| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Centre for Human Growth and Maturation at the London Centre for Paediatric Endocrinology and Metabolism (P.J.P., S.K.-K., P.C.H.) and Department of Obstetrics and Gynaecology (C.H.R.), University College London, London W1T 3AA, United Kingdom; Department of Obstetrics and Gynaecology (M.P.P.G.), Rotunda Hospital, Dublin 1, Ireland; and Program in Development and Fetal Health (J.C.P.K.), Samuel Lunenfeld Research Institute, Mount Sinai Hospital, University of Toronto, Toronto M5G 1X5, Canada
Address all correspondence and requests for reprints to: Dr. P. C. Hindmarsh, Centre for Human Growth and Maturation, Cobbold Laboratories, Middlesex Hospital, Mortimer Street, London W1T 3AA, United Kingdom. E-mail: p.hindmarsh{at}ucl.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Methods: We related smoking status in 1650 low-risk, singleton Caucasian pregnancies delivering at term to measures of fetal growth, uterine and umbilical artery blood flow, placental appearance, birth size, and cord concentrations of IGF-I and -II and IGF binding protein (IGFBP)-3.
Results: Mothers who smoked in pregnancy were younger (P < 0.001) and shorter (P = 0.03) and from lower socioeconomic groups (P < 0.001). Mean umbilical artery blood flow at 20 wk gestation was not associated with smoking status but was significantly higher in smokers at 30 wk (P = 0.006). Uterine artery blood flow was unaffected. Smoking was associated with an increase in the percentage of abnormal placentas in a dose-dependent manner and with a 3.1-fold increased risk (odds ratio 3.1, 95% confidence interval 1.3–7.6) of abnormal umbilical artery blood flow (P = 0.009). Smoking was associated with a reduction in fetal femur length (P = 0.005) and abdominal circumference as well as birth weight, length, and head circumference but not skinfold thickness. Cord plasma concentrations of IGF-I and IGFBP-3 were lower in the babies of mothers who had smoked (P = 0.02 and P = 0.01, respectively).
Conclusion: We concluded that maternal smoking is associated with an altered placental appearance on ultrasonography, increased umbilical artery blood flow resistance, and a reduction in longitudinal and intraabdominal organ growth. Circulating concentrations of IGF-I and IGFBP-3 along with measures of birth size but not markers of body fat are reduced, suggesting smoking results in a reduction in organ size and function.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Despite the detrimental effect of maternal smoking during pregnancy, little is known of the underlying mechanism(s) constraining growth. Reductions in uterine or umbilical artery blood flow have followed acute or chronic cigarette smoke inhalation (8, 9, 10). Clinical experience suggests that the placentas of mothers who smoke are small and gritty, whereas epidemiological studies indicate that placental weight may be increased (4), decreased (8), or unchanged (5, 11, 12). These discrepancies probably relate to the proportion of heavy smokers in the cohort, a factor known to influence placental angiogenesis (11, 13). Chronic or acute hypoxia and the presence of carbon monoxide in the maternal circulation could result in altered oxygen delivery to the fetus (14). All these factors may contribute to smaller babies born to mothers who smoke during pregnancy.
Size and shape at birth is a composite of several factors. IGF-I and -II are important for fetal growth as evidenced by animal (15) and clinical studies (16, 17, 18, 19, 20). It is likely that the effects of smoking on growth are mediated in part by the IGF system, which represents a final common pathway for growth in fetal life. To address this, we studied the plasma concentrations of the IGF family in cord blood from the pregnancies of smokers and nonsmokers and controlled for confounding factors such as chronic uteroplacental hypoxia-ischemia.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Patients were consecutive mothers booked at University College London Hospitals between April 1996 and July 1997 (5). A total of 1790 mothers fulfilled the entry criteria and 1650 (92%) agreed to participate. They did not differ in terms of clinical or pregnancy outcomes from those who refused. Inclusion criteria were first prenatal visit before 20 wk, ultrasound examination demonstrating a structurally normal single fetus, and a Caucasian mother. Exclusion criteria were increased nuchal translucency or evidence of a major malformation in the ultrasound scan (n = 6) or maternal steroid use for chronic inflammatory or thrombotic disorders (n = 16). Menstrual dates were used to assign gestational age unless the first ultrasound measurement differed by more than 7 d. Current cigarette consumption was categorized as never smoked (group 1) (n = 870), stopped on becoming pregnant (group 2) (n = 117), less than 10 (group 3) (n = 115), 10–20 (group 4) (n = 90), or more than 20 (group 5) (n = 26) cigarettes per day. Pregnancy complications (21) were noted along with mode of delivery.
