Associations between hypothyroidism and adverse obstetric and neonatal outcomes: a study of a population database including over 184,000 women with hypothyroidism

Abstract Background Maternal hypothyroidism has been associated with multiple adverse pregnancy outcomes. These findings have not been confirmed in a large population database study. Therefore, a large population-based cohort study was established to study the associations between maternal hypothyroidism and pregnancy and perinatal complications. Methods This is a retrospective population-based cohort study utilizing data from the Healthcare Cost and Utilization Project-Nationwide Inpatient Sample (HCUP-NIS) over 11 years from 2004 to 2014. A cohort of all deliveries between 2004 and 2014 inclusive, was created. Within this group, all deliveries to women with hypothyroidism were identified as part of the study group (n = 184,869), and the remaining deliveries were categorized as non-hypothyroidism births and comprised the reference group (n = 8,911,919). The main outcome measures were pregnancy and perinatal complications. Results Maternal hypothyroidism is associated with several pregnancy and perinatal complications, including gestational diabetes mellitus (aOR 1.43, 95%CI 1.38–1.47), gestational hypertension (aOR 1.17, 95%CI 1.11–1.22) and preeclampsia (aOR 1.21, 95%CI 1.16–1.27) (all p < 0.001). These patients are more likely to experience preterm premature rupture of membranes (aOR 1.19, 95%CI 1.09–1.29, p < 0.001), preterm delivery (aOR 1.12 95%CI 1.08–1.17, p < 0.001), are more likely to deliver by cesarean section (aOR 1.21, 95% CI 1.18–1.24, p < 0.001), and suffer from postpartum hemorrhage (aOR 1.07, 95%CI 1.01–1.13, p = 0.012), disseminated intravascular coagulation (aOR 1.20, 95%CI 1.00-1.43, p = 0.046), and undergo hysterectomy (aOR 1.42, 95% CI 1.13–1.80, p = 0.003). As for neonatal outcomes, small for gestational age and congenital anomalies are more likely to occur in the offspring of women with hypothyroidism (aOR 1.20, 95% CI 1.14–1.27 and aOR 1.34, 95% CI 1.22–1.48, both p < 0.001). Conclusions Women with hypothyroidism are more likely to experience pregnancy, delivery and neonatal complications. We found an association between hypothyroidism and hypertensive disorders, postpartum hemorrhage, transfusions, infections, preterm delivery and hysterectomy, among other problems. This data from a population sized database confirms the findings of smaller previous studies in the literature.


Introduction
Thyroid disorder is common among women of reproductive age and is the most common endocrine disease during pregnancy [1].Overt, or clinical hypothyroidism is diagnosed by an elevated thyroidstimulating hormone (TSH) concentration and a low serum free thyroxine (FT4) concentration, known to affect up to 1-2% of the women [2] and 0.2-0.5% of pregnant women [3].Subclinical hypothyroidism is defined as elevated TSH whereas the FT4 remains in the normal range and is affecting 2-13.7% of pregnant women [4].
Maternal hypothyroidism, either clinical or subclinical, has been associated with multiple adverse pregnancy outcomes.In a systematic review of clinical impact of thyroid disorders before and during pregnancy, 43 studies were included, in which the number of participants varied between dozens to a couple of hundred [5].This review and several cited subsequent studies, questioned the association with spontaneous miscarriages [5-7], gestational diabetes mellitus [6,[8][9][10][11][12], gestational hypertension or preeclampsia [6,8,9,13,14], placental abruption [15], induction of labor [8], preterm delivery [6,9,15], cesarean section [(7-9,12], low birth weight [11], neonatal intensive care unit admission [9] and fetal death [10].However, none of the studies exceeded 10,000 participants with hypothyroidism. In addition, there are some population-based studies in the literature that investigated the association between hypothyroidism and adverse pregnancy outcomes.The largest one was conducted by Turunen et al. and included 16,364 hypothyroid mothers.However, postpartum complications were not evaluated in this study [16].In another study by Chen et al, 184,611 pregnancies were included, with 7140 women with hypothyroidism (TSH > 4. 29).In this study, they found an association to spontaneous abortions and stillbirth, preterm delivery, cesarean section and large for gestational age infants.However, other adverse events, such as pre-eclampsia, gestational diabetes or preterm premature rupture of membranes were not evaluated in this study [17].
The objective of this study, therefore, was to assess the prevalence of a variety of adverse pregnancy and delivery-associated maternal and neonatal outcomes in women with hypothyroidism, in a large database, while accounting for confounding effects.

