Surgical treatment of papillary thyroid cancer diagnosed during pregnancy: a retrospective, single-center, case-control study

Abstract

Purpose

Papillary thyroid cancer (PTC), a common malignancy among women of reproductive age, is increasingly common during pregnancy because of routine prenatal imaging. However, the optimal timing for surgical treatment remains controversial. This study aimed to assess the safety and practicality of performing thyroid surgery during pregnancy through a comparative analysis of perioperative and pathological outcomes between pregnant and nonpregnant women.

Methods

We conducted a retrospective case-control study of 100 female patients aged 20–39 years who underwent thyroid surgery for PTC between January 2019 and December 2024. Patients were grouped into pregnant (n = 14) and nonpregnant (n = 86) cohorts. Clinical, surgical, and pathological data were compared.

Results

Most pregnant patients underwent surgery during the second trimester. The operative duration was significantly shorter in the pregnant group than in the nonpregnant group (71.4 ± 26.35 minutes vs. 87.2 ± 5.17 minutes, P = 0.025), and there was a trend towards a lower drain volume on postoperative day 1 (25.2 ± 15.23 mL vs. 32.7 ± 0.18 mL, P = 0.077). No significant differences were observed in terms of complication rates. Despite a higher frequency of extrathyroidal extension (P = 0.003) and advanced T stage (P = 0.038) in the pregnant group, the surgical outcomes were favorable. One neonate experienced mild intraventricular hemorrhage that resolved without sequelae.

Conclusion

Thyroid surgery during pregnancy, particularly in the second trimester, appears feasible in selected patients. Given the small sample and retrospective nature of this study, larger prospective studies are needed to validate these findings.

INTRODUCTION

Thyroid cancer occurs in 10.2 per 100,000 people under the age of 50 years, making it the second most common cancer in this age group worldwide. Papillary thyroid carcinoma (PTC) accounts for approximately 87.3% of all thyroid malignancies and is generally considered to have an indolent course and is often associated with favorable outcomes [1]. It is prevalent among women of childbearing age, and it is also likely to be detected during pregnancy [1,2,3,4,5,6].

Despite its generally slow progression, the diagnosis of PTC during pregnancy presents a unique clinical dilemma. According to the 2017 American Thyroid Association (ATA) guidelines, surgery is not strongly recommended during pregnancy for PTC, in contrast to more aggressive histologies such as follicular, medullary, or anaplastic thyroid cancer. If the tumor remains radiologically stable or is identified in the second or third trimester, the guidelines suggest deferring surgery until after delivery [7].

Nevertheless, there are situations in which surgical intervention becomes necessary during pregnancy—for example, when tumors exhibit interval growth, when sonographic features become suspicious, or when patients experience considerable psychological distress related to the cancer diagnosis [7,8,9]. Moreover, the accumulating evidence supporting the safety of nonobstetric surgery during pregnancy, including breast surgery, has reinforced the feasibility of thyroid surgery in select cases [10,11].

However, data regarding the perioperative outcomes and characteristics of pregnant patients undergoing thyroid surgery remain limited. Furthermore, the role of patient-centered decision-making—including emotional and psychological considerations—has not been well examined in the literature.

In this study, we aimed to evaluate the safety and feasibility of thyroid surgery during pregnancy by comparing the perioperative and pathological outcomes between pregnant and nonpregnant patients diagnosed with PTC.

METHODS

Ethics statement

The ethics of this study were approved by the Institutional Review Board of Inha University Hospital (No. 2024-09-014). This study was performed in accordance with the Declaration of Helsinki and written informed consent was waived due to its retrospective nature and minimal risk to participants.

Study design

This was a retrospective, single-center, case-control study conducted at Inha University Hospital (Incheon, Korea). Medical records were reviewed for all female patients aged 20–39 years who underwent thyroid surgery for PTC between January 2019 and December 2024.

