Abstract
Context:
Thyroid diseases are inconsistently reported to increase risk for pregnancy complications.
Objective:
The objective of this study was to study pregnancy complications associated with common and uncommon thyroid diseases.
Design, Setting, and Participants:
We analyzed singleton pregnancies (N = 223 512) from a retrospective US cohort, the Consortium on Safe Labor (2002–2008). Thyroid diseases and outcomes were derived from electronic medical records. Multivariable logistic regression with generalized estimating equations estimated adjusted odds ratios (ORs) with 99% confidence intervals (99% CI).
Main Outcome Measures:
Hypertensive diseases, diabetes, preterm birth, cesarean sections, inductions, and intensive care unit (ICU) admissions were analyzed.
Results:
Primary hypothyroidism was associated with increased odds of preeclampsia (OR = 1.47, 99% CI = 1.20–1.81), superimposed preeclampsia (OR = 2.25, 99% CI = 1.53–3.29), gestational diabetes (OR = 1.57, 99% CI = 1.33–1.86), preterm birth (OR = 1.34, 99% CI = 1.17–1.53), induction (OR = 1.15, 99% CI = 1.04–1.28), cesarean section (prelabor, OR = 1.31, 99% CI = 1.11–1.54; after spontaneous labor OR = 1.38, 99% CI = 1.14–1.66), and ICU admission (OR = 2.08, 99% CI = 1.04–4.15). Iatrogenic hypothyroidism was associated with increased odds of placental abruption (OR = 2.89, 99% CI = 1.14–7.36), breech presentation (OR = 2.09, 99% CI = 1.07–4.07), and cesarean section after spontaneous labor (OR = 2.05, 99% CI = 1.01–4.16). Hyperthyroidism was associated with increased odds of preeclampsia (OR = 1.78, 99% CI = 1.08–2.94), superimposed preeclampsia (OR = 3.64, 99% CI = 1.82–7.29), preterm birth (OR = 1.81, 99% CI = 1.32–2.49), induction (OR = 1.40, 99% CI = 1.06–1.86), and ICU admission (OR = 3.70, 99% CI = 1.16–11.80).
Conclusions:
Thyroid diseases were associated with obstetrical, labor, and delivery complications. Although we lacked information on treatment during pregnancy, these nationwide data suggest either that there is a need for better thyroid disease management during pregnancy or that there may be an intrinsic aspect of thyroid disease that causes poor pregnancy outcomes.
Thyroid diseases affect up to 4% of all pregnancies with primary hypothyroidism being the most prevalent disease (1). Thyroid hormone requirements increase in pregnancy, possibly leading to hypothyroidism among those with limited thyroidal reserve (1). Many women with levothyroxine treatment need dose increases in early pregnancy (2), with women without residual thyroid tissue requiring higher doses (3). Up to 60% of women treated with levothyroxine had elevated TSH during pregnancy (4), further evidence of the need for dose adjustment as well as the potential for periods of inadequate treatment.
Inadequately treated or subclinical hypothyroidism increases the risk for miscarriage and fetal death (5, 6), anemia (7), postpartum hemorrhage (7), placental abruption (7, 8), cardiac dysfunction (7), gestational hypertension/preeclampsia (9–11), gestational diabetes (12, 13), and preterm births (8, 14, 15), whereas adequately treated hypothyroidism only increases the risk for cesarean sections (16). A register-based study without estimates of treatment adequacy found increased risk for preeclampsia, diabetes, preterm births, cesarean sections, and labor inductions among those with levothyroxine use (17). However, not all studies have found an increased risk of adverse outcomes with hypothyroidism (18–22). No previous studies have evaluated the risks of adverse pregnancy outcomes associated with the iatrogenic causes of hypothyroidism, although women without residual thyroid tissue may be at higher risk of inadequate treatment during pregnancy.
Poor control of hyperthyroidism during pregnancy is also associated with increased risk of miscarriage and stillbirth (23), hypertension in pregnancy (24), preterm births (23), and maternal heart failure (23, 25). However, no large contemporary study has evaluated the effect of diagnosed hyperthyroidism on pregnancy outcomes in detail.
Race/ethnicity may be important in modifying the potential risks thyroid diseases pose during pregnancy because the risk of hyperthyroidism and hypothyroidism in the population and predisposition to adverse outcomes in pregnancy varies by race/ethnicity (26, 27). To date most studies examining the association of hypothyroidism or hyperthyroidism and pregnancy outcomes have had limited racial/ethnic diversity.
Our analysis addresses these gaps in the existing literature by using data from a large, racially/ethnically diverse US cohort.
Materials and Methods
The Consortium on Safe Labor (CSL) was an observational, nationwide cohort including 12 clinical centers with 19 hospitals (2002–2008; 87% of births occurred in 2005–2007). Clinical sites were chosen because of their geographic distribution and the availability of electronic medical records. The CSL was designed to provide a comprehensive description of contemporary labor and delivery in the United States (28). A comparison between electronic data and chart review demonstrated good agreement (28). A total of 228 668 deliveries with 233 844 newborns (including multiples) delivered at ≥23 weeks of gestation were included in the main CSL study (28). A total of 106 deliveries were excluded because of errors in identification and the final CSL consisted of 228 562 deliveries with 233 736 newborns. We restricted the present analysis to 223 512 singleton pregnancies among 204 180 women. Most women (185 785 [91.0%]) contributed only 1 pregnancy.
Electronic medical record data included maternal demographic characteristics, medical, reproductive, and prenatal history, labor and delivery summary, and postpartum information and were classified to predefined categories. Electronic maternal discharge summaries were linked to each delivery. The CSL was approved by the institutional review boards of all participating institutions. Data linkage, cleaning, recording, and validation were described previously (28).
