Effects of levothyroxine therapy on pregnancy outcomes in women with subclinical hypothyroidism

Gebruikersavatar
laura
Berichten: 3600
Lid geworden op: 11 sep 2013, 22:42
Contacteer:

Effects of levothyroxine therapy on pregnancy outcomes in women with subclinical hypothyroidism

Bericht door laura »

Effects of levothyroxine therapy on pregnancy outcomes in women with subclinical hypothyroidism. Thyroid. May 16, 2016 [Epub ahead of print]
Maraka S, Singh Ospina NM, O'Keeffe DT, Rodriguez-Gutierrez R, Espinosa De Ycaza AE, Wi C, Juhn YJ, Coddington C, Montori V, Stan MN.

Tim I.M. Korevaar

SUMMARY

Background
There is a large body of evidence that suggests that subclinical hypothyroidism during pregnancy is associated with a higher risk of adverse pregnancy outcomes. In current guidelines, subclinical hypothyroidism is defined according to population-based reference ranges for TSH and FT4. In the absence of such reference ranges, an upper cut-off value of 2.5 mU/L for TSH in the first trimester and 3.0 mU/L in the second and third trimesters is recommended. The Endocrine Society guidelines advocate treatment of subclinical hypothyroidism while the ATA guidelines advocate treatment only for women who are also TPOAb-positive. However, data on the effects of treatment of subclinical hypothyroidism during pregnancy are sparse.

Methods
Data on women with subclinical hypothyroidism, identified using a case-finding approach at the Mayo Clinic (Rochester, MN) between January 2011 and December 2013, were retrospectively collected using the laboratory data and diagnostic indexes of the hospital. These women were subsequently divided into a group that received treatment and a group that did not (dosages of levothyroxine were not reported; the treatment aim was <2.5 mU/L in the first trimester and <3.0 mU/L in the second). The records of all women identified using a search query for TSH >2.5 mU/L in this system, recorded within 100 days before or after an ICD-9 code for normal pregnancy (V22.nn), supervision of high-risk pregnancy (v23.nn) at a first clinical visit, a missed abortion (632.nn), or spontaneous abortion (634.nn) were reviewed. Subclinical hypothyroidism was defined as a serum TSH >2.5 mU/L in the first trimester or >3 mU/L for the second and third trimesters, but ≤10 mU/L. TPOAb and FT4 levels were not routinely measured. Women with twin pregnancies or use of thyroid-interfering medication, but not women with preexisting thyroid disease, were excluded. The primary outcome was pregnancy loss (no prior power calculations were performed); secondary outcomes included preterm delivery, placental abruption, gestational diabetes, gestational hypertension, preeclampsia, eclampsia, premature rupture of membranes, intrauterine growth restriction, low birth weight (<2500 g), low 5-min Apgar score (≤7), NICU admission, neonatal death, and duration of hospital stay. Multivariable logistic-regression models were used to study the association of L-T4 treatment with adverse outcomes. Analyses were mostly, but not consistently, adjusted for TSH, BMI, education level, and history of pregnancy loss or preterm delivery.

Results
The final study population comprised 366 women with subclinical hypothyroidism of which 82 received L-T4 (22.4%). Treated women had a higher BMI (the weight in kilograms divided by the square of the height in meters; 29 vs. 27), a higher mean serum TSH (4.9 vs. 3.5 mU/L; median levels not given), and a higher likelihood of having pregestational thyroid disease (21% vs. 7%) and were more likely to be TPOAb-positive (46% vs. 29%; TPOAb measurements were available for only 24% of the study population). L-T4 therapy was started at a median gestational age of 9.1 weeks (interquartile range [IQR], 7.7–11.5). Women who received treatment had a mean (±SD) of 2.5±1.6 additional TSH measurements during pregnancy, and the trimester-specific TSH goal was met in 69.4% (range, 50–72), partly met in 16.6% (range, 12–72), and not met in 13.9% (range, 10–72). One patient required a L-T4 dose reduction.

As compared with no treatment, treatment was associated with a 59% lower risk of pregnancy loss (P = 0.12), a 67% lower risk of preterm delivery (P = 0.06), and a 70% lower risk of gestational diabetes (P = 0.07). Children from treated women were less likely to have an Apgar score below 8 (0% vs. 7.0%; P<0.001). Although treatment was associated with a 94% lower risk of having a birth weight <2500 g (1.3% vs. 10.0%, P<0.001), there was no difference in intrauterine growth restriction between the groups (1.2% vs. 1.8%, P =0.99). There was no association of treatment with any of the other outcomes, including gestational hypertension, preeclampsia, and NICU admission.
Conclusions

In this study, the authors show favorable effects of treatment with L-T4 in women who were found to have a TSH >2.5 mU/L in the first trimester, or >3.0 mU/L in the second trimester using a high-risk-case–finding approach.

ANALYSIS AND COMMENTARY
Data on the harms and benefits of L-T4 treatment in subclinical thyroid disease during pregnancy are sparse. However, before we can perform good randomized controlled trials and subsequently make proper evidence-based decisions based on the effects of treatment of subclinical hypothyroidism, we need to define when gestational thyroid-function measurements are actually abnormal. The best first step in defining what is abnormal is to select women on the basis of the Gaussian distribution, in the outskirts of which nonbiologic variation is likely (outside ±1.96 SD from the mean). Based on this principle, all international guidelines advocate the use of population-based reference ranges. Unfortunately, such reference ranges cannot be calculated in every hospital; therefore, international guidelines advocate the use of fixed upper limits for TSH of 2.5 mU/L in the first trimester and 3.0 mU/L in the second.

