer to 391 more) | Very Low |
Mean change in IOP from baseline (follow up 12 months) | 397 | 388 | not applicable | MD −4.92 (−6.93 to −2.91) | Low |
Mean change in IOP from baseline (follow up 1 to 5 years) | 326 | 285 | not applicable | MD −2.04 (−2.85 to −1.23) | Low |
Mean change in IOP from baseline (follow up >5 years) | 257 | 229 | not applicable | MD −2.15 (−3.10 to −1.19) | Low |
Number of patients with an unacceptable IOP | 7/46 (15.2%) | 17/53 (32.1%) | 0.47 (0.22 to 1.04) | 170 fewer per 1000 (from 250 fewer to 13 more) | Low |
Complications: Cataract formation | 57/403 (14.1%) | 24/406 (5.8%) | 2.45 (1.55 to 3.87 | 82 more per 1000 (from 32 more to 166 more) | Not estimable (a) |
- a
Figures taken from the systematic review15. Data not provided for individual studies consequently no forest plot is provided in this guideline’s appendices.
8.6.1.2. Economic evidence
We found a cost analysis comparing early trabeculectomy (within 4 weeks of diagnosis) to medical management. See economic evidence table in Appendix D for details.
We also constructed an original model to compare various strategies for the first-choice treatment of COAG patients, including trabeculectomy and pharmacological treatment with beta-blockers and prostaglandin analogues. This was based on clinical evidence comparing trabeculectomy to beta-blockers (see 8.6.1.1). See Appendix F – 1.3 for methods and results.
Table 8-136Trabeculectomy vs. pharmacological treatment - Economic study characteristics
View in own window
Study | Limitations | Applicability | Other Comments |
---|
Ainsworth19913(a) | Serious limitations (b) | Partially applicable (c) | Early trabeculectomy was compared to conventional management: up to a maximum of three different topical or systemic drugs and late trabeculectomy if medical therapy has failed. |
NCC-AC model | Minor limitations | Directly applicable | |
- a
- b
- c
Average length of stay after surgery was 7.6 days and therefore longer than the current average.
Table 8-137Trabeculectomy vs. pharmacological treatment - Economic summary of findings
View in own window
- a
In bilateral COAG patients.
8.6.1.3. Patient views evidence
No studies were identified.
8.6.1.4. Evidence statements - Trabeculectomy vs. pharmacological treatment
8.6.2. Trabeculectomy plus pharmacological augmentation versus trabeculectomy
Evidence Table 17, Appendix D and Forest Plots in to
8.6.2.1. Clinical evidence
Table 8-138Trabeculectomy + pharmacological augmentation vs. trabeculectomy - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline | 0 | | | | | |
Number of patients with an unacceptable IOP (follow up 12 months)
26,39,49,94,113,118,123, 147 | 8 | RCT (a) | Serious limitations (b) | No serious inconsistency | No serious indirectness | No serious imprecision Additional notes (d) |
Complications: Cataract Formation (follow up 9–18 months)26,39,49,88,94,118,123,147 | 8 | RCT (a) | Serious limitations (b) | No serious inconsistency | No serious indirectness | Serious imprecision (c) Additional notes (d) |
Complications: Persistent hypotony (follow up 9–18 months)26,39,49,88,94,118,147 | 7 | RCT (a) | Serious limitations (b) | No serious inconsistency | No serious indirectness | Serious imprecision (c) Additional notes (d) |
Complications: Wound leak (follow up 9–18 months)26,39,49,88,118,147 | 6 | RCT (a) | Serious limitations (b) | No serious inconsistency | No serious indirectness | Serious imprecision (c) Additional notes (d) |
Complications: Corneal epithelial defects (follow up 9–18 months)39,49,88,113, 147 | 5 | RCT (a) | Serious limitations (b) | No serious inconsistency | No serious indirectness | Serious imprecision (c) Additional notes (d) |
- a
Studies are supplemented by data from the Cochrane systematic reviews Wilkins 2005161 and Wormald 2001162.
