Ophthalmic Surgery, Lasers and Imaging Retina

Clinical Science 

Two-Year Outcomes Comparing Anti-VEGF Injections to Laser for ROP Using a Commercial Claims Database

Michael H. Zhang, BS; Michael P. Blair, MD; Sandra A. Ham, MS; Sarah H. Rodriguez, MD, MPH

Abstract

BACKGROUND AND OBJECTIVE:

To report ocular and neurodevelopmental outcomes among infants treated for retinopathy of prematurity (ROP) in a nationwide health insurance claims database.

PATIENTS AND METHODS:

Retrospective cohort study of 298 infants treated with laser or anti-vascular endothelial growth factor (VEGF) injection identified in the MarketScan database (2011–2017) with 2-year follow-up.

RESULTS:

A review of claims data found 298 patients with International Classification of Diseases and Common Procedural Technology codes for ROP treatment and 2 years of continuous insurance coverage. Of these, 63 infants received injections and 235 received laser. Overall, the anti-VEGF group had higher rates of underlying neurological comorbidities (35% vs. 23%; P = .05) and thrombocytopenia (17% vs. 8%; P = .02). Most ocular outcomes were similar, including retinal detachment (P = .87). There were higher rates of second procedures after injection (44% vs. 10%; P < .001). Rates of language, motor, and cognitive delays were similar. Rates of cerebral palsy were higher with injections but were not statistically significant after adjusting for comorbidities (odds ratio = 1.88; P = .10).

CONCLUSIONS:

The prevalence of retinal detachment after 2 years was similar comparing anti-VEGF to laser. Despite the higher rates of underlying neurologic comorbidity in the injection group, there were no differences in language, motor, or cognitive delays.

[Ophthalmic Surg Lasers Imaging Retina. 2020;51:486–493.]

Abstract

BACKGROUND AND OBJECTIVE:

To report ocular and neurodevelopmental outcomes among infants treated for retinopathy of prematurity (ROP) in a nationwide health insurance claims database.

PATIENTS AND METHODS:

Retrospective cohort study of 298 infants treated with laser or anti-vascular endothelial growth factor (VEGF) injection identified in the MarketScan database (2011–2017) with 2-year follow-up.

RESULTS:

A review of claims data found 298 patients with International Classification of Diseases and Common Procedural Technology codes for ROP treatment and 2 years of continuous insurance coverage. Of these, 63 infants received injections and 235 received laser. Overall, the anti-VEGF group had higher rates of underlying neurological comorbidities (35% vs. 23%; P = .05) and thrombocytopenia (17% vs. 8%; P = .02). Most ocular outcomes were similar, including retinal detachment (P = .87). There were higher rates of second procedures after injection (44% vs. 10%; P < .001). Rates of language, motor, and cognitive delays were similar. Rates of cerebral palsy were higher with injections but were not statistically significant after adjusting for comorbidities (odds ratio = 1.88; P = .10).

CONCLUSIONS:

The prevalence of retinal detachment after 2 years was similar comparing anti-VEGF to laser. Despite the higher rates of underlying neurologic comorbidity in the injection group, there were no differences in language, motor, or cognitive delays.

[Ophthalmic Surg Lasers Imaging Retina. 2020;51:486–493.]

Introduction

Big data extract and analyze information from datasets that are too large and complex to be dealt with by traditional methods.1 Recent advances in analytical software have led to a revolution in making big data accessible.1 One example of big data is the IBM MarketScan Research Database (IBM, Armonk, NY), a nationwide database of insurance claims. This has previously been used to describe glaucoma, strabismus, and retina surgical outcomes.2–6 This database presents an unique opportunity to evaluate retinopathy of prematurity (ROP) outcomes, since existing national databases specific to ophthalmology lack data on patients with ROP.

ROP is a leading causes of pediatric blindness.7 For the past few decades, the standard of care for type 1 ROP has been laser peripheral retinal photocoagulation.8,9 In 2011 the Bevacizumab Eliminates the Angiogenic Threat for Retinopathy of Prematurity (BEAT-ROP) study reported improved structural outcomes and lower rates of retinal detachment (RD) with intravitreal bevacizumab (IVB) (Avastin; Genentech, South San Francisco, CA) compared to traditional laser.10 Significantly lower rates of myopia have also been reported.11–13

Nonetheless, use of anti-vascular endothelial growth factor (VEGF) agents in the treatment of type 1 ROP remains controversial, in part due to concerns about adverse neurodevelopmental effects. In premature infants, serum VEGF levels are known to be suppressed for two months after intravitreal bevacizumab, with unknown systemic effects.14 In addition, possible ocular complications include late reactivation and the risk of RD.15–17

Long-term efficacy and safety data regarding anti-VEGF in neonates are sparse. The purpose of this study is to evaluate two-year ocular and neurodevelopmental outcomes among infants treated with anti-VEGF injection or laser following the initial publication of BEAT-ROP in 2011.

Patients and Methods

This retrospective cohort study used the IBM MarketScan Research Database from the years 2011 to 2017. The MarketScan Commercial Database includes approximately 25 to 60 million employees, retirees, and their dependents each year with employer-sponsored health insurance. The MarketScan Inpatient Services table includes demographic, provider, and insurance information. As these data are de-identified and conformed to the requirements of the United States Health Insurance Portability and Privacy Act, the University of Chicago Institutional Review Board determined the study to be exempt.