Patients were divided into 2 groups: those who underwent surgery during pregnancy (pregnant group) and those who were not pregnant at the time of surgery (nonpregnant group). Pregnancy status was verified through obstetric documentation at the time of thyroid cancer diagnosis. For pregnant patients, surgical decisions were guided by the 2015 ATA criteria for sonographic progression. Surgery was considered if there was a 20% increase in nodule diameter in at least 2 dimensions, a minimal increase of 2 mm, or more than a 50% increase in volume [12,13]. Additionally, even in the absence of these criteria, surgery was performed upon patient request following detailed counseling with the care team. Prior to surgery, all pregnant patients were referred to obstetricians for coevaluation during outpatient visits to ensure maternal and fetal safety and to coordinate perioperative monitoring. Although complete nonstress test (NST) records were not available due to technical limitations, partial information on fetal heart rate and uterine contractions was retrieved from nursing notes during the perioperative period. All pregnant patients provided informed consent for surgery after being counseled about the potential maternal and fetal complications that may arise from undergoing surgery during pregnancy. All surgeries were performed by the same experienced endocrine surgeons, which ensured consistency in surgical technique and decision-making throughout the study period.

The inclusion criteria were as follows: (1) female sex, (2) age between 20 and 39 years, and (3) a diagnosis of PTC with surgical treatment during the study period. All eligible patients who met the inclusion criteria during the study period were consecutively included without selective exclusion.

The exclusion criteria were as follows: (1) lateral neck lymph node metastasis, (2) distant metastasis, (3) prior thyroid surgery, and (4) incomplete clinical or pathological data. Because no pregnant patients in our cohort had lateral neck or distant metastases, we excluded such cases from the control group as well to enable balanced comparison of perioperative outcomes. In cases involving advanced disease, the surgical approach often differs substantially—requiring wider dissection, longer operative time, and occasionally serving a palliative role rather than a curative one. Inclusion of such cases could have confounded the analysis. Accordingly, our study focused on low- to intermediate-risk patients undergoing definitive surgery and is not generalizable to those with advanced-stage disease.

The age range was restricted to 20–39 years in both groups to reduce potential confounding factors related to age-dependent variations in tumor biology, hormonal influences, and perioperative risks. However, no formal individual-level matching was performed.

Data collection

Data, including demographics (e.g., age, body mass index), obstetric history, thyroid comorbidities, operative details (e.g., extent of surgery, operative time, blood loss, drain output), hospitalization period, and biochemical markers (e.g., serum calcium, ionized calcium, parathyroid hormone [PTH]), were extracted from electronic medical records. Pathologic data included tumor size, extrathyroidal extension, margin status, lymphovascular invasion, lymph node metastasis, and genetic mutations (BRAF, TERT). Neonatal outcomes such as gestational age at delivery and neonatal complications were also recorded.

Statistical analysis

All the statistical analyses were performed using R software (ver. 4.3.1, R Foundation for Statistical Computing). The normality of continuous variables was tested using the Shapiro-Wilk test. Normally distributed variables were analyzed with the Student t-test, whereas nonnormally distributed variables were analyzed using the Mann-Whitney U-test. Categorical variables were analyzed using the Pearson chi-square test. When the expected cell frequency was less than 5, the Fisher exact test was applied to ensure statistical validity. A P-value of less than 0.05 was considered statistically significant. A complete-case analysis was conducted, excluding subjects with missing values for any of the variables analyzed.

RESULTS

This study included 100 patients: 14 in the pregnant group and 86 in the nonpregnant group. The mean ages were 33.0 and 33.6 years, respectively, with no significant difference (P = 0.307). Underlying thyroid diseases (Graves disease, hyperthyroidism, or hypothyroidism) were present in 3 pregnant patients (21.4%) and 16 nonpregnant patients (18.6%). A family history of thyroid disease was reported in 1 pregnant patient (7.1%) and 11 nonpregnant patients (12.8%). Two pregnant patients (14.3%) developed thyroid dysfunction during pregnancy, which required medical management. Surgical extent, tumor laterality, and fine-needle aspiration cytology findings were not significantly different between the 2 groups. The detailed baseline characteristics are shown in Table 1.