- b
For the antimetabolite MMC: 3 studies do not report details of randomisation method26,123,147. 3 studies do not report details of allocation concealment94,118,147. 3 studies do not report masking of outcome assessment26,118,147. Only 2 studies were placebo controlled26,147. For the antimetabolite 5-FU: 2 studies do not report details of randomisation method39,113. 3 studies do not report details of allocation concealment, masking of outcome assessment and are not placebo controlled39,49,113. One study88
is a placebo controlled double blind design.
- c
Wide confidence intervals making estimate of effect uncertain.
- d
Although there is no statistical heterogeneity observed other differences between studies are noted in type of antimetabolite (MMC or 5-FU) used and dosage, delivery method of 5-FU (intraoperative or postoperative injections), IOP failure criteria, length of follow up, reporting of complications, proportion of patients with closed-angle glaucoma of <50%, mean baseline IOP and whether patients received previous laser treatment. One study39 is exclusively in Afro-Caribbean patients and one study123 is exclusively in patients from the Indian sub-continent.
Table 8-139Trabeculectomy + pharmacological augmentation vs. trabeculectomy - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Number of patients with an unacceptable IOP | 35/337 (10.4%) | 82/218 (37.6%) | 0.33 (0.23 to 0.47) | 252 fewer per 1000 (from 199 fewer to 290 fewer) | Moderate |
Complications: Cataract Formation | 56/335 (16.7%) | 19/210 (9.0%) | 1.61 (0.96 to 2.70) | 55 more per 1000 (from 4 fewer to 153 more) | Low |
Complications: Persistent hypotony | 12/169 (7.1%) | 3/155 (1.9%) | 2.60 (0.97 to 6.97) | 30 more per 1000 (from 1 fewer to 113 more) | Low |
Complications: Wound leak | 26/139 (18.7%) | 11/125 (8.8%) | 2.02 (1.06 to 3.84) | 90 more per 1000 (from 5 more to 250 more) | Low |
Complications: Corneal epithelial defects | 32/125 (25.6%) | 6/111 (5.4%) | 3.75 (1.76 to 7.99 | 149 more per 1000 (from 41 more to 337 more) | Low |
8.6.2.2. Economic evidence
No studies were identified.
8.6.2.3. Patient views evidence
No studies were identified.
8.6.2.4. Evidence statements - Trabeculectomy + pharmacological augmentation vs. trabeculectomy
8.6.3. Trabeculectomy plus antimetabolite drug MMC versus antimetabolite drug 5-FU
Evidence Table 18, Appendix D and Forest Plots in to
8.6.3.1. Clinical evidence
Table 8-140Trabeculectomy + antimetabolite drug MMC versus antimetabolite drug 5-FU - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline | 0 | | | | | |
Number of patients with an unacceptable IOP (follow up 12 months)138,165 | 2 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | Serious imprecision (b) Additional notes (c) |
Complications: Cataract Formation IOP (follow up 12 months)138 | 1 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | Serious imprecision (b) Additional notes (c) |
Complications: Persistent hypotony IOP (follow up 12 months)138,165 | 2 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | Serious imprecision (b) Additional notes (c) |
Complications: Wound leak IOP (follow up 12 months)138,165 | 2 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | Serious imprecision (b) Additional notes (c) |
Complications: Corneal epithelial defects IOP (follow up 12 months)165 | 1 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | Serious imprecision (b) Additional notes (c) |
- a
- b
Wide confidence intervals make estimate of effect uncertain.
- c
Table 8-141Trabeculectomy + antimetabolite drug MMC versus antimetabolite drug 5-FU - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Number of patients with an unacceptable IOP | 5/54 (9.3%) | 13/47 (27.7%) | 0.34 (0.13 to 0.88) | 183 fewer per 1000 (from 33 fewer to 241 fewer) | Low |
Complications: Cataract Formation | 3/44 (6.8%) | 3/37 (8.1%) | 0.84 (0.18 to 3.92) | 13 fewer per 1000 (from 66 fewer to 237 more) | Low |
Complications: Persistent hypotony | 2/54 (3.7%) | 3/47 (6.4%) | 0.63 (0.13 to 3.11) | 24 fewer per 1000 (from 56 fewer to 135 more) | Low |
Complications: Wound leak | 2/54 (3.7%) | 2/47 (4.3%) | 1.00 (0.17 to 5.77) | 0 fewer per 1000 (from 36 fewer to 205 more) | Low |
Complications: Corneal epithelial defects | 0/10 (0%) | 3/10 (30%) | 0.14 (0.01 to 2.45) | 258 fewer per 1000 (from 297 fewer to 435 more) | Low |
8.6.3.2. Economic evidence
No studies were identified.
