British Journal of Anaesthesia 106 (4): 512–21 (2011)

Advance Access publication 8 February 2011 . doi:10.1093/bja/aer002

Pharmacokinetics of ropivacaine in patients with chronic

renal failure
P. J. Pere 1*, A. Ekstrand 2, M. Salonen 1, E. Honkanen 2, J. Sjövall 3, J. Henriksson 3 and P. H. Rosenberg 1,4
1
Department of Anaesthesiology and Intensive Care Medicine and 2 Department of Nephrology, Helsinki University Hospital, Helsinki,
Finland
3
Clinical Pharmacology and Biostatistics, AstraZeneca R&D, Södertälje, Sweden
4
University of Helsinki, Helsinki, Finland
* Corresponding author. E-mail: pertti.pere@hus.fi

Editor’s key points
† Ropivacaine and its
metabolites are excreted
by the kidneys.
† The pharmacokinetics of
ropivacaine is not
affected by renal failure.
† Metabolites may
accumulate in plasma
during long-term
postoperative infusions;
however, non-renal
elimination seems to
compensate for reduced
renal clearance in most
patients.
Background. As ropivacaine and its metabolites are excreted by the kidneys, we studied
their disposition in subjects with renal dysfunction.
Methods. Twenty patients with moderate or severe renal insufficiency and 10 healthy
volunteers received ropivacaine 1 mg kg21 i.v. over 30 min. The concentrations of
ropivacaine and its main metabolites, pipecoloxylidide (PPX) and 3-hydroxy-ropivacaine,
were measured in plasma and urine for 16 –48 h. The relationship between
pharmacokinetic parameters and creatinine clearance (CLCR) was assessed. A model for
estimating non-renal clearance of a metabolite of ropivacaine is described.
Results. Renal dysfunction had little or no influence on the pharmacokinetics of
ropivacaine. The median plasma concentrations of unbound ropivacaine were similar in
uraemic and non-uraemic subjects. Renal clearance of PPX correlated significantly with
CLCR (R 2 ¼0.81). Lack of correlation between total PPX exposure, expressed as area under
the total plasma concentration –time curve from zero to infinity, and CLCR suggests that
the clearance of PPX also includes non-renal elimination. However, in two uraemic
patients, there was increased exposure to PPX resulting from low non-renal elimination.
Conclusions. The pharmacokinetics of ropivacaine is not affected by renal failure. Although the
renal clearance of PPX correlates with CLCR, non-renal elimination seems to compensate for
reduced renal clearance in most patients. PPX may accumulate in plasma during long-term
postoperative infusions, in particular in patients with co-existing low non-renal elimination.
Systemic toxicity is still unlikely because PPX is markedly less toxic than ropivacaine.
Keywords: kidney, failure; local anaesthetics; pharmacology, pharmacokinetics; toxicity, local
anaesthetics
Accepted for publication: 27 December 2010

Ropivacaine, S(2)-1-propyl-2′ ,6′ -pipecoloxylidide, is com-
monly used in various regional anaesthetic techniques for
surgical anaesthesia and obstetric analgesia and in continu-
ous infusions for obstetric or postoperative pain. It is metab-
olized to 3-hydroxy-ropivacaine (3-OH-ropivacaine) mainly by
cytochrome P-450 (CYP) 1A2 and to an N-dealkylated metab-
olite, 2′ ,6′ -pipecoloxylidide (PPX), mainly by CYP3A4 in the
liver.1 These main metabolites, 3-OH-ropivacaine and PPX,
are excreted into urine and account for about 37% and 3%,
respectively, of the administered dose of ropivacaine in
healthy volunteers.2 3-OH-ropivacaine is virtually non-toxic,
whereas unbound PPX has been shown in rats to exert sys-
temic CNS toxicity 1/12th of that of unbound ropivacaine
(M.M. Halldin, MSc Pharm, PhD; personal communication
and AstraZeneca unpublished data).
A very low fraction of ropivacaine (1%) is excreted
unchanged into urine and, therefore, no major changes in
the systemic exposure to ropivacaine are expected in
uraemic patients in comparison with non-uraemic patients.
However, the renally excreted metabolites may accumulate
in renal insufficiency and, in addition, uraemia may impair
the metabolic functions of both the liver and the kidneys.3
In the present study, we evaluated the relationship
between the degree of renal impairment and the pharmaco-
kinetics of ropivacaine, with a special focus on the production
and elimination of PPX.
Methods
We performed an open, parallel-group, single-dose pharma-
cokinetic study in 20 patients suffering from moderately to
severely impaired renal function and 10 healthy volunteers,
that is, three groups with 10 study subjects each. The study
protocol was approved by the Coordinating Ethics Committee

