The present study involved 445 women diagnosed with beta-thalassemia major, who delivered over a four-year period in USA. The control groups was composed of 8,900 participants. We chose a 1:20 case to control ratio to sufficiently power our study and allow us to better adjust for confounding variables. In USA, beta-thalassemia major remains a rare disease, with 341 expected affected newborn per year in North, Central and South America combined [14]. During our study, the yearly prevalence increased significantly from less than 5 per 100,000 births in 2011 to 20 per 100,000 births in 2014. The ICD-9 code for beta-thalassemia major was introduced in the year 2011, the start of our study period, and has been used since in American databases to successfully identify beta-thalassemia major cases [15]. The increasing prevalence over the study period, at a glance, is most likely the result of the utilization of the code following its introduction as seen with a steep slope in Fig. 1. However, this theory was not supported when we re-calculated the p-value (< 0.0001) after excluding the first year (2011). Another plausible explanation is the addition of carriers through migration from endemic regions and more recent movements of refugees into USA [1, 16]. Our study is the first of its magnitude in a region where beta-thalassemia major is not endemic. It demonstrates the association of beta-thalassemia major diagnosis and risks of adverse delivery and neonatal outcomes.
First, our cohort of women diagnosed with beta-thalassemia major had higher rates of pregestational thyroid disease. Beta-thalassemia major patients are susceptible to iron overload due to the combination of chronic hemolytic anemia and repeated transfusions [1]. Endocrine complications such as hypothyroidism are explained by the endocrine tissue’ sensitivity to iron toxicity and have been also well reported in recent reviews [3, 5, 17, 18].
This study also, demonstrated that patients in our cohort had higher rates of prior C-section and an increased risk of undergoing C-section during the observed pregnancy when compared to the control group. This finding was held truth even after adjusting for previous C-section as a confounding variable. These findings can be partially attributed to the short stature and the bone deformity observed in beta-thalassemia patients, causing concerns of cephalopelvic disproportion [3, 17]. Short stature in patients with beta-thalassemia major can be explained by the disturbances in the parathyroid function and calcium metabolism as well as hypogonadism secondary to excessive iron deposition [3]. Moreover, chronic anemia leading to extramedullary hematopoiesis and bone marrow expansion can cause remodeling to the pelvic bones, therefore increase the risk of cephalopelvic disproportion [1]. The adjusted risk of delivery by C-section was 30% more likely for beta-thalassemia major patients compared to the control group, with a prevalence of 39.6%. Our findings are concordant with the literature. In a recent narrative review, the increased risk of delivery by C-section for beta-thalassemia patients is described qualitatively[17]. In a study in France including 37 women with beta-thalassemia major and intermedia, C-section was performed in 53.6% of pregnancies, which was significantly more often than in non-carriers (p = 0.002). [13]. In all smaller case series studying patients with beta thalassemia major and intermedia, the prevalence ranges from 24% to all deliveries being done by C-section [7, 8, 19–22].
Our cohort did not find an association between beta-thalassemia major and increased risks of adverse pregnancy outcomes, including gestational hypertension, preeclampsia and placenta previa among others. A recent review reports similar risks of preeclampsia compared to the background population [17]. In our cohort, rates of gestational diabetes tended to be higher among the beta-thalassemia major patients (9.7% VS 8.6%), though the difference was not statistically significant (p > 0.05). The pooled prevalence of gestational diabetes is estimated at 14% globally and 7.1% in North America and Caribbean [23]. Certain reviews describe a clear association related to iron toxicity, with a prevalence as high as 15% [18, 24]. An observational study in China reported a prevalence of gestational diabetes as high as 17.9% [6] Two case series on six and four beta-thalassemia major pregnancies respectively reported no incidence of gestational diabetes [5, 25]. The literature does not provide a clear consensus regarding the association between beta-thalassemia and gestational diabetes, and this might be explored further in future studies.
During normal pregnancy, there is a state of accelerated erythropoiesis, with expansion of the total red blood cell volume with concurrent expansion of plasma volume, leading to dilutional anemia [2]. In the presence of ineffective erythropoiesis, beta-thalassemia major mothers require more transfusions during pregnancy [2, 18]. In our cohort, among beta-thalassemia major pregnancies, 5.9% required transfusions during delivery compared to 1.3% in the control group (p < 0.0001). In a review of reproductive health issues in beta-thalassemia major patients, a qualitative report of increased transfusion regimen during pregnancy is described [18]. In a case series, all beta-thalassemia major women achieving pregnancy required transfusions, with increasing red blood cells consumption during pregnancy [7]. In most studies, all beta-thalassemia major cases are transfused during the pregnancy course. However, intrapartum blood transfusion were not studied before. Our study, although demonstrated a more than threefold increased risk of transfusion (aOR 4.69 95%CI 3.02–7.28), this rate was less than previously reported as not all participants were transfused. The difference in the rate of blood transfusion among patients with beta-thalassemia major in our study compared to the literature is most likely because we address only transfusions received during the admission for delivery. Despite knowing that these patients were at risk of anemia, they required blood transfusion intrapartum, suggesting their hemoglobin level was not optimized at the time of delivery. Therefore, this study highlights the recommendation to optimize hemoglobin levels before entering labour.
Women with beta-thalassemia were 70% more likely to give birth to a small for gestational age neonate. The presence of anemia of any etiology during the first trimester is a risk factor for neonates being born small for gestational age [4]. A confounding factor that has been described is that due to subfertility, more women benefit from assisted reproductive technology, and have higher rates of multiple gestations and preterm births. Sayani and colleagues found that among women with transfusion dependent thalassemia, the rates of growth restricted fetuses is similar to that of the general population when corrected for multiple gestations [17]. A study from Thailand included 597 pregnant women with beta-thalassemia traits, the prevalence of low birth weight (< 2500g) was higher, but not the rates of small for gestational age compared to the control group [10]. A study based in China including 228 beta-thalassemia carriers, with 84.48% of mothers suffering from anemia, 5.95% of infants had low birth weight [6]. As mentioned in the previous paragraph, we recorded an increased rate of intrapartum blood transfusion among women with beta-thalassemia. Therefore, we can hypothesize that some of the patients in our studies had poorly controlled anemia, which in return increased their risk of having SGA newborns.
There are several limitations to our study, including its retrospective nature, lacking information on hemoglobin and transferrin levels as well as frequency of blood transfusion throughout the pregnancy. Moreover, the code for beta-thalassemia was introduced at the beginning of our study and this might have affected the real number of cases and masked other effects of beta-thalassemia major on outcomes. However, our findings would have a higher specificity and stronger correlation if the aforementioned is true. Furthermore, given that the database does not list different medications taken, we cannot control for the effect of iron chelating therapy on outcomes. On the other hand, this study has several strength advantages such as assessing large number of pregnancies in a non-endemic area for beta-thalassemia major. In addition, excluding other hemoglobinopathies prior to selecting the cases has led to mitigate any chance that these findings could be as a result of a concurrent hemoglobinopathy other than beta-thalassemia major.