Skip to main content
NCBI home page
Toxicological Sciences logoLink to Toxicological Sciences
. 2017 Jan 23;156(2):520–526. doi: 10.1093/toxsci/kfx013

Editor’s Highlight: In Utero Exposure to Gadolinium and Adverse Neonatal Outcomes in Premature Infants

Radhika Amin *,†, Thomas Darrah ‡, Hongyue Wang §, Sanjiv Amin ¶,1
PMCID: PMC6074875  PMID: 28201627

Abstract

Gadolinium is a toxic rare earth element that is used as a contrast enhancement agent for diagnostic medical imaging. However, because of safety concerns to the developing fetus derived from preclinical studies, gadolinium can only be used during pregnancy if the potential benefits justify the potential risks to a fetus. Because there are no previous well designed safety studies on the developing fetus, we aimed to evaluate the potential adverse effects of in utero gadolinium exposure in high-risk premature infants. We performed a prospective dose (cord blood gadolinium concentration) – response (outcomes) study involving 104, 24–33 weeks gestational age (GA) infants. The mean (range) cord blood gadolinium concentration of infants measured using Inductively Coupled Plasma Mass Spectrometry was 191 (3.4–3729.6) pg/ml. The association between cord blood gadolinium concentration and each neonatal outcome was evaluated using linear or logistic regression analysis. The GA, race, gender, and antenatal steroid exposure were considered priori confounders. Recent adult human studies have shown that gadolinium exposure may be associated with nephrotoxicity. However, we found no adverse effects on renal function or other common outcomes including degree of prematurity, small for GA, respiratory distress syndrome, hyperbilirubinemia, intraventricular hemorrhage, necrotizing enterocolitis, patent ductus arteriosus, chronic lung disease, retinopathy of prematurity, and osteopenia of prematurity during the neonatal period with an increase in cord blood gadolinium concentration. None of the infants had clinically evident congenital malformations. In conclusion, gadolinium use during pregnancy is unlikely to be associated with adverse effects in infants during the neonatal period.

Keywords: dose–response, renal function, prematurity, developmental toxicology, fetus.

Gadolinium is a paramagnetic rare earth element (lanthanide series) that is increasingly used as a contrast enhancement agent for clinical and diagnostic medical imaging, including during pregnancy and in neonates (Elster, 1990; Greenberg et al., 2004; Laswad et al., 2009; Palacios Jaraquemada and Bruno, 2000; Rangamani et al., 2012; Webb et al., 2005). However, gadolinium is toxic (Broome et al., 2007; Runge, 1988) and thus there are safety concerns for the developing fetus with in utero exposure to gadolinium, as gadolinium crosses the placenta and may accumulate in the developing fetus (Novak et al., 1993). Gadolinium is also believed to diffuse across the placenta from the fetus back to the mother for excretion by the mother in urine (Garcia-Bournissen et al., 2006).

Although gadolinium used with magnetic resonance imaging (MRI) has a short half-life of ∼1.5 h and is rapidly excreted in urine, gadolinium may dissociate from the chelating agents via transmetallation reactions, which introduce toxic ionic gadolinium (Gd [III]) in vivo that may then bioaccumulate within cells and tissues (Darrah et al., 2009; Rogosnitzky and Branch, 2016). Furthermore, the gadolinium incorporated into bodily stores such as bones from prior exposures may be released back into the bloodstream circulation as observed with other metals during bone resorption and remodeling processes that are active during pregnancy and early human bone development (Gulson and Calder, 1995; Gulson et al., 1997). Thus, the potential for fetal exposure to ionic Gd (III) exists with the administration of gadolinium based contrast agents (GBCA) or during the endogenous release of ionic Gd (III) from bodily stores such as the kidneys, adipocytes, or bones during pregnancy. Therefore, cord blood gadolinium concentration is primarily reflective of endogenous release rates of gadolinium from maternal bodily stores and/or maternal exposure to GBCA within a few days prior to delivery. Free (non-bound) gadolinium is extremely toxic (Runge, 1988), disrupts cellular processes (Biagi and Enyeart, 1990), inhibits stretch-activated ion channels (Yang and Sachs, 1989), and is one of the most efficient known calcium and iron antagonists (Molgo et al., 1991).

The preclinical studies in animals have shown inconsistent toxic effects to the developing fetus from in utero exposure to gadolinium. Some animal studies have shown that in utero gadolinium exposure may be associated with intrauterine growth restrictions and congenital anomalies (Sundgren and Leander, 2011). However, other animal studies have failed to demonstrate harmful effects on the developing fetus (Morisetti et al., 1999; Rofsky et al., 1994, 1995; Soltys, 1992). The few published human neonatal case series reported no adverse neonatal outcomes from in utero exposure to gadolinium (De Santis et al., 2007; Marcos et al., 1997; Shoenut et al., 1993). However, these case series studies were small, involved predominantly term neonates, and neonatal outcomes were not comprehensively evaluated. Specific concerns for the developing fetus relate to the recent reports linking gadolinium exposure to nephrogenic systemic fibrosis and acute renal failure in human adults, and potential neurotoxic effects due to chronic retention in the brain (Abraham and Thakral, 2008; Broome, 2008; Grobner, 2006; Marckmann et al., 2007; Roberts et al., 2016; Rogosnitzky and Branch, 2016; Weigle and Broome, 2008).