& The Author [2011]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
Pharmacokinetics of ropivacaine in uraemia
of Helsinki University Central Hospital and the Finnish
National Agency for Medicines. Written informed consent
was obtained from all study subjects. The subjects were
divided into three groups according to their creatinine clear-
ance (CLCR) values, calculated at the enrolment visit 6 weeks
before the study, as follows: Group 1, CLCR .80 ml min21
(healthy volunteers); Group 2, CLCR 25 –40 ml min21; and
Group 3, CLCR ,25 ml min21. CLCR values were calculated
using the Cockroft –Gault formula.4
At the enrolment visit, the subjects underwent a thorough
physical examination. A 12-lead ECG was taken and labora-
tory tests included a full blood count, alanine aminotransfer-
ase, glutamyltransferase, plasma albumin, human
immunodeficiency virus, and hepatitis tests, and also
urinary drug screening. A urinary pregnancy test was also
performed in female subjects of childbearing potential.
Plasma creatinine was assessed at enrolment using the
enzymatic method according to the routine of the clinical
laboratory of the hospital. In addition, Jaffé’s5 method was
used to obtain a plasma creatinine value on the day of
drug administration, which served as the basis for the calcu-
lation of CLCR using the Cockroft –Gault formula in the data
analysis.
Patients who were already on dialysis or had an arteriove-
nous shunt in the arm for haemodialysis were not included in
the study. Concomitant drug therapy with any potent inhibi-
tor of CYP subenzyme 1A2 (CYP1A2) or 3A4 (CYP3A4) or
smoking that could not be discontinued at least 7 days
before the planned administration of the study drug was a
contraindication for inclusion in the study.
Experiment and follow-up
Before the administration of the experimental drug, the
urinary test for pregnancy was repeated in the female sub-
jects and another blood sample was obtained for the deter-
mination of CLCR to be used in the statistical analyses. The
subjects then received an i.v. infusion of ropivacaine hydro-
chloride (Naropinw 2 mg ml21, AstraZeneca, Södertälje,
Sweden) 1 mg kg21 over 30 min using a volume-controlled
infusion pump (B. Braun Perfusor fm, Melsungen, Germany).
A low dose and slow infusion rate were chosen to enable a
pharmacokinetic assessment with a low risk of inducing
systemic toxicity. The subjects were observed by a study
physician and two study nurses. Vital signs, ECG, and any
side-effects were monitored during the infusion and for at
least 30 min thereafter. The study subjects were confined
to the study hospital for 36 h, during which time urine was
collected and plasma samples taken. The uraemic patients
returned to the hospital for blood sampling another three
times within the next 2 days. The principal investigator
interviewed all subjects by telephone 7–14 days after the
administration of ropivacaine.
Blood and urine sampling
Blood samples for the assays of total and unbound concen-
trations of ropivacaine, a1-acid glycoprotein (AAG), which is
BJA
responsible for plasma binding of local anaesthetics, and
total and unbound concentrations of the ropivacaine metab-
olites 3-OH-ropivacaine and PPX were collected in hepari-
nized vacuum tubes from a cannula in the antecubital vein
(of the arm opposite to that used for infusion of ropivacaine)
before and 5, 15, 30, 60, and 90 min, and 2, 4, 6, 10, 12, 16,
24, and 36 h after the beginning of the ropivacaine infusion.
From the patients, samples were also obtained 48, 60, and
84 h after the infusion of ropivacaine. The blood samples
were centrifuged within 1 h and the plasma was transferred
to polypropylene tubes which were kept frozen at 2208C
until assay. Urine was collected for assays of ropivacaine,
3-OH-ropivacaine, and PPX. The subjects were instructed to
empty their bladder (control sample) before the infusion of
ropivacaine started, after which urine was collected in 4 h
fractions for 24 h followed by one 12 h fraction. Each fraction
of urine was weighed and two 5 ml samples were taken from
each fraction, transferred to polypropylene tubes and kept
frozen at 2208C until assay.
Drug assays
The total plasma concentration of ropivacaine was measured
in all plasma samples. The unbound plasma concentration of
ropivacaine was determined from a few (three to four)
selected samples from each subject as guided by the
results from the assays of the total concentrations in order
to increase the probability to select plasma samples at
times for detectable unbound ropivacaine concentrations.
Total PPX was also analysed in plasma as guided by its
urinary excretion pattern to increase the probability to
select plasma samples at times for detectable concen-
trations of PPX. The unbound plasma concentration of PPX
was analysed in a few (three to four) samples because of
its potential contribution to systemic toxicity.
The total plasma concentration of ropivacaine was
assayed using gas chromatography with a nitrogen-sensitive
detector.6 Liquid– liquid extraction was used for sample prep-
aration. The limit of quantification (LOQ) was 2.7 mg litre21 as
ropivacaine base using 100 ml plasma and the interday coef-
ficient of variation 6.8% (n¼10).
The total plasma concentration of unconjugated
3-OH-ropivacaine and PPX was assayed using coupled-column
liquid chromatography and mass spectrometric detection with
electrospray ionization. The plasma samples were prepared by
acidification followed by ultrafiltration. The LOQ was 2.9 mg
litre21 for 3-OH-ropivacaine and 2.3 mg litre21 for PPX, and
the interday coefficient of variation was 7.4% (n¼6) for 3--
OH-ropivacaine and 4.3% (n¼10) for PPX.
Unbound (free) concentrations of ropivacaine and
PPX were assayed in plasma using coupled-column liquid
chromatography and mass spectrometry detection, with
electrospray ionization after ultrafiltration of the plasma
samples.3 The quantification limit was 2.7 mg litre21 for
ropivacaine and 2.3 mg litre21 for PPX, with a coefficient of
variation of 4.5% (n¼6) for ropivacaine and 4.8% (n¼6)
for PPX.
513
BJA Pere et al.

The urinary excretion of ropivacaine, 3-OH-ropivacaine, renal function expressed as CLCR and the relevant pharmaco-
and PPX was assayed by liquid chromatography and mass kinetic variable estimates, for example, dose-normalized
spectrometry.7 Acid hydrolysis and solid-phase extraction Cmax and Cu,max, dose-normalized AUC and AUCu, and the
were used for sample preparation.8 The quantification total plasma clearance (CL) and unbound plasma clearance
limits were 27, 200, and 70 mg l21 for ropivacaine, (CLu) of ropivacaine and, where possible, of PPX, was
3-OH-ropivacaine, and PPX, respectively. calculated.
AAG concentration in plasma was measured using an Three study groups were compared pairwise through the
immunoturbidometric method with a quantification limit of ratio of average pharmacokinetic parameter estimates of
0.10 g litre21. The coefficient of variation was 3.0–5.5% at ropivacaine and PPX, respectively, for example, AUC, AUCu,
0.27 –1.39 g litre21 (n¼21). Cstop, Cmax, Cu,stop, Cu,max, CL, and CLu. This was performed
Plasma creatinine concentration, obtained on the day of after logarithmic transformation of the original pharmacoki-
drug administration, was analysed by Jaffé’s5 method. netic data and calculation of 95% confidence intervals (CIs)
including point estimates, and finally, an antilogarithmic
Pharmacokinetic analyses transformation.
The planned number of subjects in this study was esti-
The total (Cstop) and unbound plasma concentrations (Cu,stop)
mated to be sufficient to detect pharmacokinetic differences
at the end of the infusion were derived directly from the data
large enough to warrant a change in dose recommendations
and the total (Cmax) and unbound peak plasma concen-
from a safety point of view. A sample size of 10 in the group
trations (Cu,max) and the time to reach the peak plasma con-
with normal renal function and 10 in the group with severe
centration (tmax) if this occurred later than the end of
renal impairment would be able to detect a 40% difference
infusion. The area under the total plasma concentration –
in total plasma clearance with 90% power. These calcu-
time curve from zero to infinity (AUC) was calculated using
lations were based on a simple two-sided t-test, at a 5% sig-
the trapezoidal rule up to the last quantifiable sampling
nificant level, under normality assumptions, and the
point and addition of the residual area [Clast (predicted)/lz]
assumed values [mean¼400 ml min21, standard deviation
where lz is the terminal elimination rate constant estimated
(SD)¼100 ml min21] for clearance in healthy volunteers
from the individual linear regression on the terminal part of
were based on Emanuelsson and colleagues.10
the log concentration vs time curve.
In order to investigate the correlation between the
The unbound fractions ( fu) of ropivacaine and PPX in
exposure to PPX and renal function, AUC was plotted
plasma were calculated by the unbound plasma concen-
against CLCR (Fig. 1). To estimate a possible contribution of
tration divided by the total plasma concentration at the
non-renal clearance (CLNR) to the total plasma clearance
time points where both total and unbound concentrations
(CL) of PPX, 1/AUC (proportional to CL) was plotted against
were analysed from the same sample. Owing to the small
renal clearance (CLR) (Fig. 2). The equation CL¼CLR +CLNR
number of unbound plasma concentrations, the unbound
can be rewritten as 1/AUC¼CLR/dose+CLNR/dose. As a first
pharmacokinetic variables were estimated using the individ-
step, a linear relationship (Y¼ bX+ a, where b ¼1/dose and
ual mean values of fu. Unbound AUC (AUCu) was calculated
a ¼CLNR/dose) seems reasonable. Dose should be interpreted
by the total AUC times the individual mean values of fu.
as a fictitious (mean) dose of PPX, corresponding to PPX for-
The terminal half-life in plasma (t1/2) was estimated by ln
mation. CLNR/dose significantly larger than 0 indicates non–
2/lz.
The fraction of unchanged ropivacaine excreted in urine
( fe) and the fraction excreted as the major metabolites PPX
( fe,PPX) and 3-OH-ropivacaine ( fe,3-OH) were estimated by
4
the cumulative amount excreted in urine/ropivacaine dose.
AUC (h mg litre–1) (PPX)

