Effects of Loss-to-Follow-Up on Functional and Anatomic Outcomes in Patients with Neovascular Age-Related Macular Degeneration

doi:10.21203/rs.3.rs-5033887/v1

Purpose: To study the effect of loss-to-follow-up (LTFU) in patients receiving anti-vascular endothelial growth factor (VEGF) injections for neovascular age-related macular degeneration (nAMD).

Patients and Methods: Retrospective, multi-center, cohort study of patients that received intravitreal bevacuzimab, ranibizumab, or aflibercept between January 2020 and March 2020 who were lost to follow-up.

Results: A total of 7114 eyes of 5727 patients met the study inclusion criteria. At the first return visit after LTFU, logMAR VA was 0.85 (Snellen 20/142) in the LTFU group compared to 0.67 (Snellen 20/94, p<0.001) in the control group. On the final study visit, VA in the LTFU group remained decreased at 0.86 (Snellen 20/145, p<0.001). For the LTFU group, mean CFT increased when comparing the visit before LTFU (280 µm) to the return visit (295 µm, p<0.001), but there was no difference at the final study visit (277 µm, p=0.42). Despite CFT returning back to GSD values by the final study visit for those LTFU, VA did not recover.

Conclusions: Patients with nAMD being treated with anti-VEGF who were lost to follow-up during the COVID-19 pandemic experienced irrecoverable loss in VA during the duration of the study, despite resumption of anti-VEGF therapy.

Anti-vascular endothelial growth factor

macular degeneration

pandemic

This study evaluates the effect on vision and other parameters in patients with wet age-related macular degeneration who had lapses in their treatment.

Age-related macular degeneration (AMD) is the most common cause of visual impairment in patients greater than 60 years of age in the United States, and the developed world’s leading cause of irreversible blindness.¹As life expectancy continues to rise, the global prevalence of AMD has been reported as high as 9%.² Patients with neovascular AMD are at risk of profound visual impairment with a detrimental effect on quality of life.³ The use of anti-vascular endothelial growth factor (VEGF) for the treatment of neovascular AMD (nAMD) revolutionized the management of this disease. Currently, there are three US Food and Drug Administration (FDA) approved anti-VEGF agents available for the treatment of nAMD: aflibercept (Eylea; Regeneron Pharmaceuticals, Inc, Tarrytown, NY), brolucizumab (Beovu; Novartis, East Hanover, NJ), and ranibizumab (Lucentis; Genentech, Inc., South San Francisco, CA), while bevacizumab (Avastin; Genentech, Inc., South San Francisco, CA) has also been used off-label.

Ranibizumab, a humanized monoclonal fab fragment, was found to be effective in the treatment of nAMD in two landmark studies—the Minimally Classic/Occult Trial of the Anti-VEGF Antibody Ranibizumab in the Treatment of Neovascular AMD (MARINA) and Anti-VEGF Antibody for the Treatment of Predominantly Classic Choroidal Neovascularization in AMD (ANCHOR).^4,5 These trials found that approximately 90% of patients receiving monthly intravitreal treatment with ranibizumab lost < 15 letters of best-corrected visual acuity (BCVA) after 2 years.^6,7 Aflibercept, a recombinant fusion protein that serves as soluble decoy receptors for VEGF,⁷ has demonstrated similar efficacy in the treatment of nAMD with administration monthly or every two months.⁸ Bevacizumab has shown similar efficacy to ranibizumab in some studies.^9–11

Moreover, studies such as the Lucentis Compared to Avastin Study (LUCAS) found that a treat-and-extend regimen with bevacizumab or ranibizumab in patients with nAMD was associated with an improvement in BCVA of 7.9 letters and 8.2 letters, respectively, after one year of treatment.¹² The Seven-Year Outcomes in Ranibizumab-Treated Patients in ANCHOR, MARINA, and HORIZON: A Multi-center Cohort Study (SEVEN-UP) demonstrated the visual benefit of anti-VEGF injections was more nuanced, 37% of eyes showed a BCVA of at least 20/70 vision after a mean of 7.3 years of intensive ranibizumab treatment, and another 37% of with BCVA ≤20/200.¹²

