Original Paper
Received: November 28, 2003
Nephron Clin Pract 2004;97:c83–c89
DOI: 10.1159/000078635
Pharmacokinetics of Bupropion and Its
Metabolites in Haemodialysis Patients
Who Smoke
A Single Dose Study
Simon P.R. Worrall a Michael K. Almond a Soraya Dhillon b
a Southend Hospital NHS Trust, and b The School of Pharmacy, University of London, London, England
Key Words
Bupropion W Hydroxybupropion W Threohydrobupropion W
Metabolites W Haemodialysis W Pharmacokinetics
Abstract
To date, no study has investigated the effects of bupro-
pion (BP) in renal-impaired humans. This study aims to
identify the pharmacokinetics of BP and metabolites in
haemodialysis patients who smoke, determine whether
haemodialysis affects BP and metabolite clearance, and
suggest the BP dose in haemodialysis. The pharmacoki-
netics of BP and two of its major metabolites, hydroxybu-
propion (HB) and threohydrobupropion (TB) were stud-
ied in 8 smokers with ESRD receiving haemodialysis.
Following a single oral dose of 150 mg bupropion hydro-
chloride sustained-release, blood samples were taken
over 7 days, which were assayed using HPLC-mass spec-
trometry. Pharmacokinetic analysis was undertaken by
non-linear regression using MWPharm. The BP results
were similar to those for individuals with normal renal
function. The metabolites demonstrated increased areas
under the curve, indicating accumulation. Dialysis clear-
ance of HB is unlikely. The results suggest significant
accumulation of the metabolites in renal failure. Clarifi-
cation of the clinical importance of the metabolites and
© 2004 S. Karger AG, Basel
ABC 1660–2110/04/0973–0083$21.00/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/nec
Accepted: March 1, 2004
toxic plasma levels is required. The effects of haemodial-
ysis on BP and metabolites require further study. A dose
of 150 mg bupropion every 3 days in patients receiving
haemodialysis is more appropriate than the current
manufacturer’s recommendation (in renal impaired pa-
tients) of 150 mg daily. A multi-dose study is required.
Copyright © 2004 S. Karger AG, Basel
Introduction
Bupropion is the first licensed non-nicotine pharmaco-
logical therapy for smoking cessation in the UK. Smoking
is the single largest cause of preventable illness and pre-
mature death in the United Kingdom [1]. Over 120,000
people die each year in the UK due to smoking-related
diseases. Smoking cessation is one of the most cost-effec-
tive healthcare interventions. Smoking further increases
the cardiovascular risk, which is already raised in renal
patients. Data from the USA shows smokers on haemodi-
alysis (compared with non-smoking haemodialysis pa-
tients) have an increased risk of developing heart failure
(adjusted hazards ratio 1.59), peripheral vascular disease
(1.68) and of dying (1.37) [2].
Whilst bupropion is a well-tolerated drug, it can have
significant side effects in predisposed individuals, such as
Simon Worrall
Renal Unit
Southend Hospital NHS Trust, Prittlewell Chase
Westcliff-on-Sea SS0 0RY Essex (UK)
Tel. +44 1702 221098, Fax +44 1702 221099, E-Mail simon.worrall@southend.nhs.uk
Table 1. Inclusion/exclusion criteria
Inclusion
Adult
Haemodialysis patient at study hospital
Established on haemodialysis (1 3 months)
Regular smoker
No previous use of bupropion
No current/past history of mental health
illness/eating disorder
No allergy to Zyban excipients
those where accumulation can occur. The major concern
with bupropion is the risk of seizures, probably related to
peak plasma concentrations. The incidence of seizures is
approximately 0.1% [3].
There are no published studies into the kinetics of
bupropion in renal impaired patients. In renal failure the
manufacturer suggests a reduced maximum dose of
150 mg per day, whilst others have recommended no dose
reductions [4]. However, the suggestion is that accumula-
tion of the bupropion and metabolites may occur in the
presence of impaired renal function [5–7].
Different formulations of bupropion exist throughout
the world; we used the prolonged release preparation of