Urine was collected over 36 h. Consequently, during four

half– lives of 9 h9 on average, 94% fe,PPX was expected to 3
be collected in the healthy volunteers. To assess the degree
of underestimation of fe,PPX at the end of the urine collection 2
period (t), % recovery (i.e. amount excreted from zero to time
t, Aet/amount excreted from zero to infinity, Aeinf ) was esti-
1
mated as [12(1/2)n], where n¼t/t1/2. Enumerated values of
fe,PPX were calculated by Aeinf/ropivacaine dose.
0
Statistical analysis 0 20 40 60 80 100 120 140
Creatinine clearance on day 1 (ml min–1)
All variables were evaluated by means of descriptive stat-
istics, frequency tables, or graphs as appropriate.
Fig 1 Total AUC of PPX (best possible estimates) vs CLCR
Linear models were used for the exploratory regression
(R 2 ¼0.0606, n¼28) after i.v. infusion of ropivacaine 1 mg kg21
analyses modelling the relationship between CLCR and the in healthy volunteers and in patients with renal impairment.
pharmacokinetic variables. The relationship between the

514
Pharmacokinetics of ropivacaine in uraemia BJA
where b is the ordinary LS estimator. An estimator of a is
5 a′ ¼mean of Y2b′ ×mean of X.
The standard error of b′ is adjusted with (1+q 2)1/2, that is,
1/AUC (litre h–1mg–1) (PPX)

′ 2 1/2 ′
SE(b )¼(1+q ) SE(b), and the standard error of a is
2 
4
′ 2 1/2
(m¼mean of X ) SE(a )¼ s[1/n+(1+q )(m) / (X2m)] .
3 It was assumed in the study that PPX is not further metab-
olized, that is, CL¼CLR. The alternative hypothesis is CL¼
2 CLNR +CLR, where CLNR .0. Total PPX clearance can be
written as dose/AUC or 1/AUC¼CLNR/dose+CLR ×1/dose
1 where the intercept (CLNR/dose)¼0 would mean lack of non-
renal clearance.
0 A 95% CI for the intercept a (¼CLNR/dose) can be used to
0 1 2 3 5 6 test whether CLNR/dose¼0 under normal (Gaussian) distri-
Renal clearance for PPX (litre h–1)
bution assumptions. A combination of the lower limit of
the CI for the intercept a and the upper limit of the CI for
Fig 2 1/AUC of PPX (best possible estimates) vs CLR of PPX b can be used to estimate CLNR. The individual estimates of
(Y¼1.70370+0.19800X, R 2 ¼0.0413, n¼28) after i.v. infusion of CLR (PPX renal clearance) have non-ignorable standard
ropivacaine 1 mg kg21 in healthy volunteers and in patients
errors, i.e. CLR is observed with error.
with renal impairment.
Individual estimates of AUC of PPX, CLR, and SEs of CLR are
available. All parameter estimates of interest can thus be cal-
renal elimination. A statistical model for the calculation of culated. In particular, q¼ sd/sj can be estimated through the
approximate 95% CIs is described below. relation Var(X) = s2j + s2d , where estimates of Var(X ) and s2d
When using ordinary least-squares (LS) estimation of the are available.
parameters in the common linear regression model The statistical analyses were performed using SAS 8.2 for
Windows.
Y = a′ + b′ X + 1
Results
it is assumed that the variable X is observed without error,
The characteristics of the patients in each study group are
whereas Y is subject to the error 1. Suppose now that both
given in Tables 1 and 2. The infusion of ropivacaine was
X and Y are subject to error. Let
well tolerated by all subjects except one patient in the mod-
Y =h+1 erate renal impairment group, in whom the infusion was dis-
continued after 5 min because of fatigue, nausea,
hyperventilation, tachycardia, and hypertension. In this
X =j+d patient, the total plasma concentration of ropivacaine at
this point was similar to that of the others.
The errors 1 and d are assumed to be normally distributed
with expected values 0 and SDs s and sd, respectively. The Ropivacaine
errors are supposed to be uncorrelated. In addition, it is None of the pharmacokinetic parameter estimates of ropiva-
assumed that j and d are uncorrelated as well. caine (AUC, AUCu, Cstop, Cu,stop, CL, CLu, CLR, volume of distri-
Let bution at steady state Vss, fu, fraction of unbound ropivacaine
excreted in urine fe, and terminal half-life t1/2) was related to
h = a + bj
renal function in terms of creatinine clearance (Tables 3
and 4). Three patients in both renal impairment groups had
denotes the variance of js2j . If we estimate b in this model
a higher Cstop than the healthy volunteers. One of the sub-
using the ordinary LS estimator b when X is observed with
jects in the severe renal impairment group had by mistake
error, we have
continued to take diltiazem until 4 days before the ropiva-
  caine dose, and this subject had the slowest decline in the
1 − q2
E(b) = b total plasma concentration of ropivacaine. On average,
(1 + q2 )
,1% of the ropivacaine dose was excreted unchanged in
where E(b) is the expected value of b,11 which means that b the urine in all groups.
is underestimated by a factor q 2/(1+q 2), where q¼ sd/sj.
An unbiased estimator of b when X is observed with error, 3-OH-ropivacaine
but q is known, is thus The total plasma concentrations of unconjugated
3-OH-ropivacaine were low and similar in the three groups
b (Fig. 3), with the exception of one patient in the severe impair-
b′ = = (1 + q2 )b
[1 − q2 /(1 + q2 )] ment group who had an exceptionally slow decline in plasma

