Glaucoma research update
Dr Divya Perumal

Glaucoma research update

February 28, 2019 Dr Divya Perumal

The following are some of the most interesting recent journal publications relevant to optometrists and ophthalmologist involved in glaucoma management. 

Risk factors associated with missed diagnosis of narrow angles by Van Herick technique, Thompson AC, Vu, DM, Cowan LA, et al.
Ophthalmology 2018; 1 (2): 108-114

Design
This retrospective review aimed to identify factors associated with the deep-appearing anterior chamber on slit-lamp examination by the Van Herick (VH) technique in eyes with a diagnosis of narrow angle (NA) on gonioscopy.
1314 eyes from 696 patients were included and demographic/clinical factors were assessed.

Outcome
13.7% of eyes with narrow angles were misclassified as deep using VH techniques.
Primary angle closure glaucoma (PACG; OR2.43; p<0.001) and primary angle closure (PAC; OR1.38; p=0.006) were significantly more likely to be misclassified by VH compared to primary angle closure suspects (PACS).
Male (OR 2.22; p<0.001), myopia (OR 1.4; p=0.048), black (OR 4.11; p<0.001) and Asian race (OR 2.24; p= 0.044), are at greater risk of being misclassified.

Limitations
VH and gonioscopy are subjective assessments, only conducted by one unmasked specialist and potentially subject to bias.

Significance
Optometry referrals with concerns of narrow angles are frequently made on VH ratios, with a potential risk of missing 10% of at-risk cases.
Gonioscopy is essential in the diagnosis of angle closure and further aids in establishing the nature of angle closure (e.g. plateau iris, neovascular and synechial causes).
Central corneal thickness can also impact VH accuracies with thinner peripheral cornea causing limbal cornea to appear falsely deeper than it is. Afro-Caribbean patients are associated with thinner corneal thickness.


Transscleral Diode Laser Cyclophotocoagulation: A comparison of slow coagulation and standard coagulation techniques, Duerr E, Sayed M, Moster SJ et al.
Ophthalmology Glaucoma 2018; 1 (2): 115-112

Design
Retrospective case series.
52 eyes were treated with slow-coagulation (slow) and 26 eyes with pop-titrated (standard) methods.
In the “standard” group, the setting was 1.75W x 2 sec with power increased to minimum required to produce a pop. In the “slow” group, the setting was 1.25W x 4-4.5 sec for brown iris and 1.5W x 3.5-4 sec for others.

Outcomes
Presenting VA and IOP was comparable between the groups (“slow” logMAR 1.94 vs “standard” 1.71 logMAR; IOP 37 mmHg in “slow” vs 39 mmHg in “standard”.
Comparable outcomes (see Table 1) include:
VA of better than light perception: “slow” = 71.1% vs “standard” = 65.0% (p=0.599).
IOP was < 20mmHg: “slow” = 46% vs “standard” = 44% (p=0.870).
Need for a second procedure: “slow” = 28.8% vs “standard” = 23.1% (p=0.588).
Maximum medications needed to control IOP after surgery (p=0.771).
Complications were higher in the standard group (1.46) vs slow group (0.62; p=0.002).

Limitations
Retrospective nature.
Results cannot be generalised to early or moderate glaucoma.
Underpowered to detect small but potentially significant changes.

Significance
Cyclodiode was traditionally used in the treatment of eyes with poor prognosis but evidence now shows potential in patients with moderate-severe glaucoma especially with limited resources.
Hypotony and conjunctival scarring is rare (important when considering the possibility of future drainage surgery).
This paper is useful in that it provides evidence that using the slow technique may reduce the incidence of postprocedural complications while maintaining similar VA and IOP outcomes in comparison with the standard technique.

Table 1. Comparison of the rates of complication associated with cyclodiode treatment

 

Clear lens extraction for the management of primary angle closure glaucoma: surgical technique and refractive outcomes in the EAGLE cohort, Day AC, Cooper DC, Burr J et al. Br J
Ophthalmol 2018; 102:1658-1662

Design
Prospective review from the EAGLE (Effectiveness in angle-closure glaucoma of lens extraction) study, a multicentre randomised controlled trial comparing clear lens extraction (CLE) and laser peripheral iridotomy (LPI) in angle closure.
Patients included were ≥50 years and with (1) PAC with IOP > 30 mmHg or (2) PACG.