In December of 2019, a novel severe acute respiratory syndrome coronavirus-2 (Sars-Cov-2) emerged causing a global pandemic.¹³ Given risks associated with face-to-face examinations, many patients deferred seeking medical care out of fear of exposure to the virus, and many practices limited patient encounters to only urgent and emergent visits.¹⁴ Such changes were noted not only across medicine overall, but also within the field of retina. A study by Breazzano et al¹⁵ found an over 35% decrease in the number of intravitreal injections administered at the height of the pandemic across various institutions in the US. This was further corroborated by Leng et al¹⁶ in a IRIS registry report, demonstrating a decrease in the total visit volume from 2.9 million to 1.1 million (62%) and an 8% decline (330,000 to 304,000) in the number of anti-VEGF injections for nAMD, retinal vein occlusion (RVO), and diabetic macular edema (DME) from March to April 2020. In this study, the mean BCVA was stable during the baseline period through July 2020 among all three cohorts.

As the COVID-19 pandemic greatly affected the delivery of care to patients across all medical specialties, we sought to determine the functional and anatomic outcomes in patients with nAMD that were LTFU. This study provides novel insights to the effect of LTFU during the COVID-19 pandemic and beyond.

Study design

This was a multi-center retrospective cohort study. The approval of the Institutional Review Boards (IRB) at New York University Langone Health and Wills Eye Hospital was obtained prior to the commencement of the study. This study was conducted in accordance with the Health Insurance Portability and Accountability Act and adhered to the tenets of the Declaration of Helsinki. Consent to participate was deemed unnecessary by the IRB as this is a retrospective study. Patients with a history of neovascular age related macular degeneration (nAMD) receiving intravitreal injections of anti-vascular endothelial growth factor agents (VEGF) between January 1, 2020 and March 1, 2020 were identified using International Classification of Disease-10 and Current Procedural Terminology billing codes.

Patient characteristics

Patients were stratified into those with loss-to-follow-up (LTFU) and those who continued on the intended interval schedule (controls). Only patients with nAMD that received intravitreal injection with bevacizumab, ranibizumab, and aflibercept between January 2020 and March 2020 were included. The diagnosis at each injection was obtained to ensure the patient was not receiving anti-VEGF therapy for a diagnosis other than nAMD. Best available Snellen visual acuity (VA) measurements were obtained and the VA was converted to the logarithm of the minimum angle of resolution (logMAR) for analysis.

Definition of LTFU

The recommended injection interval was obtained from the electronic medical record at the last documented visit prior to March 1, 2020. Patients were considered LTFU if the visit interval exceeded the designated follow-up time by 30 days. For visits that had a recommended interval for follow-up (e.g., “follow-up within 4–6 weeks”), the latest date recommended was utilized. For the control group, a standardized date, closest to March 1, 2020, was used. The date of LTFU and the standardized date are collectively referred to as the global standard date (GSD).

Temporal analysis

Clinical data, including visual acuity, Snellen lines gained/lost, intraocular pressure, and central foveal thickness (CFT) on optical coherence tomography (OCT), were collected at set time points. These time points were 6 months prior to GSD, 3 months prior to GSD, the GSD, first return visit, 3 month follow-up visit, 6 month follow-up visit, and the final study visit. The GSD was used as the reference point for the temporal analysis in both the control and LTFU groups. The visits after the GSD (3 month and 6 month) used the standardized date as a reference for control patients and the return date for patients LTFU.

Statistical analysis

All statistical analyses were conducted using SPSS 24 (Armonk, NY: IBM Corp.). Categorical variables were evaluated using a chi-square test. Continuous variables were converted into categorical variables based on distribution and clinical relevance. A student t-test was used to analyze continuous variables. A p-value of < 0.05 was considered statistically significant.