515
BJA Pere et al.

Table 1 Subject characteristics, plasma creatinine (Jaffe), and calculated CLCR in healthy volunteers and patients with renal impairment. Mean
(SD) (min – max). *At enrolment. †On the experimental day. CLCR, creatinine clearance

Healthy volunteers (CLCR >80 ml Moderate renal impairment (CLCR 25 – 40 Severe renal impairment (CLCR 10– 25 ml
min21), n510 ml min21), n510 min21), n510
Age (yr) 39 (9.5) (26 – 51) 45 (11.8) (27 –62) 53 (7.9) (41 – 65)
Gender (M/F) 2/8 5/5 2/8
Weight (kg) 67 (7.1) (56 – 77) 73 (11.6) (56 –92) 65 (12.1) (50 –86)
Height (cm) 169 (8.7) (161 –184) 169 (6.9) (157 –176) 166 (8.2) (153 –182)
BMI (kg m22) 23 (2.5) (19 – 27) 25 (3.8) (19– 30) 23 (2.8) (18 – 27)
P-creatinine (mmol 74 (11) (58– 94) 252 (71) (170 – 359) 373 (100) (234 –523)
l21)a*
CLCR (ml min21)* 100 (15.6) (78 –127) 32 (6.5) (26– 46) 18 (5.7) (9 –25)
P-creatinine (mmol 71 (11) (60– 95) 269 (61) (178 – 351) 358 (108) (232 –503)
l21)†
CLCR (ml min21)† 103 (13.5) (82 –122) 30 (5.6) (23– 41) 19 (6.7) (9 –27)

Table 2 Patient background, clinical chemistry, and haematology in healthy volunteers and patients with renal impairment. Mean (SD) (min –
max)

Healthy volunteers (CLCR >80 ml Moderate renal impairment (CLCR 25 –40 Severe renal impairment (CLCR 10 –25
min21), n510 ml min21), n510 ml min21), n510
Diabetes 0 9 4
Diabetic nephropathy 0 6 5
Hypertension 0 10 10
Nephrotic syndrome 0 1 0
Nephrosclerosis 0 1 0
P-albumin (g litre21) 41 (3.5) (36 –46) 33 (3.9) (28 –38) 37 (1.5) (34 – 39)
B-haemoglobin (g 136 (9.1) (121 – 151) 126 (12) (108 –152) 121 (4.7) (114 –130)
litre21)
B-haematocrit (%) 40 (2.9) (35 –45) 38 (4.0) (32 –46) 37 (1.7) (35 – 41)

concentrations. The fraction of 3-OH-ropivacaine excreted into
the urine was lower in the groups with impaired renal function
than in the healthy volunteers (Table 4) (P,0.001).
Pipecoloxylidide
The total plasma concentrations were low and in general
similar between the healthy volunteers and the patients
with renal impairment (Table 3 and Fig. 4). However, one
patient in the moderate renal impairment group (Subject
222, Cmax 0.054 mg litre21, CLCR ¼26 ml min21) and two
patients in the severe impairment group (Subject 220, Cmax
0.055 mg litre21, CLCR ¼25 ml min21; and Subject 111, Cmax
0.073 mg litre21, CLCR ¼0.24 ml min21) had a Cmax higher
than the highest Cmax of the healthy volunteers. The fraction
of PPX excreted in the urine ( fe,PPX) over 36 h was lower in the
moderate and in the severe renal impairment groups than in
the healthy volunteers (Table 4).
Renal elimination of PPX was clearly related to renal func-
tion. Of the variation in CLR, 81% was accounted for by CLCR
(Fig. 5) and for enumerated fraction of PPX excreted ( fe,PPX)
(Fig. 6A) and cumulative amounts of PPX excreted over 36 h
(Fig. 6B), 29% and 43% of the variation was related to the
variation in CLCR. Low correlations were observed between
the other pharmacokinetic variables of PPX and CLCR.
Low correlations between the AUC of total PPX and CLCR
(R 2 ¼0.061) (Fig. 1) and between AUC and CLR (R 2 ¼0.041)
(Fig. 7) suggest that the CL of PPX, in addition to renal
excretion of unchanged PPX, also includes non-renal elimin-
ation of PPX. On the basis of the model 1/AUC¼CLNR/
dose+1/dose×CLR, the adjusted point estimate of CLNR
was 6.34 litre h21 and the lower CI limit was 1.15 litre h21
(Table 5). The point estimate of PPX dose was 3.94 mg. The
intercept in the model (CLNR/dose) was significantly larger
than 0, which indicates non-renal elimination.
Two subjects in the moderate and severe renal impair-
ment groups had outlying and high AUC values for PPX
(Fig. 7) and long elimination half-lives (20.3 and 23.6 h). In
these patients, non-renal elimination of PPX is less apparent.
This is further supported by a higher fraction of ropivacaine
excreted as PPX (4.4 –7.5%) (Fig. 6A) and a low expected
urinary recovery of PPX (65 –71%) in these subjects. The
unbound plasma levels of PPX were low and similar in the
different study groups, but the mean fraction of unbound

