Highlights
· Cord blood thyroid-stimulating hormone (TSH) levels were evaluated in 1,088 neonates, of whom 5.9% had levels above 20 μIU/mL.
· Higher cord blood TSH levels were associated with primiparity, gestational diabetes, gestational hypertension, mode of delivery, and birth time.
· Maternal TSH showed a modest positive correlation with cord blood TSH, while effect sizes for maternal and perinatal factors were generally small.
Introduction
Thyroid hormones are essential for brain development, especially during the first 3 years of life. These hormones are necessary for the regulation of axonal and dendritic growth, neuronal migration, synaptogenesis, and myelination [1]. Therefore, thyroid abnormalities can cause neuronal damage leading to severe intellectual disability.
Congenital hypothyroidism (CH) is the most common preventable cause of intellectual disability and developmental impairment in neonates [2,3]. The global prevalence of CH is estimated to be 1 in 3,000–4,000 live births. Moreover, the prevalence of CH is much higher in India, estimated at approximately 1 in 1,000 [4,5].
It is caused by the permanent or transient deficiency of thyroid hormones at birth. However, most infants with CH appear unaffected at birth, probably due to placental transfer of maternal thyroid hormones. The clinical manifestations may not appear until a median age of 4 months, contributing to delayed diagnosis [6].
Thyroid therapy should be initiated within 2 weeks of birth to prevent or minimize brain damage [7].
Owing to the high prevalence of CH among newborns, neonatal screening programs for the early detection of CH were introduced worldwide in the 1970s [8]. In 2007, a neonatal CH screening program was introduced by the Indian Council of Medical Research in India [9,10]. Cognitive outcomes have improved substantially following the implementation of these screening programs [11,12].
Traditionally, dried blood spots obtained from newborns by heel prick after the second day of life were used. Subsequently, umbilical cord blood sampling for thyroid-stimulating hormone (TSH) estimation was adopted as an alternative screening method because it is less invasive, facilitates sample collection before early discharge, and offers high sensitivity, a lower recall rate, and cost-effectiveness [13]. TSH was found to be more sensitive than free thyroxine (T4) for detecting CH [14].
Currently, cord blood TSH estimation is considered less reliable than serum thyroid hormone measurement because neonatal thyroid function can be affected by stress-related factors. Nevertheless, cord blood TSH estimation remains useful for CH screening in newborns. Elevated cord blood TSH levels might indicate thyroid dysfunction, necessitating further management.
Associations of maternal and perinatal factors with cord blood TSH levels have been widely studied; however, findings remain inconsistent. The present study aimed to evaluate the association of maternal and perinatal factors with cord blood TSH levels.
Materials and methods
This study was a hospital-based descriptive, cross-sectional study conducted by the Department of Biochemistry in collaboration with the Department of Paediatrics at Velammal Medical College Hospital and Research Institute, Madurai, Tamil Nadu, India, from January to December 2022. All consecutive live births during the study period were included in the study. Newborns with congenital malformations, maternal thyroid disorders, and inadequate or hemolyzed samples were excluded.
Of 1,220 newborns enrolled, 129 were excluded due to maternal hypothyroidism, and 3 cord blood samples could not be processed. Finally, 1,088 consecutive newborns delivered at our hospital were included in the study. The sample size was determined by the number of consecutive live births during the study period, representing a convenience sample.
According to hospital protocol, 2 mL of umbilical cord blood was collected immediately after delivery from the clamped cord under sterile conditions as part of routine neonatal screening and promptly transported to the laboratory for TSH estimation. A thorough physical examination was performed by a pediatrician, and Apgar scores at 1 and 5 minutes were recorded. Demographic data, birth weight, gestational age, and maternal history of gestational diabetes mellitus (GDM), gestational hypertension (GHTN), or maternal anemia were recorded.
GDM was diagnosed according to the Diabetes in Pregnancy Study Group India criteria, defined as a 2-hour plasma glucose value ≥140 mg/dL following a 75 g oral glucose load. GHTN was defined as systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg occurring after 20 weeks of gestation in previously normotensive women, consistent with the American College of Obstetricians and Gynecologists (ACOG) criteria. Maternal anemia was defined as hemoglobin concentration <11 g/dL during pregnancy according to the World Health Organization criteria.
Maternal serum TSH values were obtained from the mothers’ antenatal case records. In our hospital, thyroid function testing is routinely performed during the first trimester as part of antenatal screening.
Both maternal serum TSH and cord blood TSH were measured using an electrochemiluminescence immunoassay on a Cobas e 411 analyzer (Roche Diagnostics, Germany). The TSH reference range specified by the assay kit was 2.3–13.2 μIU/mL. Newborns with cord blood TSH levels >20 μIU/mL were recalled on the third postnatal day for measurement of serum TSH, free triiodothyronine, and free T4 using peripheral venous blood samples.
All data were analyzed using the statistical software IBM SPSS Statistics ver. 22.0 (IBM Corp., USA). The primary outcome variable was cord blood TSH level.