The fetus, due to physiologically immature organs, may be more sensitive to the developmental toxicity from exposure to gadolinium., Due to safety concerns derived from preclinical studies, gadolinium is currently classified as a category C drug by the US Food and Drug Administration (FDA) and can only be used during pregnancy if the potential benefits justify the potential risks to the fetus or mother. The available FDA-approved GBCAs differ in structure and stability. The two main categories of GBCAs are linear and macrocyclic (Rogosnitzky and Branch, 2016). The macrocyclic GBCA is more stable and releases far less gadolinium than linear GBCA (Rogosnitzky and Branch, 2016). The tissue deposition of gadolinium is far less after administration of macrocyclic agents than after linear gadolinium agents in animals (Lancelot, 2016). The safety of in utero gadolinium exposure to a fetus has not been previously evaluated in well-designed studies (Sundgren and Leander, 2011). Premature infants with immature organs provide a sensitive population for evaluating developmental toxicity due to in utero exposure to gadolinium. Therefore, our objective was to evaluate the harmful health effects of in utero gadolinium exposure as a function of cord blood gadolinium concentration in high-risk premature infants. The significance of a prospective evaluation of a gadolinium concentration – neonatal outcome relationship in high-risk premature infants is that it will provide much needed, evidence-based human safety data relevant to the developing fetus.

MATERIALS AND METHODS

Study Design

A prospective nested dose–response observational exploratory study was performed to evaluate neonatal outcomes from in utero exposure to gadolinium in a cohort of premature infants. In this dose–response study, the cord blood gadolinium concentration was the dosage exposure, and common neonatal outcomes were the response evaluations. The study was approved by the Institutional Research Review Board. Parental consent was obtained for each subject enrolled in the study.

Study Population

All premature infants 24–33 weeks gestational age (GA), born at the University of Rochester Medical Center (URMC), and admitted to the Neonatal Intensive Care Unit (NICU) were eligible for this study if cord blood was collected at the time of delivery. Infants with chromosomal disorders were excluded as these conditions are usually associated with abnormal outcomes including death. Only premature infants ≤ 33 weeks GA were included as they are admitted to the NICU and are carefully monitored. Moreover, preterm infants have a higher probability of adverse outcomes than late preterm and term infants, which improves the statistical power of the study and does not require a large sample size.

Measurement of Gadolinium – Exposure Variable

The institutional policy is to collect cord blood of all premature infants delivered at URMC for evaluation of the infants’ blood groups. The cord blood collected is routinely stored at 4 °C in a refrigerator by the institutional blood bank for 2 weeks. Soon after obtaining the consent for the study, the cord blood of each subject was transferred from the blood bank to the research laboratory. The cord blood samples were then transferred overnight in batches on dry ice to the reference analytic laboratory for measurements of gadolinium concentration.

Cord blood levels of gadolinium were measured in whole blood and reported as picograms per gram (pg/g; 1ng/g = 1000pg/g = 1025pg/ml) using a Perkin Elmer Dynamic Reaction Cell axial field Inductively Coupled Plasma Mass Spectrometry (ICP-MS) with modifications to the Environmental Protection Agency (EPA) 6020A methodologies (Sprauten et al., 2012; USEPA, 1991, 1994).

All cord blood samples were prepared for gadolinium concentration analysis by ICP-MS in a Class 100 trace metal-free clean room by standard methods (Darrah et al., 2009). Samples were spiked with internal standards consisting of known quantities of indium, bismuth, and iridium (obtained from SCP Science, Quebec, Canada), to correct for instrumental drift, diluted using water purified to 18.2 MΩ/cm resistance (by a Milli-Q water purification system, Millipore, Bedford, MA, USA), and acidified using ultra-pure 12.4 mol/l hydrochloric acid. All standards, including aliquots of the certified National Institute for Standards and Technology (NIST) 955c and procedural blanks were prepared by the same process.

Calibration standards used to determine metal concentrations in blood included aliquots of 18.2 MΩ cm resistance H2O, NIST 955c standard reference materials (SRM), and NIST 955c SRM spiked with known quantities of each metal in a linear range from 0.001 to 10 ng/g (1–10 000 pg/g). Limits of detection, limits of quantification, and method detection limits (MDL) were calculated according to the two-step approach using the t99SLLMV method at 99% CI (t = 3.71) (USEPA, 1993). Because MDL values incorporate matrix effects and interferences associated with the complex placental tissue matrix, they provide a more robust and accurate limit of detection for these analyses. All reported values were above the calculated method detection limits (0.0021 ng/g= 2.1pg/g).

Maternal and Neonatal Clinical Characteristics

Data was collected on (1) maternal demographics, (2) maternal medical conditions during pregnancy such as diabetes, thyroid disorders, and incompetent cervix, (3) pregnancy complications such as preeclampsia, abruption, oligohydramnios (based on prenatal ultrasound report), chorioamnionitis, and preterm labor, and (4) use of tocolytics such as indomethacin and magnesium sulfate. Data were also collected on conditions which may influence neonatal outcomes such as antenatal steroids. To evaluate the source of gadolinium exposure, information on maternal exposure to MRI with GBCA, during pregnancy and within 3 years prior to delivery, was collected from medical charts and radiology reports of all subjects and was also confirmed with maternal interviews. The information on type, dose, indication, and timing of GBCA exposure within 3 years prior to delivery was collected.