516
Pharmacokinetics of ropivacaine in uraemia BJA

Table 3 Pharmacokinetic parameters after 1 mg kg21 i.v. ropivacaine in healthy volunteers and patients with renal impairment. Mean (SD) (min –
max). Cstop, total plasma concentration at the end of infusion; Cu,stop, unbound plasma concentration at the end of infusion; AUC, area under the
total plasma concentration– time curve from zero to infinity; AUCu, area under the unbound plasma concentration– time curve from zero to
infinity; CL, total plasma clearance; CLu, unbound plasma clearance; Vss, volume of distribution at steady state; t1/2, terminal half-life; fu,
unbound fraction in plasma; PPX, pipecoloxylidide; Cmax, maximum total plasma concentration; Cu,stop, unbound plasma concentration at the
end of infusion; tmax, time to reach maximum plasma concentrations; AUCPPX, AUC as defined for PPX; AUCPPX,u, AUC as defined for unbound PPX;
fu,PPX, unbound fraction of PPX in plasma; C2 h, plasma concentration 2 h after infusion

Healthy volunteers (CLCR >80 ml Moderate renal impairment (CLCR 25– 40 Severe renal impairment (CLCR 10– 25
min21), n510 ml min21), n510 ml min21), n510
Ropivacaine n¼10 n¼8 –9 n¼10
Cstop (mg litre21) 1.15 (0.27) (0.63 –1.44) 1.32 (0.39) (0.83 – 2.03) 1.27 (0.32) (0.61 –1.70)
Cu,stop (mg litre21) 0.05 (0.01) (0.03 –0.07) 0.05 (0.01) (0.04 – 0.07) 0.03 (0.01) (0.04 –0.08)
AUC (h mg litre21) 2.59 (0.77) (1.65 –3.84) 3.36 (1.60) (1.41 – 5.67) 3.85 (3.09) (1.65 –12.0)
AUCu (h mg litre21) 0.09 (0.03) (0.07 –0.15) 0.11 (0.04) (0.06 – 0.20) 0.11 (0.06) (0.05 –0.22)
CL (ml min21) 404 (106) (253 –555) 383 (154) (202 –689) 324 (127) (95.3 – 471)
CLu (ml min21) 11.5 (3.1) (6.58 – 16.4) 10.6 (2.9) (5.76 –15.0) 10.1 (3.9) (4.91 – 15.4)
Vss (litre) 59.6 (15.5) (36.3 –88.4) 57.7 (12.8) (42.5 – 80.2) 52.8 (22.7) (31.7 –109)
t1/2 (h) 2.46 (0.85) (1.04 –3.98) 2.25 (0.70) (1.31 – 3.36) 2.59 (1.15) (1.40 –4.60)
fu (%) 3.59 (0.71) (2.68 –4.87) 3.54 (0.70) (2.46 – 4.59) 3.27 ( 1.00) (1.87 –5.43)
PPX n¼9 –10 n¼8 –9 n¼10
Cmax (mg litre21) 0.03 (0.01) (0.02 –0.04) 0.03 (0.01) (0.01 – 0.05) 0.04 (0.02) (0.02 –0.07)
Cu,stop (mg litre21) 0.02 (0.01) (0.01 –0.03) 0.02 (0.01) (0.01 – 0.03) 0.02 (0.01) (0.01 –0.04)
tmax (h) 5.40 (2.99) (2.00 –12.0) 5.78 (2.91) (2.00 – 12.0) 5.40 (4.22) (2.00 –16.0)
t1/2 (h) 7.87 (3.46) (4.40 –13.7) 9.19 (3.59) (5.74 – 16.0) 11.3 (6.06) (5.60 –23.6)
AUCPPX (h mg 0.48 (0.22) (0.23 –0.79) 0.65 (0.32) (0.22 – 1.34) 1.07 (1.04) (0.23 –3.31)
litre21)
AUCPPX,u (h mg 0.27 (0.12) (0.11 –0.46) 0.41 (0.24) (0.14 – 0.96) 0.51 (0.49) (0.14 –1.76)
litre21)
fu,PPX (%) 55.5 (5.6) (48.1 – 65.4) 60.7 (9.0) (44.6 –71.2) 52.1 (9.9) (31.7 – 66.2)
3-OH-ropivacaine n¼10 n¼10 n¼10
C2 h (mg litre21) 0.03 (0.01) (0.01 –0.05) 0.03 (0.01) (0.01 – 0.04) 0.04 (0.02) (0.01 –0.08)
AAG n¼10 n¼9 n¼10
Cstop (mmol l21) 16.8 (2.74) (10.9 –21.7) 19.1 (3.57) (14.6 – 26.4) 22.1 (6.40) (13.8 –36.7)

Table 4 Pharmacokinetic parameter estimates of ropivacaine, PPX, and 3-OH-ropivacaine based on urinary excretion during 36 h in healthy
volunteers and patients with renal impairment. Mean (SD) (min –max). t1/2, terminal half-life; fe, fraction of unchanged ropivacaine excreted into
urine; CLR, renal clearance; fe,PPX, fraction of administered dose of ropivacaine excreted as PPX; fe,3-OH-ropi, fraction of administered dose
of ropivacaine excreted as 3-OH-ropivacaine. aP,0.003; bP,0.001; cP,0.001; dP,0.05; eP,0.001; fP,0.001; b, c, e, and f all indicate that
P,0.001

Healthy volunteers (CLCR >80 ml Moderate renal impairment (CLCR 25 –40 ml Severe renal impairment (CLCR 10 –25 ml
min21), n510 min21), n510 min21), n510
Ropivacaine n¼4– 10 n¼6 –10 n¼5– 10
t1/2 (h) 4.99 (3.51) (2.18 –10.1) 2.59 (0.87) (1.59 –3.80) 3.42 (1.28) (1.80 – 4.82)
fe (%) 0.53 (0.68) (0.03 –2.36) 0.87 (0.51) (0.27 –2.07) 1.04 (1.59) (0.12 – 5.28)
CLR (ml min21) 2.33 (2.89) (0.21 –9.20) 3.65 (2.06) (1.34 –7.46) 2.42 (2.06) (0.78 – 6.23)
PPX n¼10 n¼9 –10 n¼8– 10
t1/2 (h) 7.30 (2.18) (4.08 –10.9)a 11.4 (4.45) (6.50 –21.3) 20.1 (16.4) (7.00 – 52.8)a
fe,PPX (%) 3.47 (1.87) (1.61 –7.30) 0.92 (0.46) (0.28 –1.77) 1.33 (1.54) (0.08 – 4.88)
CLR (ml min21) 50.6 (10.7) (29.7 –68.3)b,c 17.9 (9.19) (8.37 –37.2)b 12.3 (4.42) (7.49 – 19.6)c
3-OH-ropivacaine n¼10 n¼10 n¼10
t1/2 (h) 8.21 (3.46) (4.03 –14.2)d,f 9.83 (3.72) (3.89 –14.9) 13.2 (5.58) (6.78 – 22.4)d,f
fe,3-OH-ropi (%) 31.6 (5.2) (21.8 – 37.3)e 19.1 (8.1) (6.85 – 35.5) 18.2 (3.7) (12.7 –24.3)