Infants were selected for inclusion based on International Classification of Diseases, version 9 (ICD-9) and version 10 (ICD-10) and Current Procedural Terminology (CPT) codes. The diagnosis of ROP required an appropriate ICD code (Table A, available at www.healio.com/OSLIRetina). To identify infants treated for ROP, diagnosis was combined with a CPT code for either intravitreal injections or laser. Study subjects were followed for 2 years beginning on the date of first treatment. Infants treated between 2011 and 2015 were included. Infants were excluded if they had any gaps in insurance coverage during the first 2 years after treatment or if they received simultaneous laser and injection as the initial treatment.

Baseline comorbidities included neurological complications — any diagnosis of hydrocephalus, periventricular leukomalacia (PVL), or stage 3 or 4 intraventricular hemorrhage (severe IVH); thrombocytopenia; sepsis; bronchopulmonary dysplasia (BPD); necrotizing enterocolitis (NEC); patent ductus arteriosus (PDA) repair; length of stay (LOS); gender; and geographic U.S. census region.

Ocular outcomes included RD, vitreous hemorrhage, cataract, glaucoma, strabismus, corneal scar, and endophthalmitis, as well as additional procedures by CPT code: cataract surgery, subsequent retina surgery, and retreatment more than 30 days after the initial treatment.

Neurodevelopmental outcomes were cerebral palsy (CP), speech delay, cognitive delay, motor delay, any developmental delay, blindness, and hearing loss. These were identified by ICD-9, ICD-10, and CPT codes (Table A). ROP zone, body weight, gestational age, and ethnicity were not available.

ICD-9, ICD-10, and CPT Codes Used to Identify Cases of Retinopathy of Prematurity and Assess Initial Therapies, Baseline Characteristics, Ocular Outcomes, and Neurodevelopmental Outcomes

Table A:

ICD-9, ICD-10, and CPT Codes Used to Identify Cases of Retinopathy of Prematurity and Assess Initial Therapies, Baseline Characteristics, Ocular Outcomes, and Neurodevelopmental Outcomes

Statistical Analysis

Categorical data are summarized using percentages. Median values with interquartile range (IQR) are also reported. P values are calculated using the Pearson's chi-squared test for categorical variables (Fisher's exact test was used for any comparison with n < 5) and the Wilcoxon rank-sum test for continuous variables. Since the sample of children was small, the study presents a non-trivial risk of disclosure of patient identity, which would violate confidentiality in research. In order to control the risk of statistical disclosure, table cells with an “n” value less than 11* were suppressed18 (consistent with recommendations of the Center for Medicare and Medicaid Services for limited data sets, no cell containing a value of 1 to 10 can be reported directly.19)This is consistent with recommendations of the Center for Medicare and Medicaid Services for limited data sets; this policy states no cell containing a value of 1 to 10 can be reported directly.19 The policy aims to protect the confidentiality of patients by avoiding the release of information that can identify individual beneficiaries. The multivariate logistic regression includes neurological comorbidities (severe IVH, PVL, or hydrocephalus), BPD, thrombocytopenia, and gender. Since LOS was not available for infants treated as outpatients, the analysis was repeated with and without this variable.

Since treatment was not randomized but likely based on individual characteristics evaluated by the treating physician, a propensity score analysis was also performed. Propensity scores for receiving injections were computed as the predicted probabilities from a logistic regression model with the following covariates: gender, neurological comorbidity, thrombocytopenia, BPD, and the five U.S. Census regions. LOS was not included, as this was designed to account for differences in baseline criteria that might have influenced decisions at the time of treatment. A stratified analysis using quintiles of the propensity score was performed using the Cochran-Mantel-Haenszel method. All statistical analyses were performed with Stata version 15 (StataCorp LP, College Station, TX). The threshold for statistical significance of two-sided tests was set at P values less than or equal to .05.

Results

Overall, there were 22,559 patients with a diagnosis of ROP, of whom 787 (3.5%) patients had CPT codes indicating treatment (Figure 1). Of these 787 patients, 486 (62%) were excluded for less than 2 years of continuous enrollment in the insurance claims database. Fewer than 11* patients were excluded because they received both laser and injection treatment simultaneously. Data for the main analysis were available for 298 infants diagnosed with ROP with 2 years of follow-up, including 235 patients treated with laser and 63 treated with intravitreal injections.

Infants in the MarketScan Research Database with ICD codes indicative of retinopathy of prematurity (ROP). Infants were included in this study if CPT codes indicated treatment with laser or injection. Treated infants were excluded if insurance coverage was not continuous during the 2 years following treatment, or if they received simultaneous laser and injection procedures..

Figure 1.

Infants in the MarketScan Research Database with ICD codes indicative of retinopathy of prematurity (ROP). Infants were included in this study if CPT codes indicated treatment with laser or injection. Treated infants were excluded if insurance coverage was not continuous during the 2 years following treatment, or if they received simultaneous laser and injection procedures..

Baseline characteristics are shown in Table 1. Overall, infants who received injections had significantly higher rates of neurological comorbidities at baseline (35% vs. 23%; P = .05) and thrombocytopenia (17% vs. 8%; P = .02) compared with those who received laser. There was a tendency to treat infants in the southern U.S. census region with injections; otherwise more patients nationally received laser (P = .001). Infants who received injections also had longer LOS in the neonatal intensive care unit (P = .06). Data on LOS were unavailable for 78 patients, who were treated in the outpatient setting: 14 (18%) who received injections and 64 (82%) who received laser. Data for other variables were complete.