Thyroid diseases
Maternal thyroid disease diagnoses were established using discharge data and medical record report of nonspecific “history of thyroid disease.” Medication or laboratory data were not available. Where available, International Classification of Diseases-9 (ICD-9) codes were used to categorize disease status (see Supplemental Table 1 published on The Endocrine Society's Journals Online web site at http://jcem.endojournals.org.): (1) no thyroid disease, n = 216 901; (2) primary hypothyroidism, n = 3183; (3) iatrogenic hypothyroidism (hypothyroidism due to thyroid surgery or ablation), n = 178; (4) hyperthyroidism, n = 417; (5) other thyroid diseases (simple or nontoxic goiter, n = 88; thyroiditis, n = 66; and other thyroid disorders including benign and malignant thyroid nodules, n = 109; total n = 263); and (6) unspecified thyroid disease (reported in the medical record but without a discharge diagnostic code), n = 2570. Once a thyroid disease diagnosis was recorded, it was deemed to affect all subsequent pregnancies.
Outcome data
Pregnancy outcomes were derived from both the electronic medical record and discharge summaries (Supplemental Table 1) and included prelabor cesarean section, induction, spontaneous labor, route of delivery, preterm birth (<37 gestational weeks), late preterm birth (delivery between 34 and <37 weeks), early preterm birth (<34 weeks), gestational diabetes, gestational hypertension, preeclampsia, superimposed preeclampsia, placental abruption, threatened preterm birth, placenta previa, hemorrhage in late pregnancy or postpartum, chorioamnionitis, premature rupture of membranes (PROM), preterm premature rupture of membranes (PPROM) (PROM <37 weeks), breech presentation, maternal intensive care unit (ICU) admission, and maternal death.
Covariate data
Covariate data included race/ethnicity, age, parity, insurance type, prepregnancy body mass index (BMI) (weight in kilograms/height in meters squared), smoking during pregnancy, prior surgical procedures causing uterine scars, and history of chronic diseases including diabetes, hypertension, asthma, and depression or chronic heart, gastrointestinal or renal disease (see Supplemental Table 1). Once a mother was diagnosed with a chronic disease, all subsequent pregnancies were assumed to be affected.
Statistical analyses
Pregnancy was the unit of analysis for all statistical testing. Significance testing was based on either linear (continuous or categorical variables) or logistic (binary variables) regression using generalized estimating equations with a first-order autoregressive structure correlation matrix, which accounts for multiple pregnancies contributed by the same woman. Results are presented as P values or odds ratios (ORs). To account for multiple testing, we present 99% confidence intervals (CIs) for the ORs. All analyses were adjusted for site and the full analyses were adjusted for site, maternal age, race/ethnicity, parity, insurance status, smoking, and other chronic diseases. Analyses regarding labor and route of delivery were also adjusted for the presence of a prior uterine scar. Unadjusted and adjusted results were similar and only the adjusted results are presented here. Women with chronic hypertension were excluded from the analyses of gestational hypertension and preeclampsia and women with preexisting diabetes were excluded from the analyses of gestational diabetes. In the analyses of superimposed preeclampsia, women with chronic hypertension were not categorized as having a chronic disease unless they had other underlying conditions.
Sensitivity analyses were conducted to address missing data. Missing demographic data were treated as an unknown category in the main analyses (29). Analyses with missing outcome data (threatened preterm births, 12.3% missing; and maternal ICU admissions, 21.5% missing) were restricted to sites with complete data. Data were secondly restricted to women without any missing data. Finally, we evaluated the impact of patient prepregnancy weight. Although maternal overweight and obesity are important risk factors for several outcomes studied, thyroid diseases are known to cause weight gain or loss. Thus, maternal weight was regarded as a collider in the thyroid disease outcome pathway and adjusting for weight would have introduced bias in the analyses. To assess the robustness of our findings with respect to maternal weight, we repeated the analyses restricting to women with a normal prepregnancy BMI (18.5–24.99 kg/m2). All statistical analyses were performed using SAS 9.2 software (SAS Institute Inc., Cary, North Carolina).
Results
Of all pregnancies 1.5% were complicated with primary hypothyroidism, 0.1% with iatrogenic hypothyroidism, 0.2% with hyperthyroidism, 0.1% with other thyroid diseases, and 1.2% with unspecified thyroid disease. Women with thyroid diseases were older, were more often multiparous, and had more chronic conditions than women without thyroid diseases (Table 1). Women with primary hypothyroidism and unspecified thyroid disease were more likely to be overweight/obese and smoke less often, and women with hypothyroidism and other and unspecified thyroid diseases were more often non-Hispanic white and were more likely to have private insurance (nonsignificantly in women with other thyroid diseases) and uterine scars than women without thyroid diseases (Table 1).
