New Israeli Algorithm Predicts Gestational Diabetes Making Pregnancy Safer

They set out from Beit El; but when they were still some distance short of Efrat, Rachel was in childbirth, and she had hard labor. Genesis 35:16 (The Israel Bible™)

Pregnant women naturally are nervous about undergoing a glucose challenge test, possibly followed by a glucose tolerance test. In the first, a health care professional draws the woman’s blood an hour after she drinks a sweet liquid containing glucose, followed again an hour or two later. 

If her blood glucose is too high – 140 or more – they usually have to return for an oral glucose tolerance test after fasting for eight hours except for drinking water. If the level is 200 or more, they may suffer from gestational diabetes. Gestational diabetes is usually tested for between the 24th and 28th weeks of pregnancy,

But now, thanks to researchers at the Weizmann Institute of Science in Rehovot, a new computer algorithm can predict in the early stages of pregnancy – or even before a woman even becomes pregnant – if she is at a high risk of gestational diabetes. They just reported their findings in the prestigious journal Nature Medicine

The study analyzed data on nearly 600,000 pregnancies available from Israel’s largest health maintenance organization, Clalit Health Services. Based on these predictions, it may be possible to prevent gestational diabetes using nutritional and lifestyle changes, they said. 

“Our ultimate goal has been to help the health system take measures so as to prevent diabetes from occurring in pregnancy,” said senior author Prof. Eran Segal of the computer science and applied mathematics department and the molecular cell biology department.

Gestational diabetes is characterized by high blood sugar levels that develop during pregnancy in women who did not previously have diabetes. It occurs in three to nine percent of all pregnancies and poses significant risks for both mother and baby. 

The diagnosis usually comes as a surprise to women; it also raises the risk of a pregnancy complication called pre-eclampsia, in which they develop high blood pressure and sometimes damage to the liver and kidneys. Preeclampsia usually begins after 20 weeks of pregnancy in women whose blood pressure had been normal. Babies born to mothers with poorly treated gestational diabetes are at increased risk of being too large (macrosomia), having low blood sugar after birth, and jaundice. If untreated, it can also result in a fetus being born dead (stillbirth). 

In the new study, Segal and colleagues started out by applying a machine learning method to Clalit’s health records on some 450,000 pregnancies in women who gave birth between 2010 and 2017. Gestational diabetes had been diagnosed by glucose tolerance testing in about four percent of these pregnancies. 

The work was led by graduate students Nitzan Shalom Artzi, Dr. Smadar Shilo and Hagai Rossman from Eran Segal’s lab at the Weizmann Institute of Science, who collaborated with Prof. Eran Hadar, Dr. Shiri Barbash-Hazan, Prof. Avi Ben-Haroush and Prof. Arnon Wiznitzer of the Rabin Medical Center in Petah Tikva; and Prof. Ran Balicer and Dr. Becca Feldman of Clalit Health Services.

After processing big data – an enormous dataset made up of more than 2,000 parameters for each pregnancy, including the woman’s blood test results and her and her family’s medical histories – the scientists’ algorithm revealed that nine of the parameters were sufficient to accurately identify the women who were at a high risk of developing gestational diabetes. The nine parameters included the woman’s age, body mass index, family history of diabetes and results of her glucose tests during previous pregnancies (if any).

Next, to make sure that the nine parameters could indeed accurately predict the risk of gestational diabetes, the researchers applied them to Clalit’s health records on about 140,000 additional pregnancies that had not been part of the initial analysis. The results validated the study’s findings: The nine parameters helped to accurately identify the women who ultimately developed gestational diabetes.

These findings suggest that by having a woman answer just nine questions, it should be possible to tell in advance whether she is at a high risk of developing gestational diabetes. And if this information is available early on – in the early stages of pregnancy or even before the woman has gotten pregnant – it might be possible to reduce her risk of diabetes through lifestyle measures such as exercise and diet. On the other hand, women identified by the questionnaire as being at a low risk of gestational diabetes may be spared the cost and inconvenience of the glucose testing. 

In more general terms, this study has demonstrated the usefulness of large human-based datasets, specifically electronic health records, for deriving personalized disease predictions that can lead to preventive and therapeutic measures.