Baseline Characteristics

Table 1:

Baseline Characteristics

Table 2 reports ocular outcome measures. Percentages with RD, vitreous hemorrhage, glaucoma, strabismus, and corneal scarring were not significantly different between groups. No patient had endophthalmitis. There was a higher percentage with cataract diagnosis in the injection group; however, there was no difference in the rate of cataract surgery.

Ocular Outcomes for Infants Receiving Laser or Injection

Table 2:

Ocular Outcomes for Infants Receiving Laser or Injection

Infants who received injections were significantly more likely to have a secondary procedure than those who received laser (44% vs. 10%; P = .001). Of the 28 patients in the primary injection group receiving secondary procedures, 15 received another injection and 13 received subsequent laser. The median time to a second procedure after injection was 42 days (interquartile range [IQR]: 28 to 106 days). This was significantly longer if the second treatment was laser rather than a second injection (P = .001): median time to secondary injection was 33 days (IQR: 15 to 42 days), whereas median time to secondary laser was 88 days (IQR: 58 to 131 days).

Among the 23 patients in the primary laser group who received secondary procedures, a significant majority received subsequent laser, and fewer than 11* received subsequent injections. All patients who received an injection after laser were re-treated within 30 days of the original laser.

Median time to secondary laser was significantly shorter with primary laser than with primary injections (P = .01).

Having a second procedure was strongly associated with RD (odds ratio [OR] = 3.33; confidence interval [CI], 1.47–7.52; P = .004). For patients who received primary injections, less than 11* (< 17%) with monotherapy had RD, whereas 18% with second procedures had RD (P = .08). A similar effect was seen in the laser group, where less than 11* (< 5%) with monotherapy had detachments and 26% with second procedures had detachments (P = .019).

Developmental outcomes are seen in Table 3. Comparing injection to laser, rates of any developmental delay were high in both groups: 89% versus 87% (P = .60). Other delays were motor (< 17% vs. 15%; P = .60), cognitive (29% vs. 27%; P = .83), and speech (35% vs. 40%; P = .42).

Neurodevelopmental Outcomes of Infants Receiving Laser or Injection

Table 3:

Neurodevelopmental Outcomes of Infants Receiving Laser or Injection

Percentages of infants with CP were significantly higher with injections compared to laser (25% vs. 12%; P = .01). However, the difference in CP by treatment group was not statistically significant (OR = 1.88; 95% CI, 0.89–4.02; P = .10) after adjusting for baseline neurologic comorbidities, BPD, thrombocytopenia, and gender. Having a neurologic comorbidity, however, was associated with significantly higher odds of CP (OR = 5.34; 95% CI, 2.57–10.33; P = .001). Among the 220 infants who received inpatient treatments, findings were similar when LOS was included in the model (OR = 1.83; 95% CI, 0.80–4.19; P = .15).

Results of a propensity analysis were not significantly different from results of the standard multivariate logistic regression model (OR = 1.84; 95% CI, 0.82–4.17; P = .13).

Discussion

To the best of our knowledge, this is the first study using a big data national insurance claims database from the US to evaluate patients treated for ROP. In our current study, 3.5% of patients with ROP required treatment. Although this is lower than the 8% rate reported by Gilbert previously, her analysis only included infants with birth weight less than 1,250 grams.20 Our present study included all patients screened for ROP, which also includes lower-risk infants with birth weights between 1,250 grams and 1,500 grams, infants with gestational ages less than 31 weeks, and any discretionary babies as determined by a neonatologist. In addition, patients with type 2 ROP which spontaneously regressed would have been excluded in our study.

In this MarketScan database, ocular outcomes appear similar by treatment group. Although there is a propensity to treat sicker infants with injections, developmental outcomes were not significantly different after adjusting for neonatal comorbidities. Rates of RD, however, are higher than previously reported with anti-VEGF agents and similar to laser treatment.

The BEAT-ROP study found significantly lower rates of ROP recurrence and RD with bevacizumab compared to laser.10 However, ocular outcomes in that study were followed only until 54 weeks postmenstrual age, with approximately 19 to 20 weeks of follow-up.10 Since the BEAT-ROP study, the Postnatal Growth and ROP (G-ROP) study reported retrospective ocular outcomes for 492 infants (970 eyes) treated with laser and 18 infants (34 eyes) treated with intravitreal injections, with a median follow-up of 18 weeks.21 The RAINBOW study also reported lower rates of unfavorable structural outcomes with ranibizumab (Lucentis; Genentech, South San Francisco, CA) 0.2 mg compared to laser after 24 weeks follow-up (1.4% vs. 10%).22 In contrast to these studies, the percentages with RD in this database were similar for both anti-VEGF injections and laser. The rate of RD in both groups was also similar to the percentage with ROP progression after laser in the G-ROP study (9.2%).21