However, these cutoffs have been shown to be much too low, leading to overdiagnosis in many cases (1–3). As a consequence, in the soon-to-be-released 2016 ATA guidelines, the upper limit will be increased to 4.0 mU/L (ATA satellite meeting at ENDO 2016). A recent U.S. trial in which approximately 100,000 women were screened for thyroid dysfunction, initially used the proposed fixed upper limits for TSH for the screening. However, this led to inclusion of roughly 10% of all screened women. Subsequently, they changed to population-based reference ranges, which showed an upper limit just below 4.0 mU/L (Brian Casey, ITC meeting Orlando). Unfortunately, the use of fixed upper limits for TSH in the current study, and also the methodologic issues (bias, retrospective design, residual confounding) make it quite hard to generalize or interpret its results. The majority of the methodological issues are easily spotted and also, in part, mentioned by the authors themselves. However, two other approaches of interpreting the findings of this study may make it more valuable.

When it comes to treatment, first do no harm. High gestational thyroid function has been associated with a higher risk of gestational hypertension, preeclampsia and impaired neurocognition of the offspring, indicating that overtreatment with L-T4 may be possible during pregnancy (4–7). With the knowledge we now have on reference ranges, the women treated in this study had mild subclinical hypothyroidism at best. The combination of a functioning thyroid, hCG stimulation, and treatment makes these patients a high-risk group for overtreatment, and thus a good model of assessing potential harms. With this in mind, the fact that only one patient required a dose reduction (TSH of 0.09 mU/L) and the risks of gestational hypertension and preeclampsia were not increased in the treatment group may suggest that the risks of overtreatment, in this group of women, is small. It must be noted, however, that the actual dosage is not reported in the paper—so we can only state that a L-T4 dosage necessary to decrease the TSH from anywhere between 2.5 and 10 mU/L (95% were below 7.64 based on the data in the paper) to less than 2.5 mU/L (or 3.0 mU/L) carries a very small risk of overtreatment in a clinical setting.

Bearing in mind that the majority of patients worldwide received treatment according to the 2.5 and 3.0 mU/L cutoffs over the past 5 years, these are very reassuring data. On the other hand, it is also quite worrisome that the largest study that provides this type of data comprises only 82 treated women. Why are there not more papers showing this type of treatment data in the literature? An early pregnancy TSH of >2.5 mU/L is common. Are thyroidologists not monitoring the effects of their treatment or are the outcomes often negative and therefore not published?

Setting aside all methodologic issues, consider that in the majority of cases the directions of effects of observational studies are in line with randomized controlled trial results. What can explain the favorable trends presented in the current study from a physiologic perspective? I have always been intrigued by the association of TPOAb positivity with pregnancy loss and premature birth. In papers from our group , we adjusted the association of TPOAb positivity with premature delivery or ADHD for TSH and FT4 levels but found no change. It is hypothesized that TPOAb positivity is associated with these adverse outcomes either through negatively affecting thyroid function or via a more general autoimmune effect that affects both the thyroid and other tissues. However, in the study by Maraka et al., TPOAb were measured in only 24% of all patients; 46% of all treated women were TPOAb-positive. I look forward to the outcomes of two randomized controlled trials investigating the effect of treating euthyroid TPOAb-positive women (the TABLET trial [ISRCTN, 15948785] and the T4LIFE trial [xxxxx, NTR3364 ]).

References
1. TI Korevaar, M Medici, RP Peeters. Subclinical hypothyroidism overdiagnosis in pregnant women. JAMA Intern Med 2015;175:1872-3.
2. S Maraka, DT O'Keeffe, VM Montori. Subclinical hypothyroidism during pregnancy—should you expect this when you are expecting? A teachable moment. JAMA Intern Med 2015;175:1088-9.
3. M Medici, TI Korevaar, WE Visser, TJ Visser, RP Peeters. Thyroid function in pregnancy: what is normal? Clin Chem 2015;61:704-13. Epub March 31, 2015.
4. TI Korevaar, R Muetzel, M Medici, L Chaker, VW Jaddoe, YB de Rijke, EA Steegers, TJ Visser, T White, H Tiemeier, RP Peeters. Association of maternal thyroid function during early pregnancy with offspring IQ and brain morphology in childhood: a population-based prospective cohort study. Lancet Diabetes Endocrinol 2016;4:35-43. Epub October 20, 2015.
5. T Männisto, P Mendola, J Grewal, Y Xie, Z Chen, SK Laughon. Thyroid diseases and adverse pregnancy outcomes in a contemporary US cohort. J Clin Endocrinol Metab 2013;98:2725-33. Epub June 6, 2013.
6. M Medici, TI Korevaar, S Schalekamp-Timmermans, R Gaillard, YB de Rijke, WE Visser, W Visser, SM de Muinck Keizer-Schrama, A Hofman, H Hooijkaas, et al. Maternal early-pregnancy thyroid function is associated with subsequent hypertensive disorders of pregnancy: the generation R study. J Clin Endocrinol Metab 2014;99:E2591-8.
7. LK Millar, DA Wing, AS Leung, PP Koonings, MN Montoro, JH Mestman. Low birth weight and preeclampsia in pregnancies complicated by hyperthyroidism. Obstet Gynecol 1994;84:946-9.
laura

Kijk voor meer informatie ook eens op Schildkliertje.

Raadpleeg altijd een arts als je twijfelt over je gezondheid.
Het Schildklierforum kan niet worden beschouwd als vervanging van een consult of een behandeling.
Plaats reactie