Table 1. Demographic and clinical characteristics of the patients.

Values are presented as number only, mean ± standard deviation, or number (%).

PTC, papillary thyroid cancer; NA, not applicable.

The mean gestational age at surgery in the pregnant group was 20.1 weeks (range, 14–34 weeks). Seven pregnant patients were nulliparous and 7 were parous. Twelve of the 14 pregnant patients (85.7%) underwent surgery during the second trimester, which is consistent with current clinical guidelines. Indications for surgery included sonographic tumor invasion, rapid tumor growth, or patient preference. In 2 patients (14.3%), surgery was performed in response to the patient's preference for early management, rather than clinical progression. One patient experienced recurrent laryngeal nerve injury requiring intraoperative repair. One newborn was reported to have mild intraventricular hemorrhage (IVH) and was admitted to the neonatal intensive care unit (NICU) for observation. The cause of the IVH was unknown, and the mother underwent surgery at 22 weeks and delivered at 36 weeks of gestation. No invasive intervention was required and no prolonged sequelae were noted. The neonate was delivered by cesarean section at 36 weeks of gestation, which was a preterm birth. The detailed data for each pregnant patient are shown in Table 2.

Table 2. Detailed characteristics of patients in the pregnant group.

PTC, papillary thyroid cancer; RAI, radioactive iodine; IVH, intraventricular hemorrhage; NICU, neonatal intensive care unit.

The perioperative outcomes are summarized in Table 3. The mean operative time was significantly shorter in the pregnant group than in the nonpregnant group (71.4 ± 26.35 minutes vs. 87.2 ± 5.17 minutes, P = 0.025). No statistically significant differences were observed in estimated intraoperative blood loss (27.9 ± 28.07 mL vs. 30.8 ± 22.16 mL, P = 0.772) or length of hospitalization (2.43 ± 0.59 days vs. 2.63 ± 0.19 days, P = 0.163). A trend toward lower drainage volume on postoperative day 1 was observed in the pregnant group (25.2 ± 15.23 mL vs. 32.7 ± 0.18 mL, P = 0.077). Postoperative complication rates were similar between the groups.

Table 3. Perioperative outcomes of pregnant and nonpregnant patients undergoing thyroid surgery.

Values are presented as mean ± standard deviation or number (%).

NA, not applicable.

^a)Hypoparathyroidism outcomes were calculated only for patients who underwent total thyroidectomy (pregnant, n = 7; nonpregnant, n = 36; total, n = 43).

No cases of postoperative bleeding or dyspnea were reported in the pregnant group, whereas 2 cases of bleeding and 5 cases of dyspnea occurred in the nonpregnant group. Given the low event rates, these variables were not subjected to statistical comparison. The rates of transient and permanent vocal cord palsy were comparable between the groups.

Based on the perioperative monitoring information related to NST, the mean fetal heart rate in the pregnant group was 146.6 beats per minute (bpm) preoperatively and 150.4 bpm postoperatively. Transient uterine contractions after surgery were noted in 2 patients, all of which resolved spontaneously without the need for medication or further intervention.

