The association between cytochrome P450-2D6 genotype and prescription of antiparkinsonian drugs in hospitalized psychiatric patients using antipsychotics: a retrospective follow-up study.

Full text

Page 1

serum concentration in relation to the
prescribed dose as it eases the process of
distinguishing between ultrarapid me-
tabolism and noncompliance. It should,
however, be pointed out that genotyping
does not identify all subjects who are
phenotypically ultrarapid metabolizers.
In previous studies, a relatively small
proportion of those with the ultrarapid
metabolizer phenotype is identified by
genotyping for the duplicated or multi-
duplicated allele,9,10depending on the
definition of the cutoff point between
phenotypically extensive and phenotypi-
cally ultrarapid metabolizers. However,
if the subject is identified as an interme-
diate or poor metabolizer by genotyping,
noncompliance is obviously the most
likely explanation. It would be of in-
terest to conduct a prospective study ge-
notyping all subjects with paroxetine
concentrations in the lower range, for
example, below the 5th or the 10th per-
centiles, to further elucidate this issue.
Cu ¨neyt Gu ¨zey, MD*y
Olav Spigset, MD, PhD*y
*Department of Clinical Pharmacology,
St Olav University Hospital, Trondheim,
Norway and yDepartment of Laboratory
Medicine, Children’s and Women’s Health,
Norwegian University of Science and
Technology, Trondheim, Norway
REFERENCES
1. Charlier C, Pinto E, Ansseau M, et al.
Relationship between clinical effects, serum
drug concentration, and concurrent drug
interactions in depressed patients treated with
citalopram, fluoxetine, clomipramine, parox-
etine or venlafaxine. Hum Psychopharmacol
Clin Exp. 2000;15:453–459.
2. Norman C, Ho ¨rn M, Hummel B, et al.
Paroxetine in major depression: correlating
plasma concentration and clinical response.
Pharmacopsychiatry. 2004;37:123–126.
3. Reis M, A˚berg-Wistedt A, A˚gren H, et al.
Serum disposition of sertraline, N-desmethyl-
sertraline and paroxetine: a pharmacokinetic
evaluation of repeated drug concentration
measurements during 6 months of treatment
for major depression. Hum Psychopharmacol
Clin Exp. 2004;19:283–291.
4. Sindrup SH, Brøsen K, Gram LF, et al. The
relationship between paroxetine and the
sparteine oxidation polymorphism. Clin Phar-
macol Ther. 1992;51:278–287.
5. Sawamura K, Suzuki Y, Someya T. Effects
of dosage and CYP2D6-mutated allele on
plasma concentration of paroxetine. Eur J
Clin Pharmacol. 2004;60:553–557.
6. Charlier C, Broly F, Lhermitte M, et al.
Polymorphisms in the CYP 2D6 gene:
association with plasma concentrations of
fluoxetine and paroxetine. Ther Drug Monit.
2003;25:738–742.
7. Sachse C, Brockmo ¨ller J, Bause S, et al.
Cytochrome P450 2D6 variants in a Cauca-
sian population: allele frequencies and phe-
notypic consequences. Am J Hum Genet.
1997;60:284–295.
8. Gu ¨zey C, Aamo T, Spigset O. Risperidone
metabolism and the impact of being a
cytochrome P450 2D6 ultrarapid metabolizer.
J Clin Psychiatry. 2000;61:600–601.
9. Dahl M-L, Johansson I, Bertilsson L, et al.
Ultrarapid hydroxylation of debrisoquine in a
Swedish population. Analysis of the molecu-
lar genetic basis. J Pharmacol Exp Ther.
1995;274:516–520.
10. Bathum L, Johansson I, Ingelman-Sundberg
M, et al. Ultrarapid metabolism of sparteine:
frequency of alleles with duplicated CYP2D6
genes in a Danish population as determined
by restriction fragment length polymorphism
and long polymerase chain reaction. Phar-
macogenetics. 1998;8:119–123.
The Association
Between Cytochrome
P450-2D6 Genotype
and Prescription of
Antiparkinsonian
Drugs in Hospitalized
Psychiatric Patients
Using Antipsychotics
A Retrospective Follow-
up Study
To the Editors:
Cytochrome P450-2D6 (CYP2D6)
genotype may be associated with several
adverse effects of antipsychotic drugs.
