Article pubs.acs.org/est
Temporal Variability in Urinary Phthalate Metabolite Excretion Based on Spot, Morning, and 24‑h Urine Samples: Considerations for Epidemiological Studies Hanne Frederiksen,† Selma K. Kranich,† Niels Jørgensen,† Olivier Taboureau,‡ Jørgen H. Petersen,† and Anna-Maria Andersson*,† †
Department of Growth and Reproduction, Rigshospitalet, Copenhagen University Hospital, Blegdamsvej 9, DK-2100 Copenhagen, Denmark ‡ Center for Biological Sequence Analysis, Technical University of Denmark, DK-2800 Lyngby, Denmark ABSTRACT: Urinary phthalate excretion is used as marker of phthalate exposure in epidemiological studies. Here we examine the reliability of urinary phthalate levels in exposure classification by comparing the inter- and intrasubject variation of urinary phthalate metabolite levels. Thirty-three young healthy men each collected two spot, three first-morning, and three 24-h urine samples during a 3-month period. Samples were analyzed for the content of 12 urinary metabolites of 7 different phthalates. Variability was assessed as intraclass correlation coefficients (ICC). For the metabolites of diethyl-, dibutyl-, and butylbenzyl-phthalates moderate ICCs were observed in all three sample types, albeit highest in 24-h urine (0.51− 0.59). For the metabolites of di(2-ethylhexyl) phthalate and di-iso-nonyl phthlates lower ICCs (0.06−0.29) were found. These low ICCs indicate a high risk of misclassification of exposures for these two phthalates in population studies and hence an attenuation of the power to detect possible exposure-outcome associations. The only slightly higher ICCs for 24-h pools compared to first-morning and spot urine samples does not seem to justify the extra effort needed to collect 24-h pools.
■
INTRODUCTION Phthalate diesters are present as plasticizers and solvents in many industrial and consumer products. Human exposure is consequently widespread,1−4 and of concern as adverse effects of phthalate exposure on especially reproductive development have been shown in experimental animal studies.5,6 In human studies, associations between maternal midpregnancy urinary levels of phthalate metabolites and decreased anogenital distance in male infants have been observed.7 Also reduced masculine play in boys based on a validated questionnaire used to assess sexually dimorphic play behavior has been associated with prenatal phthalate exposure.8 Both effects are likely to be related to antiandrogenic activity during fetal development. Human urinary levels of phthalates have also been associated with decreased testosterone levels in adult men,9 increased waist circumference,10 length of gestation,11,12 and decreased sperm quality.13 Human exposure−effect studies are, however, still scarce, and some conflicting results among studies exist. For example, maternal urinary metabolites of di(2-ethylhexyl)-phthalate (DEHP) have been associated with increased length of gestation,11 decreased length of gestation,12 and even preterm birth.14 Furthermore, while some studies have found associations between adult male urinary levels of phthalate metabolites and sperm quality parameters,13 others have not.15 These discrepancies may be due to differences in study setup or study populations or statistical artifacts. In all © 2012 American Chemical Society
these published studies on associations between phthalate exposure and human effect, the exposures are based on the concentration of phthalate metabolites in single-spot urine or morning urine samples. Following uptake, phthalates are rapidly metabolized and mainly excreted in urine. Approximately 67% of DEHP and 44% of di-iso-nonyl-phthalate (DiNP) were excreted in urine as various metabolites within, respectively, 24 and 48 h after a single oral dose.16,17 The majority of phthalate metabolites were excreted even within a few hours after the exposure. Also after uptake through skin the metabolites of diethyl-phthalate (DEP) were excreted in urine within 12 h, while the metabolites of dibutyl-phthalate (DBP) had a slightly slower urinary excretion.18 Due to this rapid urinary clearance of phthalates a single urine sample represents recent exposure. Significant within-person variability in urinary phthalate metabolite levels has been shown for repeated spot or first-morning urine samples19−21 supposedly reflecting significant day-to-day variations in exposures. However, spot and morning urine samples may over or underestimate the day-to-day intraindividual variation in exposure, as the concentrations measured Received: Revised: Accepted: Published: 958
September 11, 2012 December 11, 2012 December 12, 2012 December 12, 2012 dx.doi.org/10.1021/es303640b | Environ. Sci. Technol. 2013, 47, 958−967
Environmental Science & Technology
Article