Data of neonatal outcomes included information on GA at birth, perinatal asphyxia (indexed by Apgar score at 5 min), birth weight, head circumference at birth, small for GA, congenital malformations, respiratory distress syndrome defined based on chest X-ray findings, and the need for surfactant, intraventricular hemorrhage based on head ultrasound findings as reported by a pediatric radiologist, Echocardiogram proven patent ductus arteriosus, chronic lung disease defined as oxygen requirement at 36 weeks post-menstrual age, retinopathy of prematurity as evaluated by a pediatric opthalmologist, necrotizing enterocolitis with X-ray findings of pneumatosis or perforation, highest serum creatinine (mg/dl, 1mg/dl = 88.4µmol/l) during the first 3 days after birth (a marker of renal function), degree of unconjugated hyperbilirubinemia (peak total serum bilirubin concentration [mg/dl, 1mg/dl = 17.1µmol/l]), conjugated or direct hyperbilirubinemia (direct bilirubin > 2 mg/dl), and peak alkaline phosphatase (index of osteopenia of prematurity) during the NICU stay. GA was assessed by obstetrical dating criteria, or when the data was inadequate, by a Ballard exam.

Statistical Analyses

Statistical analyses were performed using Version 9.4 of the SAS System for Windows (SAS Institute Inc., Cary, NC, USA). This was an exploratory dose–response study involving multiple neonatal outcomes in a high-risk premature population. A sample size of ∼100 subjects was assumed large enough for a statistical power. Comparisons on continuous variables were analyzed using a t-test or a Wilcoxon Rank Sum Test (if non-normally distributed). Comparisons on categorical variables were analyzed using a Chi-square test or a Fisher exact test. All tests were two sided and a P < .05 was considered statistically significant. Exploratory bivariate analyses were performed to evaluate the associations between cord blood gadolinium concentration and various neonatal outcome variables using simple linear regression (for continuous outcomes) and simple logistic regression (for categorical outcomes). Variables found to be associated with cord blood gadolinium concentration in preliminary analyses (P < .1) were further examined using multiple regression analyses after controlling for priori confounding variables, including GA, gender, race, antenatal steroid (betamethasone) exposure, and maternal exposure to MRI with GBCA.

RESULTS

A total of 161 infants, 24–33 weeks GA, were born at URMC and admitted to the NICU over an eighteen month period. Of these, 135 infants had cord blood collected and were eligible for the study. Of these, the two infants with chromosomal abnormalities were excluded. Of the remaining 133 infants, 104 received parental consent and were studied. The mean GA of infants was 29.5 (standard deviation [SD] 1.5) weeks. The mean birth weight of infants was 1326 (SD 366) g. The mean ± SD (range) of cord blood gadolinium concentration of the study population was 191 ± 519 pg/ml (3.4–3729.6) (1pg/ml =3.671 pmol/l). None of the infants had gross structural malformations at birth and none died during NICU stay. The mean NICU stay of infants was 54 days (SD 27, range 15–148 days).

The associations between cord blood gadolinium concentration and maternal conditions during pregnancy, as evaluated by a simple linear regression or a logistic regression analysis, are shown in Table 1. Gadolinium concentration was not associated with maternal age, race, ethnicity, prior history of gadolinium-based imaging studies, abruption, clinical chorioamnionitis, pregnancy-induced hypertension, oligohydramnios, preterm labor, indomethacin use for tocolysis, antenatal magnesium use, antenatal betamethasone use, maternal thyroid disorders, maternal diabetes, incompetent cervix, illicit drug use, mode of delivery, or cord blood pH. Only four mothers had a recorded history of exposure to GBCA. All four mothers had a single exposure to intravenous Gadodiamide (Omniscan), a linear non-ionic gadolinium, at a dose of 0.1 mmol/kg, and the exposures were before pregnancy. Three mothers had a MRI of the head (one for unexplained ataxia, one for migraines, and one for an unexplained headache with blurred vision). The remaining mother had a MRI of lumbosacral spine for a spinal injury resulting from an accidental fall.

TABLE 1.

Unadjusted Association Between Cord Blood Gadolinium and Maternal Outcomes

Maternal outcomes Range (mean) or proportion Coefficient or OR 95% Confidence interval P
Maternal age (years) 16–39 (27.5) −0.0010a −0.0033 to 0.0012 .36
Race (% white) 72 0.997b 0.994 to 1.001 .24
Ethnicity (% Hispanic) 10 0.999b 0.998 to 1.001 .91
MRI with Gadolinium (%) 4 0.995b 0.978 to 1.012 .58
Abruption (%) 6 0.998b 0.991 to 1.006 .63
Chorioamnionitis (%) 5 0.999b 0.993 to 1.004 .7
Preeclampsia (%) 22 1.000b 0.999 to 1.001 .70
Oligohydramnios (%) 15 1.000b 0.998 to 1.001 .73
Preterm labor (%) 66 1.001b 0.999 to 1.002 .46
Indomethacin use (%) 12 0.995b 0.985 to 1.005 .36
Magnesium sulfate use (%) 26 0.999b 0.998 to 1.000 .55
Betamethasone use (%) 78 1.000b 0.999 to 1.000 .32
Thyroid disorders (%) 7 0.997b 0.989 to 1.005 .57
Maternal diabetes (%) 10 1.000b 0.999 to 1.001 .18
Incompetent cervix (%) 9 0.997b 0.989 to 1.005 .50
Illicit drug use (%) 24 0.999b 0.998 to 1.001 .45
Mode of delivery (%) 58 1.000b 0.999 to 1.001 .94
Cord blood pH 7.02–7.44 (7.28) −9.75e to 06a −0.00004 to 0.00002 .54
Open in a new tab
a

Coefficient using linear regression.

b

Odds ratio (OR) using logistic regression.

