YAG vitreolysis is a controversial treatment option for symptomatic floaters, largely due to the lack of evidence in the literature and long-term follow-up data. Our group published the first randomized, controlled trial evaluating YAG vitreolysis versus sham laser for Weiss ring floaters. At 6 months' follow-up, there were no significant adverse events, such as retinal tears or retinal detachments. Approximately half, or 53%, of patients reported significant or complete resolution of their floater symptoms. This was significantly lower than the objective improvement of 95% observed by masked grading of color photographs, suggesting a disconnect between objective and subjective outcome.1
Hahn et al. at the American Society of Retina Specialists Research and Safety in Therapeutics Committee conducted a surveillance of retinal specialists to determine potential risks of YAG vitreolysis. They reported 16 complications including glaucoma, cataract and posterior capsule defects, retinal hemorrhage, scotomas, retinal tear, retinal detachment, and increased floaters. These voluntarily reported complications were reported during a 6-month period and may be underreported.2 There was no denominator for this surveillance study and, thus, we do not know complication rates.
The aim of the present study was to report long-term complications and efficacy in patients enrolled in our original 6-month randomized controlled trial.
Patients and Methods
This study assessed the long-term safety and efficacy of a previously reported, single-center, masked, controlled, randomized clinical trial.1 The methods of the original study were previously described. The original 6-month trial included 52 patients at Ophthalmic Consultants of Boston, Boston, who were randomized to YAG vitreolysis (36 patients) or sham laser (16 patients). Patients had the option of enrolling in an open-label extension study if they received sham laser and wished to receive YAG vitreolysis treatment, or if they received YAG vitreolysis and had residual floaters that the treating surgeon (CPS) felt could be treated. A total of 15 sham and eight treated patients enrolled in the 6-month open-label extension study. Patients were treated at the time of enrollment and offered another treatment at 1 month if there were residual floaters and symptoms. The open-label extension study allowed sham-treated patients access to one or two YAG vitreolysis treatment sessions and previously treated patients access to a second or third treatment session. Patients were followed for 6 months with depressed dilated fundus examinations and intraocular pressure (IOP) measurement.
An attempt was made to call all patients in 2018, 2.2 to 3.1 years after enrolling in the original randomized, controlled trial, requesting participation in a final observational visit. A thorough interim history, survey data as below, Early Treatment Diabetic Retinopathy Study best-corrected visual acuity (ETDRS BCVA), color photography, applanation tonometry, slit-lamp exam, and indirect ophthalmoscopy with scleral depression were performed.
The primary outcome measures included subjective percentage of improvement (from 0% to 100%),3 10-point visual disturbance score as described by Singh,4 five-level qualitative scale described by Delaney et al.,3 and the 25-item National Eye Institute Visual Function Questionnaire (NEI VFQ-25). Secondary outcomes included objective change based on masked grading (by JSH) of color wide-angle fundus photography,3 ETDRS BCVA, and adverse events.
Reasons for not participating in this long-term observational visit included: one patient was deceased, five patients declined due to the inconvenience of travel to site, one patient declined due to workload, and 10 patients did not return three phone calls/voicemails.
No patients were billed for participation. All patients provided written informed consent before participating in the study and all personal patient data were removed. The study was conducted in adherence to the tenets of the Declaration of Helsinki, and the methods and execution were consistent with the International Conference on Harmonization guidelines for Good Clinical Practice. Study approval was obtained from the Sterling Institutional Review Board, Atlanta.
Continuous outcome variables were compared between treatment groups with a two-sample t-test and within treatment groups using a paired t-test (Excel; Microsoft Corporation, Redmond, WA). The five-level qualitative scale was analyzed as the proportion reporting significant or complete resolution using a two-sample test of proportion (Stata version 12.1; StataCorp, College Station, TX). A P value of less than .05 was considered to be statistically significant.
All 52 patients were contacted; 35 returned for this long-term observational visit. Twenty-seven patients were originally in the YAG vitreolysis arm and eight patients were originally in the sham arm; 24 were phakic and 11 pseudophakic. In this cohort of 35 patients, 23 patients had a single YAG session, 11 had two sessions, and one patient had three sessions. At final follow-up, the mean duration since last intervention was 2.3 years (range: 1.1–3.0; median 2.4).
One pseudophakic patient had worsening floaters and was treated with a pars plana vitrectomy. This patient was excluded in the analysis of the long-term follow-up cohort; he will be presented as a case example.
Symptoms on a Five-Level Qualitative Scale
At final follow-up, 50% of the 34 patients stated that their floaters were significantly or completely better, compared to 41.2% of this same cohort at 6 months (P = .47).
There was a 59.4% improvement in symptoms compared to baseline (range: 0% to 100%; median: 80%; 95% CI, 47.6 to 71.3%).