Approximately 5% to 10% of the white
population can be classified as poor me-
tabolizers (PM) by diminished CYP2D6
activity. These patients may be at higher
risk of dose-dependent adverse effects
because most antipsychotic drugs are at
least partly metabolized by CYP2D6.1
Oneofthemostfrequentlyoccurringdose-
dependent adverse effects in users of
antipsychotic drugs are some of the
extrapyramidal syndromes (EPS) (ie, par-
kinsonism, akathisia, and dystonia). In
several studies, an association between
CYP2D6 genotype and EPS has been
established, although results are con-
flicting.1When a patient experiences
EPS, the treating psychiatrist can decide
FIGURE 1.
mg/d, based on a total of 1482 routine therapeutic drug-monitoring samples. The
numbers of samples included were 45 for the dose 10 mg/d, 578 for the dose 20 mg/d,
159 for the dose 30 mg/d, 470 for the dose 40 mg/d, 41 for the dose 50 mg/d, and 154
for the dose 60 mg/d. Other doses were used in 35 cases. For clarity, the 50th, 75th, and
90th percentile values are omitted for the higher doses. The serum concentrations of the 2
CYP2D6 ultrarapid metabolizers are displayed with the symbols X (first patient, 2 samples)
and + (second patient, 3 samples).
Distribution of serum concentrations of paroxetine at doses of 10 to 60
212
? 2006 Lippincott Williams & Wilkins
Letters to the Editors Journal of Clinical Psychopharmacology ? Volume 26, Number 2, April 2006
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Page 2

to switch to another antipsychotic drug.
However, in a previous study published in
this journal, we did not find an association
between CYP2D6 genotype PM and
more frequent switching of antipsychotic
drugs compared with extensive metabo-
lizers (EM) for CYP2D6. As a possible
explanation, we suggested that psychia-
trists prefer to treat EPS (parkinsonism,
akathisia, and dystonia, but not tardive
dyskinesia) with antiparkinsonian drugs
rather than to switch to another drug.2The
objective of this study is to test this
hypothesis by investigating whether in a
group of antipsychotic users, PMs for
CYP2D6 are more frequently treated with
antiparkinsonian drugs compared with
EMs for CYP2D6.
METHODS
The study was conducted in 2
psychiatric hospitals with approximately
450 beds for long-stay patients with
chronic psychiatric disorders. The study
protocol was reviewed and approved by
an independent medical ethical commit-
tee (TPWO, Arnhem, The Netherlands).
A retrospective follow-up design
was used to assess the association be-
tween CYP2D6 genotype and EPS as an
adverse effect of antipsychotic drugs.
The potential study population consisted
of 138 long-stay patients of whom the
CYP2D6 genotype was determined in a
previous study.2
Patients were eligible for the
present study if at least 1 antipsychotic
drug that is metabolized by CYP2D6
was started at least 30 days after the
start of the observation period and at
least 90 days before the end of the
observation period. This resulted in the
inclusion of 129 patients. Prescription
data were collected from April 1988
until February 2005. Primary end point
of this study was the start of an
anticholinergic antiparkinsonian drug
(either biperidene or trihexyphenidyl)
at least 7 days after the start of a drug
episode of an antipsychotic drug at least
partly metabolized by CYP2D6. Biper-
idene and trihexyphenidyl are indicated
for the treatment of EPS caused by
antipsychotic drugs. The interval of
7 days was chosen to prevent inclusion
of prescriptions of antiparkinsonian drugs
for prophylactic use. Antipsychotic drug
episodes during which antiparkinsonian
drugs were started within 7 days after
the start of the antipsychotic drug were
excluded from analysis. In case of treat-
ment with 2 or more antipsychotic drugs
at the same time, only the drug epi-
sode with a starting date closest to the
start of biperidene or trihexyphenidyl
was taken into account.
We classified the CYP2D6 meta-
bolic pathway of each prescribed anti-
psychoticdrugaccordingtotheavailable
published evidence up to April 2005.