At 20 and 30 wk gestation, further ultrasound studies were conducted to measure fetal growth using the parameters of head and abdominal circumferences and femur length using an Acuson 128/Xpi (Mountain View, CA) ultrasound machine with a 5-MHz curvilinear transducer. Participation by mothers was high for both the second- (96%) and third-trimester scans (84%), and reasons for not performing the scans were mainly fetal loss or termination of pregnancy. Only 50 mothers refused both scans and their data were excluded from analysis. Each measurement was the mean of three recordings. The coefficient of variation (CV) of the ultrasound measures was 1% on the 20-wk scan. At the 30-wk scan Doppler ultrasonography of uterine and umbilical artery blood flow was conducted and expressed as the pulsatility index (22). The ultrasonographers were unaware of the smoking status of the women. All blood flow measurements represent the mean of three measurements. Placenta were graded according to Grannum (23) as: grade 0, normal; grade 1, random echogenic areas present; grade 2, basal echogenic areas and indentations in the chorionic plate; and grade 3, echo-poor areas, irregular echogenic areas, and deep indentations in the chorionic plate.
Complete perinatal outcome and neonatal anthropometric data were available in 1484 babies including Apgar scores at 1 and 5 min. Birth weight was measured using electronic self-calibrating scales (Seca, Birmingham, UK), length by Infantometer (Child Growth Foundation, London, UK), and head and midarm circumferences with a metal tape (Holtain, Crymych, UK). Triceps, subscapular and quadriceps skinfolds were measured using skinfold calipers (Holtain, Crymych, UK). Three separate measurements were taken and the mean recorded. The CV of the measurement error for length was 0.15% based on 10 infants, each measured five times by three observers.
After birth and before completion of the third stage of labor, blood was collected into EDTA from the umbilical cord vein. A sample was used to measure cord blood pH and the remainder centrifuged, separated, and plasma stored immediately at –20 C. Assays for IGF-I, IGF-II, and IGF binding protein (IGFBP)-3 were performed in batches within 3 months of collection.
The study was approved by the Research Ethics Committee of University College London Hospitals, and written informed consent for participation was obtained from the mother for herself at the commencement of the study and for her newborn child after delivery.
Hormone assays
IGF-I. IGF-I was measured by a commercial immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). This is a nonextraction method in which the IGFBPs are separated from the IGF-I by acidification of the sample and excess IGF-II is added to block the binding proteins from recombining with the IGF-I. The within-assay CVs were 4.6 and 3.3% at 61.0 and 292.5 ng/ml, respectively. The between-assay CVs were 15.5 and 11.3% at 88.6 and 240.4 ng/ml. The standards were prepared from recombinant IGF-I and were calibrated against World Health Organization First International Reference Preparation (87/518). The minimum detection limit of the assay was 6 ng/ml. Recovery of recombinant IGF-I (50–100 ng/ml) added to cord serum before acidification was estimated to be 96–98%.
IGF-II
IGF-II was measured by a commercial coated-tube immunoradiometric assay (Diagnostic Science Laboratories, Webster, TX). This is a nonextraction method in which the IGFBPs are separated from IGF-II by dilution and acidification of the sample. The within-assay CVs were 6.5, 3.4, and 4.7% at 245, 409, and 1432 ng/ml, respectively. The between-assay CVs were 14.5 and 7.2% at 273 and 785 ng/ml, respectively. The standards were calibrated against a preparation of recombinant IGF-II. The minimum detection limit of the assay was 12 ng/ml. Studies of IGF-II recovery yielded values ranging between 95 and 98%.