All infants had outcomes evaluated until at least 36 weeks post-menstrual age, by which age the neonatal outcomes studied usually manifest. The associations between cord blood gadolinium concentration and neonatal outcomes, as evaluated by simple unadjusted linear or logistic regression analysis, are shown in Table 2. In a simple unadjusted linear regression analysis, the cord blood gadolinium concentration was negatively associated with renal function as indexed by the highest creatinine level during the first 3 days (P = .03). Increase in gadolinium concentration was associated with an improvement in renal function. On correlation analysis, there was a significant negative correlation between cord blood gadolinium concentration and highest creatinine level during the first 3 postnatal days (spearman, r = −0.27, P = 0.03). On simple regression analyses, there was no significant association of cord blood gadolinium concentration with GA, birth weight, head circumference, small for GA, gender, Apgar score at 5 min, respiratory distress syndrome, patent ductus arteriosus, intraventricular hemorrhage, severe intraventricular hemorrhage (≥ grade 2), degree of unconjugated hyperbilirubinemia, necrotizing enterocolitis, conjugated hyperbilirubinemia, osteopenia of prematurity (peak alkaline phosphatase ≥ 500 IU/ml), chronic lung disease, or retinopathy of prematurity.

TABLE 2.

Unadjusted Association Between Cord Blood Gadolinium and Neonatal Outcomes

Neonatal outcomes Range (mean) or proportion Coefficient or OR 95% Confidence interval P
Gestational age (weeks) 26.2–33 (29.5) 0.0006a −0.00003 to 0.0012 .06
Birth weight (grams) 540–2330 (1311) 0.09a −0.04 to 0.23 .17
Head circumference (cm) 20.5–37 (27.2) 0.0003a −0.0005 to 0.0012 .44
Small for gestational age (%) 16 1.0005b 0.999 to 1.001 .2
Gender (% male) 50 1.00b 0.99 to 1.00 .75
Apgar Score at 5 min 3–9 (7.8) 0.0002a −0.0003 to 0.0007 .50
Respiratory distress syndrome (%) 75 0.999b 0.998 to 1.000 .09
Patent ductus arteriosus (%) 22 0.996b 0.989 to 1.003 .22
Any intraventricular hemorrhage (%) 19 0.998b 0.994 to 1.002 .35
Severe intraventricular hemorrhage (%) 3 0.980b 0.943 to 1.018 .3
Peak total serum bilirubin (mg/dl) 4–14 (9.3) 0.0005a 0.00006 to 0.0012 .07
Highest serum creatinine (mg/dl) 0.18–1.4 (0.63) −0.0001a −0.0003 to –0.00001 .03
Necrotizing enterocolitis (%) 8 0.999b 0.996 to 1.003 .66
Direct hyperbilirubinemia (%) 0–1 (0.2) 0.999b 0.995 to 1.003 .67
Osteopenia of prematurity (%) 31 0.999b 0.998 to 1.001 .41
Chronic lung disease (%) 11 0.991b 0.978 to 1.004 .20
Retinopathy of prematurity (%) 3 0.995b 0.976 to 1.015 .63
Open in a new tab
a

Coefficient using linear regression.

b

Odds ratio (OR) using logistic regression.

Degree of prematurity, renal function, degree of unconjugated hyperbilirubinemia, and respiratory distress syndrome were identified as potential outcomes associated with in utero gadolinium exposure (P < .1). The adjusted associations between cord blood gadolinium concentration and these neonatal outcomes using multiple regression analyses are shown in Table 3. Because none of the maternal conditions were associated with gadolinium exposure on a simple linear or logistic regression analyses (Table 1), no maternal conditions, other than a maternal history of MRI with GBCA within 3 years prior to delivery and betamethasone usage, were considered as confounders for the association between cord blood gadolinium concentration and neonatal outcomes. There was a trend of significant positive correlation between cord blood gadolinium concentration and GA, suggesting no harmful influence on degree of prematurity (spearman r = 0.18, P = .062). On multiple regression analysis controlling for GA, gender, race, antenatal steroids, and maternal history of MRI with GBCA, there was no significant association between cord blood gadolinium concentration and highest creatinine concentration during the first 3 days. Similarly, there was no significant association between cord blood gadolinium concentration and degree of unconjugated hyperbilirubinemia, as well as between cord blood gadolinium concentration and respiratory distress syndrome, controlling for GA, gender, race, exposure to antenatal steroids, and maternal history of MRI with GBCA. In addition, because the potential toxic effects could be non-linear, we also compared neonatal outcomes between the 26 infants with the highest quartile of cord blood gadolinium concentration and the 26 infants with the lowest quartile of cord blood gadolinium concentration using analysis of covariance and controlling for confounding variables. We found no association between cord blood gadolinium concentration and neonatal outcomes (data not shown).

TABLE 3.

