| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Drug Monitoring and Toxicology |
1 Chemistry and Drug Metabolism Section, National Institute on Drug Abuse–Intramural Research Program, National Institutes of Health, Baltimore, MD.
2 Center for Human Toxicology, Department of Pharmacology and Toxicology, University of Utah, Salt Lake City, UT.
3 Department of Biology, University of Pittsburgh at Titusville, Titusville, PA.
4 Office of the Clinical Director, National Institute on Drug Abuse–Intramural Research Program, National Institutes of Health, Baltimore, MD.
5 Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD.
6 Reckitt Benckiser Pharmaceuticals, Inc., Richmond, VA.
7 Department of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, Johns Hopkins Bayview Medical Center, Baltimore, MD.
aAddress correspondence to this author at: Chief, Chemistry and Drug Metabolism, National Institute on Drug Abuse–Intramural Research Program, National Institutes of Health, 5500 Nathan Shock Dr., Baltimore, MD 21224. Fax 410-550-2971; e-mail mhuestis{at}intra.nida.nih.gov.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: This study examined buprenorphine and norbuprenorphine concentrations in hair obtained from 9 buprenorphine-maintained pregnant women and 4 of their infants. Specimens were analyzed by liquid chromatography-tandem mass spectrometry with limits of quantification of 3.0 pg/mg. All maternal hair specimens were washed with methylene chloride before analysis, and when sufficient amounts of maternal hair were available, specimens also were analyzed without washing. Infant hair specimens were not washed.
Results: Buprenorphine concentrations were significantly greater in unwashed hair than washed hair (P = 0.031). Norbuprenorphine concentrations were significantly greater than buprenorphine concentrations in both maternal (P = 0.0097) and infant hair (P = 0.0033). There were statistically significant associations between the cumulative maternal dose of buprenorphine administered and the concentrations of buprenorphine (washed, P <0.0001; unwashed, P = 0.0004), norbuprenorphine (washed, P <0.0001; unwashed, P = 0.0005), and buprenorphine plus norbuprenorphine (washed, P <0.0001; unwashed, P = 0.0005) for both washed and unwashed maternal hair specimens. There was a significant positive association between concentrations of buprenorphine and norbuprenorphine in maternal hair (washed, P <0.0001; unwashed, P = 0.0003), a trend for this association in infant hair (P = 0.08), and an association between buprenorphine concentrations in maternal unwashed hair and infant hair (P = 0.0002). The buprenorphine:norbuprenorphine ratio increased in distal segments.
Conclusion: Buprenorphine treatment during gestation provides an opportunity for monitoring drug disposition in maternal and fetal tissues under controlled conditions.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For ethical and safety reasons, it is important to limit exposure to medications during pregnancy because of potential harm to the developing fetus. Untreated chronic illnesses, however, such as opioid dependence, are associated with increased morbidity and mortality. Buprenorphine treatment during gestation is similar to methadone in benefits for mother and child(3). Although benefits during pregnancy appear to outweigh risks, accumulating data regarding dose-concentration relationships are critical to understanding how this medication impacts maternal and neonatal outcomes.
Scheidweiler et al.(4) demonstrated the dose-related distribution of codeine, cocaine, and metabolites into human hair. However, the disposition of illicit drugs into maternal and fetal hair cannot be rigorously studied in pregnant women because of ethical and safety issues. Buprenorphine treatment during pregnancy presents an opportunity for studying the excretion of buprenorphine and its primary metabolite, norbuprenorphine, into maternal and fetal hair under controlled conditions, and for evaluating dose-concentration relationships in pregnant women and their infants.
Neonatal hair analysis has been evaluated for monitoring gestational exposure to drugs(5)(6). Hair analysis has advantages over blood and urine testing because exposure can be monitored over a period of several months(7)(8). Neonatal hair reflects drug exposure during the 3rd trimester, when hair begins to be formed, and remains positive for the drug under investigation for up to 3 months after birth(6).
Buprenorphine has an elimination half-life of about 37 h(9) and is metabolized by N-dealkylation to norbuprenorphine, which has an even longer elimination rate(10). Kintz(11) quantified buprenorphine and norbuprenorphine concentrations in hair from 3 individuals who used buprenorphine daily. Higher concentrations of parent compound than metabolite were observed in all 3 individuals. In a subsequent study of 14 buprenorphine users, Kintz et al.(12) found parent compound in the hair of all participants, but norbuprenorphine in only 11. Additionally, buprenorphine was present at concentrations approximately 2 to 18 times higher than metabolite. Maximum buprenorphine and norbuprenorphine concentrations were 0.59 and 0.15 ng/mg, respectively.