517
BJA Pere et al.

A 0.08 A
Group≥80 ml min–1 0.08 Group≥80 ml min–1

Concentration (mg litre–1)

Concentration (mg litre–1)
0.06 0.06

0.04 0.04

0.02
0.02

0.00
0 4 8 12 16 20 24 28 32 36
0 4 8 12 16 20 24 28 32 36 Actual sampling time (h)
Actual sampling time (h)
B Group=25–40 ml min–1
B 0.08 0.08

Concentration (mg litre–1)

Group=25–40 ml min–1
Concentration (mg litre–1)

0.06
0.06
0.04
0.04
0.02
0.02
0.00
0 4 8 12 16 20 24 28 32 36
0.00 Actual sampling time (h)
0 4 8 12 16 20 24 28 32 36
Actual sampling time (h) C Group=10–24 ml min–1
0.08
C 0.08 Concentration (mg litre–1)
Group=10–24 ml min–1
0.06
Concentration (mg litre–1)

0.06 0.04

0.04 0.02

0.02
0.00
0 4 8 12 16 20 24 28 32 36
Actual sampling time (h)
Fig 3 Total plasma concentrations of unconjugated 3-OH-
ropivacaine after i.v. infusion of ropivacaine 1 mg kg21 in
healthy volunteers (A, CLCR .80 ml min21), in patients with mod-
erate renal impairment (B, CLCR 25 –40 ml min21), and in patients
with severe renal impairment (C, CLCR ,25 ml min21). Number of
data points for each patient (median, range): (A) (2, 1– 4); (B) (2,
1– 4); (C) (3, 2– 6).
PPX ( fu) was more than 10 times higher than that of ropiva-
caine. The contribution of CLNR to the CL of PPX was further
supported by a regression analysis of 1/AUC (i.e. CL of PPX)
vs CLCR, including the two outliers, which indicated that
only 3% (R 2 ¼0.0287) of the variation in 1/AUC was
accounted for by variation in CLCR (Fig. 8).
The subject in the severe renal impairment group
(Subject 111, CLCR ¼24 ml min21) who had the highest
observed Cmax of PPX also had the highest unbound concen-
tration (Cu,max) of PPX that was higher than the highest
Cu.max in the healthy volunteers. In one patient, who had
taken diltiazem until 4 days before the ropivacaine
exposure, the unbound plasma concentration of PPX was
0.00
0 4 8 12 16 20 24 28 32 36
Actual sampling time (h)
Fig 4 Total plasma concentration of PPX after i.v. infusion of ropi-
vacaine hydrochloride 1 mg kg21 in healthy volunteers (A,
CLCR .80 ml min21), in patients with moderate renal impairment
(B, CLCR 25– 40 ml min21), and in patients with severe renal
impairment (C, CLCR ,25 ml min21). Number of data points for
each patient (median, range): (A) (5, 5– 5); B: (5, 4– 6); C: (6, 5–6).
still increasing 12 h after the start of the infusion. In the
other subjects, the Cu,max of PPX was measured 2 – 6 h
after the start of the infusion.
a1-Acid glycoprotein
There was a weak correlation between the mean AAG
plasma concentration and creatinine clearance (R 2 ¼0.14)
(Table 3). One healthy volunteer and one patient with
severe renal impairment had AAG plasma concentrations
below and above, respectively, the normal reference range.
Discussion
We found that the baseline creatinine clearance varied
between 82 and 122 ml min21 in the healthy volunteers
and between 9 and 41 ml min21 in the patients. This was
considered to be an adequate representation of relevant

518
Pharmacokinetics of ropivacaine in uraemia BJA

70 4

60

AUC (h mg litre–1) (PPX)

CLR (ml min–1) (PPX)

3
50

40
2 (R2= 0.0413, n=28)
30

20 (R2=0.8117, n=29) 1
10

0 0
0 20 40 60 80 100 120 140 0 1 2 3 4 5
Creatinine clearance (ml min–1) Renal clearance for PPX (litre h–1)

Fig 5 Correlation between CLR of PPX and CLCR after i.v. infusion Fig 7 Correlation of AUC of PPX and renal clearance for PPX after
of ropivacaine 1 mg kg21 to subjects with and without renal i.v. infusion of ropivacaine 1 mg kg21 in healthy volunteers and in
impairment (n¼29). patients with renal impairment. Note the two outliers with
exceptionally high AUC values.

Table 5 Partial results from a linear regression analysis of 1/AUC

A 8 on CLR of PPX. CI, confidence interval; CLNR, non-renal clearance

Estimates All subjects

6 n 28
R2 0.0413
fe (%) (PPX)

a 1.7037
4 b 0.1980
SE 0.1871
Estimate of s2d 0.2956
2 Estimate of Var(CLR) 1.3393
Estimate of s2j 1.0437
(R2= 0.2878, n=29) Estimate of q¼ sd/sj 0.5322
0 Adjusted b (b′ ) 0.2541
0 20 40 60 80 100 120 140
Adj point estimate of PPX dose (mg) 3.9358
Creatinine clearance (ml min–1)
Adj 95% CI for b′ , upper limit 0.6599
B 16 Adjusted a (a′ ) 1.6100
Adj 95% CI for a′ , lower limit 0.7611
Adj 95% CI for a′ , upper limit 2.4590
Cumulative amount excreted