The 2-year follow-up period is one explanation for the difference in findings. Although 18 weeks is likely sufficient to determine regression of ROP after laser treatment, there are several reports of late ROP reactivation and RD 2 to 3 years after initial anti-VEGF treatment.15–17 Consequently, longer follow-up intervals may be necessary to capture the ocular complications related to anti-VEGF therapy. The BEAT-ROP study had a reactivation rate of 4% for injection and 22% for laser after 19 to 20 weeks of follow-up. Although the reactivation rate for laser in BEAT-ROP study was higher than reported in the ET-ROP study,8 the G-ROP study also reported a similar trend with recurrent ROP rates of 9.2% for laser and 0% for anti-VEGF injections. In this database, the high rate of second procedures could be due to late reactivation of ROP, which would not have been observed with shorter follow-up periods.10,21

Alternatively, the high rate of second procedures may reflect a preference to treat patients who received primary injections with delayed, prophylactic laser to prevent late RD, as recommended previously.11,23,24 Unfortunately, the CPT codes for a second procedure cannot differentiate treatment for late reactivation versus prophylactic laser treatment.24 Our rate of RD after anti-VEGF therapy at 2 years was higher than previous studies with shorter follow-up periods. This serves as a reminder that the endpoint of ROP screening after anti-VEGF injections remains controversial, and these patients should be followed closely.

Finally, a preference to treat infants with zone 1 ROP with anti-VEGF might also explain the higher rate of RD. However, the limitations of CPT and ICD codes preclude a subanalysis of outcomes by zone of ROP in the present study.

In addition to the need for more long-term ocular data regarding the efficacy and safety of anti-VEGF therapy for ROP, more neurodevelopmental outcome data are also needed. Although the BEAT-ROP group did report 2-year developmental outcomes, only 16 patients from a single study center were included.11,25 Other than this small prospective study, data are limited to retrospective studies. The Morin study found higher rates of severe delay among 27 infants treated with bevacizumab compared to 98 with laser.26 However, infants treated with IVB had higher mortality risk scores and more severe ROP at baseline in this study. In addition to selection bias, imbalance in exclusion criteria likely influenced the results.26–28 Since then, other studies have failed to find a significant difference in developmental outcomes between anti-VEGF injections and laser.25,28–30

More recently, Natarajan et al. reported higher mortality and worse cognitive outcomes among a retrospective cohort of 181 infants treated with bevacizumab compared to 224 treated with laser.31 However, infants who received bevacizumab were smaller and sicker, with lower median birth weight (P = .02), gestational age (P = .05), longer median time on ventilation (P = .04), on supplemental oxygen (P = .01), and had a higher odds of death (OR = 2.54; P = .002). Yet none of these factors were considered for statistical adjustment. There was also a significant difference within the subanalysis of patients who expired: 50% of those treated with anti-VEGF had 5-minute Appearance, Pulse, Grimace, Activity, and Respiration test results less than five, whereas only 15% of those treated with laser had such scores. Although not reported, this is indeed significant (chi-squared P = .038). Their study is the largest to-date with 405 infants but harbors significant bias when considering the heterogenous baseline characteristics and the lack of adequate adjustments for these discrepancies.

As with previous studies, similar baseline differences were also noted in this study population, with more frequent neurological comorbidities and thrombocytopenia and longer lengths of stay were higher among patients who received injections. As a result, there appears to be a propensity to treat sicker infants with injections, which is similar to the Morin and Natarajan studies. Physicians may prefer to treat fragile infants with anti-VEGF to avoid sedation or general anesthesia which would be required with laser.

Unlike the Morin and Natajaran studies, however, after adjusting for confounding factors, the association between anti-VEGF injections and CP failed to reach statistical significance in this study population. This finding is consistent with the results of numerous studies.25,26,28–30 These baseline neonatal conditions likely represent important confounding factors, all of which have been shown to independently correlate with poorer neurodevelopmental outcomes.32–39 Given the wide confidence interval, however, larger studies are necessary to rule out an association between anti-VEGF use and developmental delays. At the same time, the impact of visual impairment on neurodevelopment must not be underestimated,28,40,41 and studies which do not account for selection biases should not guide treatment.

Of note, the rate of CP in the laser group appears lower than reported previously among patients with severe ROP. The Trial of Indomethacin Prophylaxis in Preterms study reported a 24% (19/80) rate of CP in patients with severe ROP.34 While there is no gold standard for the expected rate of CP among infants with ROP, an underdiagnosis of CP in children at age two is possible.42

This study has several limitations. Most notably, 62% of patients were excluded for lack of continuous coverage over 2 years. Although outside the scope of this paper, insurance instability creates additional challenges in this vulnerable population. This is a small sample of a national database of private insurances, which does not include Medicaid. A sample of Medicaid patients would likely be much larger.43 There may be selection bias because MarketScan is a database of employer-sponsored health insurance, and the Medicaid population differs from the MarketScan population. In addition, majority of injections are in Southern states, and most lasers are in other regions, which represents a significant regional bias.

Finally, important characteristics such as zone of ROP at treatment, birth weight, gestational age, maternal education, and ethnicity are unknown, all of which have been shown to correlate with neurodevelopmental outcomes.32,44–46 Information about birth weight and gestational age are the two most critical risk factors for ROP development. Although ICD codes exist for ranges of birth weights and gestational ages, these were inconsistently and infrequently applied to patients in our study sample and were unfortunately not useful for baseline comparisons between study groups. Nonetheless, this study adds to the growing body of evidence that anti-VEGF injections for ROP may be safe despite the concerns about systemic absorption.