8.6.3.3. Patient views evidence
No studies were identified.
8.6.3.4. Evidence statements - Trabeculectomy + antimetabolite drug MMC versus antimetabolite drug 5-FU
View in own window
Clinical | There were no studies which reported number of patients with visual field progression. There were no studies which reported mean change in IOP from baseline expressed as an absolute value with standard deviation. Trabeculectomy + antimetabolite drug MMC is more effective than antimetabolite drug 5-FU in reducing the number of patients with an unacceptable IOP at 12 months follow up. (LOW QUALITY) There is no statistically significant difference between trabeculectomy + antimetabolite drug MMC and antimetabolite drug 5-FU in cataract formation at 12 months follow up. (LOW QUALITY) There is no statistically significant difference between trabeculectomy + antimetabolite drug MMC and antimetabolite drug 5-FU in causing persistent hypotony at 12 months follow up. (LOW QUALITY) There is no statistically significant difference between trabeculectomy + antimetabolite drug MMC and antimetabolite drug 5-FU in causing wound leaks at 12 months follow up. (LOW QUALITY) There is no statistically significant difference between trabeculectomy + antimetabolite drug MMC and antimetabolite drug 5-FU in causing corneal epithelial defects at 12 months follow up. (LOW QUALITY) |
Economic | No studies meeting the inclusion criteria were identified which compared trabeculectomy + antimetabolite drug MMC to antimetabolite drug 5-FU. |
8.6.4. Viscocanalostomy versus deep sclerectomy
Evidence Table 19, Appendix D and Forest Plot in
8.6.4.1. Clinical evidence
Table 8-142Viscocanalostomy versus deep sclerectomy - Clinical study characteristics
View in own window
- a
- b
Confidence intervals are wide making estimate of effect uncertain.
Table 8-143Viscocanalostomy versus deep sclerectomy - Clinical summary of findings
View in own window
8.6.4.2. Economic evidence
No studies were identified.
8.6.4.3. Patient views evidence
No studies were identified.
8.6.4.4. Evidence statements - Viscocanalostomy versus deep sclerectomy
View in own window
Clinical | There were no studies which reported number of patients with visual field progression. There is no statistically significant difference between viscocanalostomy and deep sclerectomy in reducing IOP from baseline at 6 months follow up. (LOW QUALITY) There were no studies which reported number of patients with an unacceptable IOP. There were no studies which reported complications. |
Economic | No studies meeting the inclusion criteria were identified which compared viscocanalostomy to deep sclerectomy. |
8.6.5. Non-penetrating surgery versus trabeculectomy
Evidence Table 20, Appendix D and Forest Plots in to
8.6.5.1. Clinical evidence
Table 8-144Non-penetrating surgery versus trabeculectomy - Clinical study characteristics
View in own window
Outcome | Number of studies | Design | Limitations | Inconsistency | Directness | Other considerations |
---|
Visual field progression | 0 | | | | | |
Mean change in IOP from baseline (follow up 6 months)19,20,22,40,41,67,77,90,163,164 | 10 | RCT | Serious limitations (a) | Serious inconsistency (b) | No serious indirectness | Serious imprecision (c) Additional notes (d) |
Mean change in IOP from baseline (follow up 12 months)19,20,22,41,77,90,163,164 | 8 | RCT | Serious limitations (a) | Serious inconsistency (b) | No serious indirectness | Serious imprecision (c) Additional notes (d) |
Number of eyes with an unacceptable IOP (follow up 6 or 12 months)19,20,22,41,67,77,90,163,164 | 9 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | No serious imprecision Additional notes (d) |
Complications: Cataract Formation (follow up 12 – 36 months)20,22,41,77,90,163,164 | 7 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | No serious imprecision Additional notes (d) |
Complications: Persistent hypotony (follow up 12 – 36 months)19,22,41,77,90,163,164 | 7 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | No serious imprecision Additional notes (d) |
Complications: Wound leak (follow up 6 - 12 months)41,67 | 2 | RCT | Serious limitations (a) | No serious inconsistency | No serious indirectness | Serious imprecision (c) Additional notes (d) |
- a
- b
Some statistical heterogeneity is noted in mean change in IOP from baseline at 6 and 12 months which is not satisfactorily explained by subgroup analysis for type of non-penetrating surgery, use of augmentation or presence of PXF in population.
