Article Text

Download PDFPDF

Transscleral diode laser cycloablation in patients with good vision
  1. A P Rotchford1,
  2. R Jayasawal2,
  3. S Madhusudhan2,
  4. S Ho2,
  5. A J King2,
  6. S A Vernon2
  1. 1Tennent Institute of Ophthalmology, Gartnavel Hospital, Glasgow, UK
  2. 2Department of Ophthalmology, Nottingham University Hospitals NHS Trust, Queen's Medical Centre, Nottingham, UK
  1. Correspondence to Dr A P Rotchford, Consultant Ophthalmologist, Tennent Institute of Ophthalmology, Gartnavel Hospital, Great Western Road, Glasgow G12 0YN, UK; rotchford{at}


Purpose To investigate the effect of diode laser cyclophotocoagulation for glaucoma on central visual function in patients with good visual acuity (VA).

Patients and methods Patients with preoperative VA 20/60 or better who had undergone cyclodiode treatment according to a standard protocol were evaluated retrospectively. The primary outcome variable was a recorded loss of two or more Snellen lines of best corrected VA during follow-up. Successful intraocular pressure (IOP) control was defined as being between 6 and 21 mmHg inclusive without oral acetazolamide or other glaucoma surgery.

Results 49 eyes of 43 patients with a median pretreatment acuity of 20/30 were included (range 20/16–20/60). After a mean duration of follow-up of 5.0 years, median VA was 20/60 with a line loss of two or more recorded in 15 eyes (30.6%) (mean survival time 7.7 years). 67.3% (33/49) retained VA 20/60 or better, but VA deteriorated by one Snellen line or more in 31 (63.2%), and in 16.3% (8/49), final VA was <20/200. In cases experiencing a two-line loss in acuity, the main causes were glaucoma progression (nine cases) and macula oedema (four cases). Visual loss was unrelated to total treatment dose (mean 99.7 J), initial acuity or initial IOP level. IOP was controlled at final follow-up in 39/49 (79.6%) with no cases of hypotony.

Conclusions Most of these eyes with difficult to manage glaucoma retained their good VA over long-term follow-up after undergoing diode laser cyclophotocoagulation. The proportion losing two Snellen lines is in line with that reported after trabeculectomy or tube surgery. These results suggest a possible role for the use of transscleral cyclodiode in selected eyes with significant visual potential. Further controlled prospective studies are required to better define this role.

  • Diode laser
  • cyclophotocoagulation
  • cyclodiode
  • glaucoma
  • aqueous humour
  • ciliary body
  • eye (globe)
  • treatment lasers

Statistics from

Transscleral cyclophotocoagulation is a well-established treatment for glaucoma that has not responded to other interventions, and it is widely used in advanced cases with minimal visual potential.1 There is concern, however, regarding the adverse effects of this type of destructive surgery on visual function. This concern has prevented its wider acceptance as a treatment in eyes with good residual visual acuity (VA).2 However, most of the published data relate to the treatment of eyes with initially poor vision and few series include significant numbers of cases with good VA. In this study, we describe the visual outcome in patients undergoing transscleral cyclodiode with pretreatment acuity of 20/60 or better.


The study involved a retrospective case note review of all patients who underwent cycloablation with the Oculight Sx semiconductor diode 810-nm laser and the contact G-probe (Iris Medical Instruments, Mountain View, California, USA) at Queen's Medical Centre, Nottingham, UK, between April 1994 and December 2006. The authors identified patients by a review of operating theatre records in which all cases undergoing surgery are recorded. We accessed the case notes and included all patients with preoperative best corrected Snellen VA (BCVA) of 20/60 or better on at least two consecutive occasions at the time of treatment.

Indications for cycloablation included medically uncontrollable glaucoma in subjects who declined, had not responded to or were unlikely, in the judgement of the surgeon responsible at the time, to respond to filtration surgery.