Table 1.
Demographic Data of the Singleton Pregnancies of the Consortium on Safe Labor (2002–2008), by Thyroid Disease Status
| No Thyroid Disease (n = 216 901) | Primary Hypothyroidism (n = 3183) | P Value | Iatrogenic Hypothyroidism (n = 178) | P Value | Hyperthyroidism (n = 417) | P Value | Other Thyroid Diseases (n = 263) | P Value | Unspecified Thyroid Disease (n = 2570) | P Value | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Pregnancies contributing, no. (%) | |||||||||||
| 1 | 198 265 (91.4) | 2829 (88.9) | .01 | 145 (81.5) | <.001 | 357 (85.6) | .02 | 237 (90.1) | .44 | 2340 (91.0) | .03 |
| 2 | 17 726 (8.2) | 338 (10.6) | 32 (18.0) | 51 (12.2) | 24 (9.1) | 216 (8.4) | |||||
| ≥3 | 910 (0.4) | 16 (0.5) | 1 (0.6) | 9 (2.2) | 2 (0.8) | 14 (0.6) | |||||
| Mean maternal age, y (SD)a | 27.5 (6.2) | 31.5 (5.3) | <.001 | 31.0 (5.3) | <.001 | 28.9 (5.9) | <.001 | 30.2 (5.8) | <.001 | 30.4 (5.6) | <.001 |
| Prepregnancy BMI, no. (%)b | |||||||||||
| Underweight | 7804 (3.6) | 87 (2.7) | <.001 | 5 (2.8) | .41 | 14 (3.4) | .56 | 8 (3.0) | .18 | 81 (3.2) | .001 |
| Normal weight | 76 917 (35.5) | 1003 (31.5) | 53 (29.8) | 111 (26.6) | 97 (36.9) | 930 (36.2) | |||||
| Overweight | 32 358 (14.9) | 528 (16.6) | 25 (14.0) | 55 (13.2) | 41 (15.6) | 492 (19.1) | |||||
| Obese | 15 232 (7.0) | 257 (8.1) | 21 (11.8) | 27 (6.5) | 27 (10.3) | 221 (8.6) | |||||
| Morbidly obese | 11 630 (5.4) | 244 (7.7) | 17 (9.6) | 27 (6.5) | 19 (7.2) | 227 (8.8) | |||||
| Missing | 72 960 (33.6) | 1064 (33.4) | 57 (32.0) | 185 (44.4) | 71 (27.0) | 619 (24.1) | |||||
| Nulliparous, no. (%) | 87 127 (40.2) | 1118 (35.1) | <.001 | 55 (30.9) | .01 | 132 (31.7) | <.001 | 81 (30.8) | .001 | 728 (28.3) | <.001 |
| Prior uterine scars, no. (%) | 30 796 (14.2) | 672 (21.1) | <.001 | 34 (19.1) | .02 | 63 (15.1) | .64 | 52 (19.8) | .008 | 418 (16.3) | <.001 |
| Race/ethnicity, no. (%) | |||||||||||
| Non-Hispanic white | 105 802 (48.8) | 2420 (76.0) | <.001 | 127 (71.4) | <.001 | 189 (45.3) | .10 | 173 (65.8) | .001 | 1792 (69.7) | <.001 |
| Non-Hispanic black | 49 588 (22.9) | 209 (6.6) | 27 (15.2) | 144 (34.5) | 37 (14.1) | 257 (10.0) | |||||
| Hispanic | 38 380 (17.7) | 300 (9.4) | 14 (7.9) | 48 (11.5) | 34 (12.9) | 293 (11.4) | |||||
| Asian/Pacific islander | 8977 (4.1) | 109 (3.4) | 4 (2.3) | 20 (4.8) | 7 (2.7) | 94 (3.7) | |||||
| Other | 5202 (2.4) | 32 (1.0) | 2 (1.2) | 3 (0.7) | 2 (0.8) | 55 (2.1) | |||||
| Unknown | 8952 (4.1) | 113 (3.6) | 4 (2.3) | 13 (3.1) | 10 (3.8) | 79 (3.1) | |||||
| Health insurance, no. (%) | |||||||||||
| Private | 120 044 (55.4) | 2412 (75.8) | <.001 | 144 (80.9) | <.001 | 227 (54.2) | .75 | 168 (63.9) | .08 | 1972 (76.7) | <.001 |
| Public or self-pay | 73 608 (33.9) | 483 (15.2) | 29 (16.3) | 161 (38.6) | 67 (25.5) | 517 (20.1) | |||||
| Other | 310 (0.1) | 6 (0.2) | 1 (0.6) | 2 (0.5) | 0 | 0 | |||||
| Unknown | 22 939 (10.6) | 282 (8.9) | 4 (2.3) | 27 (6.5) | 28 (10.7) | 81 (3.2) | |||||
| Maternal chronic diseases, no. (%) | |||||||||||
| Any | 35 233 (16.2) | 829 (26.0) | <.001 | 47 (26.4) | .004 | 125 (30.0) | <.001 | 64 (24.3) | <.001 | 526 (20.5) | <.001 |
| Diabetes | 3031 (1.4) | 163 (5.1) | 5 (2.8) | 16 (3.8) | 13 (4.9) | 84 (3.3) | |||||
| Hypertension | 4146 (1.9) | 113 (3.6) | 10 (5.6) | 33 (7.9) | 7 (2.7) | 57 (2.2) | |||||
| Asthma | 16 476 (7.6) | 275 (8.6) | 15 (8.4) | 46 (11.0) | 19 (7.2) | 213 (8.3) | |||||
| Heart disease | 3198 (1.5) | 95 (3.0) | 11 (6.2) | 29 (7.0) | 6 (2.3) | 34 (1.3) | |||||
| Renal disease | 1734 (0.8) | 38 (1.2) | 1 (0.6) | 8 (1.9) | 0 | 19 (0.7) | |||||
| Gastrointestinal disease | 2431 (1.1) | 39 (1.6) | 6 (3.4) | 5 (1.2) | 4 (1.5) | 34 (1.3) | |||||
| Depression | 9085 (4.2) | 301 (9.5) | 16 (9.0) | 25 (6.0) | 29 (11.0) | 164 (6.4) | |||||
| Smoking, no. (%) | 14 591 (6.7) | 144 (4.5) | <.001 | 12 (6.7) | .83 | 43 (10.3) | .06 | 12 (4.6) | .18 | 131 (5.1) | .03 |
P values were obtained from pairwise logistic regression using generalized estimating equations to account for correlated data with no thyroid disease as comparison group. All P values are adjusted for site.
Maternal age is missing for 0.1% observations.
Maternal BMI is categorized as underweight if BMI is <18.5 kg/m2, normal weight if BMI is 18.5 to 24.9 kg/m2, overweight if BMI is 25.0 to 29.9 kg/m2, obese if BMI is 30.0 to 34.9 kg/m2, and morbidly obese if BMI is ≥35.0 kg/m2.
Primary hypothyroidism
Primary hypothyroidism was associated with decreased odds of spontaneous labor and increased odds of cesarean section after spontaneous labor, prelabor cesarean section, and induction. It was also associated with 1.5-fold odds of preeclampsia, 2.3-fold odds of superimposed preeclampsia, 1.6-fold odds of gestational diabetes, 1.4- and 1.3-fold odds of threatened and observed preterm births, and 2.1-fold odds of ICU admission (Tables 2 and 3).
Table 2.