IGFBP-3
IGFBP-3 was measured by a commercial coated-tube immunoradiometric assay (Diagnostic Science Laboratories). Samples were analyzed at 1:100 dilution. The within-assay CVs were 3.9, 3.2, and 1.8% at 7.35, 27.53, and 82.72 µg/liter, respectively. The between-assay CVs were 7.6 and 4.2% at 5.43 and 27.15 µg/liter, respectively. The standards were calibrated against recombinant nonglycosylated IGFBP-3. The minimum detection limit of the assay was 0.5 µg/liter. Estimates of IGFBP-3 using this methodology will include the major proteolytic fragments of IGFBP-3.
Statistics
All data were explored for normality of distribution and log transformed where appropriate. Values for birth weight, length, and head circumference were expressed as SD scores (SDS) using the 1990 British growth reference (24). Relationships between hormonal parameters and anthropometric measures and pregnancy parameters were estimated using Pearson’s correlation coefficient. In investigating the effects of smoking on the birth weight SDS-placental weight relationship analysis of covariance was used to compare differences in slope intercepts after testing for differences in regression slopes (25). Multiple linear regression analysis was used to explore effects of cigarette smoking, cord pH, and Apgar score on cord plasma IGF-I, IGF-II, and IGFBP-3 concentrations. One-way ANOVA with the Student-Newman-Keuls post hoc test was used to determine differences between mean values of the groups. Student’s t test was used to determine differences in cord plasma IGF-I, IGF-II, and IGFBP-3 concentrations between subjects with low (<7) and high (>7) Apgar scores. 
2 Analysis was used to determine the distribution of factors such as parity, socioeconomic group, placental Grannum grade, abnormality in umbilical artery blood flow, and Apgar score with respect to smoking habit. Risk estimates were expressed as odds ratios with 95% confidence intervals.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Of the 1650 mothers, 1484 delivered a live infant. The 166 who did not deliver consisted of 127 who did not complete the study because they had moved away or were lost to follow-up, 28 who had a miscarriage or termination of pregnancy, and 11 who withdrew from the study. These mothers did not differ in terms of first-attendance characteristics from those who completed the study. Analysis was confined to infants delivered at term (gestation > 37 wk) (n = 1215) in pregnancies uncomplicated by gestational diabetes and chronic uteroplacental hypoxia-ischemia-antepartum hemorrhage, pregnancy-induced hypertension, and preeclampsia.
Mothers who smoked heavily in pregnancy were younger (one-way ANOVA F = 27.4; P < 0.001) and shorter (F = 2.64; P = 0.03) and from socioeconomic groups (SEGP) 4 and 5 (SEGP 1, 96% nonsmokers vs. 56% in SEGP 5 2 157.5; P < 0.001). Body mass index was not significantly different between smokers and nonsmokers. The maternal age and SEGP effect were interrelated because mothers from SEGPs 4 and 5 were significantly younger than SEGPs 1, 2, and 3 (one-way ANOVA F = 70.1; P < 0.001: Student-Newman-Keuls SEGPs 1, 2, and 3 vs. SEGPs 4 and 5 P < 0.05). Parity did not influence smoking habit. There were no significant differences between the groups in maternal prepregnancy weight, gestation at first antenatal attendance, weight gain during pregnancy (data not shown), gestational age at delivery, blood pressure at first antenatal attendance, or maternal hemoglobin concentration (Table 1
).
|
Uterine. There was no association between smoking status and mean uterine artery Doppler pulsatility index at the 20 or 30 wk assessment.
Umbilical. Mean umbilical artery Doppler pulsatility index was not different among the groups at 20 wk but at 30 wk was significantly higher in the smokers [group 3, 1.02 (SD 0.16); group 4, 1.02 (SD 0.18); group 5, 1.02 (SD 0.15), compared with the nonsmokers (0.97, SD 0.17) or those who had stopped smoking (1.00, SD 0.16) (one-way ANOVA, F = 3.65; P = 0.006; Student-Newman-Keuls post hoc test, P < 0.05 for smokers vs. the rest)].
