Based on those cases, researchers found that regardless of why a C-section was performed - whether because of pregnancy-related complications or the mother's medical problems, for example - vaginal delivery tended to be safer.
"My suspicion is that the labor process, the contractions, that natural squeezing probably does something to clear out the lungs so that when babies are born they have a better breathing status," said Dr. Erika Werner, who led the new study at Johns Hopkins School of Medicine in Baltimore.
"If a vaginal delivery is safe, it's something that should be attempted," she said.
The rate of C-sections in the United States has been rising for all pregnancies, leading some to worry about possible complications for mothers and babies. According to the U.S. Centers for Disease Control and Prevention, C-sections accounted for almost 1 in 3 births in 2010 - up from 1 in 5 in 1996.
There are certain circumstances - such as when a woman is bleeding or a baby has a drop in heart rate - when a C-section is required, Werner said.
However, she said, there hasn't been much data on whether C-section or vaginal delivery is the generally safer method for other premature babies, who are more fragile than full-term babies to begin with.
Her team's data come from birth certificates and hospital discharge records for 20,231 babies born in New York between 1995 and 2003. All were born very early, between 24 and 34 weeks. Normal gestation is considered 37 to 42 weeks.
Just over two-thirds of the babies were born vaginally.
After taking into account a mother's age, race and underlying medical
conditions, Werner and her colleagues found that babies delivered via
C-section were more likely to be born in respiratory distress. Just over
39 percent of them had breathing problems, compared to 26 percent of
infants delivered vaginally.More infants born via C-section also scored poorly on a test that measures overall health - including heart rate, reflexes and skin coloring - 5 minutes after birth, the researchers reported Wednesday in Obstetrics & Gynecology.
Between seven and eight preemie deliveries would have to be switched from C-sections to vaginal births to avoid one case of newborn respiratory distress, they calculated.
"In general, people have thought that vaginal delivery would be preferred if there was no other indication, but it's been a question that some people have felt is still open for research," Dr. Henry Lee, from Stanford University School of Medicine's Division of Neonatal & Developmental Medicine, said.
It's still unclear whether there are certain types of babies, or certain situations, in which C-sections might be helpful, added Lee, who wasn't involved in the new research.
Werner said that because the researchers looked back at old records, they were not able to control for all differences between women who had C-sections and vaginal deliveries - a "huge limitation" of the study.
"I was struck by the fact that there does seem to be an increased risk of some things with C-section," she said. However, "I don't know if you can make a definitive statement that it is higher risk to have a C-section based on this study."
But Lee agreed that the evidence points toward vaginal deliveries in many scenarios.
"When there is no clear maternal medical reason or indication for the baby's health (to do a C-section), then the vaginal route would be the preferred route of delivery," he said.