Despite the flaws inherent to MarketScan data, existing national databases specific to ophthalmology do not include patients with ROP at this time, and important information about ROP treatment options can be gleaned from this database. While rates of RD were similar by treatment group, they are higher than previously reported for anti-VEGF injections and may reflect the chance of late RD. In addition, there appears to be a propensity to treat sicker infants with injections even after the publication of BEAT-ROP. Retrospective studies have inherent biases and should be interpreted with caution regarding the potential link between CP and developmental delay.

References

  1. Clark A, Ng JQ, Morlet N, Semmens JB. Big data and ophthalmic research. Surv Ophthalmol. 2016;61(4):443–465. doi:10.1016/j.survophthal.2016.01.003 [CrossRef] PMID:26844660
  2. Vaziri K, Schwartz SG, Kishor KS, et al. Rates of Reoperation and Retinal Detachment after Macular Hole Surgery. Ophthalmology. 2016;123(1):26–31. doi:10.1016/j.ophtha.2015.09.015 [CrossRef] PMID:26505804
  3. Vaziri K, Schwartz SG, Kishor KS, et al. Incidence of postoperative suprachoroidal hemorrhage after glaucoma filtration surgeries in the United States. Clin Ophthalmol. 2015;9:579–584. doi:10.2147/OPTH.S78359 [CrossRef] PMID:25897196
  4. Vaziri K, Kishor K, Schwartz SG, et al. Incidence of bleb-associated endophthalmitis in the United States. Clin Ophthalmol. 2015;9:317–322. doi:10.2147/OPTH.S75286 [CrossRef] PMID:25709395
  5. Leffler CT, Vaziri K, Cavuoto KM, et al. Strabismus Surgery Reoperation Rates With Adjustable and Conventional Sutures. Am J Ophthalmol. 2015;160(2):385–390.e4. doi:10.1016/j.ajo.2015.05.014 [CrossRef] PMID:26002082
  6. Leffler CT, Vaziri K, Schwartz SG, et al. Rates of Reoperation and Abnormal Binocularity Following Strabismus Surgery in Children. Am J Ophthalmol. 2016;162:159–166.e9. doi:10.1016/j.ajo.2015.10.022 [CrossRef] PMID:26548808
  7. Kong L, Fry M, Al-Samarraie M, Gilbert C, Steinkuller PG. An update on progress and the changing epidemiology of causes of childhood blindness worldwide. J AAPOS. 2012;16(6):501–507. doi:10.1016/j.jaapos.2012.09.004 [CrossRef] PMID:23237744
  8. Early Treatment For Retinopathy Of Prematurity Cooperative Group. Revised indications for the treatment of retinopathy of prematurity: results of the early treatment for retinopathy of prematurity randomized trial. Arch Ophthalmol. 2003;121(12):1684–1694. doi:10.1001/archopht.121.12.1684 [CrossRef] PMID: 14662586
  9. Ng EYJ, Connolly BP, McNamara JA, Regillo CD, Vander JF, Tasman W. A comparison of laser photocoagulation with cryotherapy for threshold retinopathy of prematurity at 10 years: part 1. Visual function and structural outcome. Ophthalmology. 2002;109(5):928–934. doi:10.1016/S0161-6420(01)01017-X [CrossRef] PMID:11986099
  10. Mintz-Hittner HA, Kennedy KA, Chuang AZBEAT-ROP Cooperative Group. Efficacy of intravitreal bevacizumab for stage 3+ retinopathy of prematurity. N Engl J Med. 2011;364(7):603–615. doi:10.1056/NEJMoa1007374 [CrossRef] PMID:21323540
  11. Anand N, Blair MP, Greenwald MJ, Rodriguez SH. Refractive outcomes comparing primary laser to primary bevacizumab with delayed laser for type 1 ROP. J AAPOS. 2019;23(2):88.e1–88.e6. doi:10.1016/j.jaapos.2018.10.013 [CrossRef] PMID:30797978
  12. Geloneck MM, Chuang AZ, Clark WL, et al. BEAT-ROP Cooperative Group. Refractive outcomes following bevacizumab monotherapy compared with conventional laser treatment: a randomized clinical trial. JAMA Ophthalmol. 2014;132(11):1327–1333. doi:10.1001/jamaophthalmol.2014.2772 [CrossRef] PMID:25103848
  13. Hwang CK, Hubbard GB, Hutchinson AK, Lambert SR. Outcomes after Intravitreal Bevacizumab versus Laser Photocoagulation for Retinopathy of Prematurity: A 5-Year Retrospective Analysis. Ophthalmology. 2015;122(5):1008–1015. doi:10.1016/j.ophtha.2014.12.017 [CrossRef] PMID:25687024
  14. Wu W-C, Lien R, Liao P-J, et al. Serum levels of vascular endothelial growth factor and related factors after intravitreous bevacizumab injection for retinopathy of prematurity. JAMA Ophthalmol. 2015;133(4):391–397. doi:10.1001/jamaophthalmol.2014.5373 [CrossRef] PMID:25569026
