F loaters are hair- or fly-like obscurations in vision at times appearing as cobwebs, but occasionally having a “glass noodle” appearance due to remnants of fetal blood vessels. ^, Studies have found that vitreous floaters can significantly impact visual quality-of-life (QoL). ^,,, Posterior vitreous detachment (PVD) is the most common etiology of vitreous floaters in older individuals, ^,,,, while the most common cause in younger people is myopic vitreopathy, ^,, where fibrous liquefaction with aggregation of collagen fibrils and PVD occur earlier in life. ^,,,

Patients complaining of bothersome vitreous floaters usually have degradation in contrast sensitivity (CS) ^,,,, that correlates with increased vitreous density as measured by quantitative ultrasonography (QUS). ^, Since QUS has recently been shown to correlate with vitreous light scattering, this test is an accurate clinical surrogate for measuring light scattering in vivo . The term “ Vision Degrading Myodesopsia ” (VDM) refers to clinically significant cases which are diagnosed on the basis of these objective quantitative metrics. ^, However, since people have varying perceptions of disturbances to their health it is difficult to accurately determine the impact of VDM from vitreous floaters on QoL based solely upon a patient’s history. This impediment may have colored the experience of some ophthalmologists to regard patient complaints of vitreous floaters as overblown.

Recently, an informative method of assessing subjective patient disturbance from vitreous floaters was developed and validated, called the Vitreous Floaters Functional Questionnaire (VFFQ-23). The present study was undertaken to compare the VFFQ-23 to the NEI Visual Function Questionnaire (VFQ-25) ^, in terms of the relative correlations of the 2 questionnaires with vision (visual acuity (VA) and CS), vitreous structure assessed by quantitative ultrasonography, and patient management choice (observation (OBS) vs. limited refractive vitrectomy (LRV)). Questionnaire findings in the LRV group quantified Q-o-L changes post-operatively. Since the VFFQ-23 is specifically designed to evaluate floaters and the VFQ-25 assesses general vision, it is hypothesized that the VFFQ-23 will be superior to the VFQ-25 in several, if not all, of these comparisons.

This study was conducted in patients with bilateral symptomatic vitreous floaters evaluated at the VMR Institute for Vitreous Macula Retina in Huntington Beach, California (JS). Institutional Review Board (IRB) approval was granted by the Providence St. Joseph Hospital IRB and written informed consent was obtained from each subject. This study adhered to the principles of the Health Insurance Portability and Accountability Act (HIPAA) and was conducted in accordance with the ethical principles of the Declaration of Helsinki.

Correlations between the questionnaires and visual function (VA and CS) and vitreous structure (QUS) were evaluated retrospectively. Changes in the composite scores of the 2 questionnaires in patients who underwent LRV surgery were analyzed prospectively.

The study population for retrospective analyses consisted of all patients who presented to the VMR Institute for Vitreous Macula Retina in Huntington Beach, California between 2013 and 2024 with the chief complaint of bilateral “floaters”, met the inclusion/exclusion criteria, and who completed both questionnaires at study entry. For the prospective analyses, operated patients had to be evaluated with all testing prior to any intervention, had to have undergone bilateral (on separate occasions) limited refractive vitrectomy (JS), and completed both questionnaires at study entry as well as on 4 subsequent occasions over a 3-year post-op period to be included in the study. Meeting these criteria were 193 subjects (99 males and 94 females; age range 19 to 95 years) with a subjective duration of floaters prior to study entry = 57.9 +/- 68.7 months. Exclusion criteria were exudative maculopathy, diabetic retinopathy, glaucoma, retinal detachment, keratopathy, or any co-morbidities that could affect contrast sensitivity or alter vitreous structure. Patients with a history of YAG laser vitreolysis or vitrectomy (for any related or unrelated condition(s)) were excluded. Patients with multifocal intraocular lenses were excluded due to potential effects on contrast sensitivity. There were 74 asymptomatic control subjects (39 males and 35 females) who completed both questionnaires and underwent contrast sensitivity testing as well as quantitative ultrasonography. Repeatability was tested in 24 OBS subjects who had all testing done on 3 separate occasions: study entry, 1 year, and 2 years later.

Patients electing LRV underwent sutureless 25-gauge pars plana vitrectomy (Constellation, Alcon, Fort Worth, Texas) performed under local anesthesia, as previously described. ^,,, In cases where the posterior vitreous was still attached to the fundus, PVD was not intentionally induced to reduce the risk of iatrogenic retinal breaks and detachment, ^,,, as well as to minimize post-operative elevation of intravitreal oxygen levels that can promote cataractogenesis. In phakic patients, 3-4 mm of retrolental vitreous was left intact to further mitigate cataract formation. ^,,,

Since all experimental subjects had bilateral floaters, the results of visual function testing and vitreous structure assessments with ultrasound were averaged for the 2 eyes and results presented per subject (see tables and graphs).

Multiple variable logistic regression analysis (SPSS; IBM) was used to determine differences in demographics (gender: male or female) and clinical findings (myopia, PVD, lens status) between OBS and LRV groups. Test-retest reliability was determined using single-measurement, absolute-agreement, 2-way, mixed-effects intraclass correlation coefficient analyses (SPSS, IBM). Two sample t-tests were used to determine differences in quantitative values of age, CS, QUS, VFFQ-23, and VFQ-25 composite scores between OBS and LRV subjects. Pearson’s correlation analyses evaluated the relationship between the 2 questionnaires (VFFQ-23 and VFQ-25) as well as between each questionnaire and VA, CS, and QUS. In patients undergoing LRV, a paired t-test compared pre-operative CS, QUS, VFFQ-23, and VFQ-25 to post-operative measures at 6, 12, 24, and 36 months. In subjects who had bilateral LRV, post-operative evaluations were performed only after both eyes underwent surgery. The differences between the 2 questionnaires with respect to the changes in post-operative findings were evaluated using difference-in-differences calculations.

To evaluate the relative ability of the 2 questionnaires to predict a patient’s management choice (OBS vs. LRV), sensitivity, specificity, positive predictive value (PPV), negative predictive value, and accuracy were calculated using standard definitions. ^, A receiver operating characteristic (ROC) curve was generated to illustrate the trade-off between sensitivity and specificity of the questionnaire composite scores in distinguishing patients who chose OBS from those electing LRV surgery. Accuracy calculations were used to explore cutoff values of each questionnaire composite score vis à vis the likelihood of surgery. Lastly, to evaluate potential clinical usefulness in identifying patients likely to elect surgery, Decision Curve Analyses (DCA) were performed across a range of clinically relevant threshold probabilities as previously described. ^, DCA incorporates the trade-off between true positives and false positives and provides a more comprehensive assessment of net clinical benefit than positive predictive value alone. The analyses included calculation of the VFFQ-23 predictive value using a logistic regression model with a dichotomous outcome (limited refractive vitrectomy vs. observation), as described by Vickers and colleagues. ^,