Antipsychotic drugs for which an in vivo
relationship between CYP2D6 genotype
andpharmacokineticpropertieshas been
documented were classified as primary
CYP2D6 drugs (haloperidol, perphena-
zine, risperidone, sertindol, thioridazine,
and zuclopenthixol). Antipsychotic drugs
for which either in vivo evidence exists
that these are partly metabolized by
CYP2D6 or in vitro evidence of metabo-
lism by CYP2D6 was available were
classifiedaspartlyCYP2D6drugs(chlor-
promazine, clozapine, pimozide, olanza-
pine, and quetiapine). Episodes of other
antipsychotic drugs were not included.
Primarydeterminant
CYP2D6 genotype. The CYP2D6 geno-
type was determined by polymerase
chain reaction (PCR)–restriction fragment
length polymorphism
CYP2D6*4(B),CYP2D6*6(T),CYP2D6*7(E),
and CYP2D6*8(G) were investigated
using a long-distance and multiplex
PCR as described by Stuven et al.3The
presence of gene duplication that may
lead to ultrarapid metabolism was ana-
lyzed by an allele-specific PCR and was
performed as described by Lovlie et al.4
Patients were defined as PM if they were
homozygous or heterozygous for non-
coding alleles. Patients were defined as
ultrarapid metabolizers (UM) if gene
duplication was detected and mutant
alleles were absent. All other patients
were classified as EM.
The relative risk of treatment with
biperidene or trihexyphenidyl was esti-
matedwithalogisticmixed-effectsmodel
and expressed as an odds ratio (OR) with
a 95% confidence interval (CI) using EM
asthe reference group.Themixed-effects
model was used because there was more
than 1 drug episode per patient, and
therefore, the data were not independent.
Mixed-effects models make adjustments
for multiple measurements in the same
wasthe
CYP2D6*3(A),
subject. Data were analyzed for all
included episodes of antipsychotic drug
treatment (primary CYP2D6 drugs and
partly CYP2D6 drugs) and separately for
antipsychotics primarily metabolized by
CYP2D6. Finally, data were investigated
for the potential confounders age, sex,
and prescribed daily dose. A P value
of 0.05 or less was regarded as signifi-
cant. Data were analyzed using S-plus
6.2 CorrelatedData Library (insightful),
www.insightful.com.5,6
RESULTS
In our study population (N = 129),
we found poor or ultrarapid metabolism
in 12.4% of the included patients (10.1%
PM, n = 13, and 2.3% UM, n = 3). The
included patients were mainly white
(>95%), with a mean age of 54 years
(SD, 16.3), and 46.5% (n = 60) were
men. In total, there were 267 drug
episodes of antipsychotic drugs at least
partly metabolized by CYP2D6 during
the observation period. After exclusion
of drug episodes with prophylactic use,
164 drug episodes were available for
analysis. Antiparkinsonian drugs were
started in 32 (20%) drug episodes of
antipsychotic drugs (primary CYP2D6
drugs, 69%, n = 22; partly CYP2D6
drugs, 31%, n = 10). Furthermore,
antiparkinsonian drugs were started in
41% (9/22), 16% (22/139), and 33% (1/3)
of the observed drug episodes in PMs,
EMs, and UMs, respectively.
Table 1 shows that for antipsy-
choticdrugs,CYP2D6genotypePMwas
associated with a more frequent pre-
scription of antiparkinsonian drugs (OR,
3.68; 95% CI, 1.40–9.65) in patients
treated with antipsychotics primary or
partly metabolized by CYP2D6. Analy-
sis restricted to antipsychotics primarily
metabolized by CYP2D6 showed a non-
significant association with more fre-
quent prescription of antiparkinsonian
drugs (OR, 2.38; 95% CI, 0.69–8.28).
The number of drug episodes in patients
with CYP2D6 genotype UM was inade-
quate to allow analysis. Adjustments for
age,sex,andprescribeddailydosedidnot
significantly change the results obtained.
DISCUSSION
In this retrospective follow-up
study, we found that in daily clinical
? 2006 Lippincott Williams & Wilkins
213
Journal of Clinical Psychopharmacology ? Volume 26, Number 2, April 2006Letters to the Editors
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Page 3

practice, PMs for CYP2D6 treated with
antipsychoticsuseantiparkinsoniandrugs
more frequently compared with EMs for
CYP2D6. In psychiatric patients, anti-
parkinsonian drugs are almost exclu-
sively used for the treatment of EPS
(parkinsonism, akathisia, and dystonia,
but not tardive dyskinesia) and can
therefore be considered a valid marker
of EPS.