- c
For mean change in IOP from baseline from baseline at 6 and 12 months the lower confidence interval is clinically insignificant. For complications: wound leak wide confidence intervals make estimate of effect uncertain.
- d
Other differences between studies are noted in non-penetrating surgery type (viscocanalostomy or deep sclerectomy with or without implant); use of augmentation; study design where 3 studies20,77,164 randomised fellow eyes to treatment; IOP failure criteria; length of follow up from 6 months to 2 years; reporting of complications and mean baseline IOP. 5 studies19,22,40,90,164 included a proportion of patients diagnosed with PXF and one study164 included some CACG patients but <50%.
Table 8-145Non-penetrating surgery versus trabeculectomy - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Mean change in IOP from baseline (follow up 6 months) | 222 | 226 | not applicable | MD 2.57 (1.35 to 3.80) (e) | VERY LOW |
Mean change in IOP from baseline (follow up 12 months) | 202 | 204 | not applicable | MD 2.45 (1.46 to 3.44) | VERY LOW |
Number of eyes with an unacceptable IOP | 88/208 (42.3%) | 52/210 (24.8%) | 1.70 (1.30 to 2.23) | 174 more per 1000 (from 74 more to 305 more) | MODERATE |
Complications: Cataract Formation | 4/177 (2.3%) | 31/179 (17.3%) | 0.20 (0.09 to 0.44) | 138 fewer per 1000 (from 97 fewer to 157 fewer) | MODERATE |
Complications: Persistent hypotony | 8/184 (4.3%) | 39/187 (20.9%) | 0.25 (0.13 to 0.48) | 157 fewer per 1000 (from 109 fewer to 182 fewer) | MODERATE |
Complications: Wound leak | 1/49 (2%) | 4/49 (8.2%) | 0.33 (0.05 to 2.02) | 55 fewer per 1000 (from 78 fewer to 84 more) | LOW |
- e
One study40 included 3 arms, viscocanalostomy, deep sclerectomy and trabeculectomy. The data for trabeculectomy is added twice meaning there is some double counting. The overall effect to the weighted mean difference is around 0.1mmHg.
8.6.5.2. Economic evidence
No studies were identified.
8.6.5.3. Patient views evidence
No studies were identified.
8.6.5.4. Evidence statements - Non-penetrating surgery versus trabeculectomy
View in own window
Clinical | There were no studies which reported number of patients with visual field progression. Trabeculectomy is more effective than non-penetrating surgery in reducing IOP from baseline at 6 months follow up but the effect size may be too small to be clinically significant. (VERY LOW QUALITY) Trabeculectomy is more effective than non-penetrating surgery in reducing IOP from baseline at 12 months follow up but the effect size may be too small to be clinically significant. (VERY LOW QUALITY) Trabeculectomy is more effective than non-penetrating surgery in reducing the number of eyes with an unacceptable IOP at either 6 or 12 months follow up. (MODERATE QUALITY) Trabeculectomy is more likely to cause cataract formation than non- penetrating surgery at 12 to 36 months follow up. (MODERATE QUALITY) Trabeculectomy is more likely to cause persistent hypotony than non- penetrating surgery at 12 to 36 months follow up. (MODERATE QUALITY) There is no statistically significant difference between trabeculectomy and non- penetrating surgery in causing wound leaks at 6 to 12 months follow up. (LOW QUALITY) |
Economic | No studies meeting the inclusion criteria were identified which compared non- penetrating surgery to trabeculectomy. |
8.6.6. Non-penetrating surgery plus pharmacological augmentation versus non-penetrating surgery
Evidence Table 21, Appendix D and Forest Plot in
8.6.6.1. Clinical evidence
Table 8-146Non-penetrating surgery + pharmacological augmentation vs. non-penetrating surgery - Clinical study characteristics
View in own window
- a
- b
Wide confidence intervals make estimate of effect uncertain.
