Acta Anaesthesiol Scand 2014; 58: 11991213

Printed in Singapore. All rights reserved

2014 The Acta Anaesthesiologica Scandinavica Foundation.

Published by John Wiley & Sons Ltd
ACTA ANAESTHESIOLOGICA SCANDINAVICA

doi: 10.1111/aas.12377

Review Article

A systematic review and meta-analysis of ketamine for

the prevention of persistent post-surgical pain
E. D. McNicol1, R. Schumann2 and S. Haroutounian3

1
Department of Anesthesiology and Pharmacy, Tufts Medical Center, Boston, MA, USA, 2Department of Anesthesiology, Tufts Medical
Center, Boston, MA, USA and 3Department of Anesthesiology, Washington University in St Louis, St Louis, MO, USA

While post-operative pain routinely resolves, persistent postsurgical pain (PPSP) is common in certain surgeries; it causes
disability, lowers quality of life and has economic consequences.
The objectives of this systematic review and meta-analysis were
to evaluate the effectiveness of ketamine in reducing the prevalence and severity of PPSP and to assess safety associated with
its use.
We searched the Cochrane Central Register of Controlled
Trials, MEDLINE and EMBASE through December 2012 for articles in any language. We included randomized, controlled trials
in adults in which ketamine was administered perioperatively
via any route.
Seventeen studies, the majority of which administered
ketamine intravenously, met all inclusion criteria. The overall
risk of developing PPSP was not significantly reduced at any
time point in the ketamine group vs. placebo, nor did comparisons of pain severity scores reach statistical significance. Sensitivity analysis of exclusively intravenous ketamine studies

ost-operative pain routinely resolves after

healing of the surgical site; however, in some
patients, pain persists long after surgery. Persistent
post-surgical pain (PPSP, also referred to as
chronic post-surgical pain) is common, causes disability, lowers quality of life and has economic consequences.1 Reviews of its prevalence report that
2267% of persons who underwent thoracotomy,
3081% of persons who had a limb amputated,
1151% of women who had breast surgery, 356%
of persons who had a cholecystectomy and 037%
of persons who had an inguinal hernia repair
developed PPSP.14 Variations in estimates are in
part due to differences in the definitions of PPSP. A
common definition is that PPSP develops after a
surgical procedure, is not from a pre-existing condition, is of at least 2 months duration, and that
other causes of pain have been excluded.5 While
this definition is comprehensive and was devel-

included in this meta-analysis demonstrated statistically significant reductions in risk of developing PPSP at 3 and 6 months
(P = 0.01 and P = 0.04, respectively). Adverse event rates were
similar between ketamine and placebo groups.
The study data from our review are heterogeneous and demonstrate efficacy of intravenously administered ketamine only in
comparison with placebo. Highly variable timing and dosing of
ketamine in these studies suggest that no unifying effective
regimen has emerged. Future research should focus on clinically
relevant outcomes, should stratify patients with pre-existing
pain and possible central sensitization and should enroll sufficiently large numbers to account for loss to follow-up in longterm studies.
Accepted for publication 30 June 2014
2014 The Acta Anaesthesiologica Scandinavica Foundation.
Published by John Wiley & Sons Ltd

oped with the aim of accurately identifying PPSP,

its applications in research are challenging, with
some studies reporting its use, but not applying all
criteria.6 Risk factors for developing PPSP include
patient (demographic, psychosocial and genetic)
and periprocedural or perioperative factors (duration and type of surgery, extent of intraoperative
nerve damage, and intensity and duration of postoperative pain).1,7
Specific perioperative analgesic interventions
may reduce the incidence of PPSP, but it is unclear
which regimen is of most benefit. Evidence from the
use of regional anesthesia and preemptive or
multimodal analgesia has presented mixed results.1
While the development of PPSP is multifactorial,
one of the major suggested mechanisms for the transition from acute to persistent post-operative pain is
central sensitization.8 During surgery, repeated
painful stimulation of primary afferent neurons