Placental ultrasound
Figure 1 shows a dose-dependent effect of smoking on placental Grannum grade. None had Grannum 3 placentas (2 310; P < 0.001). The presence of a placenta with Grannum grade 1 or 2 was associated with 3.1-fold increased risk (odds ratio 3.1 95% confidence interval 1.3–7.6) of an abnormal high umbilical artery pulsatility index (2 6.90; P = 0.009).
|
In the nonsmoking group, placental weight was related to birth weight SDS (r = 0.56; P < 0.001) and described by the equation: birth weight SDS = 0.004 (placental weight in grams) – 2.58 (Fig. 2).
|
The intercepts for these two regression equations was significantly different (0.87 SDS or approximately 400 g) (P = 0.001), whereas there was no significant difference between the regression equation slopes.
Fetal growth and birth size
There was no difference in fetal biometric measurements (head and abdominal circumferences and femur length) at 20 wk gestation between those who smoked and those who did not. At 30 wk of gestation there was a significant reduction in femur length (one-way ANOVA, F = 3.73; P = 0.005; nonsmokers significantly greater than other groups, Student-Newman-Keuls, P < 0.05) and to a lesser extent abdominal circumference (one-way ANOVA, F = 3.08; P = 0.015; nonsmokers significantly greater than other groups, Student-Newman-Keuls, P < 0.05). Head circumference was not significantly different as was biparietal diameter (Table 2).
|
). There was no difference in the frequency of babies with a birth weight SDS lower than the third centile between those who smoked (4.8%) and those who did not (2.4%) (2 6.37; P = 0.17). However, there was an increase in the frequency of babies with a birth weight SDS less than the 10th centile between those who smoked (19%) and those who did not (7%) (2 33.2; P < 0.001). No significant effects could be observed on skinfold thickness and midarm circumference. Males were affected to the same extent as females.
|
Cord plasma IGF-I and IGFBP-3 concentrations were significantly lower in babies that had been exposed to cigarette smoking than in those whose mothers did not or had ceased smoking. Cord plasma concentration of IGF-II was unaffected by smoking status (Fig. 3).
|
2 0.81; P = 0.37). At 5 min the Apgar score was not related to any of the cord hormone concentrations. Cord pH was related to cord plasma IGF-I concentration (r = 0.13; P < 0.001) but not cord plasma concentrations of IGF-II and IGFBP-3. Subjects with low Apgar scores at 1 min had lower cord pH (r = 0.29; P = 0.001). In multiple regression analysis, cord pH and number of cigarettes smoked were important determinants of cord plasma IGF-I concentration but explained only 3% of the variance.
When analysis was confined to patients in whom causes of chronic uteroplacental hypoxia-ischemia were excluded along with pregnancies with Grannum stage 2 placental appearance, abnormal umbilical artery Doppler studies at 30 wk of gestation, and Apgar score less than 7 at 1 min, the relationship of cord plasma IGF-I (one-way ANOVA, F = 3.56; P = 0.007) and IGFBP-3 (one-way ANOVA, F = 4.07; P = 0.003) (n = 852) concentrations with smoking was maintained.
Cord plasma IGF-I concentration was strongly related to all measures of birth size. In multiple linear regression analysis, IGF-I, cigarette smoking, placental weight, maternal height and weight at first presentation, and parity were the main significant determinants of birth size, explaining 52.3% of the variance of birth weight, 20.2% of birth length, and 20.8% of head circumference (Table 4).
