The efficacy of Viscocanalostomies and combined phacoemulsification with Viscocanalostomies in the treatment of patients with glaucoma: a non-randomised observational study

This was a non-randomised observational study including all patients undergoing VC or PV at the Stanley Eye Unit, Abergele Hospital, UK between 09/11/09 and 18/12/12. They were identified from an intranet database retrospectively, in accordance with the tenets of the Declaration of Helsinki. Verbal consent is obtained from all patients for the use of their anonymised clinical data for research purposes within the clinic. Patient-focused posters are also displayed throughout the clinic explaining our use of anonymised data for research undertaken by the department with contact information for anyone requesting further information about our research database. The local research ethics committee (Wales REC 5, Bangor) and local Research and Development committee have approved this research and the consent procedure.

Patients were offered VC if, despite being on maximum medical therapy, they showed evidence of progression of visual field defects on two consecutive Humphreys field assessments, and/or IOP was poorly controlled and out of the target range for the patient. These findings were supported with evidence of corresponding loss of retinal nerve fibre layer (RNFL) and Ganglion cell layer (GCL) on optical coherence tomography (OCT). Patients were also offered surgery if they were intolerant to topical glaucoma medication. The patient’s target IOP was tailored according to age and rate of progression of glaucoma. Patients were excluded if they were aphakic, had silicone oil in the eye, had multiple previous glaucoma procedures or showed evidence of neovascular glaucoma. If cataracts were found on clinical examination at the time of listing, then patients were offered the option of a combined surgery. All patients were included, regardless of glaucoma subtype. A control arm was not included as part of the study.

VC was performed as described in Stegmann et al. [2] A small trabeculo-Descemet window was exposed, thereby avoiding the juxtacanalicular tissue which is the presumed area of resistance. This allows aqueous to percolate through the newly created window allowing decompression of the eye in a controlled fashion. Viscoelastic is injected into the ostia of the Schlemm’s canal thus dilating it and the collector channels. This causes focal microscopic ruptures in Schlemm’s canal endothelial wall and the adjacent juxtacanalicular trabecular membrane. Aqueous can egress via various outflow canals, but ultimately into episcleral vasculature or the uveoscleral route. The intrascleral lake formed by removal of deep scleral tissue allows fluid to escape via the conjunctiva and its lymphatics as well as newly formed aqueous outflow vessels. If the superficial flap developed any button hole, then the excised bit of sclera was used as a patch graft and placed over the superficial flap to avoid bleb formation. The patch graft was secured with 10–0 vicryl. When VC was combined with cataract surgery (PV), the phacoemulsification incision was placed between the superficial and deep flap once the Schlemm’s canal was identified. Phacoemulsification of cataract was completed using a standard approach and then the remainder of the viscocanalostomy was completed. A single surgeon with previous experience of VC and PV performed all procedures.

IOP was measured using Goldmann applanation tonometry (GAT) pre-operatively, and then on postoperative days 1 and 7, week 1, months 1, 3, 6, 9 and 12. Patients were then reviewed every 6 months up to year 3. Measurements were taken during routine outpatient clinics and not at consistent times of day or adjusted for diurnal variations.

The primary outcome of the study was IOP. Secondary outcomes were the number of anti-glaucoma drops used and the frequency of complications. Visual acuity, visual fields and optic nerve head appearance were not assessed.

Success criteria were defined as:

Patients with IOP of ≥22 mmHg, or who underwent repeat glaucoma surgery were noted as surgical failure. These criteria were used to allow comparison with other studies that have used similar end points [4‐8]. Kaplan–Meier survival models were used to indicate success rate over the three-year study period, taking censoring into account as some patients transferred to another unit or were lost to follow-up.

The pre- and post-operative IOPs were compared using paired t test. A comparison of pre- and post-operative IOPs was made between PV and VC surgeries using unpaired t test. The numbers of anti-glaucoma medications used pre- and post-operatively were analysed. Statistical analyses were performed using Microsoft Excel 2013 and SPSS 17.0. A p value of < 0.05 was considered statistically significant. Figures were given as mean ± standard deviation (SD).

A total of 458 patients and 620 eyes were included in the study. 323 left and 297 right eyes were operated upon.

Two hundred twenty-three were male with an age range of 18–93 (74.5 ± 10.9) and 235 were female with an age range of 43–97 (76.7 ± 8.8).