  15. Ittiara S, Blair MP, Shapiro MJ, Lichtenstein SJ. Exudative retinopathy and detachment: a late reactivation of retinopathy of prematurity after intravitreal bevacizumab. J AAPOS. 2013;17(3):323–325. doi:10.1016/j.jaapos.2013.01.004 [CrossRef] PMID:23607977
  16. Snyder LL, Garcia-Gonzalez JM, Shapiro MJ, Blair MP. Very Late Reactivation of Retinopathy of Prematurity After Monotherapy With Intravitreal Bevacizumab. Ophthalmic Surg Lasers Imaging Retina. 2016;47(3):280–283. doi:10.3928/23258160-20160229-12 [CrossRef] PMID:26985803
  17. Hajrasouliha AR, Garcia-Gonzales JM, Shapiro MJ, Yoon H, Blair MP. Reactivation of Retinopathy of Prematurity Three Years After Treatment With Bevacizumab. Ophthalmic Surg Lasers Imaging Retina. 2017;48(3):255–259. doi:10.3928/23258160-20170301-10 [CrossRef] PMID:28297039
  18. Willenborg L, De Waal T. Elements of Statistical Disclosure Control. Vol 155. Springer Science & Business Media; 2001.
  19. ResDAC. https://www.resdac.org/. Accessed July 1, 2019.
  20. Gilbert C. Retinopathy of prematurity: epidemiology. Community Eye Health. 1997;10(22):22–24.
  21. Morrison D, Shaffer J, Ying G-S, et al. G-ROP Study Group. Ocular complications following treatment in the Postnatal Growth and Retinopathy of Prematurity (G-ROP) Study. J AAPOS. 2018;22(2):128–133. doi:10.1016/j.jaapos.2017.12.005 [CrossRef] PMID:29548840
  22. Stahl A, Lepore D, Fielder A, et al. Ranibizumab versus laser therapy for the treatment of very low birthweight infants with retinopathy of prematurity (RAINBOW): an open-label randomised controlled trial. Lancet. 2019;394(10208):1551–1559. doi:10.1016/S0140-6736(19)31344-3 [CrossRef] PMID:31522845
  23. Toy BC, Schachar IH, Tan GSW, Moshfeghi DM. Chronic Vascular Arrest as a Predictor of Bevacizumab Treatment Failure in Retinopathy of Prematurity. Ophthalmology. 2016;123(10):2166–2175. doi:10.1016/j.ophtha.2016.06.055 [CrossRef] PMID:27506484
  24. Garcia Gonzalez JM, Snyder L, Blair M, Rohr A, Shapiro M, Greenwald M. Prophylactic peripheral laser and fluorescein angiography after bevacizumab for retinopathy of prematurity. Retina. 2018;38(4):764–772. doi:10.1097/IAE.0000000000001581 [CrossRef] PMID:28267112
  25. Kennedy KA, Mintz-Hittner HABEAT-ROP Cooperative Group. Medical and developmental outcomes of bevacizumab versus laser for retinopathy of prematurity. J AAPOS. 2018;22(1):61–65.e1. doi:10.1016/j.jaapos.2017.10.006 [CrossRef] PMID:29223789
  26. Morin J, Luu TM, Superstein R, et al. Canadian Neonatal Network and the Canadian Neonatal Follow-Up Network Investigators. Neuro-developmental Outcomes Following Bevacizumab Injections for Retinopathy of Prematurity. Pediatrics. 2016;137(4):e20153218. doi:10.1542/peds.2015-3218 [CrossRef] PMID:27244705
  27. Blair MP, Shapiro MJAd Hoc Group Concerning Neurodevelopment and anti-VEGF. Re: Good: Bevacizumab for retinopathy of prematurity: treatment when pathology is embedded in a normally developing vascular system (Ophthalmology. 2016;123:1843–1844). Ophthalmology. 2017;124(10):e74–e75. doi:10.1016/j.ophtha.2017.05.007 [CrossRef] PMID:28938930
  28. Rodriguez SH, Peyton C, Lewis K, et al. Neurodevelopmental Outcomes Comparing Bevacizumab to Laser for Type 1 ROP. Ophthalmic Surg Lasers Imaging Retina. 2019;50(6):337–343. doi:10.3928/23258160-20190605-01 [CrossRef] PMID:31233150
  29. Lien R, Yu M-H, Hsu K-H, et al. Neurodevelopmental Outcomes in Infants with Retinopathy of Prematurity and Bevacizumab Treatment. PLoS One. 2016;11(1):e0148019. doi:10.1371/journal.pone.0148019 [CrossRef] PMID:26815000
  30. Chen TA, Schachar IH, Moshfeghi DM. Outcomes of Intravitreal Bevacizumab and Diode Laser Photocoagulation for Treatment-Warranted Retinopathy of Prematurity. Ophthalmic Surg Lasers Imaging Retina. 2018;49(2):126–131. doi:10.3928/23258160-20180129-07 [CrossRef] PMID:29443362
  31. Natarajan G, Shankaran S, Nolen TL, et al. Neurodevelopmental Outcomes of Preterm Infants With Retinopathy of Prematurity by Treatment. Pediatrics. 2019;144(2):e20183537. doi:10.1542/peds.2018-3537 [CrossRef] PMID:31337693