There are limitations to our find-
ings. First, we could only study 3 patients
classified as UM and 13 patients classi-
fied as PM for CYP2D6. Despite these
low numbers, significant results were
obtained for PMs, but these need confir-
mationinalargernumberofpatients.The
number of UMs and the corresponding
records of antiparkinsonian drugs were
too low to allow formulation of any
conclusions. Second, the reason for
prescribing antiparkinsonian drugs was
not recorded. Antiparkinsonian drugs can
be prescribed for prophylactic use after
prescription of antipsychotic drugs with a
high potential of EPS. We prevented the
inclusion of prophylactic use of antipar-
kinsonian drugs in the analysis by
exclusion of drug episodes of antiparkin-
sonian drugs within 7 days after starting
an antipsychotic drug. Third, we could
only find a significant association be-
tween CYP2D6 genotype and prescrip-
tion of antiparkinsonian drugs in patients
using antipsychotics primarily or partly
metabolized by CYP2D6. The analysis
that was restricted to antipsychotics
primarily metabolized by CYP2D6 only
showedthesameelevatedrisk,butdidnot
havesufficientpower tofinda significant
association.Futurestudiesshouldinclude
more patients (~250–300) to allow ade-
quateanalysisofdatathatarerestrictedto
antipsychotics primarily metabolized by
CYP2D6.
The findings of previous studies
investigating the association between
CYP2D6 genotype and EPS are inconsis-
tent. Several studies found a statistically
significantassociation,1,7,8butotherstud-
ies could not replicate these results.1,9,10
Schillevoort et al8found an association
between antiparkinsonian
CYP2D6 genotype only in patients using
antipsychotics primarily metabolized by
CYP2D6. However, this result was estab-
lished in a population screened for
CYP2D6 upon admission, and therefore,
the treating psychiatrist could consider
this information. In that study, it was pos-
sible that the treating psychiatrist chose
to treat EPS with antiparkinsonian drugs
in PMs in the case of antipsychotics pri-
marily metabolized by CYP2D6, but to
switch to another antipsychotic in the case
of antipsychotics not primarily metabo-
lized by CYP2D6 because the cause of
EPS is les likely, thereby underestimating
the real risk in PMs. In our population, the
CYP2D6 genotype was unknown to the
treating psychiatrist.
The increased prescription rates
of antiparkinsonian drugs in PMs for
CYP2D6 can be seen as a determinant
for unsatisfactory response in these pa-
tients. In an earlier study, we investigated
whether PMs for CYP2D6 switched more
often to other antidepressant drugs or anti-
psychotic drugs compared with EMs for
CYP2D6. We found an association be-
tween CYP2D6 genotype PM and more
frequent switching of antidepressants, but
we could not find this association for
switching of antipsychotic drugs.2These
results suggested that the treating psy-
chiatrist probably chose to treat EPS with
antiparkinsonian drugs, instead of switch-
ing to another drug in patients using anti-
psychotic drugs.
Taken together, the CYP2D6 ge-
notypePMisassociatedwithanincreased
risk of EPS as measured by a more
frequent prescription of antiparkinsonian
drugs compared with EMs for CYP2D6.
Genotyping psychiatric patients for
CYP2D6 before starting pharmacother-
drugsand
apy may identify patients at higher risk
for dose-dependent adverse events. If
CYP2D6 genotype information is avail-
able before starting pharmacotherapy, an
individualized advice for the choice of
drug and dosage is possible, thereby
reducingtheriskofEPSandthetreatment
of antiparkinsonian drugs thereof.
Hans Mulder, PharmD*y
Frederik W. Wilmink, MD, PhDz
Svetlana V. Belitser*
Antoine C.G. Egberts, PhD*
*Department of Pharmacoepidemiology and
Pharmacotherapy, Utrecht Institute for
Pharmaceutical Sciences, Utrecht
University, Utrecht; yDepartment of Clinical
Pharmacy, Wilhelmina Hospital, Assen and
zPsychiatric Hospital GGZ Drenthe, Assen,
The Netherlands
hans.mulder@wza.nl
ACKNOWLEDGMENTS
This study was financially sup-
ported by the Psychiatric Hospital GGZ
Drenthe, Assen, The Netherlands.