Table 8-147Non-penetrating surgery + pharmacological augmentation vs. non-penetrating surgery - Clinical summary of findings
View in own window
Outcome | Intervention | Control | Relative risk | Absolute effect | Quality |
---|
Number of patients with an unacceptable IOP (follow up 12 months) | 0/13 (0%) | 2/13 (15.4%) | 0.2 (0.01 to 3.80) | 123 fewer per 1000 (from 152 fewer to 431 more) | Low |
Number of patients with an unacceptable IOP (follow up 24 months) | 1/13 (7.7%) | 1/13 (7.7%) | 1.00 (0.07 to 14.34) | 0 fewer per 1000 (from 72 fewer to 1000 more) | Low |
Complications: Persistent hypotony | 0/13 (0%) | 0/13 (0%) | Not estimable | Not estimable | Low |
Complications: Wound leak | 0/13 (0%) | 0/13 (0%) | Not estimable | Not estimable | Low |
8.6.6.2. Economic evidence
No studies were identified.
8.6.6.3. Patient views evidence
No studies were identified.
8.6.6.4. Evidence statements - Non-penetrating surgery plus pharmacological augmentation vs. non-penetrating surgery
View in own window
Clinical | There were no studies which reported number of patients with visual field progression. There were no studies which reported mean change in IOP from baseline expressed as an absolute value with standard deviation. There is no statistically significant difference between non-penetrating surgery + pharmacological augmentation and non-penetrating surgery alone in reducing the number of patients with unacceptable IOP at 12 months follow up. (LOW QUALITY) There is no statistically significant difference between non-penetrating surgery + pharmacological augmentation and non-penetrating surgery alone in reducing the number of patients with an unacceptable IOP at 24 months follow up. (LOW QUALITY) There were no studies which reported number of patients with cataract progression. There is no statistically significant difference between non-penetrating surgery + pharmacological augmentation and non-penetrating surgery alone in causing persistent hypotony at 24 months follow up. (LOW QUALITY) There is no statistically significant difference between non-penetrating surgery + pharmacological augmentation and non-penetrating surgery alone in causing wound leaks at 24 months follow up. (LOW QUALITY) There were no studies which reported corneal epithelial defects. |
Economic | No studies meeting the inclusion criteria were identified which compared non-penetrating surgery + pharmacological augmentation to non- penetrating surgery alone. |
8.7. Patients with COAG or OHT associated with pseudoexfoliation or pigment dispersion
Patients with COAG or OHT associated with pseudoexfoliation or pigment dispersion were included in the scope for this guideline. We searched for evidence of effectiveness of treatments but no studies were found either in these groups alone, or as part of subgroup analysis within the comparisons listed above. Therefore, the GDG decided not to make a specific recommendation regarding these patients. Patients should be treated according to the recommendations used for COAG patients.