Independent variables included maternal factors (age, parity, GDM, GHTN, anemia, maternal TSH) and neonatal/perinatal factors (gestational age, birth weight, mode of delivery, Apgar score). Categorical variables were expressed as percentages and quantitative variables as median (interquartile range [IQR]). Cord blood TSH levels were compared across groups using Mann-Whitney U-test or Kruskal-Wallis test, as appropriate.
Effect sizes were calculated for nonparametric tests and expressed as rank-biserial correlation (r) for Mann-Whitney U-tests and epsilon squared (ε²) for Kruskal-Wallis tests, interpreted according to Cohen criteria. Spearman correlation analysis was performed to assess associations between quantitative variables that did not follow a normal distribution. Multivariable linear regression analysis with log-transformed cord blood TSH as the dependent variable was performed to identify independent predictors while adjusting for potential confounders.
A P-value <0.05 was considered statistically significant.
The study protocol was reviewed and approved by the Institutional Ethics Committee of Velammal Medical College Hospital and Research Institute, Madurai, Tamil Nadu, India (IEC No: VMCIEC/35/2022, dated January 2, 2022). Written informed consent was obtained from all participating mothers prior to enrollment of their newborns in the study.
Results
Of 1,088 newborns enrolled in the study, 554 (50.9%) were male and 534 (49.1%) were female. Additional descriptive data for the categorical study variables are presented in Table 1.
Table 2 shows that the median cord blood TSH level was 7.22 μIU/mL (IQR, 5.39–10.28 μIU/mL). Of 1,088 newborns, 64 (6%) with cord blood TSH levels >20 μIU/mL underwent repeat thyroid function testing on the third postnatal day. The manufacturer-specified reference range for cord blood TSH was 2.3–13.2 μIU/mL.
In this study, cord blood TSH levels were higher among neonates born to primiparous mothers and mothers with GDM or GHTN, but not among those with maternal anemia. Our study showed no significant differences in cord blood TSH according to neonatal sex, gestational age, weight for gestational age, presentation, blood group, or Rh type. However, our study showed significantly higher cord blood TSH levels in neonates with perinatal depression.
We evaluated the cord blood TSH levels in different modes of delivery, which differed significantly (Table 3A). Effect size analysis indicated small effect sizes for parity while birth time, gestational diabetes mellitus, and gestational hypertension showed negligible effects. Mode of delivery also showed a small effect (Table 3B).
Neonates born by normal vaginal delivery had higher cord blood TSH levels than those born by cesarean section. Additionally, neonates born by vacuum-assisted delivery had higher cord blood TSH levels than those born by forceps-assisted delivery. Similarly, neonates born by emergency lower-segment cesarean section (LSCS) had higher cord blood TSH levels than those born by elective LSCS (Fig.
1).
We also evaluated correlations between cord blood TSH and quantitative study variables. Apgar scores at 1 and 5 minutes showed significant negative correlations. Of 1,088 newborns, 64 (6%) had elevated cord blood TSH levels >20 μIU/mL.
Significant positive correlations were observed between cord blood TSH and postnatal day-3 serum TSH and free T4 levels (Table 4A and B; Fig. 2). These correlations were derived only from the selected subgroup of neonates with elevated cord blood TSH levels who underwent repeat testing.
Among the maternal parameters, only maternal TSH levels correlated significantly with cord blood TSH levels (Table 5).
To identify independent determinants of cord blood TSH levels, a multivariable linear regression analysis was performed with log-transformed cord blood TSH as the dependent variable. Covariates included maternal age, parity, GDM, GHTN, anemia, birth weight, gestational age, and mode of delivery.
After adjustment for potential confounders, primiparity (B=-0.053, P=0.003) and mode of delivery (B=-0.015, P=0.002) were significantly associated with lower log-transformed cord blood TSH levels. In contrast, maternal complications such as GDM (B=0.118, P=0.006) and GHTN (B=0.064, P=0.047) were independently associated with higher log-transformed TSH levels (Table 6). Maternal age, maternal anemia, birth weight, and gestational age were not independently associated with cord blood TSH levels in the adjusted model.
The final regression model was statistically significant (F=8.364, P<0.001), explaining approximately 5.2% of the variability in cord blood TSH levels.
Discussion
Cord blood TSH levels are useful for screening for CH, a preventable cause of intellectual disability in children. The transient postnatal surge in newborn TSH levels is considered a normal physiological adaptation to extrauterine life and is attributed to stress-induced alpha-adrenergic stimulation and an abrupt increase in hypothalamic thyrotropin-releasing hormone. TSH levels progressively decline over the first 3–5 postnatal days [1].
Elevated levels can indicate potential thyroid dysfunction, necessitating further investigation. This study was conducted to analyze associations between various maternal and perinatal factors and cord blood TSH levels.
This study identified several maternal and perinatal factors associated with cord blood TSH levels, with maternal thyroid status, parity, and pregnancy complications showing significant associations. In our study, the sex distribution was fairly equal, with slightly more male neonates. The majority of the neonates were born at term to multiparous mothers and were appropriate for gestational age.