Outcomes
208 from 419 participants were randomised to CLE.
Baseline CDVA was ETDRS 77.9 letters (SD 12.4), which did not change significantly at 36 months.
Mean preoperative SE were +1.7D and +0.08 D at 36 months.
59% and 85% eyes were within ±0.5D and ±1.0D of predicted refraction at 36 months.
Eyes with axial length of <22mm were more likely to have >1D error from predicted outcome (see Table 2).
Mean postoperative target refraction was -0.1D spherical equivalent (SD 0.6D).
Complication rates were: 1% had posterior capsule tear, one patient had malignant glaucoma.

Limitation
Only SRK-T, Hoffer Q, Haigis and Holladay formulae were used. It may be beneficial to establish the accuracy with Barrett.
Subjects who had undergone clear lens extraction were not masked.

Significance
The refractive outcomes of cataract surgery in PAC are less predictable due to anatomical variations including shallow anterior chamber, anteriorly positioned lens and short axial length.
The outcomes were within the Royal College of Ophthalmologists (RCOphth) benchmark of 85% within ±1.0D targets.
The predictive power was under-corrected with Hoffer Q and overcorrected with SRK/T (myopic outcome), see Table 3.
Patients should be warned about the risk of refractive surprise, particularly if axial length is <22mm.

Table 2.  The table shows the mean proportion eyes within ±0.25, ±0.50 and ±1.0 dioptres by axial length

 

Table 3. The table shows the mean prediction errors, absolute prediction errors and proportion eyes within ±0.25, ±0.50 and ±1.0 dioptres of predicted refraction

 

Treatment outcomes of Mitomycin C-augmented trabeculectomy, sub Tenon's injection versus soaked sponges, after three years of follow-up, Esfandiari H et al.
Ophthalmology Glaucoma 2018; 1:66-74

Design
82 patients with POAG, were randomized either intra-Tenon injection of 0.1ml of 0.01% MMC (TI) or 0.02% subconjunctival MMC-soaked sponges (TS).
In T1: MMC was injected in the superior fornix and spread diffusely with a blunt spatula. A Weck-Cel sponge was used to prevent anterior migration. Peritomy and Tenons’ dissection was performed one minute after MMC injection, followed by BSS irrigation.
In TS: MMC-soaked sponges applied for 1-3 mins in the sub Tenon's space prior to scleral flap creation.

Outcome
The mean preoperative IOP was 22.4±4.6 mmHg with an average of 3±1 medications.
Kaplan-Meier plots are provided (see Fig 1).
Comparable results between groups included
Success at three years: 72.2%(T1) vs 65.1%(TS) (p=0.30).
Final IOP at three years was 15.3±3.7mmHg (T1) and 16.4±3.5mmHg (TS) (p= 0.55).
Mean medication was 0.9± 1.1(T1) vs 1.1±1.1 (TS) (p= 0.54).
VA and endothelial cell change (p= 0.47 and p=0.94, respectively), which is reassuring due potential limbal stem cell failure.
Blebs were more diffuse (p= 0.032), less vascularised (p=0.013), and shallow (p=0.012) in T1.

Limitations
Only European patients with POAG were included.
The delivery of MMC sponges could range from 1-3 minutes.

Significance
Study suggests equal efficacy between the two procedures with better bleb morphology with T1.
Potential issue with T1 is difficulty in determining the equivalent potency of sub Tenon's MMC to sponge concentrations. Advantages include delivering a fixed MMC concentration over a larger diffuse area, reduced operating time and elimination of sponge-related complications (lost sponge and conjunctival trauma from sponge manipulation).

Fig 1. Postopertive view of Istent inject

Dr Divya Perumal is a former optometrist and is a New Zealand-trained ophthalmologist. She has advanced training in complex glaucoma and cataract surgery, and practises at Auckland District Health Board and Eye Institute.