8.8. Recommendations and link to evidence
View in own window
Recommendation | Offer people newly diagnosed with early or moderate COAG, and at risk of significant visual loss in their lifetime, treatment with a prostaglandin analogue. |
|
|
Relative values of different outcomes | Prevention of blindness is the most important outcome. Cosmetic side effects of treatment with prostaglandin analogues may be unacceptable to some patients who may prefer an alternative treatment. |
Trade off between clinical benefits and harms | Prostaglandin analogues are effective at lowering IOP. They may affect the pigmentation of the iris and periorbital skin and cause lash growth but rarely have systemic side effects |
Economic considerations | The cost-effectiveness of trabeculectomy is dependent on a rapid progression in visual field loss. Therefore in the absence of any evidence of progression, pharmacological treatment is cost-effective. Among the pharmacological treatments PGA are the most cost- effective. |
Quality of evidence | Clinical evidence was generally of low quality. The economic evidence has minor limitations but direct applicability. |
Other considerations | Patient preference (see Relative values of different outcomes above). |
View in own window
Recommendation | Offer surgery with pharmacological augmentation (MMC or 5FU)*
as indicated to people with COAG who are at risk of progressing to sight loss despite treatment. Offer them information on the risks and benefits associated with surgery. |
|
|
Relative values of different outcomes | Progression is the most important outcome. |
Trade off between clinical benefits and harms | There is a balance to be found. On the one hand there is a higher risk of progression to blindness if the target pressure is not achieved. On the other hand there is a higher risk of side effects with more aggressive interventions. For example the risks of surgery are greater than the risks from medical treatment. |
Economic considerations | Trabeculectomy is cost-effective in cases of a detectable progression despite topical treatment. |
Quality of evidence | Clinical evidence was generally of low quality. The economic evidence has minor limitations but direct applicability. |
Other considerations | Patients may not be fit for surgery or may not wish to proceed to surgery because of anxiety or other issues. Where this situation arises alternative attempts at IOP lowering may be necessary. Options which may need to be considered include laser treatments, or multiple topical pharmacological treatments. |
- *
At the time of publication (April 2009, MMC and 5-FU did not have UK marketing authorisation for this indication. Informed consent should be obtained and documented. Both drugs should be handled with caution and in accordance with guidance issued by the Health and Safety Executive.
View in own window
Recommendation | Offer people with advanced COAG surgery with pharmacological augmentation (MMC or 5FU)*
as indicated. Offer them information on the risks and benefits associated with surgery. |
|
|
Relative values of different outcomes | Surgery is the most potent treatment for lowering IOP and can save remaining sight. If there are complications of surgery sight could be lost more quickly than if there had been persistence with pharmacological treatment. If surgery is successful the risk of losing further sight and progressing to complete blindness is reduced. |
Trade off between clinical benefits and harms | There is a risk of progression to complete blindness if COAG is not adequately treated. Although surgery has a higher risk than pharmacological treatment in the short term of causing blindness, it reduces this risk in the long term. If pharmacological treatment causes a satisfactory fall in IOP, surgery may be deferred. |
Economic considerations | Trabeculectomy is cost-effective for this group of patients even if the progression rate is very low. Blindness has a large personal and social cost (see calculation of cost of blindness in Appendix F – 1.3) |
Quality of evidence | Clinical evidence was generally of low quality. The economic evidence has minor limitations but direct applicability. |
Other considerations | There were no trials due to the ethical implications of not treating patients with severe COAG. |
- *
At the time of publication (April 2009, MMC and 5-FU did not have UK marketing authorisation for this indication. Informed consent should be obtained and documented. Both drugs should be handled with caution and in accordance with guidance issued by the Health and Safety Executive.
View in own window
Recommendation | Consider offering people with COAG who are intolerant to a prescribed medication:
After trying two alternative pharmacological treatments consider offering surgery with pharmacological augmentation (MMC or 5FU)*
as indicated or laser trabeculoplasty. |
|
|
Trade off between clinical benefits and harms | Prescribing an alternative medication should reduce the risk of progression to blindness. If there is intolerance, allergy or an inadequate IOP lowering effect surgery should be offered as an alternative treatment. |
Economic considerations | Offering a more costly BB (preservative-free preparation) is still more cost-effective than no treatment in patients with COAG. |
Quality of evidence | There was no clinical evidence. The economic evidence has minor limitations but direct applicability. |
Other considerations | Patients may not be fit for surgery or may not wish to proceed to surgery because of anxiety or other issues. In such instances laser treatment may be helpful in improving IOP control. |
- *
At the time of publication (April 2009, MMC and 5-FU did not have UK marketing authorisation for this indication. Informed consent should be obtained and documented. Both drugs should be handled with caution and in accordance with guidance issued by the Health and Safety Executive.