Zyban (GlaxoSmithKline), which is the only licensed for-
mulation of bupropion in the UK.
The aims of this study were to identify the pharmaco-
kinetic profile of bupropion and metabolites in a small
cohort of smokers on haemodialysis. We also wanted to
determine the extent to which haemodialysis affects the
clearance of bupropion and metabolites. With this infor-
mation, our final aim was to determine the appropriate
dose of bupropion for haemodialysis patients.
Bupropion is extensively metabolized to hydroxybu-
propion, threohydrobupropion and erythrohydrobupro-
pion. The metabolites, especially hydroxybupropion, are
active and probably influence efficacy and toxicity [8–
10].
Bupropion hydrochloride is metabolised, by the cyto-
chrome P450 isoenzyme CYP2B6, to the major active
metabolite, hydroxybupropion.
Whilst not metabolised by the CYP2D6 isoenzyme,
bupropion and hydroxybupropion are inhibitors of the
CYP2D6 pathway. Drugs, which have the capacity to
induce or inhibit metabolism, such as carbamazepine
should be used cautiously with bupropion as they may
adversely affect its efficacy or safety.
c84 Nephron Clin Pract 2004;97:c83–c89
Exclusion
Acute illness
Unable to give consent
Liver disease
History of head trauma/epilepsy/CNS tumour
Acute drug withdrawal
Medication likely to lower seizure threshold
Alcohol abuse
Poorly controlled diabetes
Use of stimulants/anorectic agents
Method
From a haemodialysis population of 110 attending the renal unit
at Southend Hospital NHS Trust, 25 smokers were identified. Of
these, 2 were excluded: one due to dementia and the second due to
concomitant medication. The remainder was invited to take part
after informal discussion and provision of an information leaflet. 8
adult end-stage renal failure patients (1 female, 7 male) on haemodi-
alysis, who fulfilled the entry criteria, were recruited.
The study protocol was reviewed and approved by the local
research and ethics committee. Inclusion/exclusion criteria for the
study can be seen in table 1. Each patient who participated in the
study was free to withdraw at any point. Written consent was
obtained from patients after they had received written and oral infor-
mation.
Patients received their usual prescribed medication during the
study. Patients were permitted to smoke cigarettes as they wished
during their involvement in the study. All patients had undergone a
recent check of their electrolytes, liver function and blood count prior
to the study as part of their general hospital care.
On the morning of day 1, the fasted patient attended the renal
unit. Unless the patient had central line access, a physician inserted a
peripheral 18G venous cannula. A baseline blood sample was ob-
tained. A single bupropion hydrochloride-sustained release 150 mg
tablet (Zyban, GlaxoSmithKline) was taken orally by the patient
with a drink of water or tea. Blood samples were then taken via the
venous cannula at the following times (in hours): 0.5, 1.0, 1.5, 2.0,
2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 6.0, 8.0 and 12.0. The patient was encour-
aged to fast until lunchtime. If there was a venous cannula in situ
after the 12-hour sample, the cannula was removed.
On day 2, the patient presented to the unit for dialysis. Once the
fistula was needled, a pre-dialysis sample was taken. All patients in
the study were given a 4-hour dialysis session, using a 250 ml/min
pump speed on a DCA 170G dialyser (Dicea high-performance cellu-
lose diacetate hollow fibre dialyser, Baxter Healthcare Corporation,
Renal Division, Mcgaw Park, Ill., USA). Approximately half way
through the session, a blood sample was taken directly from the
arterial and venous ports on the dialyser machine. At the end of dial-
ysis, a post-dialysis sample was taken.
On day 3, the patient had a venous sample taken approximately
24 h after dialysis.
On day 4, the patient presented to the renal unit for dialysis. The
same process as on day 2 was followed, except no samples were taken
from the arterial and venous ports during dialysis.
Worrall/Almond/Dhillon
 Table 2. Patient demographic details

