Welcome to Eurotimes

Daniel Keller
A NEW study by Johns Hopkins University researchers indicates a surprising and hitherto unexpected link between hyperglycaemia and the risk for retinopathy of prematurity in at risk infants.

In a report in the May issue of the Journal of Perinatology, the researchers report that hyperglycaemia during the first month of life increases the risk of ROP. The retrospective chart review revealed that the risk nearly doubled with each 10 mg/dL increase in the average monthly glucose level.

Christoph Lehmann MD and colleagues studied 16 premature infants born at the Johns Hopkins Hospital, Baltimore, US between 1992 and 1997 who developed ROP. They compared these patients with 31 healthy infants matched for gestational age, birth weight, APGAR and SNAP scores, race, sex and antenatal steroids. The SNAP score is an indication of severity of illness at the time of delivery.
All infants with a birth weight of less than 1,500g or a gestational age less than 29 weeks received an initial ophthalmological screening within six weeks of birth. Follow-up examinations were done at the discretion of the ophthalmologist or the neonatologist. Infants were categorised by their most severe ROP examination, which was staged according to the International Classification of ROP.

Based on an average five daily glucose measurements for each infant (more than 7,000 values in all), the researchers found a significant difference in the daily glucose values between the ROP and the healthy infants.
"Patients who developed ROP had higher daily glucose levels... the medians, the maximums, the average. They also had wider glucose swings over the course of the day," Dr Lehmann said.

A logistic regression analysis showed that the risk of ROP increased 1.96-fold for each 10 mg/dL increase in mean glucose level. A multiple regression model factoring in the known effects of oxygen exposure yielded an odds ratio of 2.7 for the development of ROP for each 10 mg/dL.

He reported that the infants who developed ROP were more likely to be on insulin as an indicator of significant hyperglycemia and that they received less glucose through an infusion.
"So despite that, they had higher glucose values. They had significantly altered glucose tolerance compared to the other [control] group," Dr Lehmann said.

Some of the control patients were noted to have had episodes of hyperglycaemia and many did not. But every patient in the ROP group did have episodes of hyperglycemia.
Hyperglycaemia a, defined as insulin levels greater than 150 mg/dL in a neonate, is very common in low birth weight infants, occurring in 45% of infants weighing less than 1,000g and in 80% of infants weighing less than 750g.
The incidence of ROP has not decreased over the years because lower birth weight infants are surviving more often, even those born after as little as 23 weeks of gestation.

Other factors also increased the risk of ROP. Based on the simple logistic regression analysis, a history of dopamine infusion or grades 3 or 4 intraventricular haemorrhage raised the risk 5.4 and 7.3-fold respectively. Other risk factors included blood transfusions, frequent blood gas measurements and ventilator use.
The study also confirmed the link between high oxygen levels and the development of ROP. Vitamin E supplementation decreased the risk of ROP, with an odds ratio of 0.37 for each IU/kg/day in the multiple regression analysis.

Dr Lehmann noted his surprise at finding, after a review of 50 years of research reports, that nobody looked at glucose levels as they affect the development of ROP. He said fungal sepsis, postnatal steroids and intraventricular haemorrhage had all been reported as risks for hyperglycaemia. And these factors are all associated with the development of ROP. In retrospect, these could be seen as indirect markers.
He noted some limitations of the study. For example, the study only looked at glucose values in the first 30 days of life. Moreover, this was a retrospective review rather than a prospective protocol.

While the study shows a correlation between elevated daily glucose levels in the first month of life and the development of ROP, glucose could be a secondary marker for other causes. It remains to be determined by a prospective trial if the elevated glucose levels are just a factor that indicates how severely sick the kids were or if the levels were themselves causative.

"If glucose is truly connected to the pathogenesis of ROP, as our study finds, then we may be able to reduce the risk using tighter glucose control in the first few months of life," he said.

Christoph U Lehmann MD, FAAP
Johns Hopkins University, Baltimore, US
clehmann@jhmi.edu

Top