We performed cyclophotocoagulation under peribulbar, retrobulbar or general anaesthesia using a standard treatment protocol that we have previously described.3 Briefly, this involves the application of 14 burns of 2 W power and 2 s duration, spaced half the width of the G-probe apart around 270° of the ciliary body, unless ocular comorbidity (eg, scleral thinning) dictates otherwise. Laser energy is focused at a point corresponding to the pars plicata, 1–2 mm behind the limbus with the heel of the probe aligned at the limbus. In the contexts of congenital glaucoma, high myopia and distortion of limbal anatomy secondary to previous surgery, we identified the position of the ciliary body before treatment by transillumination and adjusted the probe position accordingly if necessary.

Postoperatively, we prescribed topical steroids four times daily for 2–4 weeks. We discontinued glaucoma medications during follow-up using clinical discretion based on the intraocular pressure (IOP) reduction achieved. The decision to repeat the treatment was dependent on the magnitude of the hypotensive response and the drug load necessary to achieve a satisfactory IOP.

Study parameters included demographic details, BCVA, IOP, visual field loss, antiglaucoma medications, diagnosis, details of previous ocular surgery and duration of follow-up. We documented BCVA and IOP at the preoperative assessment, postoperatively on day 1, at 1 week, 1 month, 3 months and each subsequent clinic visit (every 3–6 months). We defined final follow-up as the last recorded clinic visit in the time period of the study.

As the primary end point for the study, we used a reduction in BCVA by two or more lines on two successive visits. We defined successful pressure control as a final IOP ≥6 and ≤21 mmHg irrespective of the need for topical antiglaucoma medication but without the need for oral acetazolamide or further glaucoma surgery.

Visual field testing was performed by Humphrey threshold standard automated, Goldmann or Friedmann perimetry. For the purposes of this group of patients, we classified visual field loss as advanced in the presence any of the following: (1) Humphrey 24-2 mean deviation ≥15 dB; (2) deviation of ≥15 dB affecting any part of the central 10° on Humphrey 24-2; (c) encroachment of the I4e isopter in the central 10° of a Goldmann field; or (d) any loss of the central four target points on Friedmann field.

We performed data analysis using SPSS V10 (SPSS Inc, Chicago, Illinois, USA). We used the χ2 test for determining the statistical significance of differences between discrete variables (or Fisher's exact test when a cell value less than five was expected), Student t test for parametric continuous data and Mann–Whitney for non-parametric continuous. We performed Kaplan–Meier survival analysis to investigate the loss of VA after treatment.


Of 351 patients recorded as undergoing cyclodiode treatment, we identified 51 eyes of 45 patients with initial VA of 20/60 or better. Two patients were lost to follow-up within the first 12 months (one died and one was discharged back to the referring centre). For the remaining 49 eyes of 43 patients, the distribution of glaucoma types and other baseline characteristics are shown in table 1. Median pretreatment BCVA was 20/30 (figure 1).

Table 1

Baseline characteristics

Figure 1

Snellen visual acuity distribution pre-cyclodiode and at final follow-up. CF, counting fingers; HM, hand movements; LP, light perception; NLP, no light perception.

Reliable visual fields before cyclodiode were available for 36/49 eyes (73.5%). Of these, 25 were automated Humphrey threshold fields, eight were Goldmann and three were Friedmann fields. Mean deviation (SD) for the Humphrey fields was 13.8 (9.7) dB (range 1.6–30.4). Advanced loss was present in 19/36 (52.7%) of fields.

We performed a total of 85 treatment episodes on these 49 eyes using a power setting of 2 W for a duration of 2 s for all shots. We applied a mean of 14.4 shots with a mean energy of 57.6 J per treatment episode.

We performed a mean of 1.73 treatment episodes per eye (range 1–6; median 1.0) with more than one episode given in 18/49 eyes (36.7%). During the course of treatment, we applied a total mean energy of 99.7 J per eye (range 40–336; median 60 J). After the first treatment, we followed subjects for a mean of 263 weeks (range 62–578), or 5.1 years.