Cephalic presentation and normal vaginal delivery were the most common presentation and mode of delivery, respectively. The most common blood groups were B-positive and O-positive. GDM, GHTN, and maternal anemia were present in 2.7%, 5.1%, and 3.3% of mothers, respectively (Table 1).
The cord blood TSH reference range specified in the manufacturer’s kit insert was 2.3–13.2 μIU/mL. In our study, only 5.9% of newborns had cord blood TSH levels >20 μIU/mL, which was lower than the proportion reported in previous studies [15]. However, a study conducted in Bangalore also reported a lower percentage (3.1%) of newborns with elevated cord blood TSH levels >20 μIU/mL [16].
In our study, preterm infants had slightly higher TSH levels than term infants, but this increase was not significant. However, the findings vary across studies [17-20]. Our study found no significant associations of cord blood TSH with maternal age, neonatal sex, or gestational age, similar to previous studies [15,21].
Our findings showed that newborns born to primiparous mothers had higher cord blood TSH levels than those born to multiparous mothers. This may be attributed to more challenging labor conditions in first deliveries, which may increase stress hormones and affect TSH levels [15,19,22]. In our study, conditions such as GDM and GHTN were associated with significantly higher cord blood TSH levels, consistent with several studies [19,23], although some studies showed inconsistent findings [17,21,24].
Our study found that cord blood TSH values did not correlate with maternal glucose levels and glucose challenge test results, which is consistent with previous findings [21]. Infants with fetal distress during labor exhibited elevated cord blood TSH levels, reinforcing the link between perinatal stress and thyroid function [19,24].
In our study, neonates born by normal vaginal delivery had higher cord blood TSH levels than those born by cesarean section. This may reflect physiological stress responses during labor, including catecholamine surges that can affect TSH secretion. These findings were consistent with several previous studies [19,22-24].
Additionally, assisted vaginal deliveries were associated with even higher TSH levels than normal vaginal delivery, indicating increased perinatal stress, which is consistent with previous findings [15,21,22]. Neonates born by vacuum-assisted delivery had higher cord blood TSH levels than those born by forceps-assisted delivery. Likewise, neonates born by emergency LSCS had higher cord blood TSH levels than those born by elective LSCS.
Overall, cord blood TSH levels followed the order vacuum-assisted delivery > forceps-assisted delivery > normal vaginal delivery > emergency LSCS > elective LSCS. The observed AM–PM variation (Table 3A) in cord blood TSH may reflect physiological circadian variation rather than external environmental exposures.
Although these associations were statistically significant, effect size analysis indicated that the magnitude of these effects was small, suggesting that multiple maternal and perinatal factors may collectively influence cord blood TSH levels.
Our study found no significant correlation between birth weight and cord blood TSH levels, although some studies have linked low birth weight to elevated TSH levels [15,19]. Lower postdelivery Apgar scores correlated with higher cord blood TSH levels, suggesting an association between perinatal distress and neonatal thyroid function. These findings were consistent with several previous studies that demonstrated elevated cord blood TSH in conditions associated with fetal distress, such as cord encirclement and meconium-stained liquor [15,18,22,23].
Multivariable regression analysis confirmed that GDM and GHTN were independently associated with higher cord blood TSH levels after adjusting for potential confounders. The overall explanatory power of the model was modest, suggesting that neonatal thyroid function is influenced by multiple maternal and perinatal factors.
The American Academy of Pediatrics, the American Thyroid Association, and ACOG recommend collecting TSH samples ideally between postnatal days 2 and 4, when the TSH surge subsides. However, this may be difficult because of early discharge practices [25]. Newborns with elevated cord blood TSH were recalled on postnatal day 3 for thyroid function testing using a peripheral venous blood sample.
In our study, only 6% of newborns had elevated cord blood TSH >20 μIU/mL and significant positive correlations were observed between cord blood TSH and postnatal day-3 serum TSH and free T4 levels [26].
The single-center design limits the generalizability of the findings. Because serum TSH was measured only in neonates with elevated cord blood TSH, sensitivity and specificity could not be assessed. Other limitations include unmeasured potential confounders, such as maternal diet and iodine status, that may affect thyroid hormone levels.
Environmental exposures, including exposure to endocrine-disrupting chemicals, were not documented. Although multivariable regression analysis was performed to adjust for potential confounders, the explanatory power of the model was modest, suggesting that additional biological and environmental factors may influence cord blood TSH levels. The cross-sectional study design limits the ability to establish causality.
Therefore, longitudinal studies with larger sample sizes and rigorous control of confounding variables are needed to validate our study findings.
In conclusion, our findings suggest that maternal and perinatal factors are associated with variations in cord blood TSH levels. However, the effect sizes were small, indicating a limited influence. These factors may be considered when interpreting neonatal TSH screening results, although their overall impact appears modest.
Further studies in larger and more diverse populations that include additional maternal and environmental parameters are needed to better understand the determinants of neonatal thyroid function.