The majority of eyes included were POAG, accounting for 65.0% (403 eyes). The next largest groups were NTG and PXF accounting for 17.10% (106 eyes) and 5.81% (36 eyes) respectively.

Mean follow-up duration was 31.8 (SD 9.3) months. Over 80% of subjects were followed up for the full three-year duration. Patients who had been lost to follow up were not included when calculating mean results. 118 patients had been lost to follow up by the end of the 36 months.

The mean pre-operative IOP across all procedures was 23.02 ± 5.60. The mean IOP at the first post-operative day was 12.71 ± 7.42 (p < 0.0001). After one month IOP was reduced by 36.0% (from 23.02 mmHg down to 14.73 mmHg). This is similar to the IOP reduction of 37.4% seen by the end of three-year period. Table 1 and Fig. 1 demonstrate that the IOP remained stable throughout follow-up. Complete success rate at 36 months was 65.7% with a qualified success rate of 96.0% (n = 502).

A total of 427 PV surgeries and 193 VC surgeries were performed. A significant IOP reduction of 7.92 mmHg (p < 0.0001) was achieved by PV surgery after 12 months, which was maintained up to the end of study period at 36 months. VC patients maintained a similar mean IOP throughout the post-operative follow up period with no significant difference between the groups, but had a larger reduction of 9.03 mmHg (p < 0.0001) due to a significantly higher listing mean IOP. Listing mean IOP was 22.54 ± 5.10 for PV and 24.06 ± 6.45 for VC (p = 0.0017) (Table 2).

The complete success rate at 36 months for VC was 63.1%, and qualified success rate was 94.0% (n = 149). Complete success rates were significantly higher for PV at 76.0% (p = 0.0050). There was no significant difference in qualified success with a rate of 95.9% (n = 364, p = 0.668) (Fig. 2, Fig. 3).

Throughout the study period, patients were prescribed anti-glaucoma eyedrops as pharmacological intervention to maintain IOP control, at the clinician’s discretion. If a patient was lost to follow up or did not attend an appointment they were not included in the results for that follow up year.

At 12 months, 13 patients (2.1%) required anti-glaucoma eyedrops to maintain an IOP of ≤21 mmHg. This increased to 116 (18.7%) and 152 (24.5%) at 24 and 36 months respectively.

The mean number of drops was higher for the VC group at all points throughout the study (Fig. 4).

We recorded 7 PCR/vitrectomy, 65 perforated TDW and 8 scleral defect patch graft, representing a complication rate of 12.9% across all procedures.

Comparing the post-operative IOP between surgeries with complications and overall results revealed a significant difference in IOP on day 1, where a 2.07 mmHg difference was observed (p = 0.0363). At all other time points the difference was not statistically significant (Table 3).

There were several limitations of the study. Namely, the study was performed retrospectively, and assessment of patients including IOP monitoring was not masked and therefore subject to observer bias. Visual acuity, visual field assessments and optic nerve head appearance were not used as outcomes of the study, and disease staging was not performed.

IOP was used as the primary outcome. This is an indirect measure of the disease progression but has been used as a justifiable surrogate in similar studies. IOP was measured during routine outpatient clinic appointments and therefore was not measured at a consistent time of day or measured as a mean of diurnal variation.

By including all patients during the time period of the study, multiple subtypes of glaucoma such as NTG were treated. This was because the purpose of the study was to present outcomes in real world practice, however it does create limitations interpreting the data. Especially when comparing the results to other studies, the majority of which limited the study populations to a single subtype such as POAG. Additionally the decision to offer patients surgical treatment was based on the individual patient’s target IOP range. Patients with NTG were therefore offered surgery at a lower IOP.

NTG patients achieved reduction in IOP with sustained control but, as expected, had slightly lower mean IOP when compared to POAG patients. Listing mean IOPs showed the most notable difference to POAG (NTG 18.07 +/− 2.44 mmHg and POAG 23.55 +/− 5.15 mmHg respectively), however this gap narrowed at 36 months (NTG 13.53 +/− 2.24 mmHg and POAG 14.67 +/− 3.37 mmHg). This slightly reduced the overall means, as well as the percentage decrease. It is encouraging though to see that NTG also responded well to the procedures.

Our study did not include a trabeculectomy group or other non-penetrating surgical procedures and therefore direct comparison is not possible. It must be noted however that the available literature directly comparing trabeculectomy to non-penetrating filtration surgeries has consistently found better IOP control in the trabeculectomy groups but with more frequent complications [18‐23].