A total of 7114 eyes from 5727 patients met inclusion criteria. Of these eyes, 6564 had return visits after the GSD. There were 1607 out of 7114 eyes (22.6%) that were LTFU (with or without a return visit). Five hundred and fifty of the 7114 eyes did not return after the GSD and were excluded from further analysis. There were 1057 /6564 eyes (16.1%) LTFU with a return visit after the GSD. Only those eyes with a return visit after the GSD were included in analysis. The mean number of days eyes were LTFU after the recommended date of return was 37 days (SD 33). There were 633 females (71.5%) and 2847 females (64.6%, p < 0.001) in the LTFU and control groups, respectively. The mean age of the LTFU and control groups was 84.0 years (SD 9.4) and 81.7 years (SD 9.2, p<0.001), respectively. The difference in the mean recommended treatment interval prior to the GSD for the LTFU and control group was not statistically different, 58.4 days (SD 23.6) and 59.9 days (SD 24.6), respectively (p=0.051) (Table 1).

Visual Acuity

The logarithm of minimum angle of resolution visual acuity (logMAR) 6 months and 3 months prior to the GSD was 0.60 (Snellen 20/80, SD 0.57, n=4661) and 0.62 (Snellen 20/83, SD 0.58, n=5039) for the control group, respectively, and 0.70 (Snellen 20/100, SD 0.57, n=823, p<0.001) and 0.71 (Snellen 20/102, SD 0.58, n=939, p<0.001) for the LTFU group, respectively. The VA at the GSD was 0.63 (Snellen 20/85, SD 0.59, n=5465) for the control group and 0.74 (Snellen 20/110, SD 0.60, n=1055, p<0.001) for the LTFU group. The VA at the first return for the control group was 0.67 (Snellen 20/93, SD 0.60, n=5175) and had decreased to 0.85 (Snellen 20/142, SD 0.62, n=1044, p<0.001) for the LTFU group. The VA 3 and 6 months after the standardized/return from LTFU was 0.67 (Snellen 20/93, SD 0.59, n=5205) and 0.67 (Snellen 20/93, SD 0.59, n=5020) for the control group, respectively, and 0.84 (20/139, SD 0.63, n=884, p<0.001) and 0.82 (Snellen 20/132, SD 0.63, n=734, p<0.001) for the LTFU group, respectively. The VA at the final study visit was 0.68 (Snellen 20/96, SD 0.62, n=5482) for the control group and remained decreased at 0.86 (Snellen 20/145, SD 0.65, n=1049, p<0.001) for the LTFU group (Table 2). Note the largest mean difference in logMAR VA between the control and LTFU groups occurs at the return and final study visit. There was a steady decline in vision in patients who were LTFU compared to those in the control (Figure 1).

Number of lines lost

Eyes in the control group lost 0.40 lines (SD 2.42) on return visit (n=5146) compared to 1.12 lines (SD 3.42) in the LTFU group (n=1043, p<0.001). At 6 months after the standardized visit/return from LTFU, eyes in the control group (n=4985) had lost approximately 0.57 lines (SD 3.00) compared to 1.03 lines (SD 3.75, n=733, p<0.001) in the LTFU group. At the final study visit, eyes in the control group lost 0.57 lines (SD 3.36, n=5443) compared to 1.19 lines (SD 3.90, n=1048, p<0.001) in the LTFU group.

Greater than or equal to 2 lines of vision lost

Compared to the standardized visit, 13%, 17%, and 19% of eyes in the control group lost ≥2 lines of Snellen VA at the first return visit, 6 months after GSD, and their final study visit, respectively. In the LTFU group, compared to the last visit before LTFU, 25%, 16%, and 25% of eyes lost ≥2 lines of Snellen VA at the first return visit, 6 months later, and their final study visit, respectively. Compared to the control versus LTFU groups, the percentage of patients losing ≥2 lines at the return visit and the final study visit was found to be statistically significant (p <0.001 at both time points) (Figure 2, Table 2).

Greater than or equal to 3 lines of vison lost

Compared to the standardized visit, 9%, 11%, and 14% of control eyes lost ≥3 lines of Snellen VA at the first return visit, 6 months after GSD, and their final study visit, respectively. In the LTFU group, compared to the last visit before LTFU, 20%, 11%, and 19% lost ≥ 3 lines of Snellen VA at the first return visit, 6 months later, and their final study visit, respectively. Compared to the control versus LTFU groups, the percentage of patients losing ≥3 lines at the return visit and the final study visit was found to be statistically significant (p<0.001 at both time points) (Figure 2).