Among the 43 patients who underwent total thyroidectomy (7 pregnant, 36 nonpregnant), transient hypoparathyroidism occurred in 71% of pregnant patients and 53% of nonpregnant patients (P = 0.437). Permanent hypoparathyroidism occurred in 28.6% and 11.1% of the patients, respectively (P = 0.248). Postoperative PTH levels were lower in the pregnant group on both postoperative day 1 (10.3 ± 20.85 pg/mL vs. 19.2 ± 3.05 pg/mL, P = 0.457) and day 2 (9.0 ± 16.99 pg/mL vs. 17.0 ± 0.98 pg/mL, P = 0.830), although the differences were not statistically significant. On postoperative day 2, ionized calcium levels were elevated in the pregnant group (1.09 ± 0.001 mg/mL vs. 1.01 ± 0.15 mg/mL, P = 0.047). Despite these differences, they were resolved by the first outpatient visit (1.16 ± 0.06 mg/mL vs. 1.16 ± 0.05 mg/mL, P = 0.660). Although the rate of radioactive iodine (RAI) therapy was lower in the pregnant group (42.9% vs. 77.8%), the difference was not statistically significant (P = 0.081). Among patients who underwent total thyroidectomy, the mean thyroid-stimulating hormone (TSH) level at the first postoperative follow-up was 1.1 ± 6.24 mIU/L in the pregnant group and 4.9 ± 1.58 mIU/L in the nonpregnant group. Based on these results, the mean daily levothyroxine dose was 2.2 ± 0.07 µg/kg/day and 2.0 ± 0.45 µg/kg/day, respectively. The mean TSH level at the second follow-up was 0.5 ± 9.21 mIU/L in the pregnant group and 4.1 ± 2.02 mIU/L in the nonpregnant group. The detailed data are presented in Table 4.

Table 4. Postoperative endocrine outcomes including PTH, calcium, RAI treatment, and TSH suppression therapy in the total thyroidectomy subgroup.

Values are presented as mean ± standard deviation or number (%).

PTH, parathyroid hormone; RAI, radioactive iodine; TSH, thyroid-stimulating hormone; POD, postoperative day.

The mean tumor size was 1.43 ± 0.53 cm in the pregnant group and 1.25 ± 0.90 cm in the nonpregnant group (P = 0.117). Extrathyroidal extension (57.1% vs. 17.4%, P = 0.003) and T3 stage or higher (28.6% vs. 5.8%, P = 0.038) were both significantly more common in the pregnant group. Other parameters, including surgical margin status, lymphovascular invasion, nodal metastasis, and the BRAF mutation rate, were not significantly different between the groups. No TERT promoter mutations were identified. The details are summarized in Table 5.

Table 5. Pathological characteristics of pregnant and nonpregnant patients.

Values are presented as number only, mean ± standard deviation, or number (%).

DISCUSSION

The 2017 ATA guidelines recommend deferring surgery for patients with differentiated thyroid cancer diagnosed during pregnancy unless there is clinical evidence of disease progression, such as tumor growth or newly observed suspicious lymph nodes. This cautious approach is based on the understanding that PTC typically follows an indolent course and that immediate surgical intervention offers limited benefit during gestation. The second trimester is preferred when surgery is unavoidable, as it minimizes fetal exposure during organogenesis and avoids the risk of preterm labor [7]. The Korean Thyroid Association (KTA) also provides similar recommendations. According to the KTA guidelines, newly diagnosed PTC during early pregnancy should be monitored by ultrasound rather than immediate surgery, and surgery may be considered only if there is evidence of tumor progression before 24–26 weeks of gestation. Conversely, if the tumor does not progress by 24–26 weeks or if the diagnosis is made after 20 weeks, surgery is generally deferred until after delivery [14].

Previous studies support these recommendations. Kuy et al. [15] analyzed 31,356 female patients who underwent thyroid or parathyroid surgery, including 165 pregnant patients. Pregnant women had higher rates of general and endocrine complications and longer hospital stays after thyroid or parathyroid surgery. A Japanese study compared 24 pregnant women who underwent surgery during pregnancy (mostly in the second trimester) with 21 women who underwent surgery postpartum [16]. The outcomes of both groups were comparable, and neither group had major maternal or fetal complications. These findings suggest that second-trimester surgery is safe but still support deferring surgery when feasible.

