Surprisingly, the IOM consensus panel did not issue recommendations for women who are underweight, citing a lack of evidence. I believe there is clear epidemiologic evidence, however, that women with twin pregnancies who are underweight before pregnancy and have poor gestational weight gain are at the highest risk for poor outcomes.
In a 2003 report, Dr. Luke and a team of investigators, myself included, developed BMI-specific weight gain guidelines for optimal birth weights in twin pregnancies. Our weight gain curves and recommendations, which were based on a multi-institution cohort study of 2,324 twin pregnancies, are similar to the IOM’s guidelines. We also addressed the category of underweight women, however, and advised a gain of 50-62 pounds for these women. (See box.)
Women with a pregravid underweight status who had a total weight gain within this range experienced optimal fetal growth and birth weight. (J. Reprod. Med. 2003;48:217-24).
In a separate follow-up study of the value of using these BMI-specific weight gain goals, Dr. William Goodnight and I found that in our twins clinic, women who failed to achieve their BMI-specific weight-gain goals had a lower birth weight by nearly 200 g per twin and a length of gestation that was 1 week shorter than that of women who met weight gain goals (Am. J. Obstet. Gynecol. 2006;195[suppl]:S121).
As we described in a review published in 2009, achieving twin weight-gain goals should be part of a comprehensive approach to nutrition that also includes an appropriate caloric intake (3,000-4,000 kcal/day in underweight to normal weight women); supplementation with calcium, magnesium, folate, and zinc (beyond a usual prenatal vitamin); and a nutrient-dense diet that is high in iron-rich proteins and omega-3 fatty acids (Obstet. Gynecol. 2009;114:1121-34). In our practice, we emphasize the value of meat protein and of low-mercury fish.
Dr. Luke, who has coauthored a book with Tamara Eberlein that we often recommend to our patients, titled "When Youre Expecting Twins, Triplets or Quads" (New York: HarperCollins, 2011), has best demonstrated the extent to which intensive nutritional counseling and follow-up pays off in twin pregnancies.
In a cohort study published in 2003, she enrolled 190 women with twin pregnancies in a specialized prenatal program that involved twice-monthly visits for nutritional counseling and monitoring. Women were prescribed a diet of 3,000-4,000 kcal/day, with 20% of calories from protein, 40% from carbohydrates, and 40% from fat, as well as multimineral supplementation. The women were monitored for adequate weight gain and had serial ultrasound assessments.
Compared with 339 women with twin pregnancies who were followed by their physicians at the University of Michigan but not enrolled in the program, the program participants had significantly longer gestations (about a week), higher birth weights (220 g), a 23% reduction in preterm births, and significant reductions in preterm premature rupture of membranes, preterm labor, preeclampsia, ICU admission, and other poor outcomes.
Overall, the incidence of major neonatal morbidity was 17% for the nutritional program participants, compared with 32% for those women who did not receive the specialized care (Am. J. Obstet. Gynecol. 2003;189:934-8).
Measuring Cervical Length
One of the major clinical concerns with any twin pregnancy is the prevention of preterm birth and, by extension, the identification of those women with twins who, within this broader high-risk category, are at greatest risk for preterm birth.
Since the mid-1990s, research has shown that a short cervix detected in the midtrimester (defined as 16-24 weeks) by transvaginal ultrasound is a powerful predictor of preterm birth in women with either singleton or twin gestations. Studies of twin gestations have shown that as cervical length shortens to 25 mm or less, the risk of subsequent preterm birth (defined less than 34 weeks) rises dramatically.
In a report on twin pregnancies from the Maternal-Fetal Medicine Units Network of the Eunice Kennedy Shriver National Institute of Child Health and Human Development, for instance, Dr. R.L. Goldenberg and associates demonstrated that cervical length less than or equal to 25 mm at 24 weeks was the best predictor of spontaneous preterm birth. In fact, of all 50 potential risk factors that were studied, a short cervix was the only factor that was consistently associated with preterm birth. The investigators also noted that this shorter cervical length was more common at both 24 and 28 weeks in twins, compared with singleton pregnancies (Am. J. Obstet. Gynecol. 1996;175:1047-53).
Through the 1990s and the next decade, investigators searched for a viable intervention. However, in numerous studies, the use of cerclage in mothers of twins with a short cervix was found to be of no benefit. Prophylactic tocolysis also failed to prevent preterm birth in published studies.