  32. Beligere N, Perumalswamy V, Tandon M, et al. Retinopathy of prematurity and neurodevelopmental disabilities in premature infants. Semin Fetal Neonatal Med. 2015;20(5):346–353. doi:10.1016/j.siny.2015.06.004 [CrossRef] PMID:26235349
  33. Bolisetty S, Dhawan A, Abdel-Latif M, Bajuk B, Stack J, Lui KNew South Wales and Australian Capital Territory Neonatal Intensive Care Units' Data Collection. Intraventricular hemorrhage and neurodevelopmental outcomes in extreme preterm infants. Pediatrics. 2014;133(1):55–62. doi:10.1542/peds.2013-0372 [CrossRef] PMID:24379238
  34. Schmidt B, Asztalos EV, Roberts RS, Robertson CM, Sauve RS, Whitfield MFTrial of Indomethacin Prophylaxis in Preterms (TIPP) Investigators. Impact of bronchopulmonary dysplasia, brain injury, and severe retinopathy on the outcome of extremely low-birth-weight infants at 18 months: results from the trial of indomethacin prophylaxis in preterms. JAMA. 2003;289(9):1124–1129. doi:10.1001/jama.289.9.1124 [CrossRef] PMID:12622582
  35. Weisz DE, McNamara PJ. Patent ductus arteriosus ligation and adverse outcomes: causality or bias?J Clin Neonatol. 2014;3(2):67–75. doi:10.4103/2249-4847.134670 [CrossRef] PMID:25024972
  36. Rees CM, Pierro A, Eaton S. Neurodevelopmental outcomes of neonates with medically and surgically treated necrotizing enterocolitis. Arch Dis Child Fetal Neonatal Ed. 2007;92(3):F193–F198. doi:10.1136/adc.2006.099929 [CrossRef] PMID:16984980
  37. Schlapbach LJ, Aebischer M, Adams M, et al. Swiss Neonatal Network and Follow-Up Group. Impact of sepsis on neurodevelopmental outcome in a Swiss National Cohort of extremely premature infants. Pediatrics. 2011;128(2):e348–e357. doi:10.1542/peds.2010-3338 [CrossRef] PMID:21768312
  38. Adams-Chapman I, Hansen NI, Stoll BJ, Higgins RNICHD Research Network. Neurodevelopmental outcome of extremely low birth weight infants with posthemorrhagic hydrocephalus requiring shunt insertion. Pediatrics. 2008;121(5):e1167–e1177. doi:10.1542/peds.2007-0423 [CrossRef] PMID:18390958
  39. Imamura T, Ariga H, Kaneko M, et al. Neurodevelopmental outcomes of children with periventricular leukomalacia. Pediatr Neonatol. 2013;54(6):367–372. doi:10.1016/j.pedneo.2013.04.006 [CrossRef] PMID:23727042
  40. Blair MP, Hilkert Rodriguez S, Chan RVP, Shapiro MJ. Re: Lepore et al.: Follow-up to age 4 years of treatment of type 1 retinopathy of prematurity intravitreal bevacizumab injection versus laser: fluorescein angio-graphic findings (Ophthalmology. 2018;125:218–226). Ophthalmology. 2018;125(10):e70–e71. doi:10.1016/j.ophtha.2018.04.011 [CrossRef] PMID:30243341
  41. Msall ME, Phelps DL, DiGaudio KM, et al. Behalf of the Cryotherapy for Retinopathy of Prematurity Cooperative Group. Severity of neonatal retinopathy of prematurity is predictive of neurodevelopmental functional outcome at age 5.5 years. Pediatrics. 2000;106(5):998–1005. doi:10.1542/peds.106.5.998 [CrossRef] PMID:11061766
  42. Roberts G, Howard K, Spittle AJ, Brown NC, Anderson PJ, Doyle LW. Rates of early intervention services in very preterm children with developmental disabilities at age 2 years. J Paediatr Child Health. 2008;44(5):276–280. doi:10.1111/j.1440-1754.2007.01251.x [CrossRef] PMID:17999667
  43. Markus AR, Krohe S, Garro N, Gerstein M, Pellegrini C. Examining the association between Medicaid coverage and preterm births using 2010– 2013 National Vital Statistics Birth Data. J Child Poverty. 2017;23(1):79–94. doi:10.1080/10796126.2016.1254601 [CrossRef]
  44. Tamaru S, Kikuchi A, Takagi K, et al. Neurodevelopmental outcomes of very low birth weight and extremely low birth weight infants at 18 months of corrected age associated with prenatal risk factors. Early Hum Dev. 2011;87(1):55–59. doi:10.1016/j.earlhumdev.2010.10.004 [CrossRef] PMID:21078549
  45. Patra K, Greene MM, Patel AL, Meier P. Maternal Education Level Predicts Cognitive, Language, and Motor Outcome in Preterm Infants in the Second Year of Life. Am J Perinatol. 2016;33(8):738–744. doi:10.1055/s-0036-1572532 [CrossRef] PMID:26890439
  46. Johnson S, Evans TA, Draper ES, et al. Neurodevelopmental outcomes following late and moderate prematurity: a population-based cohort study. Arch Dis Child Fetal Neonatal Ed. 2015;100(4):F301–F308. doi:10.1136/archdischild-2014-307684 [CrossRef] PMID:25834170

