Retinopathy of prematurity (ROP) is a vasoproliferative disease affecting premature infants. Risk of ROP is highest in low birth weight infants (less than 1,250 g) and infants younger than 31 weeks gestational age. The severity of the disease is characterized based on the anatomical zone of involvement (zones 1 to 3), stage of disease (stages 1 to 5), and the presence of plus disease. Plus disease is defined as venous dilation and arterial tortuosity in the posterior pole in two quadrants of the retina. Traditionally, treatment was based on the presence of threshold disease.1
Originally, the Multicenter Trial of Cryotherapy for Retinopathy of Prematurity (CRYO-ROP) study-defined threshold disease was treated with cryotherapy. Following that, the Early Treatment of Retinopathy of Prematurity (ETROP) treatment criteria allowed for cryotherapy or laser ablation. The CRYO-ROP study, published in 1988, showed that cryotherapy reduced the rate of unfavorable structural outcomes from 47% to 25% at 1-year follow-up.2 Since that time, the ETROP study defined two groups of patients with pre-threshold ROP (type 1 and type 2) with a structural failure rate of 9%.3 The treatment of type 1 prethreshold disease resulted in better structural outcomes compared with the traditional treatment of threshold disease as described in the CRYO-ROP study.2 The ETROP study helped further establish laser as early standard-of-care treatment and provided for better structural outcomes.3
The post menstrual age (PMA) of infants screened by conventional bedside examination to reach treatment warranted ROP was compared to those infants screened photographically and read remotely using telemedicine. Earlier studies have shown photographic ROP screening to be as accurate as beside examinations.4 The photoROP study5 and more recently EROP6 have shown that photographic screening can accurately determine treatment-warranted ROP. We wanted to determine if photographic screening allowed for time appropriate diagnosis and timely treatment of ROP.
Patients and Methods
This retrospective clinical study assesses whether asynchronous store-forward telemedicine screening allows for a diagnosis of treatment-warranted ROP that is as timely and accurate as bedside examination. An already assembled data set from the Vermont Registry was used for the data analysis of this study. The Vermont Registry is a national network that records data on premature infants from many centers. One of the conditions they follow is ROP. The patients studied were infants from the neonatal intensive care unit (NICU) at William Beaumont Hospital, Royal Oak, Michigan. The study includes neonates born from January 2006 to December 2014. Exclusive indirect ophthalmoscopy was used as primary screening from 2006 to 2010, and exclusive camera as primary screening was used in 2010 to 2014. In addition, the files were viewed on William Beaumont Hospital-affiliated computers to be HIPAA compliant. A SharePoint folder was utilized under the regulation of the institutional review board. Medical record numbers were used to track data instead of patient names.
Images from photographic screening were obtained by nurse practitioners on an average of once a week. During bedside examinations, physicians drew diagrams to document their findings. RetCam 120 (Clarity Medical Systems, Pleasanton, CA) was used for photographic screening of retinopathy of prematurity. Images were read on a weekly basis. Bedside exams were performed by board-certified ophthalmologists experienced in ROP.
Once the patient files were obtained, the data were sorted out to examine the number of eyes and the postmenstrual age (PMA) of the patients with treatment-warranted ROP. Both methods of diagnosis were compared to view how timely telemedicine was in comparison to bedside examinations. The PMAs at time of treatment were compared and a P value was generated. This will be a new piece of information to help NICUs determine if photographic screening is the approach they wish to adopt.
In this exploratory study, the inclusion and exclusion criteria were specific to in born neonates diagnosed with ROP prior to any treatment at William Beaumont Hospital from 2006 to 2014. In total, 65 neonates fit these criteria.
Of the 65 neonates with treatment-warranted ROP who fit the exclusion and inclusion criteria, 35 were diagnosed at bedside (2006 to 2010) and 30 were photographically diagnosed (2010 to 2014).
Gestational age for both beside examined and telemedicine-screened patients was collected and compared for statistical difference. For infants diagnosed at bedside, the range of gestational age is from 23.0 weeks to 29.4 weeks, with an average of 24.9 weeks ± 1.5 weeks. For telemedicine-screened patients, the ranges for gestation age are 23.1 to 26.4 weeks, with an average of 24.6 weeks ± 1.0 weeks. There was no statistical difference in gestational age between either group (P = .46).
