Pregnancy and Childbirth in Women with Thalassemia: Past and Present

Several studies have demonstrated that pregnancy in women with beta transfusion-dependent thalassemia (TDT) is possible with positive outcome for both the mother and the fetus if appropriately planned and followed by a multidisciplinary team. The aim of this multicenter study (AIFA Study id. 1817) is to highlight whether and how improvements in the management of the disease have influenced pregnancy in various aspects.

The Italian Society for Thalassemia and Hemoglobinopathies (SITE) collected data on 341 pregnancies from 247 women with TDT in Italy. Seventy-four women had two pregnancies, 18 women had three pregnancies, and 2 women had four pregnancies. The age at first pregnancy has significantly increased over time (Person’s correlation coefficient Rp = 0.26, p < 0.001).
Ovulation was induced by gonadotropins in 131 out of 311 cases (42.1%) and occurred spontaneously in 180 (57.9%). Conception was spontaneous in 225 out of 307 cases (73.3%), through intrauterine insemination in 30 (9.8%), and via in vitro fertilization in 45 (14.7%). The rate of spontaneous conception following spontaneous ovulation significantly increased over the years (Rp = 0.54, p = 0.011), and the risk of non-spontaneous conception correlated with higher age at conception (OR = 1.19, 95% CI = 1.12–1.27, p < 0.001). Preimplantation diagnosis was performed in 12 out of 307 cases (3.9%) for partners carrying beta-thalassemia, and heterologous gametes were used in 4.3% (13 out of 303 cases).

The rate of spontaneous abortions, excluding ongoing pregnancies (9), was 9.0% (25 out of 278), and 8 patients (2.9%) opted for voluntary pregnancy termination, one due to maternal toxoplasmosis.

The percentage of complication during pregnancy was not significantly increased compared to the non-thalassemic population (p>0.05).

Cardiac T2* measurements were available within 6 months from the start of pregnancy in 104 cases, indicating moderate iron overload in 8 and severe in 4. These values increased significantly over the years, and after 2015, all pregnancies began with a myocardial T2* >20 ms. Both pre-pregnancy and post-pregnancy T2* values increased significantly over time (Rp = 0.27, p = 0.007, and Rp = 0.33, p < 0.001, respectively). However, of the 72 women with pre-pregnancy heart T2* ≥ 20 ms for whom postpartum values were available, 11 showed moderate accumulation, and one showed severe accumulation. Of these 12 patients, 4 had pre-pregnancy liver iron concentration (LIC) <2 mg/g d.w., 6 had LIC between 2 and 7 mg/g d.w., and 1 had LIC >7 mg/g d.w. (one patient’s data not reported). Overall, pre-pregnancy LIC did not change over the years (Rp = -0.12, p = 0.27), and 19 women with known pre-pregnancy values had moderate (11 out of 88, 12.5%) or severe (8 out of 88, 9.1%) hepatic iron accumulation.
Two patients experienced heart failure immediately postpartum, one in 1990 and the other recently after a twin cesarean section following years of non-adherence to iron chelation therapy. There were two stillbirths, and, of the 245 pregnancies resulting in one or more live births, 26 were twins, and 3 were triplets (in one case, an embryo was reabsorbed in the first trimester).

Only 17 deliveries (7.2%) were vaginal, 181 (77.0%) were planned cesarean sections, mainly for precautionary reasons, and 37 (15.7%) were emergency cesarean sections. The number of cesarean section did not diminish overtime. Ninety-seven babies (42%) were born preterm, 132 (57.1%) at term, and in 16 cases the gestational age was unknown.

Preterm births decreased over time (Rp = -0.59, p = 0.0048), as did twin births (Rp = -0.65, p = 0.0015), and birth weight increased (Rp = 0.27, p < 0.001). Thirty-six neonates had complications at birth, and 47 were admitted to the neonatal intensive care unit. Three twins, two born preterm and one at term, died in the neonatal period. Eight children had homozygous beta-thalassemia. The rate of malformations (4 out of 237, 1.7%, 95% CI: 0–5%) was not higher than that of the general population, and no correlation was found with the duration of exposure to iron chelators during pregnancy. Two patients received two doses of luspatercept during pregnancy without complications. Exclusive or mixed breastfeeding was chosen for 66 children, with an increasing trend over time (Rp = 0.40, p = 0.07).

This analysis confirms that many aspects of pregnancy in women with TDT, as well as neonatal characteristics, have changed over time along with the underlying disease.