Dr. Parag K Shah, S09271, Dr. Narendran V
Introduction
India has the highest number of premature births in the world (WHO, Born Too Soon Report, 2010). Neonatal care units earlier were restricted to only urban areas but from 2005, Government of India under the National Rural Health Mission (NRHM) planned to open Special Neonatal Care Units (SNCUs) in every district hospital. Currently about 300 SNCUs are reported to be operational in the country. Thus with this improvement in neonatal care, many more preterm babies survive even in the rural parts of India. This has led to the increase in the incidence of retinopathy of prematurity (ROP). In addition, as there are limited numbers of ophthalmologists trained in ROP screening, there is a large lacuna of providing ROP screening services especially in these rural areas.
Methods
We started a ROP tele-screening program called ROPE-SOS (Retinopathy of Prematurity Eradication – Save Our Sight). The main aim of this program was to do ROP screening in the underserved and rural areas by a trained technician (non-ophthalmologist) using a retinal camera (Retcam Shuttle, Clarity Medical Systems, Pleasanton, California, USA); identifying babies with this blinding disease real-time by transmitting the retinal images to a remote ROP expert (via broadband internet) and either immediately referring these babies requiring treatment to the base hospital or if the child is too sick to travel, prompt onsite laser treatment would be done by the ROP expert visiting the SNCU with portable diode laser within 3 days.
The team comprised of a manager who was responsible to coordinate the weekly visits with the local pediatric nurse in the SNCUs, two trained technicians who were responsible to capture the retinal images by the Retcam Shuttle camera and transmit it to the remote ROP expert via broadband internet, one mid-level ophthalmic assistants who assists in dilating the eyes and taking care of the babies during the screening, a driver who takes the entire team along with the Retcam Shuttle camera to the SNCUs in a van and a remote ROP expert to grade the images and send the reports. The team will be visited SNCUs situated in districts of Coimbatore, Tirupur, Erode, Salem, Karur, Namakkal of Tamilnadu State and Palakkad, Thrissur and Ernakulam districts of Kerala State, where ROP screening is non-existent at present.A fixed day of the week was allotted to all the SNCUs. They visited not only government SNCUs but also private ones who were willing to get enrolled in this program. Continuous medical education programs on ROP were also held in each SNCU regularly for creating awareness amongst the neonatologist, the nursing staff and families of preterm babies.
With this program we planned to reach the unreached by screening at least 4000 babies in the SNCUs situated in these smaller cities (tier 2 and tier 3 cities) of Tamilnadu & Kerala during the first year.
Results
Total 59 neonatal units were covered across 18 cities. Fixed day and time was given for each unit and the team visited them every week. From 12th August 2015 till 14th May 2017, 7299 babies were screened (4234 new + 3065 review. Any stage ROP was seen in 1431 babies.111 babies (213 eyes) were identified with vision threatening ROP. Laser was given in 129 eyes, intravitreal Avastin injection in 78 eyes and Lucentis injection in 6 eyes. Vitrectomy (in addition to laser) was performed in 4 eyes. The mean gestational age and birth weight of the treated babies was 30 weeks and 1269 g respectively. Apart from ROP, 203 babies with retinal hemorrhages, ten with congenital cataracts and one each having retinoblastoma and strabismus were also diagnosed.
Multi-color ROP information posters in English, Tamil & Malayalam languages were prepared and displayed in the patient waiting area of various NICU’s. Tamil, Malayalam & English patient information brochures (handouts) are also being distributed to all the patients.ROP Continuous Medical Education (CME) programs were held once every 6 to 8 weeks at various NICUs. From August 2015 till May 14, 2017, we have conducted 11 ROP CME programs in the various district hospitals. 718 participants were sensitized on ROP screening.
Discussion
We compared our ROPE-SOS Project with KIDROP Project (Narayana Nethralaya, Bangalore) and Vittala ROP Project (Vittala International Institute of Ophthalmology, Bangalore).All the three projects use Retcam Shuttle (Clarity Medical Systems, Pleasanton, California, USA) camera for ROP telescreening. In all the three projects, Retcam is transported across the various NICUs every week in rural India through a mobile van and images were taken by trained non-ophthalmologists. The images are uploaded on a secure server and graded remotely by a trained ROP specialist from base hospital. The reports are sent immediately to the NICU and the family is counseled by the project manager or technician.