Adjusted Association Between Cord Blood Gadolinium and Neonatal Outcomes Using Multiple Regression Analyses

Outcome Coefficient or OR 95% Confidence interval P
Peak total serum bilirubin (mg/dl) 0.00026a −0.00032 to 0.00082 .39
Highest creatinine (mg/dl) −0.00007a −0.00020 to 0.00005 .24
Respiratory distress syndrome 0.999b 0.998 to 1.000 .18
Gestational age (weeks) 0.00062a −0.00003 to 0.00128 .06
Open in a new tab
a

Coefficient using multiple linear regression analysis.

b

Odds ratio (OR) using multiple logistic regression analysis; race, gender, antenatal steroids, maternal history of magnetic resonance imaging with gadolinium, and gestational age were confounders for outcomes peak total serum bilirubin, highest creatinine, and respiratory distress syndrome; maternal history of magnetic resonance imaging with gadolinium was the confounder for the outcome gestational age.

DISCUSSION

Findings in animal and human adult studies of harmful effects with gadolinium exposure raise concerns about the safety of gadolinium exposure on developing fetus (Abraham and Thakral, 2008; Broome, 2008; Sundgren and Leander, 2011). Based on existing literature, gadolinium is currently classified as a category C drug by the US FDA and can only be used during pregnancy if the potential benefits justify the potential risks to the fetus or mother. Our findings strongly suggest that in utero gadolinium exposure is not associated with adverse neonatal outcomes. We believe that this is the first prospective study evaluating the influence of in utero gadolinium exposure on neonatal outcomes as a function of cord blood gadolinium concentration.

Preclinical studies in animals have shown conflicting findings of the harmful effects of in utero gadolinium exposure. The harmful effects of in utero exposure to gadolinium were seen at much higher dosages of gadolinium than the FDA approved clinical dosage of 0.1 mmol/kg (Morisetti et al., 1999; Okuda et al., 1999; Rofsky et al., 1994, 1995; Soltys, 1992; Sundgren and Leander, 2011). One study examined the teratogenic impacts of 0.3–2.0 mmol/kg and repeated dose exposures on rabbits for a 14-day period (Okuda et al., 1999). This study found decreased fetal body weight, increased incidence of microphthalmia, vertebral anomalies, and intrauterine death at 2 mmol/kg dosage, but not at a 0.3 mmol/kg dosage. However, in another study, no evidence of an increase in spontaneous abortions, still births, or grossly detectable eye, ear, facial, limb, or extremity defects was found among 136 mice fetuses exposed to gadolinium at 9.5 days of gestation (Rofsky et al., 1994). Our findings of no grossly detectable structural deformities among premature infants are in agreement with the findings of preclinical studies in mice.

Recent reports on chronic retention of gadolinium in various tissues such as bones, the brain, and kidneys have raised concerns for long-term harmful effects (Rogosnitzky and Branch, 2016). Our findings on neonatal outcomes are consistent with the few reported neonatal case series that showed no adverse health effects from in utero exposure to gadolinium (De Santis et al., 2007; Marcos et al., 1997; Shoenut et al., 1993). However, our study differed from previous neonatal studies in study design, subject population, size of the study, neonatal outcome measures, and the length of follow-up. Compared with our study, previous studies were mainly case series reports that precluded meaningful evaluations of the association between gadolinium exposure and neonatal outcomes in the absence of a control group. Although there were no control subjects with non-detectable cord blood gadolinium concentrations, our study used a dose–response relationship to evaluate adverse neonatal outcomes as a function of cord blood gadolinium concentration. In addition, the actual numbers of cases included in previous case series reports were relatively small and mainly involved term neonates, further limiting meaningful evaluations and conclusions. Moreover, previous studies were mainly limited to growth outcomes at birth and failed to evaluate other clinically important neonatal outcomes such as renal function, jaundice, respiratory distress, chronic lung disease, or osteopenia of prematurity over a longer period of NICU stay.

In a human neonatal study involving exposure to gadolinium during the second and third trimester, no adverse health effects on growth were observed at birth in 11 term newborns (Marcos et al., 1997). We also found no adverse effects of in utero gadolinium exposure on weight or head circumference at birth. In a small neonatal case series, two out of six infants exposed to gadolinium in the last trimester had fetal distress, but no related consequences (Tanaka et al., 2001). It was difficult to evaluate any association of fetal distress to gadolinium exposure in the absence of control subjects in this case series study (Tanaka et al., 2001). Although we did not evaluate fetal distress, we found no increased rate of cesarean section, a common mode of delivery for persistent fetal distress. In addition, we found no association of cord blood gadolinium concentration with either cord blood pH or Apgar score at 5 min (index of perinatal asphyxia), indirectly suggesting that gadolinium exposure is not associated with fetal distress. Another prospective case series study that evaluated first trimester exposure to gadolinium in 26 subjects reported no maternal complications, observed one case with hemangioma, two cases with low birth weight, and two cases with miscarriages (De Santis et al., 2007). However, we found no association between in utero gadolinium exposure and small for GA. In our comprehensive evaluation of maternal conditions or pregnancy complications, we also found no increased risks of maternal conditions or pregnancy complications with gadolinium exposure. Our study involved premature live births and therefore we were unable to evaluate the association with abortions, miscarriages, or stillbirths.