Among the 26 patients originally randomized to the YAG vitreolysis arm, floater symptoms improved nonsignificantly by 9.6% at the final visit compared to 6 months (range: −30 to 100; median: 1; P = .10, 95% CI, −1.4 to 20.6), supporting that the treatment effect was maintained between month 6 and a mean of 2.3 years. Three of these patients reported diminishment of treatment effect (−30% to −15%) at final visit.
Visual Disturbance Score on a 10-Point Scale
The improvement in the visual disturbance score compared to baseline was −3.1 (range: −10 to 2; median: −2; 95% CI, −4.1 to −2.1; P < .001).
Additional YAG Laser Treatments
There were nine patients who were in the original YAG laser group who had a second treatment session. Eight of these patients had a second treatment during the open-label extension study (one had two additional sessions, for a total of three), and one had a second treatment session with another physician. Of these nine patients, there was an additional nonsignificant 17.8% improvement in symptoms at final visit compared to month 6 (range: −30 to 100; median: 10; 95% CI, −2.3 to 37.9; P = .17).
NEI VFQ-25 Scores
There was improvement in the following parameters compared to baseline: near vision activities (baseline: 78.7, final: 85.8, difference: 7.1; range: −25 to 50; median: 0; 95% CI, 1.6 to 12.6; P = .016), distance vision (baseline: 84.6, final: 89.0, difference: 4.4; range: −16.7 to 41.7, median: 0; 95%CI 0.3 to 8.5; P = .042), mental health (baseline: 71.7, final: 85.7, difference: 14.0; range: −6.3 to 56.3; median: 12.5; 95% CI 9.1 to 18.8; P < .001), and role difficulties (baseline: 81.3 baseline, final: 93.4 final, difference: 12.1; range −12.5 to 75; median: 12.5; 95% CI, 5.6 to 18.7; P < .001).
There was no change in ETDRS BCVA (difference from baseline: 0.53; 95% CI, −1.3 to 2.3; P = .57).
The IOP was unchanged from baseline (difference: −0.09, range: −7 to 7; median: 0; 95% CI, −1.2 to 1.0; P = .88).
Objective Grading by Masked Reviewer
A masked grader (JSH) reviewed baseline and final wide-angle fundus photos and found the floaters were worse in one eye, unchanged in one eye, significantly improved in seven eyes, and completely resolved in 25 eyes. Of note, the eye receiving vitrectomy had complete resolution in floaters on objective grading of photos. Of the 34 eyes treated only with YAG vitreolysis, 94% were significantly or completely better at the final follow-up visit. This objective improvement was significantly higher than the subjective significant or complete improvement reported by 50% of patients (P < .001).
Retinal Tears: In three patients, retinal tears were not evident at the 6-month visit but were evident at final visit. The first had a retinal tear with a cuff of fluid and a demarcation line that developed sometime between month 6 and 2.8 years after YAG; it was observed. This patient had a remote history of retinal detachment in the fellow eye and had vitrectomy to repair it; thus, she may have been predisposed to tears. The second had a horseshoe tear developing between month 6 and 2.07 years that was laser demarcated. She reported new floaters within the last year of her final follow-up visit. The third had several small pseudophakic breaks noted at the time of vitrectomy (which was done 1.4 years after YAG vitreolysis, with no tears evident during his 6-month exam) that were lasered intraoperatively. He had worsening of floaters after 1 YAG vitreolysis session at 6 months, subjectively and objectively. He was 100% improved (complete resolution on the 5-level scale) 1.5 years after vitrectomy without any postoperative complications.
Worsening of Floaters: In addition to the patient described above, another patient reported worsening of floaters over time, reporting a 30% improvement at 6 months, but worsening to baseline at 2.42 years after her second YAG vitreolysis session. A third patient had worsening of floaters due to uveitis, described below.
Uveitis: A male patient was diagnosed with anterior and intermediate vitritis 0.9 years after a single session of YAG vitreolysis OD. His workup was negative, and he responded to topical Durezol (Alcon Laboratories, Fort Worth, TX) under management by a uveitis specialist.
Ocular Hypertension: One patient was diagnosed with bilateral ocular hypertension 0.6 years after 1 session of unilateral YAG vitreolyis.
In conclusion, half of those treated with YAG vitreolysis reported significant or complete resolution in their floater symptoms 2.3 years after final treatment, similar to satisfaction rates noted at 6 months.1 This long-term follow-up study suggests durability of treatment effect. Other long-term metrics were consistent with 6-month results, including the visual disturbance score and the self-reported percentage of improvement. Patients also reported improvements in near activities, distance activities, mental health, and role difficulties.
The improvement in mental health is intriguing, given that we feel many patients bothered by floaters tend to be anxious at baseline. The possibility of a laser procedure improving one's psychology requires further investigation.