1199
bs_bs_banner

E. D. McNicol et al.

results in the release of excitatory neurotransmitters,

including glutamate, which binds to the N-methylD-aspartate (NMDA) receptor on the post-synaptic
membrane in the dorsal horn. This event alters
calcium influx into the second-order neuron, ultimately
producing
progressively
increased
depolarization of the post-synaptic membrane.9
Perioperative opioid administration may also contribute to hyperalgesia via similar mechanisms.10
The increased excitability and synaptic efficacy of
neurons in central nociceptive pathways secondary
to peripheral tissue damage and repeated
nociceptive input is known as central sensitization;
it reduces mechanical thresholds, exaggerates the
response to noxious stimuli and facilitates a spread
of sensitivity to normal tissue.8 Patients may report
both hyperalgesia (increased sensitivity to painful
stimuli) and allodynia (pain in response to normally
nonpainful stimuli).9,11
Glutamatergic excitation occurs acutely and subacutely following peripheral injury to primary afferent neurons in animal models, and NMDA receptor
blockade has been shown to attenuate spinal
NMDA-mediated sensitization of dorsal horn
neurons.1216 Although the role of NMDA receptors
in long-term central sensitization is unknown, the
concept of perioperative NMDA receptor blockade
to attenuate or prevent central sensitization and subsequent development of PPSP has generated
increasing clinician interest.
Ketamine, a noncompetitive NMDA receptor
antagonist, may have a role in reducing the incidence of PPSP, when administered perioperatively.17
Ketamine is a general anesthetic when used in high
doses. Even low (subanesthetic) doses appear to
reduce both pain and analgesic consumption in
acute post-operative pain.1820 Limited evidence also
describes its effectiveness in both chronic noncancer
and cancer pain populations.21,22 Ketamine may be
administered intranasally, orally or subcutaneously,
particularly when used for chronic pain, but is typically administered intravenously and occasionally
epidurally in the perioperative setting. Patients may
receive a bolus and/or continuous infusion, pre-,
intra- and/or post-operatively. It may also be
administered in combination with an opioid as a
component of patient-controlled analgesia (intravenous or epidural) or with opioids and nonopioids in
the context of a multimodal analgesic regimen.
Analgesic doses employed clinically vary widely,
but intravenous bolus doses in the range of 0.2
0.75 mg/kg and infusions of 27 mcg/kg/min have
been recommended.18,22,23 Common adverse reac-

1200

tions include cardiovascular (hypertension, tachycardia) and central nervous system (vivid dreams,
visual hallucinations, emergence delirium) events,
which are thought to be less common or severe at
subanesthetic doses and with coadministration of a
benzodiazepine.24 Ketamine is a chiral molecule and
is typically administered as a racemic 50 : 50
mixture of the two enantiomers S-ketamine and
R-ketamine. Both enantiomers bind to NMDA
receptors, but S-ketamine is a more potent antagonist and a more powerful analgesic.25
The objectives of this systematic review and metaanalysis are to evaluate the effectiveness of ketamine
administered perioperatively (pre-, intra- and postoperatively) in reducing the prevalence and severity
of PPSP in adult patients and to assess both shortand long-term adverse events associated with its
use in this setting.

Methods
Inclusion and exclusion criteria
We included randomized, placebo- or activecontrolled trials (RCTs), described as doubleblinded, in adults (aged 18 years and older)
undergoing any type of surgical procedure involving incision and associated with PPSP. Recognizing
that persistent pain maybe more easily prevented
than reversed, we nonetheless included studies
where patients may have had pre-existing pain, as
we judged that stipulating patients who did not
have pain pre-operatively would severely limit the
number of eligible studies.
We included studies administering any dose of
ketamine by any route. Both single and multiple
dose studies and studies employing continuous
infusions were assessed. Studies where ketamine
was administered in addition to a traditional analgesic (opioid and/or nonopioid) regimen in one
study group and compared with a group receiving
the same basic regimen (but without ketamine) were
also included.
For published studies, peer-reviewed journal
publication was required; abstracts were not
included unless they were less than 3 years old. We
excluded studies with fewer than 10 participants to
overcome random play of chance on estimation of
treatment effect.26 We excluded studies of experimental pain, case reports and clinical observations.