Change in visual acuity in eyes with mild vision loss (endpoint of 20/40 or better)

The VA at 6 and 3 months prior to the GSD was 0.23 (Snellen 20/34, SD 0.17, n=2045) and 0.23 (Snellen 20/34, SD 0.15, n=2185) for the control group, respectively, and 0.26 (20/36, SD 0.19, n=266, p=0.02) and 0.25 (Snellen 20/35, SD 0.14, n=295, p=0.009) for the LTFU group, respectively. The VA at the GSD was 0.20 (Snellen 20/32, SD 0.09) for the control group (n=2349) and 0.21 (Snellen 20/32, SD 0.09) for the LTFU group (n=323, p=0.001). The VA at first return visit was 0.25 (Snellen 20/36, SD 0.17, n=2194) for the control group and 0.33 (Snellen 20/43, SD 0.26, n=319, p<0.001) for the LTFU group. The VA at 3 and 6 months after standardized/return from LTFU visit was 0.26 (Snellen 20/37, SD 0.17, n=2224) and 0.28 (20/38, SD 0.21, n=2192) for the control group, respectively, and 0.32 (Snellen 20/42, SD 0.24, n=270, p<0.001) and 0.34 (Snellen 20/44, SD 0.29, n=231, p=0.001) for the LTFU group, respectively. The VA at the final study visit was 0.27 (Snellen 20/37, SD 0.23, n=2345) for the control group and 0.33 (Snellen 20/43, SD 0.28, n=322, p<0.001) for the LTFU group.

Central Foveal Thickness

The CFT at 6 months prior to the GSD was 278 mm (SD 97, n=4353) for the control group and 282 mm (SD 108, n=798, p=0.28) for the LTFU group. The CFT at the GSD was 275 mm (SD 97, n=5129) for the control group and 280 mm (SD 104, n=1011, p=0.19) for the LTFU group. The CFT at return visit was 270 mm (SD 97, n=4635) for the control group and 295 mm (SD 119, n=1003,p<0.001) for the LTFU group. The CFT at 6 months after the standardized/return from LTFU visit was 270 mm (SD 96, n=4900) for the control group and 279 mm (SD 111, n=711, p=0.04) for the LTFU group. The CFT at the final study visit was 267 mm (SD 97, n=5357) for the control group and 277 mm (SD 112, n=1011, p=0.008) for the LTFU group (p=0.008) (Table 2 and Figure 3).

The COVID-19 pandemic greatly affected timely delivery of care to patients with nAMD. The associated municipal and statewide shutdowns imposed an unprecedented delay in care for neovascular AMD (nAMD) in the anti-vascular endothelial growth factor (VEGF) era. This study sought to provide real world data on the impact of loss to follow-up (LTFU) on functional and anatomic outcomes in patients with nAMD during this time. The results of this study demonstrated meaningful functional and anatomic decline in eyes with nAMD that were LTFU which did not seem to improve during the timeline of the study.

Regularly scheduled intravitreal anti-VEGF injections are essential for the improvement or maintenance of visual acuity (VA) in patients with nAMD. The MARINA⁴ and ANCHOR⁵ studies demonstrated that monthly administration of anti-VEGF agents in patients with nAMD improved visual outcomes. Shortly after, the PIER study¹⁷ demonstrated that quarterly administration of anti-VEGF resulted in vision loss. In the management of nAMD, reducing treatment burden while optimizing visual outcome, often implemented as a treat-and-extend approach, is fundamental.

In 2017, Kim et al¹⁸demonstrated the impact of treatment discontinuation in nAMD patients previously receiving anti-VEGF. The authors found that mean VA at the visit prior to treatment discontinuation (Snellen 20/209) was significantly worse than at 24 months (Snellen 20/796, p <0.001). Going further, Soares et al¹⁹ found that among 93 eyes of 77 patients with nAMD that were LTFU for a mean of 346 days, 53.7% of patients had worse median VA by the final study visit. The median VA decreased from 0.60 (Snellen 20/80) at the visit before LTFU to 1.00 (Snellen 20/200) at the return visit (p < 0.001). Despite a mean follow-up of 346 days, and 5 anti-VEGF injections, median VA remained 1.0 (Snellen 20/200, p < 0.001). This effect was regardless of the restoration of central foveal thickness (CFT). This prompted investigators, such as by Ramakrishnan et al²⁰, to study the impact of visit adherence on visual outcomes. The authors found that patients that were considered “very late” in their adherence had significantly worse visual outcomes compared to patients who were “on time” to their appointments (Table 3).