Patient Age Gender Height Dry weight 24-hour urine Cause of renal failure
No. years cm kg output, ml

1 65 male 172 67 1,600 diabetes

2 45 male 170 56 1,200 congenital renal dysplasia
3 67 male 172 69 180 glomerulonephritis
4 58 male 180 71 1,800 polycystic kidney disease
5 43 female 165 57 0 diabetes
6 59 male 183 86.5 2,200 diabetes
7 62 male 180 86 50 hypertension
8 45 male 177 57.5 1,400 focal segmental glomerulosclerosis

On the day of the next dialysis session, a sample was taken prior
to the commencement of dialysis. The patients resumed their usual
dialysis prescription (duration, pump speed, dialyser type).
In summary, patients were dialysed on days 2 and 4 of the study.
The final blood sample was taken prior to a third dialysis session.
After a dialysis session, patients were requested to undergo a 24-hour
urine collection to determine approximate urine output. The patients
were monitored for side effects during the study.
A minimum of 3.5 ml whole blood was collected at each blood
sampling time. Immediately after collection, each sample was gently
mixed and placed on chipped ice. Within 20 min of collection, each
blood sample was centrifuged for 10 min at 2,500 g at 5 ° C. The plas-
ma was then stored at or below –70 ° C. This was imperative to
ensure accuracy of the samples; whilst bupropion appears heat and
acid labile, the metabolites are stable [11].
The 176 samples were sent as a single batch in cold chain condi-
tions to GlaxoSmithKline’s Department of Bioanalysis, USA, for
analysis using high-performance liquid chromatography (HPLC) and
mass spectometry (MS). Using peak area ratios with 1/¯2 weighted
linear regression, the Department of Bioanalysis, GlaxoSmithKline,
determined the following lower limits of quantification: bupropion:
0.500 ng/ml, hydroxybupropion: 1.00 ng/ml and threohydrobupro-
pion: 0.500 ng/ml.
At a bupropion concentration of 2.0 ng/ml, the within batch pre-
cision and accuracy were 2.5 and 100% B 10.5%. The coefficient of
variation (CV) was 9.1%. At higher concentrations the CV was 8.1%
(50 ng/ml), 8.3% (100 ng/ml) and 8.0% (150 ng/ml). For a hydroxy-
bupropion concentration of 2.0 ng/ml, the within batch precision and
accuracy were 0.8% and 100 B 15.7%. The coefficient of variation
(CV) was 11.0%. At higher concentrations the CV was 3.6% (50 ng/
ml), 3.8% (100 ng/ml) and 3.1% (150 ng/ml). At a threohydrobupro-
pion concentration of 2.0 ng/ml, the within batch precision and accu-
racy were 1.5% and 100 B 9.0%. The coefficient of variation (CV)
was 7.4%. At higher concentrations the CV was 3.4% (50 ng/ml),
3.1% (100 ng/ml) and 3.4% (150 ng/ml).
The results from the HPLC-MS analyses were assessed using the
‘MWPharm’ computer program (MediWare B.V., Groningen, The
Netherlands), at the School of Pharmacy, University of London,
England. The KinFit program (using the Simplex method) was used
for non-linear curve-fitting. For the kinetic analysis, a bioavailability,
F, of 1 was chosen.
Graphs were produced using the ‘GraphPad Prism’ computer
program (GraphPad Software Inc., SanDiego, Calif., USA). Non-
Table 3. Patient medication details
Patient Prescribed medication
No.
1 Amlodipine, aspirin, calcium carbonate, doxazosin,
enalapril, erythropoietin, isosorbide mononitrate,
ranitidine
2 Alpha-calcidol, erythropoietin
3 Aspirin, calcium carbonate, carbamazepine (neuralgia
pain), erythropoietin, lisinopril, simvastatin
4 Amiodarone, aspirin, erythropoietin, furosemide,
lisinopril
5 Alpha-calcidol, erythropoietin, insulin, lansoprazole,
lisinopril
6 alfa-calcidol, aspirin, calcium carbonate, doxazosin,
gliclazide, lansoprazole, ramipril
7 Amlodipine, aspirin, erythropoietin, perindopril,
sevelamer
8 Amlodipine, calcium carbonate, losartan
linear regression curves were produced detailing exact XY coordi-
nates. The results were compared against a previous pharmacokinet-
ic study of bupropion, with a similar dosing schedule and analysis
method [12].
Results
Judging by the 24-hour urine collections, the patients
appeared to have very different levels of residual func-
tion. This and other demographic details can be seen in
table 2. Patient 3 was the only participant on medication
that may have affected the kinetics of bupropion, namely
carbamazepine: this may have increased the metabolism
of the parent compound. The patients’ medication during
the study is summarised in table 3.

