Preventing blindness in premature infants is a pretty easy concept to champion but a much more difficult thing to achieve in underserved populations in the U.S. and the developing world. Known as "retinopathy of prematurity" or ROP, this form of blindness is an ironic affliction made possible by our modern age's extraordinary advances in
Photo: Hoag Levins
Images taken with a special eye camera indicate which premature infants may have the abnormal blood vessel growth that, if not treated, could blind them for life. Larger image
biomedical technology. Throughout most of human history infants weighing less than 1,500 grams or born before 31 weeks, routinely died. Today, they can survive and flourish, but premature exposure to the outside world and the difficulties of maintaining life in the premature baby can also promote the growth of abnormal blood vessels in their retinas and high risk of blindness.
Early ROP epidemic This condition first emerged in epidemic proportions in the 1940s and 50s when hospitals began using supplemental oxygen in incubators. After scientists and clinicians realized that oxygen increased the risk of ROP, they administered oxygen more carefully and the incidence of ROP declined. But it did not disappear as advances in neonatal care improved survival for even smaller and more premature infants.
If severe ROP is caught early, laser therapy can prevent a significant percentage of such blindness. The key is screening all babies at risk and identifying the ones needing treatment. Current U.S. guidelines call for all infants weighing less than 1500 grams to be screened serially between 4 and 6 weeks after birth by a pediatric ophthalmologist trained in ROP.
Today, the blindness burden of ROP has largely shifted to moderately developed areas such as Latin America and the former socialist economies, where about 60% of the 50,000 children gone blind from ROP reside.
"The vast majority of kids going blind nowadays are in these developing countries," explains Dr. Graham Quinn, a pediatric ophthalmologist at the Children's Hospital of Philadelphia and a longtime ROP specialist. In these areas, more and more premature babies are being saved, but ophthalmologic expertise
Photo: Hoag Levins
Alternative content
Pediatric ophthalmologist Dr. Graham Quinn of Children's Hospital in Philadelphia is conducting a study of how telemedicine might be used to extend ROP-detecting eye scans to larger populations of underserved children in both the U.S. and developing countries. Larger image
in ROP and expertise in neonatal care lags behind. ROP is becoming an increasing problem in underserved areas of the U.S. and Canada as well, because there aren't enough pediatric ophthalmologists experienced in ROP to screen all babies at risk.
Ophthalmologist shortage "In Montana, there are just three pediatric ophthalmologists in the whole state, and one of them doesn't do ROP," Dr. Quinn noted.
The question now generating debate is whether a pediatric ophthalmologist must be the one to perform the initial screening, given that more than 90% of babies will need no treatment at all. In 2003, Canadian ROP specialist Anna Ells published validated standards for "referral-warranted ROP," which could identify about 14% of the at-risk babies who would benefit from a full exam by an experienced ophthalmologist. In developed counties, using pediatric ophthalmologists for this initial screening might represent inefficient use of expensive, well-trained, but scarce human resources; in developing countries, the human resources simply aren't there.
ROP telemedicine A possible solution? Dr. Quinn points to a "telemedicine paradigm," in which trained non-specialists take digital images of the eye, electronically transmit the image for remote reading and identify babies with referral-warranted ROP. In other retinal disorders such as diabetic retinopathy and age-related macular degeneration, digital photography with remote evaluation has been used to reliably identify and manage disease. The paradigm is "taking a high-tech world into a low-tech environment, with potentially high yield," Dr. Quinn says.
In the past two decades, telemedicine programs for ROP have been launched around the world and in a few neonatal intensive care units (NICUs) in the US. The spread of this technology is a study in the
Photo: Hoag Levins
Manufactured by Clarity Medical Systems, the RetCam can be used by clinical technicians to take photos of a premature infant's retinas that can be emailed to remote ophthalmologists for analysis. Larger image
diffusion of innovation.
Dr. Quinn notes that, "there are two ways in which technology is accepted. Either, you have a big study that says here are the specific problems you're going to bump into, and here is the sensitivity, specificity and the economy of it all. Or, you have generally adopted things and you find out about the sensitivity and specificity over time." He thinks the big study approach is more efficient, but more "painful."
Definitive study He and international colleagues spent three years refining the research questions, visiting centers in Latin America, and seeking funding for a definitive study. In 2010, he received funding from the National Eye Institute for a 4-year study of 2,000 high-risk babies in 10 centers in the US and Canada, but the NEI did not fund the international centers. According to Dr. Quinn, the NEI was concerned that the centers have not had the capacity to conduct the research, a decision he called, "shortsighted." In 2006, Dr. Quinn and his colleagues had successfully conducted a telemedicine feasibility study in Peru and Brazil with seed funding from the University of Pennsylvania. The study demonstrated that NICU nurses could be trained to obtain and transfer high-quality retinal images for screening premature babies.