KIDROP Project used the TeleCare Software (i2i Telesolutions and Telemedicine Pvt. Ltd, Bangalore) to upload and grade the images. In KIDROP study 1601 babies were screened from 2007 till 2013 involving 81 NICUs from 18 districts (IJO 2014; 62:41-9). In another publication, it mentions that 7106 babies were screened from 2011 to 2015 involving 33 NICUs from 13 districts (Semin Fetal Neonatal Med 2015; 20:335-45). They covered a population of ~23 million and an area of 89,000 km2. Of the 7106 babies screened, 1591 (22.39%) had any stage of ROP, of whom 254 (15.96%) or 3.57% of all those screened required treatment. 11 eyes had unfavorable outcomes and the remaining 497 eyes had favorable structural outcomes accounting for 97.83%. 50 babies (19.69%) were having birth weight >1500 g and gestational age >30 weeks and all these would have been missed if American ROP screening guidelines were implemented. Vittala ROP Project (Br J Ophthalmol. 2013; 97:687-9) used the LAMP software (Linux, Apache, MySQL and PHP) to upload and grade the images. During the period between April 2010 and October 2011, 512 infants were screened in 6 districts of Karnataka. Any stage of ROP was seen in 78 babies (15.6%) and 28 babies (35.9%) or 5.6% of all those screened required treatment. Our ROPE-SOS Project used the indigenous modified ADRES (Aravind Diabetic Retinopathy Screening) Telescreening Software to upload and grade the images. We covered a population of ~47 million and an area of 59,000 km2 involving 56 NICUs. 5017 were screened from Aug 2015 to Dec 2016. Any stage of ROP was seen in 854 babies (17%) and 74 babies (8.6%) or 1.47% of all those screened required treatment. Of the 74 babies (139 eyes), 3 babies (6 eyes) developed stage 4A, of which four eyes needed vitrectomy and two stabilized only with laser. There were no cases of stage 4B or stage 5 seen. Thus the favourable outcome was seen in 95.7%. 16 babies (21.6%) had birth weight >1500 g birth and gestational age >30 weeks and would have been missed if American ROP screening guidelines were implemented.
In KIDROP Project, instead of uploading all images to be graded by the ROP expert, technicians are trained to diagnose and decide follow-up. In this program the technicians upload only the images they feel are severe enough to warrant treatment, for ROP expert opinion. The advantage of technicians self-reporting is that it can be less time consuming, cheaper as internet cost is reduced and the load on ROP expert is reduced. In the same study, they reported that the erroneous reporting by the technicians ranged from 4.7 to 0.9%. This has medico legal implications, especially if a child goes blind on erroneous reporting by a non-ophthalmologist. For ROP expert reporting remotely, the turnaround time for each baby is only 10 to 15 minutes and with the availability of 3G network in rural India, the internet speed is good while cost is less than INR 3000 per month. Hence it is not time consuming and not costly. Training multiple doctors on remote ROP reporting (which we have done in our project) would also reduce the reporting load on a single ROP expert. ROP expert also has the knowledge of the natural history of the disease which will decide the next follow up. With a shift towards anti VEGF treatment, wherein the entire natural history of ROP changes, reporting by technicians can be tricky.
In ROPE-SOS Project Continuous Medical Education (CME) programs on ROP were held every 6 to 8 weeks wherein the NICU staff is educated about the preventive aspect of ROP. Both the KIDROP and Vittala ROP Project reported more number of severe ROP (especially AP-ROP) in NICUs from rural India. We too saw the same, but after educating the NICU staff regarding preventive aspect of ROP (stressing on usage of blended oxygen) through the CME programs, we have seen significant reduction in AP-ROP cases from those NICUs. In one NICU from our project district, we saw 4 babies with AP-ROP from Feb to Aug 16. After educating the NICU staff regarding harmful effects of unblended oxygen overuse, not a single baby with AP-ROP was seen from Sept 16 till date. Having regular ROP CME programs could be a reason why in our project the babies requiring treatment was the least (1.47%) compared to KIDROP (3.57%) and Vittala ROP Project (5.6%). In ROPE-SOS project, universal eye screening is done from taking pictures of anterior segment of all babies followed by posterior segment imaging. Amongst anterior segment diseases we could pick up seven cases with congenital cataract, one case each with strabismus, corneal opacity and hordeoleum externum. Amongst posterior segment diseases we could pick retinal haemorrhages in 154 babies and retinoblastoma in one child.
Conclusion
Telescreening for ROP is a feasible and replicable option for developing countries, where there are no ophthalmologists to screen. We recommend regular ROP CME programs to be incorporated in the future projects, which may help in preventing ROP.
References?
Leave a Comment