Outcome measures
Primary outcome: prevalence of PPSP. Our knowledge of the literature suggested that very few

Ketamine for prevention of PPSP

studies were rigorous enough to identify incidence

and that assessment of prevalence varied between
studies, particularly in regard to follow-up periods.
We extracted incidence/prevalence of PPSP based
on its definition within each study. Where data were
presented in several formats, we chose the most
liberal definition of PPSP in an effort to employ the
most uniform possible outcomes reported across
otherwise heterogeneous studies. For example,
where a study reported number of patients with
pain intensity > 3, number of patients with pain
worse than before surgery and number of patients
with any pain at a relevant time point, we chose the
last definition as we expected this would be the
most commonly reported. We meta-analyzed prevalence at 3 months, 6 months and 1 year or later.
Secondary outcomes
1. Intensity or severity of PPSP as measured by
either visual analog or verbal rating scales,
respectively
2. Number of participants experiencing any serious
adverse event
3. Number of participants experiencing specific
adverse events
4. Severity of adverse events
5. Withdrawals due to adverse events
We extracted adverse event data from the acute
perioperative period and from subsequent longterm follow-up to the end point of each study.

Search methods for identification of studies

We searched the following databases through
December 2012: The Cochrane Register of Controlled Trials (CENTRAL), MEDLINE (via Ovid)
and EMBASE (via Ovid). We combined search terms
for ketamine, PPSP and randomized, controlled
studies, without applying language restriction. The
search strategy for MEDLINE was as follows:
1. exp Ketamine/
2. (persist* or chronic).mp. or exp Postoperative
Complications/
3. 1 and 2
4. randomized controlled trial.pt.
5. controlled clinical trial.pt.
6. randomized.ab.
7. placebo.ab.
8. drug therapy.fs.
9. randomly.ab.
10. trial.ab.
11. groups.ab.
12. or/411

13. [animals not (humans and animals)].sh.

14. 12 not 13
15. 14 and 3
This strategy was adapted for each database.
Additional studies were sought from the reference
lists of retrieved articles and reviews. No attempt
was made to assess reporting bias; however, we
searched the clinical trial registry http://
www.clinicaltrials.gov in an attempt to minimize
publication bias.

Data collection and analysis

Data extraction was divided among the review
authors, with each extraction being independently
duplicated, using a standard form. Data suitable for
pooling were entered into RevMan 5 software27 and
double-checked by another author. Extracted data
included information about the type of surgery and
number of participants treated, drug and dosing
regimen, study design (placebo or active control),
study duration and follow-up, outcome measures
and results, withdrawals and adverse events.

Assessment of risk of bias in included studies

Two review authors independently assessed the risk
of bias of all included studies. The review authors
made critical assessments for each of the following
domains: sequence generation (randomization),
allocation concealment, blinding, incomplete
outcome data and selective outcome reporting.
Review author judgment for each domain was
entered into a risk of bias table, as either low risk,
high risk or unclear risk (indicating either lack of
information or uncertainty over the potential for
bias).

Measures of treatment effect

Discrete events such as the proportion of participants experiencing PPSP, or the proportion of participants reporting adverse events, were used to
perform meta-analyses of risk ratio (RR) with 95%
confidence intervals (CIs).28 When a statistically significant RR occurred between interventions, we also
calculated the numbers needed to treat to benefit
(NNTs) or harm (NNHs), which were derived from
the absolute risk reduction (also known as risk difference). Meta-analyses were also undertaken when
comparable data were available from continuous
outcomes, such as severity of PPSP, using weighted
mean differences. We used a fixed effect model for
meta-analysis of both dichotomous and continuous
outcomes. We performed sensitivity analysis to

1201

E. D. McNicol et al.

assess the robustness of our results by alternately

undertaking meta-analysis using a random effects
model.
We performed a per-protocol rather than an
intention-to-treat (ITT) analysis. In intervention
trials for the treatment of existing conditions, participants lost to follow-up in an ITT approach are
assumed to be nonresponders (i.e., still suffer from
the condition). In this study, we assess the prevention of a potential event (PPSP) in patients receiving
a brief perioperative intervention with ketamine.
ITT analysis is not relevant in this case, as patients
lost to follow-up cannot be assumed to experience
or not to experience the event. Therefore, perprotocol analysis was used, and discrepancies
between number of participants enrolled and
number of participants in whom outcomes were
reported are noted in Table 1.