Program in India But the rapidly developing world is not waiting for the results of Dr. Quinn's larger study. Dr. Anand Vinekar, a colleague in Bangalore, India has been using a mobile camera called the RetCam to screen premature babies in parts of India, which has many tiny NICUs spread across remote villages. A "really savvy" technician travels around to these NICUs in an SUV, takes digital images of the eye, reads them himself and transmits any concerning images to the ROP specialist in Bangalore
Photo: NNP Institute, Bangalore
Anand Vinekar, MD, explains the eye scanning procedure to mothers of premature infants in Bangalore. Larger image
on his iPhone. "The technician says, 'Anand, I think you need to come look at this kid.' They've done 15,000 kids. Isn't that great?" Dr. Quinn remarks.
Despite his enthusiasm for the idea, Dr. Quinn believes in the need for rigorous research before widespread use of the technology. The ongoing study will evaluate the validity, reliability, feasibility, safety and relative cost-effectiveness of an ROP telemedicine system. The system involves training non-physicians to take digital images of the retina using a RetCam Shuttle, training non-specialists to read the image remotely, and referring babies with signs of serious ROP to an on-site ophthalmologist for evaluation and treatment. To directly compare the telemedicine system with usual care, all babies will be screened by both retinal imaging and by the standard pediatric ophthalmologic exam on the same day.
Problems in the field Even the most thoughtful, well-designed studies encounter problems in the field, and the telemedicine study is no exception. Study Director Agnieshka Baumritter notes that recruitment of study participants has been slow in some sites. "Remember, these are tiny infants, and the parents and NICU nurses are very protective of them, and justifiably so," she explains. Although the RetCam is considered non-invasive, it does involve using a metal speculum to keep the baby's eye open for imaging. To overcome some of these barriers, the researchers
Photo: Hoag Levins
As Karen Karp, RN (right), holds an infant, ophthalmic photographer Denise Pearson scans an image of his retina into the RetCam in the Neonatal Intensive Care Unit at Children's Hospital of Philadelphia. Larger image
conduct educational sessions with the NICU nurses, and show parents examples of the digital images.
Each center is selecting non-physicians to train as "Certified ROP Imagers (CRIs)". Some CRIs are nurse practitioners, NICU nurses or technicians. Baumritter comments on the ongoing debate about the most important skills that imagers should have before training. "Does the person need to be technology savvy, or NICU-trained, someone who's not afraid of handling the baby?" In unison, Dr. Quinn and Ms. Baumritter said, "It's the NICU part" that's more important, and "more scary". But Ms. Baumritter laughs and notes that she has been trained as a CRI and had neither skill previously, which bodes well for the system's adaptability across settings.
The telemedicine system might actually improve upon screening by ophthalmologists. There is some evidence, Dr. Quinn says, that ROP can be detected earlier through digital imaging than by pediatric ophthalmologists. The system standardizes the images and the interpretation of the images, thereby removing some of the more subjective interpretations of human experts. It might be more convenient as well. If the camera is kept in the NICU, the exam can be conducted when babies are stable and available, rather than when the ophthalmologist arrives.
$70,000 RetCam cost If the telemedicine system turns out to be effective in the research settings, what are its prospects for widespread adoption? The RetCam, the only retinal camera on the market, now costs $70,000, which could limit its purchase by small hospitals in the U.S. and make it unaffordable throughout much of the rapidly developing world.
But Dr. Quinn is hopeful that if the system is successful, a market will develop for retinal imaging and the price will go down. Looking even further into the future, he envisions a system marrying consumer electronics and ROP screening. "Where I think we're going is... a little camera like an iPhone that can be assigned to a baby, has digital images done... and actually analyzed on that camera. Isn't that cool?"
~ ~ ~
Janet Weiner, MPH, is Associate Director for Health Policy at the Leonard Davis Institute of Health Economics within the University of Pennsylvania. She holds a journalism degree from Northwestern University, a masters degree in public health from Johns Hopkins University, and is currently a doctoral candidate in City and Regional Planning at Penn. weinerja@mail.med.upenn.edu
EDITOR'S NOTE: This is the first in a series of new Simply Better Health Care columns by Janet Weiner that will highlight policy and practice initiatives that promise to make our health care more responsive to human needs and more reflective of our shared values. It will explore the practical ways in which creative health care professionals are making more effective use of what we already know; reaching more people with proven treatments; and overcoming time, distance, financial and cultural barriers to care.