12 Adj point estimate of PPX CLNR (litre h21) 6.3367

Adj 95% lower CI for PPX CLNR (litre h21) 1.1533
(µmol) (PPX)

subjects with various degrees of renal impairment. A group of

4 patients with mild renal impairment (CLCR 40–80 ml min21)
was initially considered, but was rejected due to the minor
(R2= 0.4274, n=30) effects of renal function on the anticipated pharmacokinetics
0 of ropivacaine and practical difficulties in the enrolment of
0 20 40 60 80 100 120 140 such subjects.
Creatinine clearance (ml min–1) As expected, due to the low fe of ropivacaine, no relation-
ships were found between CL or any of the other pharmaco-
Fig 6 Correlation between enumerated fraction excreted ( fe) as kinetic variables estimated for ropivacaine and renal function
PPX and CLCR (A) (n¼29) and between cumulative amount of in terms of creatinine clearance.12 On the other hand, the
PPX excreted and CLCR (B) after i.v. infusion of ropivacaine 1 mg
elimination of PPX ( fe,PPX and CLR) was strongly related to
kg21 to subjects with and without renal impairment (n¼30).
CLCR. However, the correlation between the exposure to PPX

519
BJA
5 the healthy volunteers. The covariation between the AUC of
1/AUC (litre h–1mg–1) (PPX)
4 the sense that R 2 is too high (0.061).
3 function, the PPX exposure in plasma is increased. It is conceiva-
2 lation of PPX to 3-hydroxy-pipecoloxylidide (3-OH-PPX) and
1 CYP1A2 and CYP3A4, respectively. Trace amounts of both
0 single dose to healthy volunteers.2 This hypothesis is further
0 20 40 60 80 100 120
Creatinine clearance (ml min–1)
Fig 8 1/AUC of PPX (best possible estimates) vs CLCR (R 2 ¼0.0287,
n¼28) after i.v. infusion of ropivacaine 1 mg kg21 in healthy vol-
unteers and in patients with renal impairment.
(AUC) and creatinine clearance was low, which suggests that
the CL of PPX also includes non-renal elimination.
The pharmacokinetics of metabolites is usually obtained
after administration of the parent drug but may require sep-
arate administration of the metabolite. The latter will require
approval for human use by regulatory bodies. To get infor-
mation on the clearance (CL) of a metabolite, it is generally
recognized that it must be given i.v. and separately from
the parent drug in order to estimate the dose, that is, total
clearance¼dose/AUC.12
To further evaluate this mechanism in a supplementary
analysis, we developed a pharmacokinetic model for non-
renal elimination of a metabolite that to our knowledge
has not previously been published. A point estimate and a
lower confidence limit of non-renal clearance of a metabolite
were estimated by the regression of 1/AUC on renal clear-
ance of the metabolite after administration of the parent
drug (ropivacaine) without the need for separate i.v. admin-
istration of the metabolite. Using this, we were able to show
that the estimated non-renal clearance was larger than zero
and statistically significant. The contribution of CLNR to the CL
of PPX was further supported by a regression analysis, indi-
cating that only 3% of the variation in 1/AUC (proportional
to CL of PPX) was accounted for by variation in CLCR (Fig. 8).
Fraction of dose excreted as PPX ( fe,PPX) was probably
somewhat underestimated due to, for example, the too
short urine collection interval in relation to the t1/2 of PPX
especially in the renal impairment groups. This was reflected
by the lower correlation between enumerated fraction of
administered dose of ropivacaine excreted as PPX ( fe,PPX),
estimated on the basis of theoretical excretion of PPX to
infinity, and creatinine clearance (R 2 ¼0.29) (Fig. 6A), than
between fe,PPX over 36 h and creatinine clearance (R 2 ¼0.43).
The period analysed for PPX plasma concentrations (16 h
for the healthy volunteers and 36 h for the patients with
renal impairment) possibly resulted in an underestimate of
t1/2 for the healthy volunteers, which may also imply an
520
Pere et al.
underestimate of AUC of plasma concentration of PPX for
PPX and creatinine clearance may thus be overestimated in
Owing to a reduced CLR of PPX in patients with impaired renal
ble that this gives an opportunity for increased further hydroxy-
4-hydroxy-pipecoloxylidide (4-OH-PPX), presumably by
metabolites have previously been detected in the urine after a
140
supported by a slower rate of formation of PPX than 3--
OH-ropivacaine. In pooled human liver microsomes, the appar-
ent Michaelis constant (Km) value for 3-OH-ropivacaine was 16
mM and for 4-OH-ropivacaine and -PPX about 400 mM.1 When
Reynolds gave a single i.v. dose of PPX 43 mg to two healthy vol-
unteers, 46% was recovered unchanged in the urine after 24
h,13 which indicates further metabolism of PPX, but no
attempt was made at the time to identify these metabolites.
The fraction excreted of 3-OH-ropivacaine was reduced and
its half-life increased with a decrease in renal function. A
similar increased exposure to unconjugated 3-OH-ropivacaine
in plasma could therefore lead to increased possibilities for
metabolism to 3-OH-PPX by CYP3A4.
One of the two outliers took diltiazem, a CYP3A4 inhibi-
tor,14 15 up to 4 days before ropivacaine administration. An
inhibitory effect of diltiazem on the CYP3A4 metabolism of
ropivacaine seems likely, as this subject had the lowest
total CL, the highest AUC, the highest fe, and longest half-life
of ropivacaine. As CYP3A4 is likely to be involved in a further
metabolism of PPX, concomitant treatment with a CYP3A4
inhibitor is also a potential explanation of why this subject
showed a high AUC of PPX (Fig. 7) (2.45 h mg litre21).
Non-renal clearance of PPX can to a large extent compen-
sate for the reduction in CLR in renal impairment. The two
outliers had an AUC of PPX up to five to six times higher
than the other subjects. Assuming that a postoperative con-
tinuous epidural infusion of ropivacaine may result in a
six-fold higher exposure of unbound PPX plasma concen-
tration in patients with renal impairment, the consequences
are reduced as the central nervous systemic toxicity of PPX is
significantly less than that of unbound ropivacaine. Further-
more, there is no extra contribution of unbound ropivacaine