Recent adult human studies have shown that exposure to gadolinium may be associated with nephrogenic systemic fibrosis and renal failure in subjects with impaired renal function (Abraham and Thakral, 2008; Broome, 2008; Grobner, 2006; Marckmann et al., 2007; Weigle and Broome, 2008). As a corollary, premature infants with immature kidneys may be at increased risk for renal failure from gadolinium exposure; however, this has not been previously investigated. We found that in utero gadolinium exposure does not affect renal function after adjusting for confounder variables in premature infants. We also found that in utero gadolinium exposure is not associated with an increased risk for common neonatal outcomes, after adjusting for confounding variables. In our study, none of the mothers had exposure to GBCA during pregnancy, with only a few having exposure within three years prior to delivery. This highly suggests an alternate source of gadolinium or a history of gadolinium based imaging studies earlier than 3 years before delivery. One potential source for gadolinium in the mothers with past history of GBCA exposure is the endogenous release of gadolinium from bones, which has been suggested previously and observed for other metallodrugs (Darrah et al., 2009). The potential alternate source is the presence of elevated gadolinium in public water supplies around the world (Badruzzaman et al., 2012; Kulaksiz and Bau, 2011). The use of gadolinium in the electronics industry and medical field has dramatically increased in recent years, with associated several-fold increases in water contamination of gadolinium, resulting from waste water discharges (Hatje et al., 2016; Idee et al., 2006). Although our study did not investigate water contamination of gadolinium, it raises the question if the source could be the drinking water supply in the Greater Rochester area and the surrounding catchment area served by the URMC.

The strength of the study is a prospective comprehensive evaluation of neonatal health effects as a function of cord blood gadolinium concentration in highly vulnerable premature infants. As randomized studies are impractical and identification of the control group is difficult, a prospective observational study using cord blood gadolinium concentration allows evaluation of the dose–response relationship and provides the best possible method to evaluate the harmful fetal effects of in utero gadolinium exposure available to date. Our findings of no harmful effects on neonatal outcomes could be due to Type II errors and inadequate power. However, our study had an 80% power to detect an R-squared of 0.07 when evaluating an association with continuous variables such as creatinine and total serum bilirubin concentration using a linear regression model. For the outcome, respiratory distress syndrome, the sample size used had an 80% power to detect an odds ratio of 1.88 with respect to one standard deviation change in gadolinium with the estimated incidence of 75% for respiratory distress syndrome. Furthermore, there was a trend of significant beneficial association with GA; therefore, the possibility of a type II error due to inadequate sample size is unlikely. There is a potential selection bias by studying only the premature population; however, we studied this population because the premature population is generally more vulnerable to adverse effects than the term population. It is conceivable that the background risk of GA dependent neonatal outcomes in this premature population may mask any adverse effects of gadolinium. To prevent any masking effect, the analyses were performed controlling for GA. We believe that the findings on most neonatal outcomes which are more common in premature infants than in term infants, such as respiratory distress syndrome, unconjugated hyperbilirubinemia, renal function, etc. can be generalized to infants > 33 weeks GA, at least during the neonatal period. However, because the likely duration of in utero exposure to gadolinium will be longer with more mature and term infants compared with preterm infants ≤ 33 weeks GA, the findings on growth outcomes such as small for GA may not be evident in premature infants and, therefore, may not be generalizable to more mature and term infants. In addition, our study does not exclude the possibility of increased risk of spontaneous abortions and fetal demise, as cord blood gadolinium levels were only evaluated on live premature births. Our study was also not designed to evaluate the mutagenic effect. Furthermore, the possibility of adverse neonatal effects at much higher gadolinium concentrations than that observed in our population cannot be excluded. Finally, neurological harmful effects are difficult to assess during the neonatal period and normal head ultrasound findings do not rule out any possible neurotoxic effects in vulnerable premature infants (Ray et al., 1996). Nonetheless, our dose–response study involving vulnerable premature infants provides useful safety data for during the neonatal period.

In summary, there appears to be no major harmful effects from in utero gadolinium exposure to premature infants during the neonatal period. It is possible that harmful effects, specifically neurotoxic and carcinogenic effects, of medicinal or environmental toxins such as gadolinium in vulnerable premature infants may manifest at a later age. A prospective longitudinal study is warranted to evaluate the potential long-term harmful effects of in utero gadolinium exposure.

ACKNOWLEDGMENTS

We are thankful to obstetric nurses for collection of cord blood. We are also thankful to blood bank technicians for organizing cord blood collection. We are grateful to the parents for participation in this study.

FUNDING

University of Rochester Clinical and Translational Science Award (UL1 TR000042) from the National Center for Advancing Translational Sciences of the National Institutes of Health.