Assessment of heterogeneity
We visually assessed heterogeneity by studying
forest plots and quantified statistical heterogeneity
using the I2 statistic. An I2 value of greater than 50%
is considered to indicate substantial heterogeneity.29
Where possible, we performed the following predetermined subgroup analyses in an attempt to
explain heterogeneity: type of surgery, dosage
regimen, route of administration, definition of PPSP
and follow-up periods.

Results
Results of the search
Our literature search yielded 918 records
(CENTRAL, 281; MEDLINE, 580; EMBASE 57). Our
search of http://www.clinicaltrials.gov produced
five studies (NCT 01017393, NCT00313378,
NCT00726258, NCT01017393, NCT00354029), all of
which were also published articles and were
included in our literature search records.3034 Fortythree articles were selected for full-text review
(Fig. 1). Seventeen studies met all inclusion
criteria.3046
All included studies were published since 2000,
with the majority being less than 5 years old, reflecting the relative novelty of using ketamine in this
arena. All were placebo controlled. One study42 also
included an active control group in which participants received gabapentin. In total, 1015 participants
were randomized to receive ketamine and 785 to
receive placebo, although a small percentage of participants withdrew from their respective studies
before receiving their assigned intervention. The

1202

dose of ketamine administered varied both within

and between studies. Some studies administered
only a single bolus dose of ketamine, whereas the
majority administered a bolus plus a subsequent
continuous infusion. Boluses were typically administered immediately before surgery, with infusions
started directly afterward and continued for up to
72 h post-surgically. The majority of studies administered ketamine intravenously (Suppa et al.43
administered an intramuscular bolus in addition to
an intravenous infusion). Two studies employed the
epidural route; both administered ketamine added
to patient-controlled epidural analgesia.33,46 De Kock
et al.36 included separate arms receiving either epidural or intravenous ketamine. Lastly, four studies
that administered S-ketamine, either intravenously34,40,43 or epidurally,33 reported variously to be
between two to four times as potent as the racemic
mixture.47 Table 1 details the types of surgery performed and duration of follow-up.

Risk of bias in included studies

All but three30,31,45 included studies described satisfactory methods of randomization; however, only
nine studies adequately described methods of allocation concealment. Only 8 of the 17 studies used
adequate methods to ensure blinding. In the other
nine studies, the risk of bias was unclear because
attempts at blinding participants, at blinding assessors or both were not adequately described. Two
studies were assessed as being at high risk of attrition bias (incomplete reporting of outcome data).43,45
In both studies, there were a large number of participants in both arms lost to follow-up, with no
description of the reasons for withdrawal. In the
majority of studies, due to their long-term nature,
there were considerable losses to follow-up, but
these were generally balanced between arms along
with similar reasons for withdrawal. For assessment
of selective reporting, most studies reported on all
of the outcomes described in their Methods sections
and most of these we considered clinically relevant
for this analysis. However, Sen et al.42 reported incidence of PPSP as being significantly lower in the
gabapentin group but did not include supporting
data, Suppa et al.43 evaluated residual pain at 6
months but did not report it and one study did not
report adverse events.46 We were able to confirm
that all outcomes were reported in the five studies
also published on the trial registry http://
www.clinicaltrials.gov. Other potential sources of
bias included the considerable variability in how
each study defined PPSP, although we attempted to

Ketamine for prevention of PPSP

Table 1
Summary of included studies, ketamine vs. placebo.
Study

Surgery, time points when

PPSP assessed, number of
participants (enrolled/at each
time point for selected
ketamine regimen)

Ketamine regimen(s)

Definition of PPSP used in

analysis

Secondary outcomes
measured at 3 months or later

Results of secondary
outcomes

Bilgen et al.35

C-Section under general

anesthesia
6 months and 1 year.
K: 35/35/35
P: 35/35/35
Radical mastectomy with
axillary lymph node
dissection
3 months
K: 13/12
P: 18/18
Surgical resection of rectal
adenocarcinoma
6 months and 1 year.
K: 20/19/19
P: 20/18/17

B: 0.25, 0.5* or 1 mg/kg IV

before induction of
anesthesia

Any pain

Pain intensity (VRS): number

of patients with mild,
moderate, severe or no pain
at all.

