Ophthalmic Surgery, Lasers and Imaging Retina

Editorial Open Access

Playing With Fire

C. Armitage Harper III, MD; Kinley D. Beck, MD; Emmanuel Chang, MD, PhD; Ryan Young, MD; Darius M. Moshfeghi, MD

  • Ophthalmic Surgery, Lasers and Imaging Retina. 2020;51(10):542-544
  • https://doi.org/10.3928/23258160-20201005-01
  • Posted October 28, 2020


The authors present their concerns surrounding data presented in studies from 2018 and 2020 regarding very low dose bevacizumab for the treatment of retinopathy of prematurity.

[Ophthalmic Surg Lasers Imaging Retina. 2020;51:542–544.]


The authors present their concerns surrounding data presented in studies from 2018 and 2020 regarding very low dose bevacizumab for the treatment of retinopathy of prematurity.

[Ophthalmic Surg Lasers Imaging Retina. 2020;51:542–544.]

We appreciate the collaborative effort and initiative of Wallace et al.1,2 to investigate the lowest effective dose of intravitreal bevacizumab (IVB) (Avastin; Genentech, South San Francisco, CA) for the treatment of type 1 retinopathy of prematurity (ROP). Given the biological plausibility of systemic effects from IVB, dose de-escalation studies are important to establish the safety and efficacy of its use in the treatment of ROP. These studies and the authors' conclusion that retinal structural outcomes are “very good” with low doses of IVB have brought several concerns to our attention that should be addressed. First, the above premise of using low-dose IVB is predicated upon the theoretical conclusion that higher-dose IVB results in lower serum levels of vascular endothelial growth factor (VEGF) with subsequent systemic complications. Were serum VEGF levels measured? If so, were the measurements proportional to IVB dose? If not, this lack of data seems to obviate the hypothesis of the study. In addition, recent evidence from the RAINBOW trial indicates that there were no differences in the plasma-free concentrations of VEGF between the ranibizumab 0.1 mg group, ranibizumab 0.2 mg group, or laser group over time.3 Given this information and the aforementioned limitations, intravitreal ranibizumab (IVR) would be theoretically safer based on molecular structure, and lowering the dose would not be necessary.

In the 2018 de-escalation study, 25 of the 61 study eyes (41%) required additional treatment.2 The authors categorized the three groups into early failure (no improvement 3 to 5 days after injection, or recurrent within 4 weeks), late recurrence (after 4 weeks), and persistent avascularity.2 We would like to highlight that the groups are likely distinctively, physiologically different. The early recurrence indicates an overabundance of VEGF that was inadequately controlled during the initial treatment. Late recurrence represents a resurgence of VEGF after the initial dose of anti-VEGF has dissipated. The persistent avascular group represents the majority of type 1 ROP that has been previously treated with anti-VEGF.4 We also recommend routine use of fluorescein angiograms (FAs) in all neonates previously treated with anti-VEGF agents with persistent avascular retina, because these eyes may have recurrences with subsequent retinal detachments up to 5 years after treatment5 or even into adulthood.6 The use of FA was at the investigator's discretion in this study,2 which may have affected the choice of retreatment and may have also underestimated the true occurrence of persistent avascularity. Additionally, we recommend treating all infants with laser in the avascular zone at approximately 60 weeks using FA as guidance, or if recurrence necessitates earlier treatment, unless the FA at 60 weeks does not demonstrate active leakage and the retinal vasculature has grown to anterior zone III. This can prevent late recurrences that become more challenging to detect as the children become older and more difficult to appropriately examine in clinic. Also, it may have long-term protective benefits in lowering the lifetime risk of retinal detachments as adults.

The 2020 de-escalation study did not disclose the anatomical description of the failures despite six out of 23 failing in the 0.002 mg group.1 It also only reports results in a 4-week period. This short period was likely chosen to investigate whether the lowest dose was safe and efficacious. However, it may promote a false sense of security among examiners given most recurrences and complications occur after this short interval. Wong et al. reported reactivation at 5.9 weeks after IVR.7 Two other papers reported reactivation at 7 to 8 weeks after initial IVR.8,9 Chen et al. reported reactivation at 13.6 weeks after IVB, and the risk of reactivation correlated with the degree of avascular retina.10 This emphasizes the importance of carefully and accurately documenting zones both at the time of treatment and the time of recurrence.

It is also important to re-examine the infants 48 to 72 hours post-injection to look for endophthalmitis and cataract formation. Then, examine the infants every week to follow the course of ROP closely. Furthermore, the use of low-dose IVB requires a compounding lab to formulate such lose dosages. Consequently, contamination may occur leading to endophthalmitis. The advantage of 0.625 mg IVB or 0.2 mg IVR is that no manipulation of the original formulation is required, thereby eliminating these risks.

Lastly, and most importantly, success in the treatment of type 1 ROP should be defined as no progression to stage 4 or 5 retinal detachment. The authors' 2018 study had four retinal detachments in 61 infants (four eyes out of 112 that received treatment) and one cataract, which they conclude is a “very good” outcome.2 Although the studies cannot be directly compared due to differences in study design, in comparison, Cernichiaro et al. had no progression to retinal detachment, no cataracts, or other adverse effects in 220 treated eyes.11 Large multicenter studies that involve heterogeneous experience levels and practice patterns in their participating centers may not necessarily have the best outcomes, in addition to the possibility of lower doses being less efficacious.

Regarding the potentially injection-related cataract observed in the study, Wright et al. demonstrated that the half-inch 30-gauge needle can anatomically reach the nasal side of the eye when injected temporally.12 An 8-mm needle, as used in this study, is also long enough to cause iatrogenic lens damage. Therefore, we recommend using a 4 mm 32-gauge needle to obviate retinal or lenticular complications.