REFERENCES

  1. Abraham J. L., Thakral C. (2008). Tissue distribution and kinetics of gadolinium and nephrogenic systemic fibrosis. Eur. J. Radiol. 66, 200–207. [DOI] [PubMed] [Google Scholar]
  2. Badruzzaman M., Pinzon J., Oppenheimer J., Jacangelo J. G. (2012). Sources of nutrients impacting surface waters in Florida: A review. J. Environ. Manage. 109, 80–92. [DOI] [PubMed] [Google Scholar]
  3. Biagi B. A., Enyeart J. J. (1990). Gadolinium blocks low-and high-threshold calcium currents in pituitary cells. Am. J. Physiol. 259, C515–C520. [DOI] [PubMed] [Google Scholar]
  4. Broome D. R. (2008). Nephrogenic systemic fibrosis associated with gadolinium based contrast agents: A summary of the medical literature reporting. Eur. J. Radiol. 66, 230–234. [DOI] [PubMed] [Google Scholar]
  5. Broome D. R., Girguis M. S., Baron P. W., Cottrell A. C., Kjellin I., Kirk G. A. (2007). Gadodiamide-associated nephrogenic systemic fibrosis: Why radiologists should be concerned. Am. J. Roentgenol. 188, 586–592. [DOI] [PubMed] [Google Scholar]
  6. Darrah T. H., Prutsman-Pfeiffer J. J., Poreda R. J., Ellen Campbell M., Hauschka P. V., Hannigan R. E. (2009). Incorporation of excess gadolinium into human bone from medical contrast agents. Metallomics: Integr. Biometal Sci. 1, 479–488. [DOI] [PubMed] [Google Scholar]
  7. De Santis M., Straface G., Cavaliere A. F., Carducci B., Caruso A. (2007). Gadolinium periconceptional exposure: Pregnancy and neonatal outcome. Acta Obstet. Gynecol. Scand. 86, 99–101. [DOI] [PubMed] [Google Scholar]
  8. Elster A. D. (1990). Cranial MR imaging with Gd-DTPA in neonates and young infants: Preliminary experience. Radiology 176, 225–230. [DOI] [PubMed] [Google Scholar]
  9. Garcia-Bournissen F., Shrim A., Koren G. (2006). Safety of gadolinium during pregnancy. Can. Family Physician Med. Famille Can. 52, 309–310. [PMC free article] [PubMed] [Google Scholar]
  10. Greenberg S. B., Bhutta S. T., Buchmann R. F. (2004). Gadolinium-enhanced magnetic resonance angiography in neonates and infants suspected of caval or aortic thrombosis. Pediatr. Radiol. 34, 948–951. [DOI] [PubMed] [Google Scholar]
  11. Grobner T. (2006). Gadolinium – a specific trigger for the development of nephrogenic fibrosing dermopathy and nephrogenic systemic fibrosis? Nephrol. Dialysis Transplant: Official Publication of the European Dialysis and Transplant Association – European Renal Association 21(4), 1104–1108. [DOI] [PubMed]
  12. Gulson B. L., Calder I. C. (1995). Lead mobilisation during pregnancy. Med. J. Aust. 163, 447.. [DOI] [PubMed] [Google Scholar]
  13. Gulson B. L., Jameson C. W., Mahaffey K. R., Mizon K. J., Korsch M. J., Vimpani G. (1997). Pregnancy increases mobilization of lead from maternal skeleton. J. Lab. Clin. Med. 130, 51–62. [DOI] [PubMed] [Google Scholar]
  14. Hatje V., Bruland K. W., Flegal A. R. (2016). Increases in anthropogenic gadolinium anomalies and rare earth element concentrations in San Francisco Bay over a 20 year record. Environ. Sci. Technol. 50, 4159–4168. [DOI] [PubMed] [Google Scholar]
  15. Idee J. M., Port M., Raynal I., Schaefer M., Le Greneur S., Corot C. (2006). Clinical and biological consequences of transmetallation induced by contrast agents for magnetic resonance imaging: A review. Fundam. Clin. Pharmacol. 20, 563–576. [DOI] [PubMed] [Google Scholar]
  16. Kulaksiz S., Bau M. (2011). Rare earth elements in the Rhine River, Germany: First case of anthropogenic lanthanum as a dissolved microcontaminant in the hydrosphere. Environ. Int. 37, 973–979. [DOI] [PubMed] [Google Scholar]
  17. Lancelot E. (2016). Revisiting the pharmacokinetic profiles of gadolinium-based contrast agents: Differences in long-term biodistribution and excretion. Investigative Radiol. 51, 691–700. [DOI] [PubMed] [Google Scholar]
  18. Laswad T., Wintermark P., Alamo L., Moessinger A., Meuli R., Gudinchet F. (2009). Method for performing cerebral perfusion-weighted MRI in neonates. Pediatr. Radiol. 39, 260–264. [DOI] [PubMed] [Google Scholar]
  19. Marckmann P., Skov L., Rossen K., Heaf J. G., Thomsen H. S. (2007). Case-control study of gadodiamide-related nephrogenic systemic fibrosis. Nephrol. Dialysis Transplant.: Official Publication of the European Dialysis and Transplant Association – European Renal Association 22, 3174–3178. [DOI] [PubMed] [Google Scholar]
  20. Marcos H. B., Semelka R. C., Worawattanakul S. (1997). Normal placenta: Gadolinium-enhanced dynamic MR imaging. Radiology 205, 493–496. [DOI] [PubMed] [Google Scholar]
  21. Molgo J., del Pozo E., Banos J. E., Angaut-Petit D. (1991). Changes of quantal transmitter release caused by gadolinium ions at the frog neuromuscular junction. Br. J. Pharmacol. 104, 133–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Morisetti A., Bussi S., Tirone P., de Haen C. (1999). Toxicological safety evaluation of gadobenate dimeglumine 0.5 M solution for injection (MultiHance), a new magnetic resonance imaging contrast medium. J. Comput. Assist. Tomogr. 23(Suppl. 1), S207–S217. [DOI] [PubMed] [Google Scholar]