A small subgroup analysis of nine patients undergoing a second YAG vitreolysis session at month 6 (one had a third session at month 7) reported an average of a 17.8% improvement in their floater symptoms compared to that at month 6. Though this trend was not statistically significant, the impact of second or third YAG vitreolysis treatment sessions on efficacy and safety should be evaluated in the future with larger trials to determine the optimal number of treatment sessions for specific floater types. Further, the present study did not utilize standardized parameters to determine which eyes should receive additional YAG vitreolysis sessions; it was largely based on the treating physician's (CPS) assessment of whether there were residual floaters that could be treated and were associated with persistent symptoms. Future studies should evaluate parameters warranting retreatment.
Similar to the results noted at 6 months, there was a significant disparity at 2.3 years of follow-up between the robust 94% objective improvement after YAG vitreolysis noted by a masked grader and the 50% subjective improvement reported by patients. It is possible that this patient population is still bothered by minor floaters not visible on wide-angle color photography, despite significant or complete objective improvement of their Weiss ring floater.
There were several adverse events noted between month 6 and the end of follow-up at 2.3 years, that were not evident within the first 6 months after YAG vitreolysis. The most significant were three eyes with delayed retinal tears that were noted 1.4 to 2.8 years after YAG vitreolysis. All of these tears were asymptomatic and detected on examination; they were not evident at month 6 on depressed fundus examination. Without a long-term control group, it is unclear if these delayed retinal tears are related to treatment or represent the inherent risk for these eyes. Interestingly, a single retinal tear through lattice degeneration occurred in a sham-treated patient during the original 6-month study, supporting some degree of inherent risk.1 The delayed tears reinforce the importance of long-term follow-up in eyes undergoing YAG vitreolysis, and patient education to return immediately with any symptoms.
There were several other minor adverse events noted during this follow-up study that might not be related to YAG vitreolysis. One patient described an initial 30% symptomatic improvement at 6 months after YAG vitreolysis, but then reported a progressive worsening of floater symptoms at 2.4 years of follow-up. This result likely relates to the progressive nature of vitreopathy in some patients. Another had worsening symptoms 6 months after treatment but was 100% satisfied after vitrectomy. One patient developed worsening floater symptoms from anterior and intermediate uveitis, 0.9 years after YAG vitreolysis. It is unlikely this is related to treatment, as there was no inflammation for the first 6 months after YAG vitreolysis. One patient developed ocular hypertension in both eyes shortly after her 6-month visit. Though YAG vitreolysis is associated with ocular hypertension and glaucoma,2,4 it is unlikely that unilateral treatment caused bilateral ocular hypertension in this case.
There are several weaknesses to this long-term follow-up study. There were no standardized guidelines dictating when additional YAG vitreolyisis should be performed. There was no control arm followed to the final follow-up of this observational study, precluding the ability to calculate the inherent rate of symptomatic variation and adverse events over time in this population. Only Weiss ring floaters were included in this study; the results are unlikely transferrable to other floater types. Two important metrics, contrast sensitivity and reading speed, were not assessed during this study; these should be included in future floater studies. Only 35 of the 52 patients enrolled in the initial randomized controlled trial voluntarily returned for the present long-term follow-up vision, allowing for the possibility of selection bias.
These long-term results show a sustained response to YAG vitreolysis after an average of 2.3 years after treatment, similar to the response reported at 6 months. The long-term adverse effect profile appears acceptable, though the finding of delayed retinal tears supports the need for a larger randomized, controlled trial with longer follow-up so that we may determine the true risks and rates of adverse events after YAG vitreolysis. Further, more work needs to be done to identify the response rates with different floater types, not just Weiss ring floaters. We need to understand how many treatment sessions are optimal, and when there are diminishing results. Finally, given that some patients had little to no response subjectively, despite a robust objective response, we need to develop tools to better screen for patient types — not just floater types — who may be more likely to respond subjectively and mirror their objective result.
- Shah CP, Heier JS. YAG Laser Vitreolysis vs Sham YAG Vitreolysis for Symptomatic Vitreous Floaters. JAMA Ophthalmol. 2017;135(9):918–923. doi:10.1001/jamaophthalmol.2017.2388 [CrossRef] PMID:28727887
- Hahn P, Schneider EW, Tabandeh H, Wong RW, Emerson GGAmerican Society of Retina Specialists Research and Safety in Therapeutics (ASRS ReST) Committee. Reported Complications Following Laser Vitreolysis. JAMA Ophthalmol. 2017;135(9):973–976. doi:10.1001/jamaophthalmol.2017.2477 [CrossRef] PMID:28750116
- Delaney YM, Oyinloye A, Benjamin L. Nd:YAG vitreolysis and pars plana vitrectomy: surgical treatment for vitreous floaters. Eye (Lond). 2002;16(1):21–26. doi:10.1038/sj.eye.6700026 [CrossRef] PMID:11913884
- Favre M, Goldmann H. [Genesis of posterior vitreus body detachment] [Article in German]. Ophthalmologica. 1956;132(2):87–97. doi:10.1159/000302922 [CrossRef] PMID:13378828