Pharmacokinetics of Bupropion and Nephron Clin Pract 2004;97:c83–c89 c85

Metabolites in Haemodialysis
 Table 4. Mean pharmacokinetic values for bupropion (n = 8)

Statistical parameter Tmax Cmax Elim t1/2· Elim t1/2ß Vd Vd Cl AUC,

h ng/ml h h litres litres/kg litres/h ng/ml/h

Mean 2.83 126.22 2.79 15.55 1,330.93 19.36 113.83 1,446

Median 2.82 127.60 2.36 12.43 1,011.00 14.71 102.00 1,470
Standard deviation 0.72 34.92 1.25 7.32 651.88 9.48 45.37 394

Table 5. Mean pharmacokinetic values for hydroxybupropion (n = 8)

Statistical parameter Tmax Cmax Elim t1/2 Vd Vd Cl AUC

h ng/ml h litres litres/kg litres/h ng/ml/h

Mean 13.81 525.40 37.85 275.30 4.00 5.66 37,530

Median 10.94 500.20 39.56 228.20 3.32 4.72 31,940
Standard deviation 5.88 251.94 12.23 135.10 1.97 3.58 24,660

Table 6. Mean pharmacokinetic values for threohydrobupropion (n = 8)

Statistical parameter Tmax Cmax Elim t1/2 Vd Vd Cl AUC

h ng/ml h litres litres/kg litres/h ng/ml/h

Mean 13.85 165.40 56.45 986.53 14.35 13.10 17,136

Median 14.05 124.90 51.65 1,026.20 14.93 13.41 11,200
Standard deviation 5.07 97.64 13.25 459.31 6.68 7.51 13,690

Table 7. Comparison of mean pharmacokinetic values for bupropion and metabolites between a study of patients
with normal renal function (n = 34) and the current study of patients with end-stage renal failure (n = 8)

Parameter Bupropion Hydroxybupropion Threohydrobupropion

normal ESRF normal ESRF normal ESRF

Tmax, h 2.94 2.83 6.97 13.81 5.14 13.85

Cmax, ng/ml 143.50 126.22 429.50 525.40 142.30 165.40
Cl, l/h 135.50 113.83 * 5.66 * 13.10
AUC, ng/ml/h 1,165 1,446 15,883 37,530 6,190 17,136
Elim t1/2, h 18.75 18.34 22.00 37.85 48.00 56.45

All 8 patients had normal liver function (as determined
by ALT and bilirubin tests). Only patients 5 and 8 had low
albumin levels (25 and 29 g/l, respectively) – all others
were within the normal range (35–50 g/l). No patient
experienced any effect, adverse or otherwise during the
study, except for patient 2, who claimed to have a reduced
desire to smoke on days 2 and 3.
The bupropion results, in table 4, show that renal im-
pairment does not have a significant impact on the parent
compound after a single dose. The elimination t1/2 is simi-
lar to the Hsyu study [12].
In other studies [13–15], clearance was generally great-
er at 140–175 liters/h, although one study had a clearance
of 90 liters/h [16]. Sweet et al. [5] quote an almost identi-