Tracqui et al.(13) analyzed hair from 6 participants treated with oral buprenorphine in a detoxification program. Concentrations ranged from 4 to 140 pg/mg and 0 to 67 pg/mg for buprenorphine and norbuprenorphine, respectively. In all cases, buprenorphine exceeded norbuprenorphine concentrations.
Wilkins et al.(14) reported buprenorphine hair concentrations for 12 participants receiving 8 mg of buprenorphine daily for up to 180 days. Buprenorphine or norbuprenorphine was detected in 11 of 12 individuals. In contrast to previous reports, norbuprenorphine was detected in higher concentrations than buprenorphine.
Vincent et al.(15) also reported higher norbuprenorphine concentrations in hair of 4 of 5 participants receiving buprenorphine. Hair specimens were prepared according to the Kintz et al.(12) solid-phase extraction and GC-MS methods. Cirimele et al.(16) attempted to resolve the discrepancy by analyzing the wash used to decontaminate specimens. In 40 of 66 washes, buprenorphine was present in higher quantities than norbuprenorphine, suggesting that the former was preferentially lost during washing.
Several additional aspects of buprenorphine therapy and pharmacokinetics remain unclear. There are no data relating the dose of buprenorphine administered in pregnancy to concentration of drug in maternal and fetal tissues and fluids. No studies have examined the association between cumulative buprenorphine dose and drug concentrations in hair. To evaluate the ability of hair analysis to predict magnitude of gestational drug exposure, we designed a study to investigate the relationship between dose of buprenorphine administered to 9 pregnant women and neonatal and maternal buprenorphine and norbuprenorphine concentrations in washed and unwashed hair.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inclusion criteria for pregnant women were age 21–40 years; estimated gestational age by sonogram of 16–30 weeks; DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, 4th Edition) diagnosis of current heroin dependence; request for maintenance pharmacotherapy; recent self-reported heroin use (use on more than 4 of the past 7 days); and an opiate-positive urine specimen at intake. Exclusion criteria were urine positive for undocumented methadone at enrollment, current DSM-IV diagnosis of alcohol abuse or dependence, self-reported regular use of benzodiazepines, current medication use for another Axis I disorder, presence of a serious concurrent medical illness, diagnosis of preterm labor, evidence of fetal malformation, positive HIV test, or positive sickle cell trait(2). Nearly 1500 women were screened, and only 57 qualified for this phase II study. Because of the experimental nature of the drug intervention, treatment was not permissible until later gestational ages. Strict inclusion and exclusion criteria were necessary for safety reasons, especially given the limited knowledge about buprenorphine during pregnancy. Of 57 consenting women, 30 were subsequently randomized to study treatment; 15 received methadone, and 15 received buprenorphine. Nine of 15 women completed the buprenorphine arm of the study. The final sample size enrolled in treatment at delivery was 9 women (8 African-American and 1 white) (Table 1
). The mean (SD) age of the participants was 30 (3.4) years (range 22–32 years). This study was part of a previously published larger trial(2).
|
urine drug screening at intake and during buprenorphine maintenance
The Dade Behring enzyme-multiplied immunoassay technique (Emit) was used at enrollment to screen participants for opiates, cocaine, benzodiazepines, cannabinoids, and methadone. During buprenorphine maintenance, urine specimens were collected 3 times per week and assayed for the above drug classes. One specimen, selected weekly at random, was sent for assay for amphetamines, cannabinoids by immunoassay and for antidepressants, barbiturates, opioids, phencyclidine, and other drug testing by thin-layer chromatography. Breath samples (Alco-Sensor III, Alcopro) were randomly obtained once weekly for monitoring alcohol ingestion.