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Maternal smoking was most common among socioeconomic groups 4 and 5 who tended to be the youngest mothers, an observation noted in part by others (4). Mothers who smoked were shorter, but there was no difference in maternal weight at presentation between the smoking groups that might suggest explanations for differences in birth size. This would imply, that in this cohort, impaired maternal nutrition at this stage of pregnancy was not a major issue. We have also inferred that both smokers and nonsmokers had similar calorie intakes during pregnancy because maternal weight gains were similar. Furthermore, late gestational studies of food intake suggest that the fetal growth restriction observed in smokers is not related to decreased maternal food intake (26). This does not imply, however, that impaired nutrition in early/midgestation is not an important factor in determining fetal growth in other populations. Maternal hemoglobin concentrations at this early stage of pregnancy were similar among the groups, suggesting that there was little effect on oxygen-carrying capacity, although a change through pregnancy cannot be excluded. An increase in maternal hemoglobin concentration in smokers has been noted in some pregnancies, which might imply a longer-term compromise of oxygen delivery (27).
There was no significant difference in uterine Doppler artery pulsatility index between smokers and nonsmokers, suggesting that altered uterine artery blood flow is an unlikely explanation for the poor intrauterine growth observed. These findings confirm acute and chronic studies of cigarette exposure in pregnant women (8, 9, 10).
Smoking was associated with an increase in the proportion of placentas with abnormal echogenicity, and this effect was dose dependent. Other causes for abnormal placental morphology such as preeclampsia had been excluded. Placental weight was not different in smokers, compared with nonsmokers, despite the fact that birth weight was lower. The relationship between birth weight and placental weight was maintained across the range of cigarettes smoked with an overall reduction of 400 g in those smoking 20 cigarettes per day. This preservation of placental weight probably reflects an increase in placental angiogenesis (13), compensating or attempting to compensate for the compromised areas observed on ultrasound. The degree of angiogenesis and cigarette consumption are presumably the explanations for the variation in placental epidemiology findings in smokers in whom placental weights may be increased (4) or unchanged (5, 11, 12). A reduction in placental weight is probably observed in very heavy smokers (8) in whom marked structural damage to the villous stroma and basement membrane of the fetal capillaries takes place (28). The presence of an ultrasonographically abnormal placenta was associated with an increased chance of detecting an increase in umbilical artery resistance. Although indirect, this suggests that either changes in placental morphology lead to a reduction in umbilical artery blood flow or the effect of smoking is manifest in both structures.
The net effect of smoking on the fetus, whether mediated through altered umbilical artery blood flow, placental function, or a direct toxic effect, was a symmetrical reduction in birth weight, length, and head circumference. Of note, in this cohort, smoking did not increase the incidence of babies that had a birth weight SDS less than the third centile but did increase the numbers with a birth weight SDS less than the 10th centile, suggesting an overall shift of the birth weight distribution in the smoking group. These observations are of particular interest, given the association of small size with increased risk of cardiovascular disease in adult life (29) and suggest that in such analyses maternal smoking status needs to be included. This is important because unless the effect of smoking on organ size is accounted for in the determination of later risk, that risk may be incorrectly ascribed to the effect of birth size per se. The reduction in birth weight did not appear to result from alterations in body fat, as evidenced by the skinfold thickness measures, confirming previous studies (30, 31). Caution needs to be exercised in this interpretation because more precise methods for estimation of body fat were not performed (32). That aside, the data might suggest that the effects of smoking on birth weight reflect changes in organ size, e.g. reduced head circumference and, by inference, brain size as well as other components of body size, e.g. bone as reflected in body length.
There was no difference in fetal anthropometric measures at 20 wk of gestation, which may reflect the sample size required at this stage of gestation to detect an effect. By 30 wk clear differences in femur length and abdominal circumference could be detected but little effect on head circumference. The greater effect on femur length, compared with abdominal circumference, might be expected, given the different growth patterns of length and liver (which must constitute the major component of abdominal circumference). Peak length velocity is attained between 26 and 28 wk of gestation, whereas peak liver growth is later in the third trimester (33). The situation with head circumference is not so easy to explain because our data would suggest that compromised growth takes place during the third trimester, whereas the greatest increase in head circumference takes place in the first trimester. A possible explanation could derive from the fact that nicotinic receptors are present in some brain regions by midgestation (34), and animal studies suggest that this is a critical time period when nicotine exposure reduces brain volume (35).