Outcomes of Singleton Pregnancies in the Consortium on Safe Labor (2002–2008), by Thyroid Disease Status
| Pregnancy Complication | No Thyroid Disease (n = 216 901) | Primary Hypothyroidism (n = 3183) | Iatrogenic Hypothyroidism (n = 178) | Hyperthyroidism (n = 417) | Other Thyroid Diseases (n = 263) | Unspecified Thyroid Disease (n = 2570) |
|---|---|---|---|---|---|---|
| Gestational hypertension | 5859 (2.7) | 110 (3.5) | 7 (3.9) | 21 (5.0) | 6 (2.3) | 77 (3.0) |
| Preeclampsia | 10 197 (4.7) | 181 (5.7) | 8 (4.5) | 32 (7.7) | 13 (4.9) | 121 (4.7) |
| Superimposed preeclampsia | 1793 (0.8) | 52 (1.6) | 5 (2.8) | 16 (3.8) | 2 (0.8) | 25 (1.0) |
| Gestational diabetes | 10805 (5.0) | 298 (9.4) | 14 (7.9) | 31 (7.4) | 16 (6.1) | 183 (7.1) |
| Threatened preterm birtha | 6550 (3.4) | 134 (5.5) | 13 (9.6) | 23 (7.2) | 14 (6.7) | 122 (4.8) |
| Placental abruption | 3497 (1.6) | 43 (1.4) | 8 (4.5) | 9 (2.2) | 4 (1.5) | 61 (2.4) |
| Placenta previa | 1517 (0.7) | 33 (1.0) | 2 (1.1) | 6 (1.4) | 5 (1.9) | 22 (0.9) |
| PROM | 15138 (7.0) | 256 (8.0) | 16 (9.0) | 26 (6.2) | 32 (12.2) | 123 (4.8) |
| PPROM | 4943 (2.3) | 92 (2.9) | 7 (3.9) | 12 (2.9) | 10 (3.8) | 48 (1.9) |
| Chorioamnionitis | 6753 (3.1) | 85 (2.7) | 5 (2.8) | 14 (3.4) | 5 (1.9) | 57 (2.2) |
| Hemorrhage | 24211 (6.6) | 206 (6.5) | 20 (11.2) | 21 (5.0) | 18 (6.8) | 239 (9.3) |
| Total vaginal deliveries | 156542 (72.2) | 1948 (61.2) | 100 (56.2) | 285 (68.4) | 183 (69.6) | 1862 (72.5) |
| Prelabor cesarean sections | 24898 (11.5) | 572 (18.0) | 31 (17.4) | 55 (13.2) | 34 (12.9) | 291 (11.3) |
| Spontaneous labors | 117276 (54.1) | 1425 (44.8) | 79 (44.4) | 187 (44.8) | 127 (48.3) | 1350 (52.5) |
| Spontaneous labor with cesarean section | 19845 (16.9) | 373 (26.2) | 26 (32.9) | 35 (18.7) | 28 (22.1) | 277 (20.5) |
| Inductions | 74727 (34.5) | 1186 (37.3) | 68 (38.2) | 175 (42.0) | 102 (38.8) | 929 (36.2) |
| Induction with cesarean sections | 15616 (20.9) | 290 (24.5) | 21 (30.9) | 42 (24.0) | 18 (17.7) | 140 (15.1) |
| Breech presentation | 9233 (4.3) | 181 (5.7) | 17 (9.6) | 25 (6.0) | 14 (5.3) | 126 (4.9) |
| Preterm birthb | 25216 (11.6) | 447 (14.0) | 31 (17.4) | 92 (22.1) | 39 (14.8) | 319 (12.4) |
| Late preterm birth | 16763 (7.7) | 291 (9.1) | 18 (10.1) | 62 (14.9) | 25 (9.5) | 213 (8.3) |
| Early preterm birth | 8453 (3.9) | 156 (4.9) | 13 (7.3) | 30 (7.2) | 14 (5.3) | 106 (4.1) |
| Maternal ICU admissionc | 940 (0.6) | 15 (0.6) | 1 (0.7) | 5 (1.6) | 3 (1.5) | 11 (0.5) |
| Maternal deaths | 18 (0.0) | 0 | 0 | 1 (0.2) | 0 | 0 |
All figures are n (%).
Threatened preterm labor is missing for 12.3% of total observations.
Preterm birth is all births occurring before 37 weeks of gestation. Late preterm birth indicates preterm births occurring at ≥34 weeks of gestation but <37 weeks of gestation. Early preterm birth indicates preterm births at <34 weeks of gestation.
Maternal ICU admission information is missing with 23.7% of total observations.
Table 3.
Odds Ratios of Adverse Pregnancy Outcomes Among Women with Thyroid Diseases in the Consortium on Safe Labor (2002–2008)
| Pregnancy Complication | No Thyroid Disease (n = 216901) | Primary Hypothyroidism (n = 3183) | Iatrogenic Hypothyroidism (n = 178) | Hyperthyroidism (n = 417) | Other Thyroid Diseases (n = 263) | Unspecified Thyroid Disease (n = 2570) |
|---|---|---|---|---|---|---|
| Gestational hypertensiona | 1.00 | 1.17 (0.90–1.52) | 1.28 (0.47–3.46) | 1.78 (0.97–3.26) | 0.81 (0.28–2.40) | 1.14 (0.84–1.55) |
| Preeclampsiaa | 1.00 | 1.47 (1.20–1.81) | 1.22 (0.48–3.08) | 1.78 (1.08–2.94) | 1.17 (0.55–2.53) | 1.11 (0.87–1.41) |
| Superimposed preeclampsiab | 1.00 | 2.25 (1.53–3.29) | 3.04 (0.68–13.55) | 3.64 (1.82–7.29) | 1.03 (0.17–6.22) | 1.14 (0.67–1.95) |
| Gestational diabetesc | 1.00 | 1.57 (1.33–1.86) | 1.34 (0.63–2.82) | 1.32 (0.81–2.15) | 1.09 (0.55–2.16) | 1.31 (1.07–1.62) |
| Threatened preterm birthd | 1.00 | 1.43 (1.13–1.82) | 2.18 (1.01–4.70) | 2.13 (1.18–3.85) | 1.71 (0.82–3.55) | 1.19 (0.93–1.53) |
| Placental abruption | 1.00 | 0.93 (0.62–1.38) | 2.89 (1.14–7.36) | 1.42 (0.59–3.44) | 1.04 (0.28–3.77) | 1.19 (0.85–1.67) |
| Placenta previa | 1.00 | 1.20 (0.76–1.91) | 1.45 (0.23–9.16) | 1.92 (0.67–5.51) | 2.55 (0.78–8.29) | 1.02 (0.58–1.78) |
| PROM | 1.00 | 1.08 (0.91–1.29) | 1.39 (0.71–2.73) | 0.82 (0.47–1.40) | 2.02 (1.23–3.30) | 0.85 (0.67–1.09) |
| PPROM | 1.00 | 1.24 (0.94–1.65) | 1.64 (0.61–4.42) | 0.99 (0.47–2.11) | 1.79 (0.78–4.13) | 0.97 (0.66–1.43) |
| Chorioamnionitis | 1.00 | 1.11 (0.82–1.50) | 1.62 (0.48–5.51) | 1.51 (0.74–3.05) | 0.84 (0.26–2.72) | 0.98 (0.68–1.40) |