dosing
The mean (SD) initial buprenorphine dose was 8.2 (2.1) mg/d (range 4–12 mg/d). Buprenorphine was administered sublingually with observation by research staff. Double-blind changes in medication dose were made through protocol-driven clinical decision criteria. Dose changes were limited to once every 2 weeks unless clinically indicated. To maintain blinding for research staff, dose changes were known only to pharmacy staff. Participants received 4–24 mg daily sublingual buprenorphine during pregnancy and up to 10 weeks after delivery for a mean (SD) period of 21.1 (4.8) weeks (range 14–30 weeks). At the time of delivery, mean (SD) buprenorphine dose was 18.7 (3.5) mg/d (range 14–24 mg/d)(2).
hair specimens
Drug disposition in hair occurs from multiple sources including blood, sweat, and sebum. In an attempt to account for total disposition of drug in hair from all sources, the entire hair specimen was analyzed. Maternal hair specimens were obtained from the posterior vertex approximately every 4 weeks. The 1st specimen analyzed from participants A–F (see Tables 2
and 4
, and Supplemental Data Table 1 that accompanies the online version of this article at http://www.clinchem.org/content/vol53/issue12 ) was obtained before dosing. The next specimen for these individuals and the 1st specimen for participants G–I were obtained, on average, 21 days after dosing was initiated. The specimens (n = 51) weighed 3.0–34.9 mg and represented up to 15 months hair growth. Hair specimens were cut into 3-cm segments and minced with scissors. Individual hair segments (n = 148) weighed 2.7–10.3 mg. Infant hair specimens (n = 4) were collected below the crown of the head within 48 h of birth.
|
|
Aliquots of each hair segment were placed into silanized glass vials. When sufficient specimen was available (n = 18), 2 aliquots of each segment were prepared to compare quantitative data for washed vs unwashed hair. Washing was performed with methylene chloride as per Cirimele et al.(16). Briefly, hair segments were washed twice with 2 mL of methylene chloride for 2 min at room temperature and allowed to evaporate to dryness. The 2nd aliquot, if available, was not washed before analysis. Infant hair specimens were analyzed unwashed because there was insufficient hair to test washed and unwashed, and it was not known if washing would remove excess amounts of drug, owing to the fine texture of hair.
digestion, extraction, and analysis by mass spectrometry
Buprenorphine-d4 and norbuprenorphine-d3 (1 ng/mg each) were added as internal standards to 3.5–10.1 mg of adult or 1.6–5.3 mg of infant hair. Hair was digested (solubilized) in 2 mol/L NaOH overnight at room temperature. Specimen pH was adjusted to 10.5 with 6 mol/L HCl, and extracted with n-butyl chloride/acetonitrile/ethylacetate (4:1:1 vol/vol). Chromatographic separation was achieved on a YMC ODS-AQ 2.0 x 150 mm S-3 120 Å reversed-phased HPLC column. The liquid chromatography–tandem mass spectrometer was operated in selected reaction monitoring mode with electrospray ionization. Four selected reactions were monitored (m/z): 468.2
396.1 for buprenorphine, 472.2400.1 for buprenorphine-d4, 414.2100.9 for norbuprenorphine, and 417.2100.9 for norbuprenorphine-d3. Analyte concentrations were determined from peak area of native drug to peak area of deuterated internal standard, and ratio comparison to the calibration curve generated from analysis of human hair fortified with known buprenorphine and norbuprenorphine and internal standard concentrations. Hair calibrators extracted in each batch were prepared by adding drug-free human hair at concentrations from 3 to 10 000 pg/mg. QC hair samples to assess accuracy and precision of each batch also were included at 25, 100, and 1000 pg/mg. Intra- and interassay imprecision were within 11% for buprenorphine and norbuprenorphine at these concentrations. QC samples fortified only with buprenorphine demonstrated no conversion to norbuprenorphine during digestion and extraction. Limits of quantification were 3.0 pg/mg of hair for buprenorphine and norbuprenorphine.
statistical analysis
Analyses were done using Statistical Analysis Systems (SAS) version 9 (SAS Institute). A Spearman correlation matrix was calculated to determine whether a correlation existed between cumulative maternal dose and buprenorphine or norbuprenorphine concentration in infant hair. The cumulative buprenorphine dose administered to the mother immediately before birth was used in each case.
Most analyses were performed using repeated measures linear regression (SAS Proc Mixed), making best use of available data. To test hypotheses of equality (comparing analyte concentrations in washed vs unwashed hair, comparing buprenorphine vs norbuprenorphine concentrations, comparing analyte ratios in maternal vs infant hair), intercept-only models were used, with difference score (washed minus unwashed, maternal minus infant) as the dependent variable. If the intercept was significantly different from zero, then a difference was considered to exist.