No participants had pain

severity > mild at 6 months
or 1 year.

B: 0.5 mg/kg IV before incision

INF: 0.25 mg/kg/h infusion until
end of surgery

NPSI and BPI used, but

unclear how PPSP was
defined

NPSI total score (0100), %

with hyperalgesia at wound
site, surface area of site of
hyperalgesia, nature of pain.

NPSI score very low in both

groups (4.9 vs. 4.45).
NS between groups for any
secondary outcome.

B: 30 min before skin incision,

INF: until end of surgery.
IV arms: B 0.25 mg/kg, INF
0.125 mg/kg/h; or 0.5 mg/kg
and 0.25 mg/kg/h*
Epidural arms: B: 0.25 mg/kg,
INF 0.125 mg/kg/h; or
0.5 mg/kg and 0.25 mg/kg/h.
B: 1 mg/kg IV at induction
INF: 1 mg/kg/h IV during
surgery, then 1 mg/kg over
24 h
B: 0.15 or 0.5 mg/kg* IV

Do you feel any pain at the

scar area?

Analgesic use in those

reporting pain, pain at any
other place, unpleasant
manifestations experienced
since operation.

Acetaminophen codeine
sufficient in all participants
experiencing residual pain.
NS between groups for all
other secondary outcomes.

NPSI > 0

NPSI total score, SF-36.

Participant rating of pain

management as excellent,
acceptable or poor.
For analysis, anyone reporting
acceptable or poor
assigned to prevalence of
PPSP group.
Patient-reported phantom or
stump pain during previous
24 h (used stump pain for
analysis)

Pain intensity (VAS 010)

during rest and movement,
patient satisfaction.

Median NPSI score = 0 in both

groups.
SF-36 scores did not differ
between groups.
No difference between groups,
but orthopedic placebo
subgroup had worst pain
(P = 0.041).

Crousier et al.30

De Kock et al.36

Duale et al.31

Dullenkopf et al.37

Elective thoracotomy
4 months
K: 42/34
P: 44/35
General and orthopedic
(unspecified)
3 months
K: 41/29
P: 33/25

Hayes et al.38

Above and below knee

amputations
6 months
K: 22/15
P: 23/17

B: 0.5 mg/kg pre-induction

INF: 0.15 mg/kg/h 72 h
post-operatively

Joseph et al.32

Thoracotomy with partial

pneumonectomy
3 months
K: 24/18
P: 27/19

B: 0.5 mg/kg epidural at

anesthesia induction
INF: 3 mcg/kg/min epidural
intraop,
1.5 mcg/kg/min 48 h
post-operatively

Limitation of daily activities

Katz et al.39

Radical prostatectomy
6 months
K: 56/36
P: 50/38

Any pain at site of surgery

Mendola et al.40

Thoracotomy with partial or full

pneumonectomy
3 and 6 months
K: 33/31/29
P: 33/30/28

B: 0.2 mg/kg IV either 10 min

before or 70 min* after
incision
INF: 2.5 mcg/kg/min IV for
total of 80 min
S (+) ketamine INF:
0.1 mg/kg/h IV via
elastomeric pump,
pre-incision 60 h

Remerand et al.41

Total hip replacement

3 and 6 months
K: 79/75/72
P: 75/72/70

B: 0.5 mg/kg IV pre-incision

INF: 2 mcg/kg/min IV 24 h

Unclear

Ryu et al.33

Thoracotomy
3 months
K: 103/65
P: 106/68

Any pain around the incision

site

Sen et al.42

Elective abdominal
hysterectomy
3 and 6 months
K: 20/20/20
P: 20/20/20
Hemorrhoidectomy
K: 43/39
P: 39/38

S (+) ketamine PCEA (+ 0.12%

levobupivacaine and
2 mcg/ml fentanyl).
B: 1.2 mg as soon as patient
arrived at operating room
INF: 1.2 mg/h, additional
boluses of 1 mg, lockout
20 min, until 3rd
post-operative day
B: 0.3 mg/kg IV 1 h before
surgery
INF: 0.05 mg/kg/h IV until end
of surgery

Incidence of incisional pain

no further details.
Not reported in Results

Pain intensity (VRS 010),

impact of pain on daily
activities.