REFERENCES
1. Dahl ML. Cytochrome p450 phenotyping/
genotyping in patients receiving antipsy-
chotics: useful aid to prescribing? Clin
Pharmacokinet. 2002;41(7):453–470.
2. Mulder H, Wilmink FW, Beumer TL, et al.
The association between cytochrome P450
2D6 genotype and prescription patterns of
antipsychotic and antidepressant drugs in
hospitalized psychiatric patients: a retrospec-
tive follow-up study. J Clin Psychopharma-
col. 2005;25(2):188–191.
3. Stuven T, Griese EU, Kroemer HK, et al.
Rapid detection of CYP2D6 null alleles by
longdistance- and
chain reaction. Pharmacogenetics. 1996;6(5):
417–421.
4. Lovlie R, Daly AK, Molven A, et al.
Ultrarapid metabolizers of debrisoquine: char-
acterization and PCR-based detection of al-
leles with duplication of the CYP2D6 gene.
FEBS Lett. 1996;392(1):30–34.
5. Pinheiro JC, Bates DM. Mixed-effects Models
in S and S-plus. Berlin: Springer; 2000.
6. Shao J, Tu D. The Jackknife and Bootstrap.
Berlin: Springer; 1995.
7. Patsopoulos NA, Ntzani EE, Zintzaras E,
et al. CYP2D6 polymorphisms and the risk of
tardive dyskinesia in schizophrenia: a meta-
analysis. Pharmacogenet Genomics. 2005;15(3):
151–158.
8. Schillevoort I, De Boer A, Van der Weide J,
et al. Antipsychotic-induced extrapyramidal
syndromes and cytochrome P450 2D6 geno-
type: a case-control study. Pharmacogenetics.
2002;12(3):235–240.
multiplex-polymerase
TABLE 1. Relative Risk for Treatment with Antiparkinsonian Drugs in Users of
Antipsychotics
Antipsychotics
Drug Episodes with Start of
Antiparkinsonian
Drug Treatment (%)
OR (95% CI) of
Treatment with
Antiparkinsonian Drugs
EM
PM
16 (22/139)
41 (9/22)
1 (reference)
3.68 (1.57–8.65)
214
? 2006 Lippincott Williams & Wilkins
Letters to the EditorsJournal of Clinical Psychopharmacology ? Volume 26, Number 2, April 2006
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Page 4

9. Scordo MG, Spina E, Romeo P, et al.
CYP2D6 genotype and antipsychotic-induced
extrapyramidal side effects in schizophrenic
patients. Eur J Clin Pharmacol. 2000;56:
679–683.
10. Andreassen OA, MacEwan T, Gulbrandsen
AK, et al. Non-functional CYP2D6 alleles
and risk for neuroleptic-induced movement
disorders in schizophrenic patients. Psycho-
pharmacology. 1997;131(2):174–179.
Treatment of
Neuroleptic-induced
Tardive Dyskinesia
with Levetiracetam
A Case Series
To the Editors:
Neuroleptic-induced tardive dys-
kinesia (TD) is a persistent, sometimes
irreversible, side effect of neuroleptic
therapy characterized by abnormal move-
ments, including lingual and orofacial
dyskinesia, grimacing, tics, choreic move-
ments, athetosis, and dystonia.1Al-
though atypical neuroleptics (eg, clo-
zapine, risperidone, olanzapine) are less
likely to produce neuroleptic-induced
TD than typical agents,2,3neuroleptic-
induced TD has been reported in ap-
proximately 1% of patients treated with
atypical agents.4,5 The pathophysiol-
ogy of TD remains poorly under-
stood; however, numerous theories have
been proposed, including dopamine re-
ceptor supersensitivity6
tion of g-aminobutyric acid–mediated
neurotransmission.7
Levetiracetam is an antiepileptic
drug (AED) with novel mechanisms of
action that have not been fully eluci-
dated.8Levetiracetam was considered a
candidate to reduce abnormal move-
ments in TD.9In addition, levetiracetam
has been extensively used as an adjunc-
tive therapy because of its favorable
safety profileand pharmacokinetic prop-
erties, including minimal hepatic me-
tabolism associated with a low risk of
drug interactions.8
In this series of patients, levetir-
acetam was added to each subject’s
usual therapeutic regimen in an open-
label, naturalistic design. A total of 17
patients were enrolled in the order of
and dysfunc-
their presentation at the DeKalb Com-
munity Service Board (Atlanta, Ga)
within a 6-month period. Patients were
provided with a complete description of
the trial, and informed consent was
obtained. Seven patients were treated on
an inpatient basis, and 10 were treated
on an outpatient basis. Eight men and
9 women, ranging in age from 24 to
70 years (mean, 50 years), were included
in this analysis. Of these 17 patients,
13 were diagnosed with schizophrenia,
3 with schizoaffective disorder, and 1
with bipolar disorder.