To determine whether there were associations between the cumulative maternal dose of buprenorphine and analyte concentrations, the cumulative dose was the independent variable in repeated measures linear regression having as dependent variables total buprenorphine, total norbuprenorphine, and total buprenorphine plus norbuprenorphine, for washed and unwashed hair. This examination was limited to maternal data only (n = 9).
Associations between concentrations of buprenorphine and norbuprenorphine in maternal (washed and unwashed) and infant hair were assessed with repeated measures linear regression with norbuprenorphine concentration as the dependent variable and buprenorphine concentration as the independent variable.
To determine whether the ratio of buprenorphine:norbuprenorphine in maternal hair changed according to segment, a repeated-measures linear regression model was fit with ratio as the dependent variable and hair segment as the independent variable. Segment was coded as a class variable to do pairwise comparisons, i.e., comparing each segment to the 1st hair segment with P values adjusted using the Dunnett-Hsu adjustment to keep overall alpha at 0.05.
Analysis of the relationship between concentrations of maternal buprenorphine, norbuprenorphine, or norbuprenorphine plus buprenorphine in hair and concentrations in infant hair was performed with repeated measures linear regression with maternal hair concentration as dependent variable and infant concentration as the independent variable.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Of the 18 unwashed specimens collected after dosing began, 17 were positive for buprenorphine [40.4 (28.7) pg/mg, range 0.9–111.0 pg/mg] and norbuprenorphine [563.1 (500.3) pg/mg, range 10.8–2018.6 pg/mg]. Concentrations of buprenorphine and norbuprenorphine were greater in unwashed hair: mean difference was 10.8 (15.6) pg/mg for buprenorphine (n = 16 pairs, t = –2.70, df = 7, P = 0.031) and 190.0 (345.7) pg/mg for norbuprenorphine (n = 17 pairs, t = –2.16, df = 7, P = 0.067). Ratios of buprenorphine and norbuprenorphine in washed [0.24 (0.44) pg/mg, range 0.04–1.93 pg/mg] and unwashed specimens [0.10 (0.10) pg/mg, range 0.04–0.49] were not significantly different (t = 0.27, df = 6, P = 0.80). All infant hair specimens were positive for buprenorphine and norbuprenorphine (Table 3
).
|
Norbuprenorphine concentrations were significantly greater than buprenorphine concentrations in maternal hair (washed, P = 0.0131; unwashed, P = 0.0097) and infant hair (P = 0.0033) (Tables 2
and 3
). Only patient G, a participant with blond hair, had buprenorphine concentrations exceeding those of norbuprenorphine. This individual had buprenorphine concentrations similar to dark-haired participants but much lower norbuprenorphine concentrations. Results from infant hair analyses should be considered tentative because these data were based on only 4 observations.
There were significant positive associations between cumulative maternal buprenorphine dose and total concentrations of buprenorphine (washed, P <0.0001; unwashed, P = 0.0004), norbuprenorphine (washed, P <0.0001; unwashed, P = 0.0005), and buprenorphine plus norbuprenorphine (washed, P <0.0001; unwashed, P = 0.0005) in whole hair specimens (all segments included), both washed and unwashed (Table 2
). These examinations were limited to maternal data only (n = 9). Significant relationships between cumulative maternal buprenorphine dose and analyte concentrations also were observed for several hair segments in washed hair (see Supplemental Data Table 1) and for the 1st segments from unwashed hair specimens.
There was a statistically significant association between concentrations of buprenorphine and norbuprenorphine in maternal hair (washed, P <0.0001; unwashed, P = 0.0003) and a trend for the association in infant hair (P = 0.08).
The buprenorphine:norbuprenorphine ratio increased as hair segments became more distal (P <0.0001) (see Table 4
and Supplemental Data Table 1). Ratios in segments 4 and 5 were significantly greater than in segment 1 (adjusted P <0.01), although segments 2 and 3 were not (adjusted P >0.05). Stratifying the data into homogeneous groups with respect to same number of hair segments (e.g., 2 segments, 3 segments) showed that the greatest increase was from segment 1 to segment 2 (the ratio more than doubled), then leveled off for subsequent segments (2 to 3, 3 to 4, etc.).