Materials and methods
We conducted a retrospective population-based study utilizing data from the HCUP-NIS over 11 years from 2004 to 2014, inclusively.The HCUP-NIS is the largest inpatient sample database in the United States of America (USA) and is comprised of hospital inpatient stays submitted by hospitals in 48 states and the District of Columbia.Each year, the database provides information relating to 7 million inpatient stays, including patient characteristics, diagnoses, and procedures.The data are representative of �20% of admissions to American hospitals and geographically represent over 96% of the population in the USA.
We evaluated deliveries by using the international classification of diseases, ninth edition, clinical modification (ICD-9-CM) codes for delivery-related discharge diagnoses (650.xx,677.xx, 651.xx-676.xxwhere the fifth digit is 0, 1 or 2), and birth-related procedural diagnosis (72.x, 73.x, 74.0-74.2).We limited our study group to the admissions that ended with delivery or maternal death in order to guarantee that multiple admissions in the same pregnancy will be excluded.Within this group, all women with diagnosis of hypothyroidism were identified using ICD-9 diagnostic codes 244 (244.0,244.1, 244.2, 244.3, 244.8, 244.9), 245.2, 246.1, and the remaining deliveries were categorized as nonhypothyroidism births and comprised the reference group.Women who had a diagnosis of congenital hypothyroidism were excluded from the study.
Hypothyroidism is usually defined as two abnormal TSH (higher than the upper limit of the assay cut offs) with or without abnormal free t3 and free t4.Presently there is controversy over the appropriate upper limit of normal for serum TSH.Some studies have suggested that the upper limit is 2.5 mU/L, whereas others suggested an upper limit of 4.1-4.5 mU/L [18].In our database, the upper limit was the one that was used in each respective lab.
Baseline clinical characteristics included: patient age, race, income, insurance type, hospital type, previous cesarean section, multiple gestation, smoking history, obesity (body mass index � 30 kg/m2), preexisting hypertension and preexisting diabetes mellitus.Pregnancy outcomes include hypertensive disorders of pregnancy (gestational hypertension and preeclampsia/eclampsia), gestational diabetes mellitus, placenta previa and venous thromboembolism (deep vein thrombosis and pulmonary embolism).Delivery outcomes include preterm premature rupture of the membranes, preterm delivery, placental abruption, cesarean section, maternal infection and chorioamnionitis, and hysterectomy.Postpartum hemorrhage, disseminated intravascular coagulation and blood transfusions were also investigated.Neonatal outcomes included small for gestational age infants, intrauterine fetal death, and congenital anomalies.
Statistical analysis: An initial analysis was performed to identify the prevalence of pregnant women with hypothyroidism over the entire duration of the study.We then compared baseline clinical and demographic characteristics between women with hypothyroidism to those without hypothyroidism.Subsequently, logistic regression analyses were conducted to explore associations between hypothyroidism and obstetrical and neonatal outcomes through the estimation of odds ratio (OR) and 95% confidence intervals (CI).The regression models were adjusted for the potential confounding effects of maternal demographic, pre-existing clinical characteristics, and concurrently occurring characteristics to generate adjusted CIs.A confounder was determined if it was part of the demographics or pre-pregnancy diseases and rates between those with hypothyroidism and the controls gave a p < 0.05.All analyses were performed using SPSS 23.0 (IBM Corporation, Chicago, USA) software for all analyses.
This study used exclusively publicly accessible, anonymized data; therefore, according to Tri-Council Policy Statement (2018), institutional review board approval was not required.