| Hemorrhage | 1.00 | 0.96 (0.79–1.17) | 1.57 (0.81–3.03) | 0.87 (0.48–1.58) | 0.92 (0.47–1.80) | 1.01 (0.85–1.22) |
| Prelabor cesarean section | 1.00 | 1.31 (1.11–1.54) | 1.71 (0.89–3.28) | 0.94 (0.56–1.58) | 0.91 (0.48–1.73) | 1.27 (1.02–1.58) |
| Spontaneous labor | 1.00 | 0.82 (0.74–0.90) | 0.73 (0.48–1.12) | 0.76 (0.58–1.00) | 0.84 (0.60–1.18) | 0.83 (0.74–0.93) |
| Spontaneous labor with cesarean section | 1.00 | 1.38 (1.14–1.66) | 2.05 (1.01–4.16) | 0.96 (0.53–1.74) | 0.97 (0.49–1.92) | 1.07 (0.86–1.31) |
| Inductions | 1.00 | 1.15 (1.04–1.28) | 1.20 (0.78–1.86) | 1.40 (1.06–1.86) | 1.27 (0.89–1.82) | 1.14 (1.02–1.28) |
| Induction with cesarean sections | 1.00 | 1.15 (0.93–1.41) | 2.23 (1.00–4.99) | 1.27 (0.75–2.14) | 0.81 (0.36–1.81) | 0.92 (0.70–1.19) |
| Breech presentation | 1.00 | 1.10 (0.89–1.35) | 2.09 (1.07–4.07) | 1.30 (0.76–2.22) | 1.12 (0.53–2.40) | 1.04 (0.82–1.33) |
| Preterm birthe | 1.00 | 1.34 (1.17–1.53) | 1.62 (0.95–2.75) | 1.81 (1.32–2.49) | 1.43 (0.92–2.24) | 1.19 (1.01–1.39) |
| Late preterm birth | 1.00 | 1.23 (1.04–1.45) | 1.30 (0.68–2.49) | 1.77 (1.23–2.54) | 1.28 (0.74–2.19) | 1.14 (0.94–1.38) |
| Early preterm birth | 1.00 | 1.48 (1.19–1.84) | 2.06 (0.95–4.45) | 1.57 (0.95–2.60) | 1.64 (0.82–3.29) | 1.25 (0.96–1.63) |
| Maternal ICU admissionf | 1.00 | 2.08 (1.04–4.15) | 2.81 (0.21–37.00) | 3.70 (1.16–11.80) | 3.93 (0.79–19.48) | 1.36 (0.59–3.13) |
All numbers are ORs with 99% CIs. ORs are obtained from multivariate logistic regression with generalized estimating equations to account for correlated data. All results are adjusted for site, insurance, parity, race/ethnicity, age, smoking, and other chronic diseases. Spontaneous labor, all cesarean section variables, and induction are also adjusted for previous uterine scars.
Women with chronic hypertension have been excluded from the analyses of gestational hypertension and preeclampsia.
Chronic hypertension was not regarded as part of the chronic conditions in the analyses of superimposed preeclampsia.
Women with preexisting diabetes were excluded from the analyses of gestational diabetes.
Threatened preterm labor is missing for 12.3% of total observations.
Preterm birth is all births occurring before 37 weeks of gestation. Late preterm birth indicates preterm births occurring at ≥34 weeks of gestation but <37 weeks of gestation. Early preterm birth indicates preterm births at <34 weeks of gestation.
Maternal ICU admission information is missing for 23.7% of total observations.
Iatrogenic hypothyroidism
Iatrogenic hypothyroidism was associated with increased odds of cesarean section after spontaneous labor and was also associated with 2.9-fold odds of placental abruption, 2.2-fold odds of threatened preterm births, and 2.1-fold odds of breech presentation (Tables 2 and 3).
Hyperthyroidism
Hyperthyroidism was associated with increased odds of induction but not for cesarean section. It was also associated with 1.8-fold odds of preeclampsia, 3.6-fold odds of superimposed preeclampsia, 2.1- and 1.8-fold odds of threatened and observed preterm births, and 3.7-fold odds of ICU admission (Tables 2 and 3).
Other thyroid diseases
Other diagnosed thyroid diseases (goiter, thyroiditis, and nodules) were associated with 2.0-fold odds of PROM (Tables 2 and 3). The distributions of goiter, thyroiditis, and other thyroid conditions among those with adverse outcome were 34.4%, 43.8%, and 21.9% for PROM.
Unspecified thyroid diseases
Unspecified thyroid disease during pregnancy was associated with a decreased risk of spontaneous labor and increased odds of induction, prelabor cesarean section, gestational diabetes, and preterm birth (Tables 2 and 3).
Sensitivity analyses
We performed 3 sensitivity analyses, restricting the data to (1) eight sites with the best outcome and exposure reporting, (2) women with no missing values, and (3) normal-weight women. Although precision was reduced because of smaller sample sizes, the estimates remained very similar to the main analyses (data not shown).
Discussion
In this large, contemporary cohort of racially and ethnically diverse US women, we found that primary hypothyroidism was associated with increased odds of preeclampsia, preterm birth, gestational diabetes, induction, cesarean section, and ICU admission. Iatrogenic hypothyroidism (hypothyroidism due to thyroid surgery or ablation) was associated with increased risk of placental abruption and cesarean delivery, whereas hyperthyroidism was associated with increased risk of preeclampsia, preterm birth, induction, and ICU admission and other thyroid diseases (goiter, thyroiditis, and nodules) were associated with increased risk of PROM.