c86 Nephron Clin Pract 2004;97:c83–c89 Worrall/Almond/Dhillon

cal clearance, at 113 liters/h, to this study. However, this
is with a bioavailability of 0.5; at F = 1 (which was used in
this study), the clearance is half this value. Consistent
with this reduced clearance, they had an extended elimi-
nation half-life, which we do not see in this present study.
Given that our patients had worse renal function than the
elderly patients in the Sweet study, the poor clearance in
the elderly may be related to an increased Vd.
With respect to the AUC of 1,446 ng/ml/h, all other
studies adjusted to the same dose have lower AUC values
of 905, 1,063, 747 and 800 (all ng/ml/h) [5, 13, 14, 16].
This suggests that bupropion may produce higher AUC
values and have a slightly reduced clearance in haemodi-
alysis patients. The results on dialysis suggest there maybe
some clearance of bupropion by haemodialysis: the mean
change from the arterial to the venous sample was –13.3%
B 8.2. However, the metabolite results are of greater
importance. Few studies have included the kinetics of
bupropion’s metabolites, to allow comparison.
Accumulation of the active metabolite, hydroxybupro-
pion, in renal impairment is very likely, as seen from the
results in table 5. Tmax is double that in the Hsyu work,
Fig. 1. Log mean plasma concentrations for bupropion and metabo-
whilst the Cmax is over 20% higher. The AUC and elimina- lites vs. time.
tion half-life are greatly increased, by 136 and 73%,
respectively, indicating a reduced clearance. The results
from the dialysis sessions suggest hydroxybupropion, the
major active metabolite, is not cleared by dialysis: the Discussion
mean venous sample was 4 B 8.2% greater than the mean
arterial sample. This is the first study to present the pharmacokinetics
For threohydrobupropion, accumulation is also a of bupropion and its major metabolites in humans with
probability, as seen from the results summarized in renal impairment. Previous work has investigated bupro-
table 6. Tmax is increased almost 3-fold, and there is a pion in renal impaired animals [6, 7]. Accumulation of
slight increase in the Cmax (16%) and elimination half-life bupropion and its metabolites has been demonstrated in
(17%). The AUC is substantially greater, by almost 3-fold. the plasma and the brain.
This is suggestive of reduced clearance. This is confirmed The effects of bupropion in elderly patients have been
by results from earlier studies [5, 14]. Clearance by hae- investigated [5]. A reduced clearance and prolonged half-
modialysis may occur based on the mean 16.8 B 7.5% life in both bupropion and the metabolites was seen. Mul-
decrease from the arterial to the venous sample. tiple dosing led to an accumulation of the metabolites rel-
The difference between the kinetics of the metabolites ative to the parent compound suggesting a saturation of
in normal and poor renal function is highlighted in ta- the elimination pathways. It is predicted that a similar
ble 7. The difference in the kinetics of the metabolites scenario would occur in haemodialysis patients, but to a
compared with the parent compound can be seen in fig- greater extent. The elimination half-lives for the metabo-
ure 1. This highlights the importance of the metabolites, lites, after chronic dosing, were very similar to those in
in terms of the higher plasma levels, and extended pres- this single-dose study. After chronic dosing, it is likely that
ence in the plasma, which will increase the risk of toxicity the half-lives of the metabolites in haemodialysis patients
should significant accumulation occur. will be even longer.
The results from samples taken during dialysis indicate
that haemodialysis will not increase the clearance of
hydroxybupropion from the plasma. This may mean that
the major method of clearance of this potentially toxic