The IGFs are important determinants of fetal growth (16, 17, 18, 19, 20). In this study the cord plasma IGF-I and IGFBP-3 concentrations were lower in the babies of mothers who had smoked, whereas the cord plasma IGF-2 concentrations were unaffected. This may reflect the fact that IGF-I and insulin are more important for late fetal growth than IGF-II or that there is a compensatory increase in the latter (16). IGF-I appears to be influenced by acute hypoxia-ischemia because it was related to Apgar score at 1 min and cord pH in a manner noted in animal studies (36). However, when analysis was confined to those pregnancies in which acute and chronic uteroplacental hypoxia-ischemia was excluded, the association of smoking with lower cord plasma IGF-I concentration remained. The cord plasma IGF-I concentration appeared to be an independent predictor of birth size, suggesting that an additional explanation for the reduced birth size of infants of mothers who smoke in pregnancy may be a direct action of smoking on IGF generation in the fetal liver.
In conclusion, we have demonstrated in a longitudinal study that smoking in pregnancy is associated with an increase in placental echogenicity, reduced umbilical artery blood, fetal production of IGF-I, and birth size without alterations in uterine artery blood flow and maternal nutritional status. The sequence of associations suggest that smoking induces microvascular changes in the placenta that are associated with an increase in vascular resistance in the umbilical artery. Fetal growth of structures such as the skeleton and liver are compromised during periods of rapid growth, and the alteration in liver size or a direct effect of cigarette smoking on liver IGF-I generation is associated with a reduction in organ growth and in particular a reduction in birth weight. How smoking influences head growth is unclear. Although we describe a sequence of associations, it is possible that the effects of cigarette smoking represent a direct toxic effect with a variable dosing effect on different structures. It would be unwise, however, to overstate the case for the latter because a lack of dose response in some measures (umbilical artery blood flow), compared with others (birth length), may be more of a reflection of the precision of these measures. It also needs to be considered that because circulating IGF-I concentrations are related to IGFBP-3 concentration, the effects of smoking could be due to primary effects on IGFBP-3. Furthermore, because the liver is the primary organ source for circulating IGF-I and IGFBP-3, it might be that a smaller liver leads to lower IGF-I/IGFBP-3 production. Finally, because in many epidemiological studies in which randomization is not possible, the possibility that smoking is a confounder and that the causal agent is not realized remains. What is clear is that smoking cessation before or at the commencement of pregnancy can lead to a normal growth outcome for the baby.
|
Footnotes |
|---|
The study was designed by P. C. Hindmarsh, J. C. P. Kingdom, and C. H. Rodeck. Laboratory analysis was undertaken by P. J. Pringle and M. P. P. Geary, and all authors contributed to analysis of the data and writing of the manuscript. All authors had access to all data in the study, and the responsibility for the decision to submit for publication was a joint one.
There are no conflicts of interest.
First Published Online February 15, 2005
Abbreviations: CV, Coefficient of variation; IGFBP, IGF binding protein; SDS, SD score; SEGP, socioeconomic group.
Received August 20, 2004.
Accepted February 3, 2005.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Chiesa, J. F. Osborn, C. Haass, F. Natale, M. Spinelli, E. Scapillati, A. Spinelli, and L. Pacifico Ghrelin, Leptin, IGF-1, IGFBP-3, and Insulin Concentrations at Birth: Is There a Relationship with Fetal Growth and Neonatal Anthropometry? Clin. Chem., March 1, 2008; 54(3): 550 - 558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. Geerts, D. E. Grobbee, C. K. van der Ent, B. M. de Jong, M. M. van der Zalm, N. van Putte-Katier, J. L.L. Kimpen, and C. S.P.M. Uiterwaal Tobacco Smoke Exposure of Pregnant Mothers and Blood Pressure in Their Newborns: Results from the Wheezing Illnesses Study Leidsche Rijn Birth Cohort Hypertension, September 1, 2007; 50(3): 572 - 578. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Saenger, P. Czernichow, I. Hughes, and E. O. Reiter Small for Gestational Age: Short Stature and Beyond Endocr. Rev., April 1, 2007; 28(2): 219 - 251. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