However, these data do not necessarily imply that surgery during pregnancy should always be avoided. When clinically indicated, second-trimester thyroid surgery has been shown to be safe in multiple studies [17]. While such cases are uncommon, there are reports of tumor progression in patients who decide to delay surgery until after delivery [18,19]. In addition, deferring surgery may lead to significant delays in initiating postoperative treatments such as RAI therapy and TSH suppression, both of which are important for long-term oncologic management [3,6,9,20]. These treatments are often postponed further by the need to complete breastfeeding or by the demands of caring for a newborn, which may limit timely access to medical services. Surgery during early pregnancy, when performed safely, may help avoid prolonged delays in comprehensive cancer management and support more timely initiation of definitive therapy after delivery [9].

In addition to clinical factors, the psychological impact of a cancer diagnosis during pregnancy must also be considered. Many patients diagnosed with thyroid cancer experience considerable anxiety, regardless of the tumor's biological behavior [21]. This anxiety may stem from uncertainty about disease progression, concern for fetal health, or fear of potential treatment-related harm. Several studies have demonstrated that cancer-related psychological stress can affect not only maternal decision-making but also obstetric outcomes such as preterm birth [8] . While such concerns may not always reflect actual clinical risk, the emotional burden is clinically meaningful and should not be underestimated in treatment planning.

In our study, 14 pregnant women underwent thyroid surgery for PTC. Most surgeries were performed in the second trimester, which is consistent with guideline recommendations [7]. Notably, 2 patients in the pregnant group elected to undergo surgery primarily owing to emotional distress, despite the absence of radiologic progression. Both patients underwent lobectomy—one in the second trimester and the other in the third trimester. The perioperative outcomes were favorable, with no maternal surgical complications and uneventful postoperative recovery in both patients.

Perioperative fetal monitoring was performed using NST. According to the available perioperative nursing records related to NST monitoring, the mean fetal heart rate remained within normal physiological ranges both before and after surgery, and no significant abnormalities were observed. Although transient uterine contractions were noted in 2 patients, they resolved spontaneously without the need for medication or additional intervention, suggesting that surgery did not pose substantial stress to the fetus. While the available data were limited to partial nursing records, we believe these findings provide supportive evidence that thyroid surgery, when appropriately timed and managed, can be performed safely during pregnancy without compromising fetal stability.

Among all pregnant patients included in the study, only one fetal complication was observed: mild IVH. This occurred in the neonate of one of the patients who underwent surgery due to emotional burden. The mother had received routine prenatal care at an external hospital, and no abnormalities had been identified during pregnancy. Regarding the neonate, we were only able to obtain details through maternal self-report. According to the mother, the baby was delivered by cesarean section at 36 weeks of gestation, which qualifies as a preterm birth, although the exact reason for the early delivery remains unclear. Hemorrhage was detected postnatally and the neonate was admitted to the NICU for observation but recovered fully without any sequelae through supportive care alone at the delivery hospital, without the need for surgical or intensive medical intervention. Unfortunately, further information such as IVH grade, underlying causes, or specific postnatal management could not be retrieved, and this reflects the inherent limitation of relying on secondhand reports in a retrospective study. Although no major neonatal complications were reported in the pregnant group, the assessment of fetal and neonatal outcomes was limited by the retrospective nature of the study and incomplete external obstetric records.

Despite more frequent extrathyroidal extension and advanced T stage in the pregnant group than in the nonpregnant control group, the perioperative results were comparable. A significantly shorter operative time was observed, and there was a trend toward lower drain output on postoperative day 1. These differences may reflect a more focused and time-efficient surgical strategy aimed at minimizing fetal exposure.