IOP and glaucoma medication use at final follow-up is shown in table 2. IOP was controlled at final follow-up without oral acetazolamide in 39/49 (79.6%). None of the 49 eyes developed sustained hypotony (IOP<6 mmHg) during follow-up, although one eye developed transient hypotony after both treatment episodes that resolved by the end of the first postoperative month.

Table 2

IOP and glaucoma medication use at final follow-up after cyclodiode

Median VA at final follow-up was 20/60 (figure 1). It improved in 8/49 eyes (16.3%), remained unchanged in 11/49 (22.5%) and deteriorated by one line or more in 30/49 (61.2%). Final VA, after a mean of 5.0 years follow-up, was reduced to <20/200 in 8/49 eyes (16.3%).

A loss of two Snellen lines or more of acuity was measured in 15/49 eyes (30.6%) in 15 different patients. The mean (SD) survival time using this primary end point was 399.7 (70.3) weeks (95% CI 330.8 to 468.6) or 7.7 years (figure 2). During the first 12 months, 18.3% (9/49) of eyes lost two or more lines of acuity.

Figure 2

Kaplan–Meier plot showing survival time to loss of two or more lines of Snellen visual acuity.

There was no difference in the mean number of treatment episodes between those losing VA (1.60) and those maintaining VA (1.79) (p=0.69). Loss of VA was also independent of the total energy applied (p=0.66), age (p=0.11), sex (p=0.14), preoperative IOP (p=0.63), preoperative VA (p=0.17) or advanced preoperative visual field loss (p=0.22). Those who lost VA had a similar final IOP (15.0 mmHg) to the remainder (16.1 mmHg) (p=0.55).

In eight cases, loss of VA was solely attributed to glaucoma progression, and it was a contributory factor in one further case (table 3). Of these nine cases, loss of acuity occurred in spite of IOP consistently below 21 mmHg during follow-up in four cases, with a mean final IOP of 14 mmHg. All had advanced disc cupping and/or visual field loss affecting the central 10°. In the other five eyes, IOP was not consistently controlled after cyclodiode treatment. In one case, rapid loss of VA from 20/60 to 20/600 occurred 6 days after treatment. This was attributed to optic nerve “snuff out” caused by an IOP of 50 mmHg.

Table 3

Details of cases losing two or more lines of VA after cyclodiode treatment

Visual loss was attributed partly or entirely to macula oedema in four cases. One had a history of proliferative diabetic retinopathy, one was aphakic and one had a history of previous vitreous loss during cataract surgery. These three eyes underwent a single standard treatment session of 56 J, and the reduction in VA was delayed between 6 and 12 months post-treatment. In the fourth case, the macula oedema developed a week after the second treatment episode.

One case was due to a rhegmatogenous retinal detachment, which developed 10 weeks after cyclodiode in an eye with uveitic glaucoma. This patient had already had a spontaneous retinal detachment in the fellow eye.

One case with no identified risk factors developed a vitreous haemorrhage 5 days after cyclodiode in an eye with primary open-angle glaucoma (POAG). By the time that this had cleared spontaneously, a dense cataract had formed reducing acuity to counting fingers. The patient declined cataract surgery.


Cycloablation is a recognised therapeutic approach in the management of complex and refractory glaucoma and appears to be an effective method of achieving a sustained reduction in IOP.1 Advantages over the main alternatives—filtration or drainage device surgery—include technical ease of the procedure, quicker recovery and none of the risks specific to intraocular surgery. However, cyclodestructive procedures have been linked to subsequent loss of VA in a substantial proportion of those treated.4 5 As a result, this approach has generally been reserved for patients who have previously undergone multiple glaucoma operations, are poor surgical candidates or who already have limited visual potential.2 6

Resistance to the wider application of laser cyclotherapy is based, for the most part, on series of cases with VA significantly compromised by the glaucoma itself or, in the case of secondary glaucomas, its underlying cause such as retinal vascular disease or uveitis. Published data on the long-term prognosis for vision after cyclodiode treatment applied to patients with significant baseline VA are few (table 4).7–11