Hypothyroidism was associated with a higher probability of developing preeclampsia, superimposed preeclampsia, and gestational diabetes and with an increased risk of preterm births in our study, consistent with previous studies (9–17). Thyroid hormones seem to be important in placentation and regulation of early pregnancy (30), which might partly explain the association between hypothyroidism and gestational hypertensive disease and preterm birth. Normal thyroid function is needed to maintain normal energy and lipid metabolism (31), and weight gain or lipid disturbances associated with hypothyroidism may also increase the risk of adverse pregnancy outcomes. Changes in energy metabolism and weight may also explain the observed association between hypothyroidism and gestational diabetes. We observed increased risks of pregnancy outcomes even among normal weight women with hypothyroidism, although with less precision because of the smaller sample size.
Our finding of increased likelihood of ICU admission among women with hypothyroidism is novel, although hypothyroidism was linked previously to cardiac failure and hemorrhage (7). These factors, along with the increased risk of cesarean section and the presence of comorbid conditions among women with hypothyroidism may increase their likelihood of being admitted to the ICU.
Women with hypothyroidism also had decreased odds of presenting in spontaneous labor. In addition to prelabor cesarean sections, cesarean delivery was increased in women with hypothyroidism, both after spontaneous onset of labor or induction. Although the association between hypothyroidism, inductions, and cesarean section was described previously (16, 17), no other study has investigated the association of hypothyroidism with data on the timing of cesarean delivery. The reasons for increased risk of cesarean delivery may be due to the associated pregnancy complications, such as hypertensive disorders, gestational diabetes, and preterm birth. Whether otherwise uncomplicated hypothyroidism increases risk of cesarean section warrants further study.
Iatrogenic hypothyroidism was associated with higher risk estimates than primary hypothyroidism for most outcomes (although with less precision because of the small sample size) and was also associated with increased risk of placental abruptions and breech presentations. Because women with iatrogenic hypothyroidism may have lost all residual thyroid tissue, they may be more difficult to treat adequately during pregnancy (3). Increased risk for placental abruption was observed previously among women with subclinical hypothyroidism or thyroid peroxidase antibody positivity (8, 32), but whether the increased risk is due to clustering of adverse outcomes such as preeclampsia or due to the hypothyroidism or thyroid autoimmunity itself cannot be determined in our study. The observed increased risk of breech presentation was higher among preterm births, suggesting that this finding might be more a function of gestation length than an effect of hypothyroidism. However, previous studies have also associated thyroid hypofunction with breech presentations in term pregnancies, suggesting that maternal hypothyroidism may have a more profound effect on fetal motor function (33–35).
Consistent with previous reports (23–25), we found an association between increased odds of hypertension, preterm birth, and ICU admissions and hyperthyroidism. Notably, 2 of the 5 women with hyperthyroidism needing ICU treatment had cardiac failure after obstetrical surgery. A novel finding was that hyperthyroid women had more inductions than women without thyroid diseases. Maternal hyperthyroidism is associated with fetal growth restriction, tachycardia, and even fetal hypothyroidism, because thyroid hormones, TSH receptor antibodies, and antithyroid drugs cross the placenta (1). Adequate management of hyperthyroidism reduces the risks of these adverse pregnancy outcomes (36), indicating that thyroid hormone excess might be the underlying cause.
Deliveries complicated with other thyroid diseases (ie, goiter, thyroiditis, and nodules) were associated with higher odds of PROM, which is a novel finding. Women with goiter and thyroiditis have a high prevalence of thyroid autoantibodies, and up to 20% of women with thyroid autoantibodies develop hypothyroidism during pregnancy. Hypothyroidism might partly explain the increased risk of adverse outcomes observed, but thyroid autoantibody positivity has also been associated with PPROM and preterm births (20). Although PROM and PPROM possibly have different etiologies, autoimmunity might have some effect on the increased risk of PROM in our study.
Women with thyroid disease noted in their medical records, who did not have specific diagnoses included in their discharge summaries (unspecified thyroid disease), experienced increased risk for gestational diabetes, prelabor cesarean section, induction, and preterm birth. Because their disease was only recorded in the medical history and not the discharge summary, this group may include women who have milder disease or whose disease was not complicated during pregnancy. In general, their risks were similar to those of the hypothyroid women although somewhat attenuated, which supports the notion their disease was less active.
The strengths of this study are its large size with comprehensive data collection from the hospital medical records from the intrapartum admission, allowing for the evaluation and adjustment for important confounding factors. The large dataset supports the study of rare outcomes such as ICU admissions, placental abruptions, and placenta previa and rare exposures such as iatrogenic hypothyroidism and other thyroid diseases. We might have missed cases of subclinical hypothyroidism but do not expect substantial bias from this potential misclassification given the large numbers of pregnancies under study. Although the study data cover a 7-year time span, during which knowledge about thyroid diseases in pregnancy has grown, the proportion of cases did not change over time, suggesting that increased interest did not lead to better case ascertainment. However, we did observe small improvements in outcomes such as preterm delivery over time. It is also possible that physician concerns for patients with thyroid disease lead to improved detection of other chronic or comorbid conditions, but it is unlikely to have influenced the rate of labor and delivery complications.
Although we lacked treatment data, we observed high risks of pregnancy complications associated with thyroid diseases, especially hyperthyroidism. If we presume that most women have been adequately managed, our risk estimates are probably conservative. A survey among obstetricians showed that at the time of the CSL study only 66% of them would have recommended antepartum thyroid function testing in their patients with hypothyroidism or hyperthyroidism (37), indicating that up to 30% of women with thyroid disease would have had no follow-up of their disease or treatment status during pregnancy. Indeed, a Scottish study found that up to 30% of women with levothyroxine treatment have elevated TSH at some point during pregnancy without changes to their levothyroxine dose (4).