In this retrospective study, we found that patients in the LTFU group had a large decrease in VA of approximately two lines of Snellen VA from the date of LTFU to the return visit. The LTFU group did not recover vision to the level prior to LTFU, which has been described previously in patients LTFU who were receiving anti-VEGF therapy (Table 3).^19,21,22 The control group also experienced a significant decrease in VA, when compared to the global standard date (GSD), despite continued treatment with anti-VEGF.

Anatomically, CFT increased dramatically from the visit prior to LTFU and the first return visit. This correlated with the decline in VA during this interval. However, despite an improvement in the CFT with resumption of treatment with anti-VEGF injections, the LTFU group did not recover VA. Moreover, after return in the LTFU group, CFT measurements steadily improved towards baseline values, although VA was not recovered. This discrepancy in anatomic and functional outcomes has been reported in prior studies.^19,22 Notably, the CFT continued to decrease in the control patients, with a significantly lower CFT at the final study visit. One hypothesis for this finding is that the decrease in vision in the control group, despite continuing treatment, may occur due to development of atrophy, which can be supported by the paralleled decrease in CFT.

A more recent paper by Stone et al²¹ found that patients with delayed anti-VEGF therapy (≥8 weeks from recommendation) with nAMD had the greatest loss of vision compared to patients with diabetic macular edema (DME) and macular edema secondary to retinal vein occlusion (RVO). The mean loss to follow-up time was 13 weeks. Additionally, a recent paper by Greenlee et al²²demonstrated that patients who were LTFU during the COVID-19 for a mean treatment lapse of five months for nAMD had irrecoverable visual decline, despite restoration of CFT upon re-initiation of anti-VEGF therapy. This finding of restoration of CFT despite persistent visual decline was also seen by Soares et al.¹⁹ However, one of the limitations of these studies was the possibility of being underpowered (Table 4).

One aspect of our study to note is there may have been some inherent differences in the LTFU group compared to the control group. For example, on average, patients in the LTFU group were approximately three years older (Table 1). Additionally, at 6 months prior to the GSD, despite a mean recommended treatment interval of 2 months (60 days) in LTFU group, these patients actually only followed up to receive injections every 3 months (90 days). On the contrary, 6 months prior to the standardized date for control patients, their recommended treatment interval was on mean approximately 2 months (58 days), and they were receiving injections approximately every 2 months (58 days). In addition, there was a significant difference in VA between the LTFU and control groups even before LTFU during the pandemic. Thus, the control group may have been more tightly adherent to their recommended treatment interval (Table 1) even prior to the government shutdown with the COVID-19 pandemic.

Limitations

This study has several limitations. First, as is the nature of retrospective chart review, some of the electronic medical records required manual review of charts and data entry by multiple research participants which may have led to errors in coding the data. Additionally, use of International Classification of Disease-10 codes may have the potential for misclassification bias in certain instances. A study on the use of coding in retrospective analysis²³supports its use in diabetic retinopathy studies, but to date no such studies exist for nAMD. Moreover, as mentioned, there may have been inherent differences in patients LTFU versus control patients. Given that Wills Eye Hospital and NYU Langone Health are tertiary referral centers, patients referred by outside ophthalmologists who were LTFU may have followed up with outside retina specialists where they required less travel, diminishing perceived risk of exposure during the COVID-19 pandemic. Further, our definition of LTFU—patients who did not follow-up within 30 days of their suggested follow-up date—may have not captured other patients who experienced vision loss despite falling into the control group (eg—a patient presenting 14 days after the recommended follow-up interval).

