Small-Gauge Pars Plana Vitrectomy for Visually Significant Vitreous Floaters

Abstract

Purpose:

This article evaluates our experience at a retina-only private practice with small-gauge pars plana vitrectomy (PPV) for visually significant vitreous floaters. We review the surgical outcomes, complication rates, and percentage of second-eye surgery for the same indication.

Methods:

A retrospective, interventional case series was conducted of consecutive patients undergoing PPV for significant vitreous floaters from September 2014 to December 2018 at a high-volume vitreoretinal surgery practice. Preoperative visual acuity (VA), complication rates, and visual outcome following surgery were evaluated.

Results:

A total of 104 eyes in 81 patients underwent PPV for visually significant floaters; 35 (43.2%) patients had PPV in both eyes. Mean preoperative VA was 0.16 ± 0.17 logMAR (∼20/29 Snellen equivalent) and improved to 0.12 ± 0.15 logMAR (∼20/26 Snellen; Wilcoxon test, P = .008) at the last follow-up after PPV. All patients had improvement in VA at the final postoperative visit, with a VA of 20/40 or better achieved in 93.3% of cases. The complication rate of vitreous hemorrhage postoperatively was 0.96%. There were no cases of postoperative retinal tears, breaks, or endophthalmitis.

Conclusions:

Small-gauge PPV in the carefully selected patient is an effective and safe procedure to eliminate symptoms. VA following PPV for vitreous floaters significantly improved. Nearly half of the patients studied (43.2%) underwent PPV in the other eye.

Introduction

The human vitreous humor is a clear matrix within the eye consisting of water and macromolecules, such as hyaluronic acid and collagen, that undergoes age-related synchsis and syneresis, which lead to a posterior vitreous detachment (PVD). ¹ These intravitreal opacities cast shadows on the retina and demonstrate a behavior often described by patients as opacities “floating” in and out of their visual field. Therefore, this clinical circumstance has been clinically described as “vitreous floaters.” ²

Over the past decades, vitreous floaters have not been considered a severe problem in most clinical settings, underestimating the adverse impact that they have on patients’ activities of daily living, quality of life, and vision. In 2011, Wagle et al, through a questionnaire, found that people were willing to take an 11% risk of death and a 7% risk of blindness to get rid of symptoms relating to floaters. ² Recently, more patients are seeking clinical treatment to improve their symptoms. Some people are bothered by floaters to such an extent that treatment is required to control their symptoms.

Neodymium-doped yttrium aluminum garnet (Nd:YAG) laser vitreolysis has been increasingly offered as a treatment of vitreous floaters. This noninvasive method of treatment uses laser to disrupt the vitreous opacities. It tends to be most effective for the treatment of central, well-defined vitreous floaters. ¹ In a study evaluating complications after laser vitreolysis, Hahn et al reported complications that included retinal tear, scotomas, increase in floaters, and posterior capsule defects. The American Society of Retina Specialists Research and Safety in Therapeutics committee recommended further studies to evaluate the efficacy and complication rates of laser vitreolysis in a prospective manner. ³

Currently, pars plana vitrectomy (PPV) seems to be the only effective surgical approach for the removal of vitreous opacities, with prior reports indicating high patient satisfaction. ⁴ Previous studies on this topic have included both phakic and pseudophakic patients, large- and small-gauge instrumentation, and also evaluated eyes that did not have a PVD at the time of vitrectomy surgery. ^{4 -10} It is likely that some of these various variables have contributed to increased complication rates and less than favorable postoperative outcomes. The surgeons involved in our study followed strict criteria, which involved evaluating the duration of preoperative symptoms, lenticular status, and presence of a PVD.

The aim of this study is to evaluate our experience with PPV for vitreous floaters and to assess the complication rate, surgical outcomes, and percentage of second eye surgery.

Methods

A retrospective review of written and electronic medical records was completed of patients undergoing surgical repair for visually significant vitreous floaters with 23- or 25-gauge PPV performed by 5 surgeons at a single vitreoretinal surgery practice from September 2014 to December 2018.

Approval for this HIPAA (Health Insurance Portability and Accountability Act)–compliant, retrospective analysis was obtained from Advarra Institutional Review Board for Human Subjects Research. All cases undergoing PPV were identified using Current Procedural Terminology codes 67036, 67039, and 67040. Inclusion criteria included the subjective complaint of significant visual disturbance due to vitreous floaters for more than 6 months, pseudophakia, and the presence of a PVD confirmed on clinical examination. Medical-record review of clinic notes, operative reports, and surgical logs were used to confirm these cases.