next morning’s first void was pooled in a 5-L polyethylene container. At 42−48 days (third visit) and again 82−88 days (fourth visit) following the first visit, 24-h urine pools were collected again, with the first-morning void at the end of the 24h collection period being collected in a separate container. The urine samples were handed in at the chemical laboratory the day the collection for each visit was completed. Time of last void prior to sampling was not recorded. One participant missed one first-morning urine and as this is part of the 24-h pool this 24-pool was also discharged for this man. One participant forgot to collect the morning spot urine (visit 2) in a separate container and the urine from this void was instead included in the pooled samples for this day. For this participant only the spot urine was missing as a separate measure but the 24-h pool was deemed correct. For one participant the first-morning urine sample was by a mistake done in the lab pooled with the rest of the 24-h pool before an aliquot was taken out for separate measurement. Thus for this person one morning urine was missing but the 24-h pool was correct. In total we obtained in 65 spot urine samples (33 at visit 1 and 32 visit at 2), 97 first-morning samples (2 missing), and 98 24-h pools (1 missing, due to a missing first-morning sample). All samples were collected in April to September 2008. In the chemical laboratory all urine samples were handled immediately after reception. From the morning spot urine (only visit 2) and first-morning urine (visit 2−4) collected in separate containers aliquots of 5 mL were decanted to scintillation vials and the remaining urine was pooled with the rest of the urine collected over the 24-h period to obtain a full 24-h pool. The whole portion was weighed and the volume was calculated based on the weight of the 24-h pool subtracting the average weight of the containers assuming a mass density of the urine of 1g/mL. Aliquots of about 15 mL of the 24-h urine pools were decanted to the scintillation vials and all samples were stored at −20 °C until analysis. Analytical Methods. All urine samples were analyzed for the total content of monoethyl-phthalate (MEP), the sum of mono-n-butyl- and monoiso-butyl-phthalate (∑MBP(i+n)), monobenzyl-phthalate (MBzP), mono(2-ethylhexyl)-phthalate (MEHP), mono(2-ethyl-5-hydroxyhexyl)-phthalate (MEHHP), mono(2-ethyl-5-oxohexyl)-phthalate (MEOHP), mono(2ethyl-5-carboxypentyl)-phthalate (MECPP), mono-n-octylphthalate (MOP), monoiso-nonyl-phthalate (MiNP), mono(hydroxy-iso-nonyl)-phthalate (MHiNP), mono(oxo-isononyl)-phthalate (MOiNP), and mono(carboxy-iso-octyl)phthalate (MOiCP) by liquid-chromatography−tandem-massspectrometry (LC-MS/MS) with preceding enzymatic deconjugation and solid phase extraction. The methods for preparation of samples, standard solutions, and quality controls as well as the instrumental analysis have previously been described24 and were used with modification as described in the following. All internal standards and their respective isotopic labeled standards were purchased from Cambridge Isotope Laboratories (Andover, MA, USA and distributed by Bie & Berntsen A/S (Rødovre, Denmark)) except MECPP, mono-(4methyl-7-hydroxyloctyl)-phthalate (MHiNP), mono-(4-methyl7-oxo-octyl)-phthalate (MOiNP), mono-(4-methyl-7-carboxyheptyl)-phthalate (MCiOP), and their respective D4-labeled standards, which were a generous gift from Prof. Jürgen Angerer (Institute of the Ruhr-Universit at Bochum (IPA), Bochum, Germany). The method for LC-separation was a 17min method and the solvent programming was as follows: 0.0−
will be highly dependent on the hours elapsed between the exposure and collection of urine. Urinary concentrations are also affected by how diluted the urine is; the methods used to adjust for variation in urinary dilution are not perfect. Thus for a true estimate of the temporal intraindividual variability in urinary phthalate metabolite excretion 24-h urine pools need to be considered. In population studies the statistical power to identify exposure−outcome associations is influenced by how reliable the measures of the studied variables are. To evaluate the between-person variation in relation to the within-person variability in phthalate exposures based on urinary excretion we analyzed twelve different phthalate metabolites of seven phthalate diesters in repeated 24-h urine pools (three pools per participant) collected in 33 young adult men over a 3month period. In addition, phthalate metabolites where analyzed in first-morning urine samples and morning spot urine samples collected on the same day as the 24-h urine pools allowing us to compare the classification of the phthalate exposure of the study subjects based on these three different types of urine sampling.