  23. Novak Z., Thurmond A. S., Ross P. L., Jones M. K., Thornburg K. L., Katzberg R. W. (1993). Gadolinium-DTPA transplacental transfer and distribution in fetal tissue in rabbits. Invest. Radiol. 28, 828–830. [PubMed] [Google Scholar]
  24. Okuda Y., Sagami F., Tirone P., Morisetti A., Bussi S., Masters R. E. (1999). Reproductive and developmental toxicity study of gadobenate dimeglumine formulation (E7155) (3)--Study of embryo-fetal toxicity in rabbits by intravenous administration. Tox. Sci. 24(Suppl 1), 79–87. [DOI] [PubMed] [Google Scholar]
  25. Palacios Jaraquemada J. M., Bruno C. (2000). Gadolinium-enhanced MR imaging in the differential diagnosis of placenta accreta and placenta percreta. Radiology 216, 610–611. [DOI] [PubMed] [Google Scholar]
  26. Rangamani S., Varghese J., Li L., Harvey L., Hammel J. M., Fletcher S. E., Duncan K. F., Danford D. A., Kutty S. (2012). Safety of cardiac magnetic resonance and contrast angiography for neonates and small infants: A 10-year single-institution experience. Pediatr. Radiol. 42, 1339–1346. [DOI] [PubMed] [Google Scholar]
  27. Ray D. E., Cavanagh J. B., Nolan C. C., Williams S. C. (1996). Neurotoxic effects of gadopentetate dimeglumine: Behavioral disturbance and morphology after intracerebroventricular injection in rats. Am. J. Neuroradiol. 17, 365–373. [PMC free article] [PubMed] [Google Scholar]
  28. Roberts D. R., Chatterjee A. R., Yazdani M., Marebwa B., Brown T., Collins H., Bolles G., Jenrette J. M., Nietert P. J., Zhu X. (2016). Pediatric patients demonstrate progressive T1-weighted hyperintensity in the dentate nucleus following multiple doses of gadolinium-based contrast agent. Am. J. Neuroradiol. doi: 10.3174/ajnr.A4891. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rofsky N. M., Pizzarello D. J., Duhaney M. O., Falick A. K., Prendergast N., Weinreb J. C. (1995). Effect of magnetic resonance exposure combined with gadopentetate dimeglumine on chromosomes in animal specimens. Acad. Radiol. 2, 492–496. [DOI] [PubMed] [Google Scholar]
  30. Rofsky N. M., Pizzarello D. J., Weinreb J. C., Ambrosino M. M., Rosenberg C. (1994). Effect on fetal mouse development of exposure to MR imaging and gadopentetate dimeglumine. J. Magn. Reson. Imaging 4, 805–807. [DOI] [PubMed] [Google Scholar]
  31. Rogosnitzky M., Branch S. (2016). Gadolinium-based contrast agent toxicity: A review of known and proposed mechanisms. Biometals: An International Journal on the Role of Metal Ions in Biology, Biochemistry, and Medicine 29, 365–376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Runge V. M. (1988). Gd-DTPA: An i.v. contrast agent for clinical MRI. Int. J. Radiat. Appl. Instrum. Part B: Nucl. Med. Biol. 15, 37–44. [DOI] [PubMed] [Google Scholar]
  33. Shoenut J. P., Semelka R. C., Silverman R., Yaffe C. S., Micflikier A. B. (1993). MRI in the diagnosis of Crohn's disease in two pregnant women. J. Clin. Gastroenterol. 17, 244–247. [DOI] [PubMed] [Google Scholar]
  34. Soltys R. A. (1992). Summary of preclinical safety evaluation of gadoteridol injection. Invest. Radiol. 27(Suppl. 1), S7–S11. [PubMed] [Google Scholar]
  35. Sprauten M., Darrah T. H., Peterson D. R., Campbell M. E., Hannigan R. E., Cvancarova M., Beard C., Haugnes H. S., Fossa S. D., Oldenburg J., et al. (2012). Impact of long-term serum platinum concentrations on neuro- and ototoxicity in Cisplatin-treated survivors of testicular cancer. J. Clin. Oncol: Off. J. Am. Soc. Clin. Oncol. 30, 300–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Sundgren P. C., Leander P. (2011). Is administration of gadolinium-based contrast media to pregnant women and small children justified? J. Magn. Reson. Imaging 34, 750–757. [DOI] [PubMed] [Google Scholar]
  37. Tanaka Y. O., Sohda S., Shigemitsu S., Niitsu M., Itai Y. (2001). High temporal resolution dynamic contrast MRI in a high risk group for placenta accreta. Magn. Reson. Imaging 19, 635–642. [DOI] [PubMed] [Google Scholar]
  38. USEPA. (1991). US EPA SW-846 Method 6020A. Analysis of soils and sediments by inductively coupled plasma-mass spectrometry.
  39. USEPA. (1993). Title 40 of the US Code of Federal Regulations, Part 136, Appendix B, Washington, DC.
  40. USEPA. (1994). EPA 200.8 Determination of trace elements in waters and waste waters by inductively coupled plasma-mass spectrometry.
  41. Webb J. A., Thomsen H. S., Morcos S. K. and Members of Contrast Media Safety Committee of European Society of Urogenital, R. (2005). The use of iodinated and gadolinium contrast media during pregnancy and lactation. Eur. Radiol. 15, 1234–1240. [DOI] [PubMed] [Google Scholar]
  42. Weigle J. P., Broome D. R. (2008). Nephrogenic systemic fibrosis: Chronic imaging findings and review of the medical literature. Skeletal Radiol. 37, 457–464. [DOI] [PubMed] [Google Scholar]
  43. Yang X. C., Sachs F. (1989). Block of stretch-activated ion channels in Xenopus oocytes by gadolinium and calcium ions. Science 243, 1068–1071. [DOI] [PubMed] [Google Scholar]

Articles from Toxicological Sciences are provided here courtesy of Oxford University Press

RESOURCES