S (+) ketamine
B: 0.35 mg/kg IV immediately
before incision
INF: 5 mcg/kg/min until 2 min
post-surgery

Any pain (NRS > 0) at rest

Pain intensity at rest, sitting

and with defecation.

Spreng et al.34

NPSI > 1

Incidence of stump and

phantom limb troublesome
pain (NRS 6), severity of
stump and phantom limb
pain (NRS highest, lowest or
usual), duration of stump
and phantom limb pain.
Pain intensity (NRS 010) at
rest and with movement,
analgesic consumption,
limitation of normal activity
induced by pain, limitation of
the abduction of the surgery
side arm, any complication.
MPQ, pain interference in daily
life, methods of pain relief
sought, medication use.

Worst pain intensity (NRS

010), pain localization, total
amount of analgesics taken
up to 6th post-operation
month. PPSP considered
mild for NRS 03, moderate
and severe if hampering
daily life.
Pain intensity (NRS 0100) at
rest, with walking, in other
locations, analgesic use,
distance and difficulty
walking.
Pain severity (VAS 0100) at
rest and with movement,
allodynia, numbness.

NS between groups for all

secondary outcomes.
Note: The development of
significant phantom limb
pain after day 3 was treated
with amitriptyline and/or
sodium valproate.
NS between groups for all
secondary outcomes.

NS in number of participants
taking medication (none in
any group).
NS between groups for all
other secondary outcomes.
NS between groups for all
secondary outcomes.

At 6 months, 10 P group vs. 3

K group NRS at rest > 3
(P = 0.04).
NS between groups for all
other secondary outcomes.
NS between groups for all
secondary outcomes.

Pain intensity low in both

groups and NS different.
Data not presented for impact
on daily activities, but
reported to be similar.
NS between groups.

1203

E. D. McNicol et al.
Table 1 Continued
Study

Surgery, time points when

PPSP assessed, number of
participants (enrolled/at each
time point for selected
ketamine regimen)

Ketamine regimen(s)

Definition of PPSP used in

analysis

Secondary outcomes
measured at 3 months or later

Results of secondary
outcomes

Suppa et al.43

Elective repeat C-section

3 years
K: 28/13
P: 28/13

Any residual pain at the scar

area

Thoracotomy, primarily for lung

cancer
3 and 6 months
K: 25/22/22
P: 25/22/22

Analgesic drugs used for

wound pain, wound
dysesthesia, pain in other
sites, analgesic drugs for
pain in other sites.
Pain intensity (NRS 010) at
baseline and worst,
unpleasant sensations on
the surgical wound, pt felt
inconvenienced by the
wound, number of pts who
received pain medication.

All dichotomous, NS between

groups.

Suzuki et al.44

S (+) ketamine
B: 0.5 mg/kg IM 10 min
post-partum
INF: 2 mcg/kg/min IV 12 h
post-operatively
INF: 0.05 mg/kg/h IV after
intubation 72 h
post-operatively

Sveticic et al.45

Major elective orthopedic

surgery
3 and 6 months
K: 176/49/17
P: 176/42/19

Pain at the same location

worse than before surgery

Pain intensity (VAS 010).

Wilson et al.46

Above and below knee

amputation
3, 6 and 12 months
K: 24/15/15/14
P: 29/19/16/15

PCA: 1.5 mg (+ 1.5 mg

morphine) IV Q8 min prn up
to 10 in 1 h, until pts
required an average of < 1
PCA bolus/h during the last
12 h
B: 0.5 mg/kg (+ 0.5%
bupivacaine) epidural before
starting surgery.
INF: 3.3 mg/kg/l (+
bupivacaine 0.125%)
epidural at 15 ml/h
intraoperatively, 1020 ml/h
post-operatively 4872 h

Any pain in the stump

Pain intensity (VAS 010) of

phantom and stump pain,
MPQ, NPSI, QST, Hospital
anxiety and depression
scale, analgesic use.