Loss to follow-up during the pandemic resulted in significant and seeming irrecoverable loss in VA in patients with nAMD who were receiving anti-VEGF therapy during the study period. Despite resumption of anti-VEGF therapy, VA did not significantly improve from the first visit after LTFU to 6 months later despite improvement in anatomic outcomes. Our study is the most well-powered LTFU study in patients with nAMD, providing novel insights on the impact of LTFU on both functional and anatomic outcomes in the general practice setting.

ACKNOWLEDGEMENTS

None

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author [VSD]. The data are not publicly available due to the fact their containing information could compromise the privacy of research participants.

CONFLICTS OF INTEREST/DISCLOSURE

Dr. Modi reported being a consultant for Allergan, Alimera, Genentech, Thea and Zeiss. Dr. Dedania reported being a consultant for Character Bio, Spark Therapeutics, Regeneron, Iveric Bio, and Apellis. The remaining authors (Pandit, Obeid, Nair, Patel, Cohen, Do, Venkat, and Choi) have no disclosures to report.

FUNDING

None

AUTHOR CONTRIBUTIONS

Dr. Dedania had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors

Acquisition, analysis, or interpretation of data: All authors

Drafting of the manuscript: Dedania, Pandit, Nair, Modi, Choi

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Obeid, Patel

Supervision: Dedania

Lim S, Mitchell P, Seddon J, Holz F, Wong T. Age-related macular degeneration. Lancet. 2012;379(9827):1728–38.
Mitchell P, Liew G, Gopinath B, Wong T. Age-related macular degeneration. Lancet. 2018;392(10153):1147–59.
Taylor D, Hobby A, Binns A, Crabb D. How does age-related macular degeneration affect real-world visual ability and quality of life? A systematic review. BMJ Open. 2016;6(12):e011504. 10.1136/bmjopen-2016-011504. Published 2016 Dec 2.
Rosenfeld P, Brown D, Heier J, Boyer D, Kaiser P, Chung C, Kim R. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med. 2006;355(14):1419–31.
Brown D, Kaiser P, Michels M, Soubrane G, Heier J, Kim R, Sy J, Schneider S. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med. 2006;355(14):1432–44.
Stewart M. Aflibercept (VEGF-TRAP): the next anti-VEGF drug. Inflamm Allergy Drug Targets. 2011;10(6):497–508.
Heier J, Brown D, Chong V, Korobelnik J, Kaiser P, Nguyen Q, Kirchof B, Ho A, Ogura Y, Yancopoulos G, et al. Intravitreal aflibercept (VEGF trap-eye) in wet age-related macular degeneration. Ophthalmology. 2012;119(12):2537–48.
Martin D, Maguire M, Fine S, Ying G, Jaffe G, Grunwald J, Toth C, Redford M, Ferris F. Ranibizumab and bevacizumab for treatment of neovascular age-related macular degeneration: two-year results. Ophthalmology. 2012;119(7):1388–98.
Chakravarthy U, Harding S, Rogers C, Downes S, Lotery A, Wordsworth S, Reeves B. Ranibizumab versus bevacizumab to treat neovascular age-related macular degeneration: one-year findings from the IVAN randomized trial. Ophthalmology. 2012;119(7):1399–411.
Stewart M, Rosenfeld P. Predicted biological activity of intravitreal VEGF Trap. Br J Ophthalmol. 2008;92(5):667–8.
Avery R, Re B, et al. Comparison of ranibizumab and bevacizumab for neovascular age-related macular degeneration according to LUCAS treat-and-extend protocol (Ophthalmology 2015;122:146 – 52). Ophthalmology. 2016;123(2):e14–6.
Rofagha S, Bhisitkul R, Boyer D, Sadda S, Zhang K. Seven-year outcomes in ranibizumab-treated patients in ANCHOR, MARINA, and HORIZON: a multicenter cohort study (SEVEN-UP). Ophthalmology. 2013;120(11):2292–9.
WorldHealthOrg. WHO COVID-19 Explorer 2021 [Internet]: World Health Organization. https://worldhealthorg.shinyapps.io/covid/; (Updated 03/25/2022, Cited 09/26/2021).
Breazzano M, Shen J, Abdelhakim A, Dagi-Glass L, Horowitz J, Xie S, Moraes G, Chen-Plotkin A, Chen R. Resident physician exposure to novel coronavirus (2019-nCoV, SARS-CoV-2) within New York City during exponential phase of COVID-19 pandemic: Report of the New York City Residency Program Directors COVID-19 Research Group. medRxiv: the preprint server for health sciences. 2020.
Breazzano M, Nair A, Arevalo J, Barakat M, Berrocal A, Chang J, Chen A, Elliot D, Garg S, Ghadiali Q, et al. Frequency of Urgent or Emergent Vitreoretinal Surgical Procedures in the United States During the COVID-19 Pandemic. JAMA ophthalmol. 2021;139(4):456–63.
Leng T, Gallivan MD, Kras A, Lum F, Roe M, Li C, Parke D, Schwartz S. Ophthalmology and COVID-19: The Impact of the Pandemic on Patient Care and Outcomes-An IRIS® Registry Study. Ophthalmology. 2021.
Regillo C, Brown D, Abraham P, Yue H, Ianchulev T, Schneider S, Shams N. Randomized, double-masked, sham-controlled trial of ranibizumab for neovascular age-related macular degeneration: PIER Study year 1. Am J Ophthalmol. 2008;145(2):239–48.
Kim J, Chang Y, Kim J. Natural course of patients discontinuing treatment for age-related macular degeneration and factors associated with visual prognosis. Retina. 2017;37(12):2254–61.
Soares R, Mellen P, Garrigan H, Obeid A, Wibbelsman T, Borkar D, Ho A, Hsu J. Outcomes of eyes lost to follow-up with neovascular age-related macular degeneration receiving intravitreal anti-vascular endothelial growth factor. Ophthalmol Retina. 2020;4(2):134–40.
Ramakrishnan M, Yu Y, VanderBeek B. Association of visit adherence and visual acuity in patients with neovascular age-related macular degeneration: secondary analysis of the comparison of age-related macular degeneration treatment trial. JAMA ophthalmol. 2020;138(3):237–42.
Stone L, Grinton M, Talks J. Delayed follow-up of medical retina patients due to COVID-19: impact on disease activity and visual acuity. Graefes Arch Clin Exp Ophthalmol. 2021;259(7):1773–80.
Greenlee T, Wang V, Kang H, Ohlhausen M, Chen A, Hom G, Conti T, Briskin I, Babiuch A, Singh R. Consequences of lapses in treatment with vascular endothelial growth factor inhibitors in neovascular age-related macular degeneration in routine clinical practice. Retina. 2021;41(3):581–7.
Lau M, Prenner J, Brucker A, VanderBeek B. Accuracy of billing codes used in the therapeutic care of diabetic retinopathy. JAMA ophthalmol. 2017;135(7):791–4.