In conclusion, our results indicate that thyroid diseases are associated with significant increases in morbidity during pregnancy, but given the fact that we lacked information on treatment during pregnancy, we were unable to explore whether inadequate treatment was the cause or whether some of the risk is due to thyroid disease itself. Previous studies have shown that adverse outcomes could be mitigated and possibly prevented by adequate management of thyroid diseases (16, 38) and that there is a need for improved education and thyroid management during pregnancy (37, 39). If we assume that most women were treated, our observed risk estimates may be conservative. We found that thyroid diseases were associated with increased risk for multiple adverse outcomes, including preeclampsia and preterm birth, and women with thyroid diseases were more likely to be admitted to the ICU. Further research is still needed to distinguish whether women with adequately treated thyroid disease have a higher risk of pregnancy complications because of the disease itself or whether treatment can truly prevent adverse outcomes.
Acknowledgments
This work was supported by the Intramural Research Program of the National Institutes of Health, Eunice Kennedy Shriver National Institute of Child Health and Human Development. The data included in this article were obtained from the Consortium on Safe Labor, which was supported by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, through Contract HHSN267200603425C. The named authors alone are responsible for the views expressed in this article, which do not necessarily represent the decisions or the stated policy of the Eunice Kennedy Shriver National Institute of Child Health and Human Development.
Institutions involved in the Consortium on Safe Labor are the following (in alphabetical order): Baystate Medical Center, Springfield, Massachusetts; Cedars-Sinai Medical Center Burnes Allen Research Center, Los Angeles, California; Christiana Care Health System, Newark, Delaware; Georgetown University Hospital, MedStar Health, Washington, DC; Indiana University Clarian Health, Indianapolis, Indiana; Intermountain Healthcare and the University of Utah, Salt Lake City, Utah; Maimonides Medical Center, Brooklyn, New York; MetroHealth Medical Center, Cleveland, Ohio; Summa Health System, Akron City Hospital, Akron, Ohio; The EMMES Corporation, Rockville Maryland (Data Coordinating Center); University of Illinois at Chicago, Chicago, Illinois; University of Miami, Miami, Florida; and University of Texas Health Science Center at Houston, Houston, Texas.
Disclosure Summary: The authors have nothing to disclose.
Footnotes
- BMI
- body mass index
- CI
- confidence interval
- CSL
- Consortium for Safe Labor
- ICU
- intensive care unit
- OR
- odds ratio
- PROM
- premature rupture of membranes
- PPROM
- preterm premature rupture of membranes.
References
- 1. Stagnaro-Green A, Abalovich M, Alexander E, et al. Guidelines of the American Thyroid Association for the diagnosis and management of thyroid disease during pregnancy and postpartum. Thyroid. 2011;21:1081–1125 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Alexander EK, Marqusee E, Lawrence J, Jarolim P, Fischer GA, Larsen PR. Timing and magnitude of increases in levothyroxine requirements during pregnancy in women with hypothyroidism. N Engl J Med. 2004;351:241–249 [DOI] [PubMed] [Google Scholar]
- 3. Loh JA, Wartofsky L, Jonklaas J, Burman KD. The magnitude of increased levothyroxine requirements in hypothyroid pregnant women depends upon the etiology of the hypothyroidism. Thyroid. 2009;19:269–275 [DOI] [PubMed] [Google Scholar]
- 4. Vadiveloo T, Mires GJ, Donnan PT, Leese GP. Thyroid testing in pregnant women with thyroid dysfunction in Tayside, Scotland: the Thyroid Epidemiology, Audit and Research Study (Tears). Clin Endocrinol (Oxf). 2013;78:466–471 [DOI] [PubMed] [Google Scholar]
- 5. Abalovich M, Gutierrez S, Alcaraz G, Maccallini G, Garcia A, Levalle O. Overt and subclinical hypothyroidism complicating pregnancy. Thyroid. 2002;12:63–68 [DOI] [PubMed] [Google Scholar]
- 6. Allan WC, Haddow JE, Palomaki GE, et al. Maternal thyroid deficiency and pregnancy complications: implications for population screening. J Med Screen. 2000;7:127–130 [DOI] [PubMed] [Google Scholar]
- 7. Davis LE, Leveno KJ, Cunningham FG. Hypothyroidism complicating pregnancy. Obstet Gynecol. 1988;72:108–112 [PubMed] [Google Scholar]
- 8. Casey BM, Dashe JS, Wells CE, et al. Subclinical hypothyroidism and pregnancy outcomes. Obstet Gynecol. 2005;105:239–245 [DOI] [PubMed] [Google Scholar]
- 9. Ashoor G, Maiz N, Rotas M, Kametas NA, Nicolaides KH. Maternal thyroid function at 11 to 13 weeks of gestation and subsequent development of preeclampsia. Prenat Diagn. 2010;30:1032–1038 [DOI] [PubMed] [Google Scholar]
- 10. Leung AS, Millar LK, Koonings PP, Montoro M, Mestman JH. Perinatal outcome in hypothyroid pregnancies. Obstet Gynecol. 1993;81:349–353 [PubMed] [Google Scholar]
- 11. Sahu MT, Das V, Mittal S, Agarwal A, Sahu M. Overt and subclinical thyroid dysfunction among Indian pregnant women and its effect on maternal and fetal outcome. Arch Gynecol Obstet. 2010;281:215–220 [DOI] [PubMed] [Google Scholar]
- 12. Tudela CM, Casey BM, McIntire DD, Cunningham FG. Relationship of subclinical thyroid disease to the incidence of gestational diabetes. Obstet Gynecol. 2012;119:983–988 [DOI] [PubMed] [Google Scholar]
- 13. Karakosta P, Alegakis D, Georgiou V, et al. Thyroid dysfunction and autoantibodies in early pregnancy are associated with increased risk of gestational diabetes and adverse birth outcomes. J Clin Endocrinol Metab. 2012;97:4464–4472 [DOI] [PubMed] [Google Scholar]