Demographic data including age and sex, initial ophthalmic presentation, ophthalmic findings, and any intraoperative or postoperative complications were recorded. All patients underwent examination on postoperative day 1, week 1, and month 1 after surgery. A minimum follow-up duration of 3 months was required to study surgical outcomes, including visual acuity (VA) improvement and its Snellen equivalent. Exclusion criteria included history of venous or arterial occlusive disease, advanced glaucoma or age-related macular degeneration, previous retinal detachment, endophthalmitis, or uveitis.

Using Microsoft Excel, the mean and SD of preoperative and postoperative VA were calculated after converting to the logMAR equivalent of each patient’s Snellen VA. Counting fingers or hand motion VA were recorded as 20/2000 (logMAR 2.00) and 20/20000 (logMAR 3.00), respectively. VA was recorded at time of decision for surgical intervention and at the most recent visit following surgery. The Wilcoxon signed-ranks test was used for statistical significance when comparing the VA between the preoperative values, specifically prior to PPV, and those obtained following surgery at the most recent visit.

Results

Table 1 demonstrates preoperative and postoperative data and demographics. A total of 104 eyes in 81 patients were included in the study; 35 patients underwent surgery in both eyes (43.2%). A total of 36 male and 45 female patients with an average age of 69 years (range, 53-82 years) were included in the study. All eyes were pseudophakic at the time of PPV. Two eyes had previous retinal tears that were treated with laser barricade in the clinic. All eyes included in the study underwent a single 23- or 25-gauge PPV.

Table 1.

Demographic and Preoperative and Postoperative Data.

	No.	%
Eyes	104
Right	51	49.0
Left	53	50.9
Patients	81
Male	36	44.4
Female	45	55.6
Age, y
Mean ± SD	69 ± 6.5
Range	53-82
Pseudophakic prior to PPV	104	100
Preoperative BCVA, mean ± SD, logMAR	0.16 ± 0.17
Postoperative BCVA, mean ± SD, logMAR	0.12 ± 0.15

Abbreviations: BCVA, best-corrected visual acuity; PPV, pars plana vitrectomy.

In 4 eyes (3.8%), ocular hypotension (intraocular pressure [IOP] < 9 mm Hg; range, 4-7 mm Hg) was noted at the first postoperative visit. This resolved spontaneously by the postoperative week 1 visit. Three of the 4 eyes subsequently developed ocular hypertension that required temporary topical treatment. In 36 eyes (34.6%), ocular hypertension (IOP > 21 mm Hg; range, 22-46 mm Hg, mean 26.5 mm Hg) was documented within the 3-month postoperative period; one of these had a mild hyphema at the first postoperative visit that resolved 9 days after surgery. Eleven eyes received topical therapy, and 1 patient also required a short course of oral acetazolamide. Five patients (4.8%) required longer-term use of 1 glaucoma drop and were still on the single drop at their last follow-up visit.

One patient (0.9%) presented with a mild vitreous hemorrhage on the first postoperative visit that resolved spontaneously within 1 month (Table 2). Neither of the 2 patients who had treated retinal tears prior to PPV developed a new retinal tear during follow-up. There were no new retinal tears, detachments, or cases of endophthalmitis.

Table 2.

Postoperative Complications.

	No.	%
Retinal detachment	0	0
Vitreous hemorrhage	1	0.9
Retinal tear	0	0
Endophthalmitis	0	0
Transient ocular hypertension (within postoperative period)	36	34.6
Ocular hypertension (at last follow-up)	5	4.8

The mean duration of follow-up was 473 days (range, 95-1300 days). VA of 20/40 or better was tested postoperatively in 97 eyes at their most recent visit. The mean preoperative VA (prior to PPV for vitreous floaters) was 0.16 ± 0.17 logMAR (Snellen equivalent, ∼20/29 ) and improved to 0.12 ± 0.15 logMAR (Snellen, ∼20/26; Wilcoxon test, P = .008) at the last known follow-up after PPV (average, 473 days after surgery; range, 95-1300 days).

Conclusions

Vitreous floaters can be visually significant and impair patient quality of life by interfering with daily activities that range from reading to driving. Few management options can be discussed with patients, including observation, Nd:YAG laser vitreolysis, and PPV. As previously discussed, Nd:YAG laser vitreolysis has known risks including but not limited to retinal tears, scotomas, increase in floaters, and posterior capsule defects, for which the American Society of Retina Specialists Research and Safety in Therapeutics committee recommended additional investigation. ³ Moreover, Delaney et al found that 61.5% of patients who underwent Nd:YAG laser vitreolysis found no symptomatic improvement. ⁵ In contrast, Shah and Heier ¹¹ found a statistically significant symptomatic improvement in the Nd:YAG vitreolysis group in a randomized clinical trial that compared Nd:YAG vitreolysis vs sham Nd:YAG vitreolysis for symptomatic vitreous floaters in 52 patients. Nevertheless, PPV appears to be a more effective option in the appropriate patient for visually significant floaters.