Birth weights for both groups were also collected and compared for statistical difference. The range of birth weights for beside examined neonates was 445 g to 900 g, and the range for the telemedicine-screened group was 447 g to 950 g. There was a statistical difference in birth weight, as bedside-screened neonates had an average of 618.9 g ± 115.9 g and telemedicine-screened had an average of 683 g ± 134 g (P = .043). This statistical difference is corrected for when PMA values are compared.
From 2006 to 2010, 35 NICU patients were found to need treatment for ROP through bedside examination, with an average PMA of 36.5 weeks ± 2.4 weeks. From 2010 to 2014, a total of 30 infants were found to need treatment for ROP through photographic screening. The average PMA at treatment for the telemedicine-screened patients was 36.4 weeks ± 2.6 weeks. There was no statistical difference in PMA of both groups after correcting for birth weight (P = .58). These data suggest that telemedicine is just as timely as bedside screening in determining treatment-warranted ROP.
Additional data collected include stage of disease, zone, plus, and number of laser spots delivered. These data were not statistically compared, as charts were not uniform in including all information.
This project is significant because it will contribute to the literature that telemedicine screening to diagnose ROP is as effective and timely when compared to bedside examinations. It is critical for physicians to be able to make a diagnosis through telemedicine as accurately and as timely as done by bedside examinations.
Photographically obtained images provide better documentation than drawings.
(A) Gestational ages (weeks) for bedside-examined, treatment-warranted retinopathy of prematurity (ROP) in neonates. The ranges are from 23 weeks to 29 weeks for bedside-examined, treatment-warranted ROP. (B) Gestational ages (weeks) for photographically screened, treatment-warranted ROP neonates. The ranges are from 23 weeks to 26 weeks for photographically screened, treatment-warranted ROP.
Comparison of postmenstrual age at the time of diagnosis for treatment-warranted retinopathy of prematurity (ROP) in bedside-examined and photographically screened neonates. This double bar graph displays the postmenstrual age of infants who were bedside examined or photographically screened for treatment-warranted ROP.
In addition, if telemedicine was determined in larger studies to be a more timely method in detecting treatment warranted ROP, then a larger percentage of patients would be eligible for the ETROP treatment, which could provide for better outcomes. Telemedicine could allow hospitals with limited access to ophthalmologists to provide quality care and preclude patient transfer for treatment that is not needed. In addition, it is expected that telemedicine screening will reduce physician time while still maintaining the quality of screening. Providing a change in protocol that reduces physician time may result in better care for the infant.
The limitations of the study mainly pertain to sample size. The study concludes that there is not a significant difference between the two populations (bedside- and telemedicine-screened neonates) in terms of appropriate timing of treatment-warranted ROP. Larger studies that include multiple hospital systems may be able to detect differences between the two populations.
This study does suggest that infants monitored by telemedicine for ROP do not have any delay of treatment and may require less NICU to NICU transfer.
- Harnett ME. Pediatric Retina. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2013.
- Cryotherapy for Retinopathy of Prematurity Cooperative Group. Multicenter trial of cryotherapy for retinopathy of prematurity. Arch Ophthalmol. 1988;106:471–479. doi:10.1001/archopht.1988.01060130517027 [CrossRef]
- Early Treatment for Retinopathy of Prematurity Cooperative Group. Revised indications for the treatment of retinopathy of prematurity: Results of the early treatment for ROP randomized trial. Arch Opthalmol. 2003;121(12):1684–1694. doi:10.1001/archopht.121.12.1684 [CrossRef]
- Ells AL, Holmes JM, Astle WF, et al. Telemedicine approach to screening for severe retinopathy of prematurity: A pilot study. Ophthalmology. 2003;110(11):2113–2117. doi:10.1016/S0161-6420(03)00831-5 [CrossRef]
- Balasubramanian M, Capone A, Hartnett ME, Pignatto S, Trese MT. The Photographic Screening for Retinopathy of Prematurity Study (Photo-ROP): Study design and baseline characteristics of enrolled patients. Retina2006;26(Suppl 7):S4–S10. doi:10.1097/01.iae.0000244291.09499.88 [CrossRef]
- Telemedicine approaches to evaluating acute-phase retinopathy of prematurity: Study design. Ophthalmic Epidemiol. 2014;21(4):256–267. doi:10.3109/09286586.2014.926940 [CrossRef]