No changes were made in the
patients’ antipsychotic treatment regi-
mens during the course of the trial.
Drugs used during the trial included
haloperidol (n = 8), risperidone (n = 4),
clozapine (n = 1), olanzapine (n = 3),
fluphenazine (n = 3), aripiprazole (n =
2), and quetiapine (n = 2). Other
concomitant drugs included AEDs (val-
proic acid, n = 2; carbamazepine, n = 2;
gabapentin, n = 1), other drugs to treat
neuroleptic-induced side effects (aman-
tadine, n = 1; benztropine, n = 4), and
antidepressants or
(bupropion, n = 2; buspirone, n = 1;
escitalopram, n = 2; mirtazapine, n = 1;
and lithium, n = 1).
Levetiracetam was added to each
patient’s baseline treatment regimen,
starting at a dose of 250 mg BID. The
dose of levetiracetam was titrated to
response based upon the Abnormal
InvoluntaryMovement
(AIMS) and its tolerability, with a more
rapid titration schedule for inpatients
than for outpatients.
AIMS was used to assess clinical
response to treatment with levetirace-
tam. Using a ranking of 0 (none) to
5 (severe), the scale evaluates facial,
oral, extremity, and trunk movements.
Global judgments of the severity, inca-
pacitation, and patient awareness of
abnormal movements are also included.
The patient’s dental status was also
assessed. The AIMS scores of treatment
with levetiracetam were measured at
baseline and at least once thereafter at
various times on days 1 to 196, the
duration of the trial. The author, trained
and experienced with the use of AIMS,
was the sole rater.
The baseline AIMS score was
also used to diagnose TD based on the
moodstabilizers
Scale 1–10
Schooler-Kane criteria, which require a
sum of 4: a score of greater than 3 on
any one of the AIMS items 1 to 7 plus a
score of greater than 1 on another cate-
gorical item.10Adverse effects were
monitored and recorded during the trial.
The average dose of levetiracetam
was2089mg/d(range,1000–4000mg/d).
All patients experienced some im-
provement in symptoms of neuroleptic-
induced TD with treatment with levetir-
acetam. The AIMS scores improved, on
average, by 80% (range, 37%–100%).
The average reduction in AIMS score,
from 18.3 to 4.0, was statistically
significant using the last observation
carried forward Wilcoxon matched pairs
signed rank test11(P = 0.00001526)
(Fig. 1). The mean time to maximum
improvement was 22.8 days (range, 3–
105 days).
Overall, levetiracetam was gener-
ally well tolerated. Three inpatients
reported ‘‘mild’’ sedation, but no dose
adjustment was necessary. Dose was
reduced from 4000 to 3000 mg/d for 1
outpatient who reported ‘‘feeling de-
pressed.’’ No patient discontinued par-
ticipation secondary to side effects. Five
outpatients were lost to follow-up.
TD is a serious side effect of neu-
roleptic therapy that can impair a pa-
tient’s quality of life and interfere with
compliancewithcontinuedtreatment—a
dangerous consequence for patients who
requirechronictherapy.Althoughtheuse
of atypical neuroleptic drugs instead of
typical agents may lower the incidence
of TD,2,4,12TD can still develop.2,4,12
At present, TD induced by neuroleptic
agents remains a serious concern,13and
well-accepted effective treatments for
this condition are lacking.
FIGURE 1.
score from baseline to maximal improve-
ment (P = 0.00001529).
Change in average AIMS
? 2006 Lippincott Williams & Wilkins
215
Journal of Clinical Psychopharmacology ? Volume 26, Number 2, April 2006Letters to the Editors
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.