as it is not affected by renal function.
Our study subjects were not exposed to surgery or major
stress and, therefore, their slightly elevated AAG levels in
plasma, which is typical for uraemic patients,16 remained
unchanged. In patients undergoing major surgery, on the
other hand, there is a stress-induced increase in the acute-
phase proteins including AAG.17 Ropivacaine is 94%
bound to AAG. As ropivacaine has an intermediate-to-low
hepatic extraction ratio,1 2 its total plasma clearance is
expected to vary with changes in the unbound fraction. An
increase in AAG results in a decrease in the unbound fraction
leading to a decrease in total clearance and an increase in
total plasma concentrations during postoperative infusion.18 19
Pharmacokinetics of ropivacaine in uraemia
On the other hand, the intrinsic (unbound) clearance remains
unchanged.18 19 This is important as it is the unbound
plasma concentration of the drug that is related to systemic
pharmacodynamic effects and toxicity.
The pharmacokinetics of ropivacaine is not altered in
patients with impaired renal function. Although unconju-
gated plasma concentrations of 3-OH-ropivacaine were rela-
tively high (similar to those of PPX), the toxic potential of this
metabolite is negligible. Although a substantial part of the
active metabolite PPX is renally excreted, there is also clini-
cally relevant non-renal elimination of PPX in patients with
impaired renal function. By using a new method, a point esti-
mate and a lower confidence limit of non-renal clearance of
the PPX were estimated by the regression of 1/AUC on renal
clearance of the metabolite after administration of the
mother drug. Owing to inter-individual variation in non-renal
clearance, some patients with impaired renal function may
show increased exposure to PPX resulting from low non-renal
clearance. As these patients also have a relatively longer
half-life of PPX, their steady-state plasma concentration
during, for example, a postoperative continuous epidural
infusion is not expected to be reached until after about
48 h. Furthermore, the clinical consequences are reduced
as the central nervous systemic toxicity of PPX is low.
Acknowledgements
We are grateful to the clinical study nurses: Marja Friman, RN,
Marcus Nilson, RN, Elisa Hurmansalo, RN, and Taina Landg-
ren, RN. We wish to acknowledge Yvonne Askemark for
being responsible for the bioanalytical work and we are
grateful to Quintiles AB, Analytical Services, Sweden, for per-
forming the bioanalyses.
Conflict of interest
J.S. and J.H. are shareholders of AstraZeneca Plc.
Funding
This study was funded by AstraZeneca R&D, Södertälje,
Sweden.
References
1 Ekström G, Gunnarsson UB. Ropivacaine, a new amide-type local
anesthetic agent, is metabolized by cytochromes P450 1A and 3A
in human liver microsomes. Drug Metab Dispos 1996; 24: 955– 61
2 Halldin MM, Bredberg E, Angelin B, et al. Metabolism and excretion
of ropivacaine in humans. Drug Metab Dispos 1996; 24: 962–8
3 Jokinen MJ, Neuvonen PJ, Lindgren L, et al. Pharmacokinetics of
ropivacaine in patients with chronic end-stage liver disease.
Anesthesiology 2007; 106: 43–55
BJA
4 Cockcroft DW, Gault MH. Prediction of creatinine clearance from
serum creatinine. Nephron 1976; 16: 31 –41
5 Slot C. Plasma creatinine determination. A new and specific
Jaffe reaction method. Scand J Clin Lab Invest 1965; 17:
381– 7
6 Engman M, Neidenström P, Norsten-Höög C, Wiklund S-J,
Bondesson U, Arvidsson T. Determination of ropivacaine and
[2H3]ropivacaine in biological samples by gas chromatography
with nitrogen-phosphorus detection or mass spectrometry.
J Chromatogr B Biomed Sci Appl 1998; 709: 57 –67
7 Abdel-Rehim M, Bielenstein M, Askemark Y. Determination of
ropivacaine and its metabolites in patient urine: advantage of
liquid chromatography-tandem mass spectrometry over liquid
chromatography-UV detection and liquid chromatography-mass
spectrometry. Anal Chim Acta 2003; 492: 253– 60
8 Arvidsson T, Askemark Y, Halldin MM. Liquid chromatographic
bioanalytical determination of ropivacaine, bupivacaine and
major metabolites. Biomed Chromatogr 1999; 13: 286– 92
9 Arlander E, Ekström G, Alm C, et al. Metabolism of ropivacaine in
humans is mediated by CYP1A2 and to a minor extent by CYP3A4:
an interaction study with fluvoxamine and ketoconazole as in
vivo inhibitors. Clin Pharmacol Ther 1998; 64: 484–91
10 Emanuelsson B-M, Persson J, Sandin S, Alm C, Gustafsson LL.
Intraindividual and interindividual variability in the disposition
of the local anesthetic ropivacaine in healthy subjects. Ther
Drug Monit 1997; 19: 126–31
11 Draper NR, Smith H. Applied Regression Analysis, 3rd Edn. John
Wiley & Sons, Inc.: New York, 1998; 89–93
12 Rowland M, Tozer TN. Clinical Pharmacokinetics, Concepts and
Applications. Williams & Wilkins: Baltimore, 1995; 257, 367
13 Reynolds F. Metabolism and excretion of bupivacaine in
man: a comparison with mepivacaine. Br J Anaesth 1971; 43:
33–7
14 Varhe A, Olkkola KT, Neuvonen PJ. Diltiazem enhances the effects
of triazolam by inhibiting its metabolism. Clin Pharmacol Ther
1996; 59: 369– 75
15 Zhou SF, Xue CC, Yu XQ, Li C, Wang G. Clinically important drug
interactions potentially involving mechanism-based inhibition of
cytochrome P450 3A4 and the role of therapeutic drug monitor-
ing. Ther Drug Monit 2007; 29: 687– 710
16 Grossman SH, Davis D, Kitchell BB, Shand DG, Routledge PA. Dia-
zepam and lidocaine plasma protein binding in renal disease. Clin
Pharmacol Ther 1982; 31: 350–7
17 Rosenberg PH, Pere P, Hekali R, Tuominen M. Plasma concen-
tration of bupivacaine and two of its metabolites during continu-
ous interscalene brachial plexus block. Br J Anaesth 1991; 66:
25–30
18 Erichsen CJ, Sjövall J, Kehlet H, Hedlund C, Arvidsson T. Pharma-
cokinetics and analgesic effect of ropivacaine during continuous
epidural infusion for postoperative pain relief. Anesthesiology
1996; 84: 834– 42
19 Burm AGL, Stienstra R, Brouwer RP, Emanuelsson B-M, van
Kleef JW. Epidural infusion of ropivacaine for postoperative
analgesia after major postoperative surgery. Anesthesiology
2000; 93: 395– 403
521