8.9. Supporting recommendations
View in own window
Recommendation | Offer people who present with advanced COAG and who are listed for surgery interim treatment with a prostaglandin analogue. |
|
|
Trade off between clinical benefits and harms | If COAG is severe when first diagnosed, treatment to lower IOP should be started immediately as any amount of progression could cause additional severe visual disability. There is a risk of progression to complete blindness if COAG is not adequately treated. |
Economic considerations | Blindness has a large personal and social cost (see NICE’s social value judgements document) |
Other considerations | None |
View in own window
Recommendation | Check that there are no relevant comorbidities or potential drug interactions before offering medication. |
|
|
Trade off between clinical benefits and harms | Some pharmacological treatments that are effective at lowering IOP may have serious systemic side effects, particularly worsening of chronic obstructive pulmonary disease and asthma by beta blocker eye drops. There are many potential drug interactions with beta-blockers and alpha receptor agonists. The patient’s general health should not be compromised by any pharmacological treatment as alternative treatments for COAG are available. |
Economic considerations | None |
Other considerations | Older people are more likely to experience adverse reactions to medications |
View in own window
Recommendation | Encourage people using the prescribed pharmacological treatment to continue with the same treatment unless:
|
|
|
Trade off between clinical benefits and harms | Persisting with medication will reduce the risk of progression to blindness. If the medication is causing harm because of allergy or intolerance a different medication can be offered. |
Economic considerations | Changes in therapy are associated with additional costs of visits. If a change is unnecessary then these costs should be avoided. |
Other considerations | None |
View in own window
Recommendation | Check the person’s adherence to their treatment and eye drop instillation technique in people with COAG whose IOP has not been reduced sufficiently to prevent the risk of progression to sight loss despite pharmacological treatment. If adherence and eye drop instillation technique are satisfactory offer one of the following:
alternative pharmacological treatment (a prostaglandin analogue, beta-blocker, carbonic anhydrase inhibitor or sympathomimetic); more than one agent may be needed concurrently to achieve target IOP surgery with pharmacological augmentation (MMC or 5FU)* as indicated If the pharmacological treatment option is chosen, after trying two alternative pharmacological treatments consider offering surgery with pharmacological augmentation (MMC or 5FU)*
or laser trabeculoplasty. |
|
|
Trade off between clinical benefits and harms | Complications of surgery may cause harm but if alternative treatments fail then surgery offers the least risk of progression to blindness. |
Economic considerations | None. |
Other considerations | Patients may not be fit for surgery or may prefer not to proceed to surgery because of anxiety or other issues. |
- *
At the time of publication (April 2009, MMC and 5-FU did not have UK marketing authorisation for this indication. Informed consent should be obtained and documented. Both drugs should be handled with caution and in accordance with guidance issued by the Health and Safety Executive.
View in own window
Recommendation | Offer people with COAG who prefer not to have surgery or who are not suitable for surgery:
pharmacological treatment (a prostaglandin analogue, beta-blocker, carbonic anhydrase inhibitor or sympathomimetic); more than one agent may be needed concurrently to achieve target IOP
|
|
|
Trade off between clinical benefits and harms | Alternative treatments to surgery are less effective but have a lower risk of immediate loss of sight. Some patients may choose a higher long term risk of sight loss to a low risk of immediate sight loss. |
Economic considerations | None. |
Other considerations | Patients may prefer certain options ahead of others. |
View in own window
Recommendation | After surgery offer people with COAG whose IOP has not been reduced sufficiently to prevent the risk of progression to sight loss one of the following:
pharmacological treatment (a prostaglandin analogues, beta-blocker, carbonic anhydrase inhibitor or sympathomimetic); more than one agent may be needed concurrently to achieve target IOP further surgery
|
|
|
Trade off between clinical benefits and harms | If surgery fails to control IOP topical medical treatment should be restarted. Repeat surgery may be required and if so should be offered. Cyclodiode laser treatment may need to be considered. |
Economic considerations | None. |
Other considerations | Patients may prefer certain options ahead of others. |
8.10. Summary of all recommendations on treatment for patients with COAG
The recommendations have been reordered to reflect the patient’s pathway.
Check that there are no relevant comorbidities or potential drug interactions before offering medication.
Offer people newly diagnosed with early or moderate
COAG, and at risk of significant visual loss in their lifetime, treatment with a prostaglandin analogue.