Baseline Characteristics

Injection (n = 63)Laser (n = 235)

n (%)n (%)P Valuea

Neurologic Comorbidity22 (35%)54 (23%).05
  Hydrocephalus13 (21%)28 (12%).07
  Periventricular leukomalacia<11* (<17%)<11* (<5%).29
  Severe intraventricular hemorrhage17 (27%)42 (18%).11

Thrombocytopenia11 (17%)18 (8%).02

Sepsis29 (46%)92 (39%).32

Bronchopulmonary Dysplasia43 (68%)151 (64%).55

Necrotizing Enterocolitis14 (22%)40 (17%).34

Patent Ductus Arteriosus Repair17 (27%)55 (23%).56

Male Gender34 (55%)135 (57%).62

Median Length of Stay, Days (IQR)115 (81–149)103 (78–128).06

U.S. Census Region< .001
  Northeast<11* (<17%)48 (20%)
  North Central<11* (<17%)56 (24%)
  South44 (70%)85 (36%)
  West<11* (<17%)>35* (>15%)
  Unknown<11* (<17%)<11* (<5%)

Ocular Outcomes for Infants Receiving Laser or Injection

Injection (n = 63)Laser (n = 235)
n (%)n (%)P Valuea
Retinal Detachment< 11* (< 17%)24 (10%).87
Vitreous Hemorrhage< 11* (< 17%)< 11* (< 5%)1.00
Cataract< 11* (< 17%)<11* (< 5%).01
Glaucoma< 11* (< 17%)< 11* (< 5%)1.00
Strabismus24 (38%)81 (34%).59
Corneal Scar< 11* (< 17%)< 11* (< 5%).51
Endophthalmitis001.00
Retina Surgery< 11* (< 17%)12 (5%).75
Cataract Surgery< 11* (< 17%)< 11* (< 5%).58
Second Procedure28 (44%)23 (10%)< .001
Median Time to Second Procedure, Days (IQR)42 (28–106)39.5 (13–63).34
Median Time to Second Laser, Days (IQR)88 (58–131)38 (15–70).01
Median time to second injection, days (IQR)33 (15–42)25 (22–27).19

Neurodevelopmental Outcomes of Infants Receiving Laser or Injection

Injection (n = 63)Laser (n = 235)
n (%)n (%)P Valuea
Cerebral palsy16 (25%)29 (12%).01
Speech delay22 (35%)95 (40%).42
Cognitive delay18 (29%)64 (27%).83
Motor delay< 11* (< 17%)36 (15%).60
Any developmental delay56 (89%)203 (87%).60
Hearing loss20 (32%)70 (30%).76
Blindness< 11* (< 17%)16 (7%)1.00

ICD-9, ICD-10, and CPT Codes Used to Identify Cases of Retinopathy of Prematurity and Assess Initial Therapies, Baseline Characteristics, Ocular Outcomes, and Neurodevelopmental Outcomes

CodeICD-9ICD-10CPT

Initial Diagnosis of ROP362.2XH35.X

Initial Treatment
  Laser67229, 67228
  Injection67028

Baseline Criteria
  Hydrocephalus331.4, 331.7, 742.3G91.9, Q03.X
  Periventricular leukomalacia779.7P91.2
  Severe intraventricular hemorrhage772.13, 772.14P52.21, P52.221
  Thrombocytopenia287.XD69.X
  Sepsis771.81P36.X
  Bronchopulmonary dysplasia770.7P27.X
  Necrotizing enterocolitis777.5XP77.X
  Patent ductus arteriosus repairV45.89Z98.8933820

Ocular Outcomes
  Retinal detachment361.XH33.X67017, 67036, 67113, 67108, 67109, 67110
  Vitreous hemorrhage379.23H43.X
  Cataract diagnosis366.X, 743.XH26.0X, H26.1X, Q12.X, Q14.0
  Glaucoma365.X, 743.2H40.X, Q13.2, Q15.X
  Strabismus378.XH50.X
  Corneal scar371.00–371.03H17.9, H17.1X, H17.81X, H17.82X
  Endophthalmitis360.00, 360.01, 360.02, 360.03, 360.04, 360.19, 364.05H44.0, H44.00X, H44.01X, H44.19, H20.05X
  Cataract surgery66850, 66852, 66982, 66984
  Retina surgery67017, 67036, 67113, 67108, 67109, 67110
  Second procedurea67028, 67228, 67229

Neurodevelopmental Outcomes
  Cerebral palsy343.XG80.X
  Speech delay315.3XF80.X
  Cognitive delay315.9F81.X
  Motor delay315.4F82.X
  Any developmental delay315.X, 783.XR62.5
  Blindness369.XH54.X
  Hearing loss389.XH91.X
Authors

From Pritzker School of Medicine, The University of Chicago, Chicago, Illinois (MHZ); the Department of Ophthalmology and Visual Science, The University of Chicago, Chicago, Illinois (MPB, SHR); Retina Consultants, Ltd., Des Plaines, Illinois (MPB); and the Center for Health and the Social Sciences, The University of Chicago, Chicago, Illinois, (SAH).

Presented at American Association for Pediatric Ophthalmology and Strabismus annual meeting, San Diego, California, March 2019, and the ROP Hot Topics meeting, Chicago, Illinois, October 2018.

Supported by the National Center for Advancing Translational Sciences of the National Institutes of Health (Bethesda, MD) through Grant No. UL1 TR000430. The funding organizations had no role in the design or conduct of this research.

The authors report no relevant financial disclosures.

Address correspondence to Sarah H. Rodriguez, MD, MPH, Department of Ophthalmology, University of Chicago, 5841 S. Maryland Avenue, MC2114, Chicago, IL 60637; email: srodriguez5@bsd.uchicago.edu.

Received: February 24, 2020
Accepted: July 08, 2020

10.3928/23258160-20200831-02

Sign up to receive

Journal E-contents