Table 4

Reported frequency of a two-line reduction in VA

In this cohort of patients, the largest we have identified to date with VA initially 20/60 or better, acuity was found to have declined from a median of 20/30 to 20/60. A two or more line reduction in Snellen acuity was seen in 30.6% of eyes and was reduced to <20/200 in 16.3%. The most frequent cause of visual loss was further progression of glaucoma, which was an attributable cause in half the cases. In a number of these, this occurred despite a significant and sustained response to treatment. The only other cause linked to more than one case of visual loss was cystoid macula oedema, which was a contributory factor in three. Notably in two of these cases, neither of whom had an intact posterior capsule, cystoid did not develop until 6 months or more after treatment. In a further case, it could be argued that cyclodiode therapy might have led to worsening of diabetic macula oedema.

In no case did persistent hypotony develop, a finding that is consistent with our previous reported experience using the relatively conservative, standardised treatment protocol used in these patients. Macular pucker, which has previously been reported as a cause of visual loss after laser cyclotherapy,8 was not seen clinically, although neither OCT nor fluorescein angiography was used.

Clearly, it is difficult to attribute the deterioration in vision in the patients in this cohort in part or wholly to the cyclodiode treatment. Some loss may have been coincidental reflecting the natural course of the underlying disease and some might have occurred whatever surgical intervention had been applied. Nevertheless, loss of two lines of VA in 30% over a mean 5-year follow-up period is concerning. However, other treatment options also carry a significant risk to vision. In the recent tube versus trabeculectomy study, patients with medically uncontrolled glaucoma who had previously undergone cataract and/or filtration surgery were randomised to insertion of a glaucoma drainage device or augmented trabeculectomy.12 Mean baseline VA was equivalent to that in our cohort before cyclodiode. A two or more Snellen line reduction in acuity was observed in 32% in the tube arm and 33% in the trabeculectomy group after 12 months. Even after excluding cataract as a cause, the figures were 27% and 24%, respectively. The equivalent 1-year figure for our cohort was 18% despite inclusion of a large number of patients with chronic uveitis, aphakia and other secondary glaucomas that were excluded from the tube versus trabeculectomy study. Similarly, in the Advanced Glaucoma Intervention Study, 40% of medically uncontrolled patents treated with either trabeculectomy or laser trabeculoplasty had a three-line reduction in VA during 5 years of follow-up.13 Loss of VA is, therefore, by no means the preserve of any single surgical modality.

The earlier use of laser cyclotherapy in eyes with good residual VA appears to be gaining support.10 11 Ansari et al11 presented a retrospective subgroup of 23 eyes with POAG and mean pretreatment VA in the range 20/20 to 20/120 who underwent a single application of transscleral cyclodiode between 60 and 160 J (mean 120 J). Mean VA did not deteriorate, although the final VA was worse in three cases (13%), two of which were due to cataract progression.

In another retrospective study of 21 eyes with prelaser VA 20/80 or better that had previously mostly undergone filtration surgery, cycloablation was applied either by diode or Nd:YAG laser.10 Although 17 of the 21 patients required additional laser and/or surgical treatment to further lower IOP, VA remained within one line of the pretreatment level in 81%.

A prospective study in Ghana even examined the effect of cyclodiode as a primary surgical treatment in one eye of patients with POAG.9 Although a reduction in VA was observed in 18 (23%) after a mean follow-up of 13.2 months, there was no significant difference in acuity between the fellow eyes treated medically and the diode-treated eye. In the 19 eyes with pretreatment VA 20/60 or better, only one had a decrease in VA.

Our study's main limitations are its retrospective nature and the absence of a control group to distinguish the adverse effects of the treatment from the natural history of the underlying disease. Although undertaking cyclodiode treatment in eyes with uncontrolled glaucoma but with good VA comes with a significant risk of visual loss, that risk does not appear to be any greater than for other, more conventional surgical interventions. This needs testing in a randomised study.

In the absence of such comparative data, this study indicates that cyclodiode treatment can be used effectively in eyes with good VA and may be a useful alternative surgical treatment in patients with difficult to manage glaucoma.



  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • At a glance
    Harminder S Dua Arun D Singh