We present a series of patients who had visually significantly vitreous floaters and underwent vitrectomy surgery over a 4-year period. All surgeons followed strict criteria to sign up patients for surgery: The duration of symptoms must have been greater than 6 months, and the presence of pseudophakic status and PVD on examination were mandatory. Patients with posterior ocular pathology including prior retinal detachment, vein occlusion, uveitis, or glaucoma were excluded. Abiding by these criteria, we found that our complication risk of postoperative retinal tears or detachments was zero. There was 1 eye (0.9%) that presented with postoperative vitreous hemorrhage on day 1 that spontaneously resolved by postoperative day 36. This finding is similar to the previously reported 1.03% rate of clinically significant vitreous hemorrhage noted after PPV for vitreous floaters. ⁸ After spontaneous resolution of vitreous hemorrhage, that patient went on to have excellent VA of 20/20 and no retinal tears or breaks documented at subsequent visits.

In a previous study in which a PVD was not present in all patients preoperatively, and a PVD was not induced during surgery, 4 of the 195 cases (2.1%) developed recurrent floaters from developing a PVD. ⁸ Those patients required additional vitrectomy surgery. By confirming preoperative PVD on examination, we did not have to induce a PVD in any of the cases in the present study. This criterion was likely the most important factor in reducing iatrogenic retinal breaks. Previous studies in which PVD was induced during surgery reported higher rates of retinal tears and detachments. ¹²

Another factor contributing to the low complication rate in our study was the exclusive use of small-gauge vitrectomy systems. Many studies have shown reduced complications with 23- and 25-gauge vitrectomy compared with 20-gauge, including reduced retinal and vitreous incarnation at sclerotomy site, fewer iatrogenic breaks, and less dialyses at the vitreous base. ^{12 -17} De Nie et al ⁶ reviewed 110 vitrectomies for vitreous floaters over 12 years, of which more than 50% underwent 20-gauge PPV with a high postoperative retinal detachment rate of 11%. The smaller vitrectomy probes allow more precise and controlled movements, resulting in safer removal of vitreous with reduced traction on the retina. Moreover, with smaller-gauge vitrectomy the transconjunctival sutureless approach provides less discomfort for the patient and increased surgical efficiency.

We had 4 cases of postoperative hypotony on postoperative day 1 that resolved by the 1-week follow-up, similar to previous reports with sutureless small-gauge valved cannulas. ^{14 -17} There were 36 eyes (34.6%) with elevated IOP that resolved, with topical therapy in most cases and oral acetazolamide in 1 eye, within the postoperative period by an average of 66 days postoperatively (range, 9-344 days). Five eyes (4.8%) required topical therapy (1 drop) to maintain normal IOP at the most recent follow-up. None of the patients in this study had a known history of elevated IOP, but it is important to monitor postoperative IOP. Carrying a diagnosis of glaucoma does not preclude a patient from undergoing PPV for floaters. However, because PPV is an elective procedure, it is important to discuss the risk of possible postoperative elevated IOP.

Unlike previous studies in which less than half of the participants’ eyes were pseudophakic, we included only pseudophakic eyes. ^7,9 We therefore had zero cases of cataract formation after vitrectomy and were able to perform a more thorough anterior vitrectomy and peripheral vitreous base shaving.

The patient cohort in this study did show improved vision in all patients. Previous studies have reported similar findings of equivalent VA previtrectomy and postvitrectomy for visually significant floaters. ⁴

There are limitations to our present study. First, this was a retrospective study. Unfortunately, there have been no randomized controlled trials on vitrectomy for vitreous floaters—only retrospective case reports and retrospective cohort studies. Second, refractions were not routinely performed, which limit the best-corrected preoperative and postoperative VA reported in this study. Finally, we did not perform a questionnaire quantifying patients’ preoperative dissatisfaction with floaters on quality of life, nor did we perform a questionnaire quantifying satisfaction postoperatively. Per our medical-record review of the patient history and subjective changes, nearly all patients reported subjective improvement in vision and overall satisfaction with the surgical outcome. As a result, nearly half (43.2%) of the patients elected to undergo PPV for visually significant floaters in the other eye. Going forward, additional evaluation including contrast sensitivity and quality-of-life measures could help improve quantification of patient satisfaction and subjective visual changes.

PPV is an effective and safe way to address visually significant vitreous opacities. With careful patient selection, the benefits of surgical intervention can greatly outweigh the risks.