In our study population, transient hypoparathyroidism occurred in 71.4% (5 of 7) and permanent hypoparathyroidism in 28.6% (2 of 7) of the pregnant patients who underwent total thyroidectomy. Vocal cord palsy occurred in 7.1% (1 of 14) for both transient and permanent cases. In the nonpregnant group, the rates were 52.8% (19 of 36) for transient hypoparathyroidism and 11.1% (4 of 36) for permanent hypoparathyroidism; transient and permanent vocal cord palsy occurred in 5.8% (5 of 86) and 2.3% (2 of 86), respectively. No cases of postoperative bleeding or dyspnea were reported in either group. Compared with a large Korean cohort study of 2,636 patients by Lee et al. [22], which reported rates of 28.4% for transient hypoparathyroidism, 0.3% for permanent hypoparathyroidism, and 0.7% and 0.2% for transient and permanent vocal cord palsy, respectively, the complication rates in our study—particularly in the pregnant group—were higher [22]. This may reflect our small sample size, making rates more susceptible to variation, and the possibility that surgery during pregnancy itself contributes to increased complication risks. Nonetheless, all complications were managed conservatively without long-term sequelae, supporting the feasibility of thyroid surgery during pregnancy when appropriately indicated and performed with multidisciplinary care.

We found no significant differences in the incidence of transient or permanent hypoparathyroidism or in postoperative PTH or calcium levels in the subgroup of patients who underwent total thyroidectomy. Although ionized calcium levels were slightly higher on postoperative day 2 in the pregnant group, this difference did not persist at the time of the first outpatient follow-up. Patients who exhibited laboratory-confirmed hypocalcemia or symptoms of hypoparathyroidism were treated with oral calcium carbonate and alfacalcidol. These medications were gradually tapered on the basis of serial monitoring of calcium and PTH levels during outpatient follow-up. These findings suggest that perioperative calcium regulation and preservation of parathyroid function are both feasible and effective in pregnant patients undergoing total thyroidectomy. The lower rate of postoperative RAI therapy observed in the pregnant group is likely attributable to its contraindication during pregnancy and lactation. Among patients who underwent total thyroidectomy, the pregnant group showed mean TSH levels that were generally close to the intended suppression target, although the small sample size and large standard deviations require cautious interpretation. In contrast, the nonpregnant group maintained mean TSH levels around 4 mIU/L at both follow-ups, indicating that consistent TSH suppression was not achieved in this group.

This study had several limitations. First, it was a retrospective, single-center analysis with a small sample size, particularly in the pregnant group (n = 14), which may have limited the statistical power. Future studies with larger cohorts or multicenter designs could enhance the generalizability and robustness of these findings. Second, selection bias is possible, as the decision to undergo surgery during pregnancy is influenced by disease characteristics and patient preferences. If additional data become available, comparisons between pregnant patients undergoing surgery and those not undergoing surgery could help assess the extent of this bias. Third, obstetric and neonatal outcomes were limited to immediate delivery-related events and were based primarily on maternal self-reports, which may be subject to recall bias or incomplete reporting. While NST was performed perioperatively for fetal monitoring, storage limitations of monitoring equipment precluded retrospective analysis of fetal data, and we had to rely on nursing records for partial information on fetal heart rate and uterine contractions. Additionally, most antenatal care and deliveries were conducted at external institutions, and corresponding obstetric records—such as birth weight, Apgar scores, and gestational complications—were not accessible. As a result, many neonatal outcomes, including preterm labor and postnatal events, were obtained secondhand, increasing the risk of information bias. Incorporating continuous follow-up data from surgery through delivery could allow for more detailed assessments of fetal safety. Fourth, we did not evaluate long-term maternal oncologic outcomes such as recurrence or survival, nor did we assess postnatal developmental outcomes in neonates. Further studies with extended follow-up are necessary to address these gaps. Finally, the applicability of our findings to other populations and health care systems may be limited. Integrating data from diverse demographic groups and international health care contexts would provide a more comprehensive understanding of the management of thyroid cancer during pregnancy.

Despite these limitations, our findings suggest that thyroid surgery may be safely performed during pregnancy when clinically indicated. In selected cases with tumor progression or significant emotional and logistical concerns, second-trimester surgery can be considered as a reasonable option. However, given the limited sample size, further large-scale studies are warranted to validate these observations.