Table 1 Baseline Characteristics of Control and Loss-to-Follow-Up Groups

Mean (SD) Age, years	81.7 (9.2)	84.0 (9.4)	<0.0001
Gender, n (%)			<0.0001
Male	1560 (35.4%)	252 (28.5%)
Female	2847 (64.6%)	633 (71.5%)
Mean Treatment Interval Recommended prior to LTFU (SD), days	58.4 (23.6)	58.1 (22.1)	0.71
Mean (SD) number of injections in the six months prior to GSD^a	3.9 (1.5)	3.5 (1.3)	<0.0001
Medication received at time of injection prior to GSD^a			<0.0001
Bevacizumab	342 (6.2%)	58 (10.2%)
Ranibizumab	3734 (67.9%)	321 (56.4%)
Aflibercept	1420 (25.8%)	190 (33.4%)
Average interval between injections 6 months prior to GSD (days)^a	60 (±23)	92 (±44)	<0.001

Notes:

^aTotal n is selected factor representative of total number of injections (n=6564)

*p-value <0.05 was considered statistically significant

Abbreviations:

GSD, global standard date; LTFU, loss to follow-up; SD, standard deviation

Table 2. Comparison of Visual Acuity (LogMAR) of Control Versus Loss-to-Follow-Up Groups

	Control (SD)	LTFU (SD)	p-value*
6 months prior to GSD	0.60 (0.57)	0.70 (0.57)	<0.001
3 months prior to GSD	0.62 (0.58)	0.71 (0.58)	<0.001
GSD	0.63 (0.59)	0.74 (0.60)	<0.001
Return visit	0.67 (0.60)	0.85 (0.62)	<0.001
6 months after GSD	0.67 (0.59)	0.82 (0.63)	<0.001
Final visit	0.68 (0.62)	0.86 (0.65)	<0.001

GSD = global standard date, LTFU = loss-to-follow-up, SD = standard deviation

*p-value of <0.05 was considered statistically significant.