- 14. Jones WS, Man EB. Thyroid function in human pregnancy. VI. Premature deliveries and reproductive failures of pregnant women with low serum butanol-extractable iodines. Maternal serum TBG and TBPA capacities. Am J Obstet Gynecol. 1969;104:909–914 [PubMed] [Google Scholar]
- 15. Schneuer FJ, Nassar N, Tasevski V, Morris JM, Roberts CL. Association and predictive accuracy of high TSH serum levels in first trimester and adverse pregnancy outcomes. J Clin Endocrinol Metab. 2012;97:3115–3122 [DOI] [PubMed] [Google Scholar]
- 16. Matalon S, Sheiner E, Levy A, Mazor M, Wiznitzer A. Relationship of treated maternal hypothyroidism and perinatal outcome. J Reprod Med. 2006;51:59–63 [PubMed] [Google Scholar]
- 17. Wikner BN, Sparre LS, Stiller CO, Källén B, Asker C. Maternal use of thyroid hormones in pregnancy and neonatal outcome. Acta Obstet Gynecol Scand. 2008;87:617–627 [DOI] [PubMed] [Google Scholar]
- 18. Agarwal MM, Dhatt GS, Punnose J, Bishawi B, Zayed R. Thyroid function abnormalities and antithyroid antibody prevalence in pregnant women at high risk for gestational diabetes mellitus. Gynecol Endocrinol. 2006;22:261–266 [DOI] [PubMed] [Google Scholar]
- 19. Ashoor G, Maiz N, Rotas M, Jawdat F, Nicolaides KH. Maternal thyroid function at 11–13 weeks of gestation and spontaneous preterm delivery. Obstet Gynecol. 2011;117:293–298 [DOI] [PubMed] [Google Scholar]
- 20. Cleary-Goldman J, Malone FD, Lambert-Messerlian G, et al. Maternal thyroid hypofunction and pregnancy outcome. Obstet Gynecol. 2008;112:85–92 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Männistö T, Vääräsmäki M, Pouta A, et al. Perinatal outcome of children born to mothers with thyroid dysfunction or antibodies: a prospective population-based cohort study. J Clin Endocrinol Metab. 2009;94:772–779 [DOI] [PubMed] [Google Scholar]
- 22. Männistö T, Vääräsmäki M, Pouta A, et al. Thyroid dysfunction and autoantibodies during pregnancy as predictive factors of pregnancy complications and maternal morbidity in later life. J Clin Endocrinol Metab. 2010;95:1084–1094 [DOI] [PubMed] [Google Scholar]
- 23. Davis LE, Lucas MJ, Hankins GD, Roark ML, Cunningham FG. Thyrotoxicosis complicating pregnancy. Am J Obstet Gynecol. 1989;160:63–70 [DOI] [PubMed] [Google Scholar]
- 24. Millar LK, Wing DA, Leung AS, Koonings PP, Montoro MN, Mestman JH. Low birth weight and preeclampsia in pregnancies complicated by hyperthyroidism. Obstet Gynecol. 1994;84:946–949 [PubMed] [Google Scholar]
- 25. Sheffield JS, Cunningham FG. Thyrotoxicosis and heart failure that complicate pregnancy. Am J Obstet Gynecol. 2004;190:211–217 [DOI] [PubMed] [Google Scholar]
- 26. Aoki Y, Belin RM, Clickner R, Jeffries R, Phillips L, Mahaffey KR. Serum TSH and total T4 in the United States population and their association with participant characteristics: National Health and Nutrition Examination Survey (NHANES 1999–2002). Thyroid. 2007;17:1211–1223 [DOI] [PubMed] [Google Scholar]
- 27. Bryant AS, Worjoloh A, Caughey AB, Washington AE. Racial/ethnic disparities in obstetric outcomes and care: prevalence and determinants. Am J Obstet Gynecol. 2010;202:335–343 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Zhang J, Troendle J, Reddy UM, et al. Contemporary cesarean delivery practice in the United States. Am J Obstet Gynecol. 2010;203:326.e1−326.e10 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Zhang J, Branch DW, Ramirez MM, et al. Oxytocin regimen for labor augmentation, labor progression, and perinatal outcomes. Obstet Gynecol. 2011;118:249–256 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Barber KJ, Franklyn JA, McCabe CJ, et al. The in vitro effects of triiodothyronine on epidermal growth factor-induced trophoblast function. J Clin Endocrinol Metab. 2005;90:1655–1661 [DOI] [PubMed] [Google Scholar]
- 31. Yen PM. Physiological and molecular basis of thyroid hormone action. Physiol Rev. 2001;81:1097–1142 [DOI] [PubMed] [Google Scholar]
- 32. Abbassi-Ghanavati M, Casey BM, Spong CY, McIntire DD, Halvorson LM, Cunningham FG. Pregnancy outcomes in women with thyroid peroxidase antibodies. Obstet Gynecol. 2010;116:381–386 [DOI] [PubMed] [Google Scholar]
- 33. Kooistra L, Kuppens SM, Hasaart TH, et al. High thyrotrophin levels at end term increase the risk of breech presentation. Clin Endocrinol (Oxf). 2010;73:661–665 [DOI] [PubMed] [Google Scholar]
- 34. Kuppens SM, Kooistra L, Wijnen HA, et al. Maternal thyroid function during gestation is related to breech presentation at term. Clin Endocrinol (Oxf). 2010;72:820–824 [DOI] [PubMed] [Google Scholar]
- 35. Pop VJ, Brouwers EP, Wijnen H, Oei G, Essed GG, Vader HL. Low concentrations of maternal thyroxin during early gestation: a risk factor of breech presentation? BJOG. 2004;111:925–930 [DOI] [PubMed] [Google Scholar]
- 36. Vissenberg R, van den Boogaard E, van Wely M, et al. Treatment of thyroid disorders before conception and in early pregnancy: a systematic review. Hum Reprod Update. 2012;18:360–373 [DOI] [PubMed] [Google Scholar]
- 37. Power ML, Kilpatrick S, Schulkin J. Diagnosing and managing thyroid disorders during pregnancy: a survey of obstetrician-gynecologists. J Reprod Med. 2004;49:79–82 [PubMed] [Google Scholar]
- 38. Negro R, Schwartz A, Gismondi R, Tinelli A, Mangieri T, Stagnaro-Green A. Universal screening versus case finding for detection and treatment of thyroid hormonal dysfunction during pregnancy. J Clin Endocrinol Metab. 2010;95:1699–1707 [DOI] [PubMed] [Google Scholar]
- 39. Rinaldi MD, Stagnaro-Green AS. Thyroid disease and pregnancy: degrees of knowledge. Thyroid. 2007;17:747–753 [DOI] [PubMed] [Google Scholar]