Results
There were 9,096,788 births between 2004 and 2014, inclusively.Of these pregnancies, 184,869 women, or 2.032 per 100 women, had a documented diagnosis of hypothyroidism.Within the 11-year study period, there was a steady rise in the incidence of hypothyroidism, from 1.336 per 100 in 2004 to 3.150 per 100 in 2014 (p < 0.0001).(Figure 1).Baseline demographic characteristics for our study population are summarized in Table 1.Women with hypothyroidism were more likely to be older than 25 years of age and Caucasian, have higher household incomes and private insurance and deliver in an urban teaching hospital, as compared with the nonhypothyroidism obstetrical subjects.Pregnant women with hypothyroidism were also less likely to smoke or use recreational drugs in pregnancy.Women with hypothyroidism were more likely to have chronic hypertension and pre-gestational diabetes, though they were less likely to be obese than the non-hypothyroidism obstetrical population.Women with hypothyroidism were also more likely to have undergone in vitro fertilization treatments and have multi-gestation pregnancies.The crude and adjusted effect measures for the association between pregnancy and delivery outcomes and their respective prevalence in each cohort are outlined in Table 2. Women with hypothyroidism were more likely to suffer from gestational diabetes mellitus, gestational hypertension, and preeclampsia (all p < 0.001), after controlling for confounding effects of age, race, income level, medical insurance type, obesity, in vitro fertilization use, previous cesarean section, chronic hypertension, pre-gestational diabetes, multiple gestation, tobacco smoking, and illicit drug use.Regarding delivery outcomes, they were more likely to experience preterm premature rupture of membranes (p < 0.001), preterm delivery (p < 0.001), and were more likely to deliver by cesarean section (p < 0.001), after controlling for confounding effects of age, race, income level, medical insurance type, obesity, in vitro fertilization use, previous cesarean section, chronic hypertension, pre-gestational diabetes, multiple gestation, tobacco smoking and illicit drug use, pregnancy-associated hypertension, gestational hypertension, preeclampsia and gestational diabetes mellitus.
After controlling for all confounding effects, women with hypothyroidism were more likely to experience chorioamnionitis in labor (p ¼ 0.019), to develop maternal infections in the postpartum period (p ¼ 0.017), postpartum hemorrhage (p ¼ 0.012), disseminated intravascular coagulation (p ¼ 0.047), require blood transfusions (p ¼ 0.009), and hysterectomy (p ¼ 0.012) compared to the control group.However, no association was found between hypothyroidism and placental abruption or venous thromboembolism.
As for neonatal outcomes (Table 3), small for gestational age infants and congenital anomalies were more likely to occur in the offspring of women with hypothyroidism (both p < 0.001).

Discussion
The present study involving 184,869 women who had a diagnosis of hypothyroidism, demonstrated that hypothyroidism was associated with a small but significant increased risk of overall adverse pregnancy and delivery outcomes.Multiple smaller studies in the literature have been published in the past, which have assessed the possible increased risk of pregnancy and delivery outcomes in women with hypothyroidism [5,6,16,17].Similar to these previous publications, we found that pregnancy complications such as preeclampsia and gestational diabetes mellitus are associated with hypothyroidism.Interestingly, a study from Finland on more than 5000 pregnancies reported that overt hypothyroidism predicts the risk of the mother developing diabetes later in life, and not just during pregnancy [hazard ratio 6.0 (95% CI) (2.2-16.4)][19].
Regarding delivery outcomes, we found a significant association between maternal hypothyroidism and preterm premature rupture of membranes, preterm delivery, chorioamnionitis and cesarean section, excluding placental abruption.A possible explanation could be provided by understanding the physiologic role of thyroid hormone in early placental and embryo development [20][21][22].Thyroid hormones are important for trophoblast migration [23] proliferation and invasion [22], and abnormal placentation may represent a direct consequence of inadequate thyroid hormone availability [24].Abnormal placentation is a wellknown contributor to placental-mediated pregnancy complications like those mentioned above [25] and may account for the higher rates of placenta previa that we found in hypothyroid mothers (p ¼ 0.009).These factors may further contribute to the understanding of our findings that hypothyroidism in pregnancy is associated with the occurrence of postpartum hemorrhage, disseminated intravascular coagulation and blood transfusions.In a recent study from Pakistan of 702 women, postpartum hemorrhage was found to be associated with hypothyroidism in the preconception and third trimester, but without discrimination between clinical or subclinical hypothyroidism [22].As for subclinical hypothyroidism per se, previous studies failed to show similar results while investigating an association between subclinical hypothyroidism and postpartum hemorrhage: Wang et al. found that the incidence of spontaneous abortions in pregnant women with subclinical hypothyroidism was increased, but no other obstetrical complications including postpartum hemorrhage were augmented (n ¼ 756) [26].In another small study, no difference was found in terms of coagulation parameters and postpartum hemorrhage between euthyroid or subclinical hypothyroid patients with thyroid antibodies (n ¼ 57) and healthy controls (n ¼ 52) [27].Our findings of increased hematologic complications in hypothyroid mothers may be related to the high rates of preterm premature rupture of membranes, chorioamnionitis and cesarean section that were observed in these subjects.In consequence, it could have a role in the high rates that were observed of hysterectomies in the hypothyroid group.
Finally, we also demonstrated a significantly increased risk of small for gestational age infants in hypothyroid mothers.In the literature, there is a controversy regarding embryo growth and infant birthweight in hypothyroid mothers [5].Some studies found an increased risk of small for gestational age or low birthweight infants for mothers who suffered from clinical, subclinical, or isolated hypothyroxinemia [6, [28][29][30][31], whereas other studies showed the opposite [16,17].Our large study, therefore, farther contributes to the positive relationship between hypothyroidism and small for gestational age infants, which has clinical implication in pregnancy follow-up.Another implication for clinical practice is the increased risk of congenital anomalies which we found in our study (aOR 1.34, 95% CI 1.22-1.48,p < 0.001).This finding correlates with previous studies, but not all.In a Swedish study which included 848,468 women, 9,866 of whom used thyroid hormone, a significantly increased risk of congenital anomalies was found (OR ¼1.14, 95% CI 1.05-1.26),however for less extent [8].Another association between maternal hypothyroidism and major congenital anomalies was published in the Finland study (OR 1.14 [CI 1.06-1.22])[16].On the contrary, other studies did not show such association [5].Therefore, farther studies are needed to evaluate and specify this risk.
There are several limitations to our study, the most important being that it was a retrospective analysis with TSH levels and information on presence of thyroid peroxidase antibodies not available.We cannot verify that the diagnosis of hypothyroidism was correctly made in all participants, as such the diagnosis of hypothyroidism may have been inflated although this would only serve to diminish any difference we detected and as such the difference found were likely real.In addition, we cannot distinguish between clinical and subclinical hypothyroidism.This could also result in findings being slightly blunted.However, this would also suggest that any significant findings were likely real.A possible bias can result from including ICD-9 diagnostic code 244.8.This code was given for acquired hypothyroidism.However, it includes secondary hypothyroidism as well, which we could not exclude from our analysis.In addition, odds ratio values were of a small magnitude.These finding highlights that most contribution to pathologies in pregnancy by hypothyroidism are likely mildly elevated in risk.Given that we do not have access to serum TSH levels in pregnancy we cannot be sure that subjects were compliant with their medications.However, this database as such likely reflects the true use of thyroid replacement in the general population and the resultant risks in pregnancy.Given that the database does not list different medications taken, we cannot control for the effect of different thyroid replacement types on outcomes.The fact that most findings in our study confirmed results from publications with smaller samples validates the outcomes of our study.It would have been nice to be able to report the type of neonatal anomalies in the study (i.e.major, minor, genetic) however, this data was unavailable.
One strength of our study is that it is the first truly large study to evaluate the risk of hypothyroidism in pregnancy, which was clearly lacking form the medical literature.Being the first true large population study, which included 9,096,788 births analyzed, and more than 184,000 women with hypothyroidism provides an evaluation of the independent risk of the disease on a variety of adverse pregnancy, delivery, and neonatal outcomes, while being powered to effectively control for confounding effects.