Table 3. Comparison of Central Foveal Thickness in Control Versus Loss-to-Follow-Up Groups

	Control, µm (SD)	LTFU, µm (SD)	p-value*
6 months prior to GSD	278 (97)	282 (108)	p=0.28
3 months prior to GSD	276 (99)	280 (99)	p=0.19
GSD	275 (97)	280 (104)	p=0.19
Return visit	270 (97)	295 (119)	<0.001
6 months after GSD	270 (96)	279 (111)	<0.001

GSD = global standard date, LTFU = loss to follow-up, SD = standard deviation

*p-value of <0.05 was considered statistically significant

Table 4. Comparisons of Prior Loss-to-Follow-Up Studies in Neovascular Age-Related Macular Degeneration Receiving Anti-Vascular Endothelial Growth Factor Therapy

Authors	n	Definition of LTFU	Control Group	Results
Kim et al. (2017)¹⁸	35 (eyes)	24 month treatment discontinuation	No	Mean VA at visit prior to treatment discontinuation (Snellen 20/209) was significantly worse than at 24 months (Snellen 20/796, p <0.001). The presence of intra-retinal fluid was predictive of visual decline (p = 0.004).
Soares et al. (2020)¹⁹	93 (eyes)	>6 months	No	Median VA worsened from visit prior to LTFU (Snellen 20/80) to return visit (Snellen 20/200, p = <0.001), and this decline persisted through 12 months. Mean CFT increased from 201 µm at the visit before LTFU to 240 µm at return (p = 0.004), and then CFT improved to 183 µm by the end of the study period. VA declined despite a restoration of CFT.
Ramakrishnan et al. (2020)²⁰	1178 (patients)	>60 days	No	Patients who were considered to be very late in their follow-up interval (>60 day follow-up from their last appointment) saw fewer letters (12.5 letters worse) than patients in the on-time group (p < 0.001).
Stone et al. (2021)²¹	537 (eyes) (nAMD: LTFU = 166, control = 518)	>8 week delay in treatment	Yes	Median VA worsened between visit prior to LTFU (Snellen 20/50) to return visit (Snellen 20/60). This interval change in vision was found to be significantly worse than controls (baseline Snellen 20/50, 20 weeks post-baseline (standard date) Snellen 20/50, p = 0.0010). Mean CFT in the LTFU group increased from 311 µm to 342 µm (p = 0.0074).
Greenlee et al. (2021)²²	482 (eyes) (LTFU = 241, control = 241)	≥3 month treatment lapse	Yes	Both groups had similar baseline VA. The LTFU experienced decreased VA on return visit (Snellen 20/100) compared to the control groups (Snellen 20/63, p <0.01). LTFU group experienced an increase in CFT compared to controls (279.4 µm versus 253.7 µm, p <0.01), with subsequent improvement on resumption of treatment (259.1 µm versus 246.8 µm, p =0.06). VA declined in the LTFU group despite restoration of CFT.

LTFU = loss-to-follow-up, VA = visual acuity, CFT = central foveal thickness.

No competing interests reported.

Effects of Loss-to-Follow-Up on Functional and Anatomic Outcomes in Patients with Neovascular Age-Related Macular Degeneration

Status:

Version 1

Abstract

Figures

PLAIN LANGUAGE SUMMARY

INTRODUCTION

METHODS

Study design

Patient characteristics

Definition of LTFU

Temporal analysis

Statistical analysis

RESULTS

DISCUSSION

CONCLUSIONS

Declarations

References

Tables

Additional Declarations

Status:

Version 1