■
MATERIALS AND METHODS Study Subjects. The subjects for this longitudinal study were recruited among young healthy Danish men from the general population who were participating in a large ongoing cross-sectional study on male reproductive health.22 The first 60 men attending the clinic within a 2-month period were asked to participate in the present extended study when they showed up for the clinical examination in the cross-sectional study and 33 men agreed to participate (55% participation rate). Mean anthropometric data (range) for the 33 men were as follows: age, 19.3 years (range: 18.3−22.3); height, 182 cm (range: 171−197), weight, 76.6 kg (range: 53.9−93.2); and body mass index (BMI), 23.2 kg/m2 (range: 16.5−31.9). They did not differ in these characteristics from the 27 men who declined to participate. All participants gave their written acceptance after having received written and oral information about the study. The study was approved by the ethical committee for the Copenhagen municipality (ref. KF 01-117/ 96 and KF 01-292/98 with amendment of January 19, 2006). Study Design. At the first visit in the clinic each participant was physically examined, answered a questionnaire, supplied a serum-, seminal plasma-, and spot urine sample23 and was furthermore enrolled to participate in the present study. During a 3-month period each of the participants visited the clinic 3 more times and in total they, for this study, delivered 2 spot urine samples, 3 first-morning urine samples, and 3 24-h urine samples according to the general outline of the study illustrated in the abstract art. Spot urine samples from the first visit were collected between 8:30 a.m. and 12:00 noon when the participants visited the clinic. The samples were collected in polyethylene cups and about 15 mL of each sample was decanted to a 20-mL glass scintillation vial with top packed with aluminum foil. Two to eight days following the first visit, each of the participants collected all urine excreted during a 24-h period starting with the first void after the first-morning urine on day 1 (which was collected in a separate 1-L polyethylene container as a morning spot urine sample) and ending with the next morning’s first void, which also was collected in a separate 1-L polyethylene container as a first-morning urine sample). The urine collected over the 24 h period between the morning spot urine and the 959
dx.doi.org/10.1021/es303640b | Environ. Sci. Technol. 2013, 47, 958−967
Environmental Science & Technology
Article
(MiNP, MHiNP, MHiOP, MOiCP) were expressed as the sum of the metabolites; ΣDEHPm and ΣDiNPm, respectively, by adding the molecular concentrations of each metabolite and subsequently multiplying with the molecular weight of the respective parent compound. Bivariate correlation analyses (Spearman's rho) were performed to analyze for correlations between the levels of different phthalate metabolites within the same sample and between the levels of individual phthalate metabolites in the three different types of urine samples. In some of the samples the concentration was below LOD. For correlation analyses all data were included and not detectable data were all set to 0.01, which was a 10-fold lower value than the lowest LOD. To assess the temporal intraindividual variability in urinary phthalate concentrations and total amount excreted, we estimated the between-person and within-person variance and calculated intraclass correlation coefficients (ICCs) for the repeated samples of each type of sampling on natural-log transformed data. The ICC is defined as the between-person variance divided by the total variance, which in our study setup was the sum of the between-person variance and the withinperson variance. The ICC ranges from 0 to 1 and the higher the ICC the lower the fraction of the total variance is contributed by within-person variance, i.e. the higher reliability between intraindividual repeated measures in relation to the total range of measurements. The Reliability analysis function in the PASW statistics software package were used to calculate the betweenand within-subject variance and ICCs based on a two-way random effect ANOVA.
1.5 min, 5% B; 1.6 min, 27% B; 6.0 min, 30% B; 6.1−10.0 min, 45% B; 10.1 min, 70% B; 12.0−15.5 min, 90% B; 15.6−17.0 min, 5% B. Because of the short LC-separation (17 min) used for this study, MnBP and MiBP could not be sufficiently separated and were therefore analyzed as one analyte, ∑MBP(i+n). A mixture of MnBP and MiBP (1:1) was used for calibration curves and 13C4−MnBP was used as internal standard for ∑MBP(i+n). The method accuracy and precision were validated by repeated (n = 5) intraday analysis (∼repeated measurements in same batch) of urine pool samples spiked with different concentrations of native phthalate standards (5, 10, and 50 ng/ mL) and by repeated interday analysis (repeated measurements from batch to batch) of urine control samples spiked in low and high concentrations (n = 24 repeated measures over a 2-month period) according to the validation method previously described.24 For ∑MBP(i+n) a 1:1 mixture of the MBP isomers