Conclusions
Our large database study, along with previous studies, demonstrated the association between maternal hypothyroidism and obstetrical complications such as preeclampsia and gestational diabetes mellitus, preterm premature rupture of membranes, preterm delivery, chorioamnionitis and cesarean section.In addition, we found that in the postpartum period there is a significant association between maternal hypothyroidism and postpartum hemorrhage, disseminated intravascular coagulation and blood transfusion, which were not previously established.Further studies should be directed to determine if compliance with thyroid replacement minimizes or even eliminates these risks.

Figure 1 .
Figure 1.Prevalence of Hypothyroidism disease among women who gave birth between 2004 and 2014.

Table 2 .
Pregnancy and delivery outcomes.Pregnancy outcome: adjust for age, race, medical insurance plan type, obesity, chronic hypertension, illicit drug use, previous cesarean section, tobacco smoking during pregnancy, human immunodeficiency virus, multiple gestation, pregestational diabetes mellitus, in vitro fertilization, hospital type and income.
a b Delivery outcomes: adjust for age, race, medical insurance plan type, obesity, chronic hypertension, illicit drug use, previous cesarean section, tobacco smoking during pregnancy, human immunodeficiency virus, multiple gestation, pregestational diabetes mellitus, in vitro fertilization, hospital type, income, pregnancy induced hypertension, gestational hypertension, preeclampsia, gestational diabetes mellitus and placenta previa.

Table 3 .
Neonatal outcomes a .Fetal outcomes: adjust for age, race, medical insurance plan type, obesity, chronic hypertension, illicit drug use, previous cesarean section, tobacco smoking during pregnancy, human immunodeficiency virus, multiple gestation, pregestational diabetes mellitus, in vitro fertilization, hospital type, income, pregnancy induce hypertension, gestational hypertension, preeclampsia, gestational diabetes mellitus and placenta previa. a