This article is more than 5 years old. Information may no longer be current.
Delayed TSH elevation associated with invasive procedures in neonatal ICU newborns
Newborns with delayed thyroid-stimulating hormone elevation may experience a greater need for surgery and other invasive procedures compared with newborns without delayed elevation, according to published findings.
“The occurrence of [delayed] TSH weeks after a normal TSH measurement at the first neonatal screening is enigmatic,” the researchers wrote. “TSH levels in the second screening may be only mildly elevated, but ... in many neonates TSH levels reached 100 IU/mL or more. Considering the importance of normal thyroid function to neurocognitive development, especially during the early years of life, a second thyroid screen is currently recommended in certain at-risk populations.”
Amnon Zung, MD, of the pediatric endocrinology unit at Kaplan Medical Center in Israel, and colleagues evaluated 100 newborns with delayed TSH (52 boys) from neonatal ICUs in Israel and 200 control neonates without delayed TSH elevation from the same neonatal ICU (104 boys) to compare risk factors for delayed TSH elevation. Forty-six variables were compared, including maternal parameters, treatment in the neonatal ICU, procedures in the neonatal ICU, and syndromes, malformations and several clinical emergencies.
Birth weight and gestational age were similar among all participants.
Cesarean section, mechanical ventilation, patent ductus arteriosus, pneumothorax and the administration of cefotaxime, vancomycin, fluconazole, dopamine, ibuprofen, furosemide, insulin and packed red blood cells were significantly more common in participants with delayed TSH compared with controls and could be considered risk factors for delayed TSH, according to the researchers. The rate of surgery or other invasive procedure was lower in controls (5.1%) compared with participants with delayed TSH (6%).
Four risk factors for delayed TSH were identified by multivariate analysis: patent ductus arteriosus and administration of vancomycin, insulin or furosemide.
“The observations made in this study strengthen our previous work in which we claimed that [delayed] TSH is not associated with [gestational age] or birth weight, but rather, is typical to the most severe cases in the [neonatal] ICU,” the researchers wrote. “Both studies support a paradigm of second thyroid screening in all [neonatal] ICU patients rather than the selective approach for certain at-risk populations. Although we found certain risk factors associated with [delayed] TSH, further studies are required to elucidate the common pathway by which these different factors contribute to its evolution.” – by Amber Cox
Disclosures: The researchers report no relevant financial disclosures.
Perspective
Back to Top
In this interesting paper, the authors investigate determinants of a delayed TSH rise that reaches above the reference range and may, therefore, lead to a delayed diagnosis of congenital hypothyroidism. Particularly from the perspective that newborns with congenital hypothyroidism might benefit from receiving treatment as early as possible, this is a relevant outcome, and new determinants would benefit risk assessments. Although the authors conclude that gestational age and birth weight are not associated with a delay in the diagnosis of elevated TSH, no association of gestational age at birth or birth weight (continuously) is presented.
There are two key factors that were not studied, but would have been an addition to the current literature. First, is the TSH or thyroxine concentration at the first measurement indicative of having a delayed diagnosis of elevated TSH? It seems to me that a newborn with an initial TSH concentration that is just below the cutoff has a much higher chance of being above the reference range at a second measurement — suggesting that we should be more worried about missing newborns that are initially in the middle or lower end of the normal range. Second, what are determinants of the change in TSH concentrations between a first and second measurement? If the results are true (which is unlikely due to the large number of statistical tests and the lack of dose or duration dependency for the medications identified as determinants) then, for example, vancomycin should raise the TSH concentration in all newborns rather than causing only some newborns to be above the reference range. In my opinion, future studies should focus on these types of analyses before investigating TSH elevation per se.