was used for spiking. All control materials, blank urine pool, and urine pool samples spiked with standards were enzymatic hydrolyzed and SPE purified and thereby treated in the same manner as the unknown samples. Both the intraday and interday variation, expressed as the relative standard deviation (RSD) were below 15% for all analytes, except ∑MBP(i+n) (21%) in the low spike level. The absolute recovery of spiked samples was above 90% for all analytes except the DEHP metabolites in the lowest spike level (MEHP 89%, MEHHP 72%, MEOHP 88%, and MECPP 83%). Limits of detections (LOD) were calculated as previously described24 (Table 1). We used urinary osmolality to adjust for urinary dilution. In contrast to urinary creatinine adjustment, which has the limitation that urinary creatinine excretion varies with sex, age, BMI, fat-free mass, and even ethnicity, and urinary specific gravity, which is not only influenced by the number of molecules in urine but also by their molecular weight and size, urine osmolality is directly related to the number of particles in solution and is unaffected by the molecular weight and size of these particles. In subjects with normal renal function osmolality thus reflects an individual’s hydration status. Urinary osmolality was measured by freezing point depression method using an automatic cryoscopic osmometer (Osmomat 030 from Gonotec, Berlin, Germany). For each 9-sample measurement, a standard urine pool was measured. Mean urinary osmolality for this standard pool (N = 42) was 0.341 Osm/kg with a relative standard deviation (RSD) of 0.68%. The median (range) osmolality of all urine samples included in this study was 0.799 (0.081−1.238) Osm/kg, which is within the normal urinary osmolality range. Data Analysis and Statistics. Phthalate levels were measured in urinary concentration (ng/mL) and the amount excreted per 24 h was calculated by multiplying the concentration in a 24-h pool with the volume (mL) of the pool. To describe the levels of the phthalate metabolites in urine, selected percentiles and the minimum and maximum concentration were computed. The urinary phthalate metabolite concentrations were also adjusted for the variation in osmolality (Osm/kg) of the urine by normalizing them to an osmolality of 0.8 Osm/kg (corresponding to the median osmolality of all the included urine samples). This was done by dividing the urinary phthalate metabolite concentration (ng/ mL) in each sample with the urinary osmolality (Osm/kg) of that sample and multiplying by 0.8. The concentrations of the DEHP metabolites (MEHP, MEHHP, MEOHP, MECPP) and the DiNP metabolites
■
RESULTS In the method used the concentrations of MiBP and MnBP, the metabolites of the two isoforms of DBP, were measured together and are here presented as the sum of MiBP and MnBP (ΣMBP(i+n)). Four different metabolites of DEHP and DiNP were measured separately but for clarification they are in the following also presented as, respectively, the sum of DEHP metabolites (∑DEHPm) and the sum of DiNP metabolites (∑DiNPm) as described in Materials and Methods. The concentrations (ng/mL) of phthalate metabolites measured in first-morning urine, spot urine, and 24-h urine samples are presented in Table 1 as range and selected percentiles. In summary, ΣMBP(i+n) and ∑DEHPm metabolites were excreted in the highest concentration followed by MEP > MBzP > ∑DiNPm metabolites. MEP was excreted in the largest range from a few ng/mL to 20 μg/mL. Except for MOP, the hydrolyzed metabolite of di-n-octyl phthalate, which was only detectable in very low amounts in few of the samples, and MiNP, the hydrolyzed metabolite of DiNP, which was only detectable in approximately 45% of the samples, all the other metabolites were detectable in all or almost all samples. Comparison of Spot, Morning, and 24-h Urine Samples. The associations among phthalate metabolites in first-morning urine, spot urine, and 24-h urine samples from the same 24-h period (samples collected for second visit in clinic) are shown in Figure 1. Correlation coefficients between unadjusted concentrations measured in morning urine and spot urine were r = 0.55 (MEP, p < 0.01), r = 0.56 ((∑MBP(i+n), p < 0.001), r = 0.66 (MBzP, p < 0.01), r = 0.46 (∑DEHPm, p < 0.01), and r = 0.72 (∑DiNPm, p < 0.01). Adjusting concentrations according to the osmolality of the samples did not change the significance of the correlations (data not shown). 960
dx.doi.org/10.1021/es303640b | Environ. Sci. Technol. 2013, 47, 958−967
Environmental Science & Technology
Article
Table 1. Urinary Phthalate Metabolite Concentrations (ng/mL) in Young Danish Men (Samples Collected in 2008) percentile morning urine MEP ∑MBP(i+n) MBzP MEHP MEHHP MEOHP MECPP MOP MiNP MHiNP MOiNP MCiOP ∑DEHPm ∑DiNPm spot urine MEP ∑MBP(i+n) MBzP MEHP MEHHP MEOHP MECPP MOP MiNP MHiNP MOiNP MCiOP ∑DEHPm ∑DiNPm 24-h urine pool MEP ∑MBP(i+n) MBzP MEHP MEHHP MEOHP MECPP MOP MiNP MHiNP MOiNP MCiOP ∑DEHPm ∑DiNPm
Na
LOD
N > LOD
minimum
5
25
50
75
95
maximum
97 97 97 97 97 97 97 97 97 97 97 97 97 97
0.24 3.94 1.26 0.31 0.60 0.14 0.43 0.04 0.62 0.31 0.16 0.08
97 97 95 97 97 97 97 10 45 94 97 97
4.99 20.8