Pharmacokinetics of Bupropion and Nephron Clin Pract 2004;97:c83–c89 c87

Metabolites in Haemodialysis
metabolite in haemodialysis patients is the residual kid-
ney function of the patient. The results for bupropion and
threohydrobupropion suggest some clearance may occur
on dialysis.
The plasma levels of the active metabolites are greater
than that of bupropion, and this has been demonstrated in
cerebrospinal fluid [8], where plasma levels correlated
with cerebrospinal fluid concentrations. If these levels are
excessive, toxic effects may occur. The metabolites have
significant neurological activity, suggesting they are re-
sponsible for the activity and the neurological side effects
[9, 10].
Whilst therapeutic drug monitoring of bupropion has
been recommended [17, 18], there is no consensus on
ideal plasma levels [8, 19–21]. Hydroxybupropion plasma
levels greater than 1,250 ng/ml were associated with poor
therapeutic efficacy whilst lower levels achieved a desired
antidepressant response. A similar effect was seen with
threohydrobupropion and erythrohydrobupropion. Al-
though logical, no proven relationship between toxicity
and plasma levels of bupropion or the metabolites exists.
Reports of toxicity, including death, with bupropion exist,
but details of plasma levels are absent [22, 23]. It appears
that efforts should be focused on the metabolites in terms
of identifying their activity and safe plasma levels for
treatment.
Ambiguity exists in the literature regarding dose altera-
tions in renal disease. In elderly patients with an esti-
mated reduced clearance (80% of that in younger pa-
tients) a 25% dose reduction was recommended [5], sug-
gesting a greater reduction in clearance will require an
even greater dose reduction to achieve the same effect.
Whilst the bupropion clearance was slightly reduced, the
strong suggestion is that the clearance of the metabolites is
greatly reduced. If the metabolites have a significant role
in the therapeutic and toxic effects, a significant reduction
in the dose of the parent compound would be required.
A larger multi-dose study needs to confirm the results
presented here, and demonstrate that accumulation of
metabolites does occur. To ethically demonstrate accu-
mulation, we urgently need information on toxic plasma
levels of the metabolites and bupropion. It would be bene-
ficial to investigate the kinetics and safety of this drug in
other patient groups within the renal community. The
results of drug clearance on haemodialysis require clarifi-
cation; dialysate drug levels also need to be measured.
More work is needed to define the activity and side effects
of the metabolites. This will help determine whether dose
reductions are required, or whether accumulation of the
metabolites is safe. Knowledge of the protein binding of
c88 Nephron Clin Pract 2004;97:c83–c89
bupropion and the metabolites in renal failure would
determine whether binding changes in uraemic and dialy-
sis-dependent patients. Bupropion is a basic drug so
would therefore bind to the acute phase reactive protein,
Alpha-1-acid glycoprotein, rather than albumin. As this
protein constitutes a small percentage of the globulin lev-
el, and can be affected by nephritis and inflammatory pro-
cesses in various ways, it is difficult to predict how protein
binding would be affected by end-stage renal failure and
the haemodialysis process.
Future work needs to determine desired plasma levels
for smoking cessation, and levels to be avoided to mini-
mise the risk of adverse effects, such as seizures. In the
longer term, efficacy and safety studies of bupropion in
renal patients, both alone and in combination with nico-
tine replacement therapy are required.
There is an apparent large variation between the pa-
tients in terms of urine volume, presenting a heterogenous
cohort. However, this reflects clinical practice. There may
also be a question relating to the accuracy of single 24-
hour urine collections. The limited size and heterogeneity
of the cohort mean that the standard deviations of the
mean kinetic parameters for bupropion and metabolites
are large, emphasizing the preliminary nature of the
results and the need for a larger study.
The timing of the samples was a compromise between
ideal sample times and the logistics of organising the
study around patient’s dialysis sessions, and obtaining
volunteers. The inclusion of patient 3 (with the concomi-
tant use of carbamazepine) may have affected the results,
especially in a small cohort, due to possible increased
bupropion metabolism. Patients were allowed to smoke
during the study. Smoking does not alter the kinetics of
bupropion and its metabolites [12]. It has been shown
there is no effect of gender on the kinetics of bupropion
[13], although this was recently challenged by Stewart et
al. [24], who also demonstrated increased levels of hy-
droxybupropion in females compared with males. Further
work in larger groups is required.
Conclusion
This study has determined the pharmacokinetics of
bupropion and its pharmacologically active metabolites,
hydroxybupropion and threohydrobupropion, for the first
time, in a small group of haemodialysis patients who
smoke, following a single 150-mg oral dose. The pharma-
cokinetic profile of bupropion was similar to that in
patients with normal renal function; however, the metab-
Worrall/Almond/Dhillon
olites demonstrated reduced clearance with the probabili-
ty of accumulation and toxicity on repeated dosing. Even
though bupropion is only used for a period of approxi-
mately 2 months as a smoking cessation treatment, inap-
propriately high doses may still cause significant accumu-
lation and severe toxicity. Further work in larger groups
with multiple dosing is urgently required to confirm these
results and help confirm an appropriate dosing schedule
in haemodialysis patients. This will require greater knowl-
edge of the activity and toxicity of the active metabolites.
The kinetics of bupropion and the metabolites were suffi-
ciently different from work in patients with normal renal
function to warrant a review of current dosage recommen-
dations. Given that the major active metabolite, hydroxy-
bupropion, is not dialysable and may rely on residual
renal function for clearance, with a t1/2 of approximately
40 h, a daily dosing schedule of 150 mg is likely to cause
toxicity. Based on the results from the current study, a
reduction in the frequency of the 150 mg dose to every 3
days would be a safe approach.
This study has advanced the knowledge of bupropion
and its metabolites, and we are nearer the safe use of
bupropion in haemodialysis patients. This will benefit
those wishing to quit smoking and reduce their risk of a
premature death due to cardiovascular disease.
Competing Interests
None.
Acknowledgements
Thanks must go to the following: The staff and patients of the
Renal Unit, Southend Hospital NHS Trust, for their assistance dur-
ing the project. The staff of GlaxoSmithKline in Middlessex, UK &
North Carolina, USA for their logistical and technical support.