Offer people with advanced
COAG surgery with pharmacological augmentation (
MMC or 5FU)
* as indicated. Offer them information on the risks and benefits associated with surgery.
Offer people who present with advanced
COAG and who are listed for surgery interim treatment with a prostaglandin analogue.
Encourage people using the prescribed pharmacological treatment to continue with the same treatment unless:
Check the person’s
adherence to their treatment and eye drop instillation technique in people with
COAG whose
IOP has not been reduced sufficiently to prevent the risk of
progression to sight loss despite pharmacological treatment. If adherence and eye drop instillation technique are satisfactory offer one of the following:
alternative pharmacological treatment (a prostaglandin analogue, beta-blocker, carbonic anhydrase inhibitor or sympathomimetic); more than one agent may be needed concurrently to achieve
target IOPsurgery with pharmacological augmentation (
MMC or
5-FU)
* as indicated
If the pharmacological treatment option is chosen, after trying two alternative pharmacological treatments consider offering surgery with pharmacological augmentation (
MMC or
5-FU)
* as indicated or
laser trabeculoplasty.
Offer surgery with pharmacological augmentation (
MMC or
5-FU)
* as indicated to people with
COAG who are at risk of progressing to sight loss despite treatment. Offer them information on the risks and benefits associated with surgery.
Consider offering people with
COAG who are intolerant to a prescribed medication:
After trying two alternative pharmacological treatments consider offering surgery with pharmacological augmentation (
MMC or
5-FU)
* as indicated or
laser trabeculoplasty.
After surgery offer people with
COAG whose
IOP has not been reduced sufficiently to prevent the risk of
progression to sight loss one of the following:
pharmacological treatment (a prostaglandin analogues, beta-blocker, carbonic anhydrase inhibitor or sympathomimetic); more than one agent may be needed concurrently to achieve
target IOPfurther surgery
Offer people with
COAG who prefer not to have surgery or who are not suitable for surgery:
pharmacological treatment (a prostaglandin analogues, beta-blocker, carbonic anhydrase inhibitor or sympathomimetic); more than one agent may be needed concurrently to achieve
target IOP
8.11. Research recommendations on treatment for patients with COAG
See APPENDIX G
8.11.1. Update of National survey of trabeculectomy
The GDG recommended the following research question:
What are the current
NHS national benchmarks for surgical success and complications in people with
COAG undergoing
trabeculectomy drainage surgery with and without pharmacological augmentation?
Why this is important
The answer to this question would provide more accurate and up-to-date evidence for surgical treatment in COAG. Surgical success and complication rates could then be used to update benchmarks for clinical audit and assist in planning service provision. It would also then be possible to inform people having surgery of the chances of success and complications. The current evidence base is the National Survey of Trabeculectomy. However, this is now 10 years old and techniques have changed. The benchmarks created from the new survey would set a standard against which newer techniques could be evaluated. The study design would be similar to the audit of 10 years ago, to allow comparison of outcomes now in the light of changes in technique and the recommendations made by that audit..
8.11.2. Laser treatment
The GDG recommended the following research question:
What is the
clinical effectiveness and cost effectiveness of initial argon, diode or selective
laser trabeculoplasty compared with prostaglandin analogues alone or laser trabeculoplasty plus prostaglandin analogues in combination in people with
COAG?
Why this is important
The answer to this question would provide data on the comparative clinical effectiveness and cost effectiveness of laser treatment versus modern ocular hypotensive agents, particularly prostaglandin analogues. Laser treatment may control IOP in some people for a time without the need for topical medications, and in others, it may offer additional benefit to topical medications. In either case there may be cost savings and improved prevention of progression. Existing trials of laser trabeculoplasty compared with pharmacological treatment use outdated pharmacological agents. Because of the lack of evidence, the role of laser trabeculoplasty in COAG management cannot be clearly defined. An RCT should be used to answer this research question, and sham laser treatment would be needed to enable double masking or at least single masking.
- *
At the time of publication (April 2009, MMC and 5-FU did not have UK marketing authorisation for this indication. Informed consent should be obtained and documented. Both drugs should be handled with caution and in accordance with guidance issued by the Health and Safety Executive.