In conclusion, although a lower dose of bevacizumab is effective for initial cessation of type 1 ROP, we believe that we cannot underestimate the potential for high recurrence rates given the short follow-up in the study, which can be potentially prevented with use of FA to assess residual vascular activity, and follow-up laser photocoagulation. The study outcomes, we believe, were less than optimal, and should be placed in context of other studies and experiences. It is critical to not only monitor for regression of initial ROP after low-dose anti-VEGF treatment but also closely monitor and document the amount of residual avascular retina to understand the long-term implications for re-activation and complications such as progression to retinal detachment to ensure excellent visual outcomes.


  1. Wallace DK, Kraker RT, Freedman SF, et al. Pediatric Eye Disease Investigator Group (PEDIG). Short-term Outcomes After Very Low-Dose Intravitreous Bevacizumab for Retinopathy of Prematurity. JAMA Ophthalmol. 2020;138(6):698–701. doi:10.1001/jamaophthalmol.2020.0334 [CrossRef] PMID:32324197
  2. Wallace DK, Dean TW, Hartnett ME, et al. Pediatric Eye Disease Investigator Group. A Dosing Study of Bevacizumab for Retinopathy of Prematurity: Late Recurrences and Additional Treatments. Ophthalmology. 2018;125(12):1961–1966. doi:10.1016/j.ophtha.2018.05.001 [CrossRef] PMID:29887334
  3. Filder M, Fleck BW, Stahl A, et al. on behalf of the RAINBOW study group; Ranibizumab Population Pharmacokinetics and Free VEGF Pharmacodynamics in Preterm Infants With Retinopathy of Prematurity in the RAINBOW Trial. Trans Vis Sci Technol. 2020;9(8):43. doi:10.1167/tvst.9.8.43 [CrossRef] PMID:
  4. Lepore D, Quinn GE, Molle F, et al. Follow-up to Age 4 Years of Treatment of Type 1 Retinopathy of Prematurity Intravitreal Bevacizumab Injection versus Laser: Fluorescein Angiographic Findings. Ophthalmology. 2018;125(2):218–226. doi:10.1016/j.ophtha.2017.08.005 [CrossRef] PMID:
  5. Yonekawa Y, Wu WC, Nitulescu CE, et al. Progressive Retinal Detachment in Infants with Retinopathy of Prematurity Treated with Intravitreal Bevacizumab or Ranibizumab. Retina. 2018;38(6):1079–1083. doi:10.1097/IAE.0000000000001685 [CrossRef] PMID:
  6. Hamad AE, Moinuddin O, Blair MP, et al. Late-Onset Retinal Findings and Complications in Untreated Retinopathy of Prematurity. Ophthalmol Retina. 2020;4(6):602–612. doi:10.1016/j.oret.2019.12.015 [CrossRef] PMID:32059986
  7. Wong RK, Hubschman S, Tsui I. Reactivation of retinopathy of prematurity after ranibizumab treatment. Retina. 2015;35(4):675–680. doi:10.1097/IAE.0000000000000578 [CrossRef] PMID:25768252
  8. Chan JJT, Lam CPS, Kwok MKM, et al. Risk of recurrence of retinopathy of prematurity after initial intravitreal ranibizumab therapy. Sci Rep. 2016;6(1):27082. doi:10.1038/srep27082 [CrossRef] PMID:27256987
  9. Hu Q, Bai Y, Chen X, Huang L, Chen Y, Li X. Recurrence of Retinopathy of Prematurity in Zone II Stage 3+ after Ranibizumab Treatment: A Retrospective Study. J Ophthalmol. 2017;2017:5078565. doi:10.1155/2017/5078565 [CrossRef] PMID:28491468
  10. Chen TA, Shields RA, Bodnar ZH, Callaway NF, Schachar IH, Moshfeghi DM. A Spectrum of Regression Following Intravitreal Bevacizumab in Retinopathy of Prematurity. Am J Ophthalmol. 2019;198:63–69. doi:10.1016/j.ajo.2018.09.039 [CrossRef] PMID:
  11. Cernichiaro-Espinosa LA, Harper CA III, Read S, et al. Report of Safety of the Use of a Short 32G Needle for Intravitreal Anti-Vascular Endothelial Growth Factor Injections for Retinopathy of Prematurity: A Multicenter Study. Retina. 2018;38(6):1251–1255. doi:10.1097/IAE.0000000000002172 [CrossRef] PMID:29689029
  12. Wright LM, Vrcek IM, Scribbick FW III, Chang EY, Harper CA III, . Technique for Infant Intravitreal Injection in Treatment of Retinopathy of Prematurity. Retina. 2017;37(11):2188–2190. doi:10.1097/IAE.0000000000001561 [CrossRef] PMID:28196051

From Austin Retina Associates, Austin, Texas (CAH, RY); Eyesight Ophthalmic Services, Portsmouth, New Hampshire (KDB); Retina and Vitreous of Texas, Houston, Texas (EC); Stanford Byers Eye Institute, Palo Alto, California (DM).

Dr. Moshfeghi received personal fees from Regeneron, Bayer, and Novartis, as well as grants from Genentech, outside the submitted work. The remaining authors report no relevant financial disclosures.

Dr. Moshfeghi did not participate in the editorial review of this manuscript.

Address correspondence to C. Armitage Harper III, MD, Retina Associates of Austin, Austin, TX; email: caharper@austinretina.com.

This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International (https://creativecommons.org/licenses/by/4.0). This license allows users to copy and distribute, to remix, transform, and build upon the article, for any purpose, even commercially, provided the author is attributed and is not represented as endorsing the use made of the work.
Received: July 30, 2020
Accepted: August 20, 2020


Sign up to receive

Journal E-contents