References

1 Royal College of Physicians: Nicotine Addic-
tion in Britain: A Report of the Tobacco Advi-
sory Group of the Royal College of Physicians.
London, Royal College of Physicians, February
2000.
2 Foley RN, Herzog CA, Collins AJ: Smoking
and cardiovascular outcomes in dialysis pa-
tients: The United States Renal Data System
Wave 2 Study. Kidney Int 2003;63:1462–
1467.
3 Aubin HJ: Tolerability and safety of sustained-
release bupropion in the management of smok-
ing cessation. Drugs 2002;62:45–52.
4 Aronoff GR, et al: Drug Prescribing in Renal
Failure – Dosing Guidelines for Adults, ed 4.
Washington, American College of Physicians,
1999, p 92.
5 Sweet RA, Pollock BG, Kirshner M, Wright B,
Altieri LP, DeVane CL: Pharmacokinetics of
single- and multiple-dose bupropion in elderly
patients with depression. J Clin Pharmacol
1995;35:876–884.
6 DeVane CL, Laizure SC, Cameron DF: The
effect of experimentally-induced renal failure
on accumulation of bupropion and its major
basic metabolites in plasma and brain of guinea
pigs. Psychopharmacology (Berl) 1986;89:404–
408.
7 Kaka JS, Al-Khamis KI, Tanira MO: Effect of
hepatic and renal dysfunction on disposition of
bupropion in rats. Eur J Drug Metab Pharma-
cokinet 1988;13:149–153.
8 Golden RN, DeVane CL, Laizure SC, Rudor-
fer MV, Sherer MA, Potter WZ: Bupropion in
depression. II. The role of metabolites in clini-
cal outcome. Arch Gen Psychiatry 1988;45:
145–149.
9 Martin P, Massol J, Colin JN, Lacomblez L,
Puech AJ: Antidepressant profile of bupropion
and three metabolites in mice. Pharmacopsy-
chiatry 1990;23:187–194.
10 Sanchez C, Hyttel J: Comparison of the effects
of antidepressants and their metabolites on
reuptake of biogenic amines and on receptor
binding. Cell Mol Neurobiol 1999;19:467–
489.
11 Laizure SC, DeVane CL: Stability of bupro-
pion and its major metabolites in human plas-
ma. Ther Drug Monit 1985;7:447–450.
12 Hsyu PH, Singh A, Giargiari TD, Dunn JA,
Ascher JA, Johnston JA: Pharmacokinetics of
bupropion and its metabolites in cigarette
smokers versus nonsmokers. J Clin Pharmacol
1997;37:737–743.
13 Findlay JW, Van Wyck Fleet J, Smith PG, Butz
RF, Hinton ML, Blum MR, Schroeder DH:
Pharmacokinetics of bupropion, a novel anti-
depressant agent, following oral administration
to healthy subjects. Eur J Clin Pharmacol 1981;
21:127–135.
14 Laizure SC, DeVane CL, Stewart JT, Dom-
misse CS, Lai AA: Pharmacokinetics of bupro-
pion and its major basic metabolites in normal
subjects after a single dose. Clin Pharmacol
Ther 1985;38:586–589.
15 Lai AA, Schroeder DH: Clinical pharmacoki-
netics of bupropion: A review. J Clin Psychia-
try 1983;44:82–84.
16 Posner J, Bye A, Jeal S, Peck AW, Whiteman P:
Alcohol and bupropion pharmacokinetics in
healthy male volunteers. Eur J Clin Pharmacol
1984;26:627–630.
17 Preskorn SH: Should bupropion dosage be ad-
justed upon therapeutic drug monitoring? Psy-
chopharmacol Bull 1991;27:637–643.
18 Goodnick PJ: Pharmacokinetic optimisation
of therapy with newer antidepressants. Clin
Pharmacokinet 1994;27:307–330.
19 Preskorn SH: Antidepressant response and
plasma concentrations of bupropion. J Clin
Psychiatry 1983;44:137–139.
20 Fogel P, Mamer OA, Chouinard G, Farrell PG:
Determination of plasma bupropion and its
relationship to therapeutic effect. Biomed
Mass Spectrom 1984;11:629–632.
21 Goodnick PJ: Blood levels and acute response
to bupropion. Am J Psychiatry 1992;149:399–
400.
22 Storrow AB: Bupropion overdose and seizure.
Am J Med 1994;12:183–184.
23 Harris CR, Gualtieri J, Stark G: Fatal bupro-
pion overdose. J Toxicol Clin Toxicol 1997;35:
321–324.
24 Stewart JJ, Berkel HJ, Parish RC, Simar MR,
Syed A, Bocchini JA Jr, Wilson JT, Manno JE:
Single-dose pharmacokinetics of bupropion in
adolescents: Effects of smoking status and gen-
der. J Clin Pharmacol 2001;41:770–778.

Pharmacokinetics of Bupropion and Nephron Clin Pract 2004;97:c83–c89 c89

Metabolites in Haemodialysis