There was no correlation between cumulative maternal dose and buprenorphine or norbuprenorphine infant hair concentrations. There was an association between buprenorphine concentrations in unwashed maternal hair and infant hair (P = 0.0002); however, this association was not present for norbuprenorphine.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A small percentage (5.6%–8.9%) of washed and unwashed hair specimens were negative for either buprenorphine or norbuprenorphine. Generally, these samples were the 1st or 2nd specimens collected after dosing began. Some contributing factors include the fact that the hair root and follicle beneath the scalp may contain additional drug, but these concentrations were not measured, and if low concentrations of drugs were present, they may have been below the method limits of quantification of 3 pg/mg. Drug incorporated into the hair root from blood generally does not reach the level of the scalp for 7 to 10 days; however, drug present in sweat may be deposited on hair within minutes to hours. Additionally, maternal plasma concentrations of buprenorphine and norbuprenorphine are unknown and may have varied considerably within and between individuals, and the minimum detectable dose in hair has not been established.
A quantitative relationship between cumulative maternal dose of buprenorphine and concentration of buprenorphine or norbuprenorphine in infant hair was not established, possibly because of small sample size (n = 4). Other factors were that infant hair forms during the 3rd trimester and as a result does not reflect the entire buprenorphine administration period, and fetal hair is exposed to drug in amniotic fluid.
Norbuprenorphine concentrations were significantly greater than buprenorphine concentrations in both maternal and infant hair. This finding is consistent with the work of Wilkins et al.(14) and Vincent et al.(15), but contrasts with the results of Kintz(11) and Tracqui et al.(13). Cirimele et al.(16) attributed this discrepancy to loss of buprenorphine to a greater extent during methylene chloride washing. A limitation of the present study was that the discarded methylene chloride wash was not analyzed for the presence of buprenorphine or norbuprenorphine. Although concentrations of buprenorphine and norbuprenorphine were higher in unwashed hair in this study, the explanation of Cirimele et al.(16) cannot account for greater concentrations of the metabolite, because statistically significant greater concentrations of norbuprenorphine than buprenorphine were found in both washed and unwashed maternal hair. Additionally, in the present study, there was no statistically significant difference between ratios of buprenorphine:norbuprenorphine in washed compared to unwashed hair. Furthermore, Kintz(11) and Tracqui et al.(13) used a methylene chloride wash before analysis, as was done in this investigation. It also is unlikely that the acidic extraction used by Vincent et al.(15), Kintz(11), and Tracqui et al.(13) could account for these discrepant results, because Vincent et al. also reported higher norbuprenorphine hair concentrations. Specimen preparation procedures in Wilkins et al.(14) and the current protocol differed from other studies by including dissolution of hair with strong base, and extraction of buprenorphine and norbuprenorphine at pH 10.5. This pH was selected to enhance recovery of norbuprenorphine. We found that acid digestion diminished the amount of norbuprenorphine recovered. Because the only participant in this study with buprenorphine concentrations exceeding norbuprenorphine concentrations was one with blond hair, the possibility that norbuprenorphine is preferentially incorporated into pigmented hair should be considered. All participants in the study by Wilkins et al.(14) had dark-colored hair; hair color was not described for the other investigations. Analysis of melanin content in the present study was not performed.
The ratio of buprenorphine:norbuprenorphine increased distally along the hair shaft. This distribution could indicate that norbuprenorphine is less tightly bound to hair than buprenorphine, and as a result there is increased loss of norbuprenorphine in distal segments from routine hygienic care. Other explanations are that there is enhanced transport of buprenorphine along the hair follicle or greater deposition and adsorption of buprenorphine from sweat.
This study demonstrated that buprenorphine and norbuprenorphine are incorporated into fetal hair in utero and that analysis of hair can be used to study dose-concentration relationships in pregnant women on an individual basis. An association between concentrations of these analytes in hair has been described for the 1st time. The fact that an association was established between the cumulative dose of buprenorphine and concentrations of drug and metabolite in hair suggests that it may be possible to estimate degree of exposure of individual women during pregnancy using analyte concentrations. Additionally, data indicate that neonatal hair concentrations could be used to qualitatively identify drug exposure during gestation, but not the degree of exposure.
|
Acknowledgments |
|---|
Financial disclosures: None declared.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
