Women with EDS also had AORs of 1.8 for antepartum hemorrhage (2.8% versus 1.6%; 95% CI, 1.2-2.7; P less than .01). Cesarean delivery was more likely in women with EDS, with an AOR of 1.6 (37.4% versus 26.9%; 95% CI, 2.0-5.1); conversely, instrumental vaginal delivery was less likely in women with EDS (AOR = 0.5; 95% CI, 0.4-0.7; P less than .01 for both), meaning that spontaneous vaginal delivery was less likely in the EDS cohort.
The higher frequency of Cesarean deliveries may be attributable to anticipatory management by physicians seeking to avoid such complications as antepartum hemorrhage, as well as to the increased rate of placenta previa seen among the EDS cohort, Dr. Nicholls-Dempsey said.
After statistical adjustment, women in the EDS cohort were more than three times as likely to have hospital stays of both more than 7 days and 14 days (5.7% versus 2.1%, AOR = 3.1 for 7 days; 2.3% versus 0.7%, AOR = 3.8 for 14 days; P less than .01 for both).
Rates of some other maternal complications, such as pre-eclampsia, eclampsia, and gestational hypertension, were not elevated in the EDS cohort. Rates of premature rupture of membranes, chorioamnionitis, uterine rupture, postpartum hemorrhage, perineal laceration, and venous thromboembolism were also similar between groups.
However, not only was the AOR for preterm birth 1.5 for infants of women with EDS, but IUGR was more common in these neonates as well (AOR = 1.7, P less than .01 for both). The latter finding was unexpected, and Dr. Nicholls-Dempsey and her colleagues currently don’t have a mechanistic explanation for the higher IUGR rate.
Dr. Nicholls-Dempsey explained that she and her colleagues used data from the United States’ Health Care Cost and Utilization Project’s Nationwide Inpatient Sample (HCUP-NIS) to compare outcomes of women with EDS with the national sample as a whole.
Between 1999 and 2013, 13,881,592 births occurred in the HCUP-NIS cohort, with 910 deliveries to women who had EDS. These women were identified by ICD-9 codes, she said.
Comparing women with EDS to the non-EDS cohort, women with EDS were more likely to be Caucasian, have a higher income, and to be smokers; the cohorts were otherwise similar.
Ehlers-Danlos syndrome is a heterogeneous disorder involving abnormalities of collagen synthesis, with 13 known subtypes not captured in the HCUP-NIS data, Dr. Nicholls-Dempsey acknowledged. She characterized this as both a limitation but also a potential strength of the study.
“I really like this study because ... we know there’s 13 types of EDS that are genetically different ... They have their overlapping symptoms, but each one is different,” she said. “In an ideal world, we would have each subtype, and we would run this type of analysis on each subtype, to really be able to say to a patient, ‘You have this mutation, and this complication is going to be a big problem for you.’” The numbers of each subtype are so small that this is infeasible, she noted.
Still, the national sample acquired over many years offers real-world outcomes that clinicians can use in shared decision-making with EDS patients who are contemplating pregnancy or are already pregnant. Also, knowing which complications are more likely in patients with EDS can help plan optimal management of labor and delivery, Dr. Nicholls-Dempsey said.
Over the study’s 14-year span, the overall arc of EDS pregnancy outcomes is well captured regardless of mutation type. “It’s very applicable to the general population” of individuals with EDS, she noted. “Because it’s not type-specific, it’s really a good overview of what you can expect in EDS patients, regardless of the type.”
Dr. Nicholls-Dempsey reported no conflicts of interest and no outside sources of funding.
SOURCE: Nicholls-Dempsey L et al. Am J Obstet Gynecol. 2019 Jan;220(1):S381-382. Abstract 574