Baseline (usual) NRS score

of 1

Baseline pain intensity lower in

K group at 3 months
(P = 0.02). Number of pts
who received pain
medication lower at 3
months (P = 0.03).
All other secondary outcomes
NS between groups.
Pain intensity NS between
groups at 3 or 6 months.

Anxiety (P < 0.001) and

depression (P = 0.003)
reduced in K group vs.
pre-operatively from 3
months until end of study,
but not in P group vs.
pre-operatively.
All other secondary outcomes
NS between groups.

*Regimen used in meta-analysis.

K, Ketamine; P, placebo; B, bolus; INF, infusion; IV, intravenous; BPI, brief pain inventory; MPQ, McGill Pain Questionnaire; NPSI, neuropathic pain symptom inventory; NS, not statistically
significant; PCA, patient-controlled analgesia; PCEA, patient-controlled epidural analgesia; QST, quantitative sensory testing; VAS, visual analog scale; VRS, verbal rating scale. PPSP,
persistent post-surgical pain; NRS, numeric rating scale; IM, intramuscular.

homogenize data as much as possible, as described

above. Also, the majority of studies were powered
to show a difference in acute outcomes rather than
for those measured at 3 months and later.

Primary outcome
Sixteen studies presented data that contributed to
the meta-analysis of prevalence of PPSP (Sen et al.42
did not report prevalence). In studies where more
than one ketamine regimen was administered, we
chose the dose closest to those used in other studies.
We were able to perform sub-analysis at three separate time points: 3 months (for one study, we used
data from 4 months31), 6 months and at 1 year or
later.
At 3 months, the overall RR of developing PPSP
when both intravenous and epidural routes were
assessed was 0.84 in the ketamine group vs. placebo;
that is, a 16% relative reduction in the probability of
developing PPSP, but this was not statistically significant (P = 0.06, 95% CI: 0.701.01) (Fig. 2). No
single study demonstrated a statistically significant
difference in PPSP risk between ketamine and
placebo. Overall, 109 of 384 (28%) participants
receiving ketamine reported PPSP at 3 months vs.
137 of 387 receiving placebo (35%).

1204

At 6 months and at 1 year or later, the overall RR

was again not statistically significant when both
intravenous and epidural routes were assessed. The
overall number of participants reporting declined at
each time point, with 260 participants from ketamine
arms reporting at 6 months and 71 reporting at 1
year or later. For the 6-month and 1-year or later
analyses only one study demonstrated a statistically
significant reduction in risk of developing PPSP (at 6
months).41 In this study of patients undergoing hip
replacement, 6 of 72 participants in the ketamine
group vs. 15 of 70 participants in the placebo group
reported PPSP (P = 0.04).
One study compared ketamine with an active
comparator, gabapentin.42 The authors reported that
the prevalence of incisional pain was reduced in the
gabapentin group at 3 and 6 months compared with
both the ketamine and placebo groups, but did not
present supporting data.

Secondary outcomes
Mean intensity of PPSP was reported in five studies
at 3 months and four studies at 6 months, using
either a visual analog or verbal rating scale (zero
indicating no pain, 10 indicating worst imaginable
pain). At 3 months, the reduction in pain intensity in

Ketamine for prevention of PPSP

918 records
identified
through database
searching

5 additional
records identified
through
clinicaltrials.gov

568 records after duplicates

removed

568 records
screened

525 records
excluded

26 full-text articles excluded:

43 full-text
articles
assessed for
eligibility

No pain outcome at 3
months: n = 20 (many studies
had additional reasons for
exclusion)
Ketamine part of multimodal
strategy not employed in
control arm: n = 1
All arms received same dose of
ketamine: n = 1

17 studies
included in
qualitative
analysis

Abstract > 3 years old: n = 1

< 10 participants in each group:
n=1
Retrospective study: n = 1

17 studies
included in
quantitative
synthesis
(metaanalysis)

Single blind: n = 1

Fig. 1. Study flow diagram.

patients receiving ketamine (n = 154) vs. placebo

(n = 147) was neither statistically nor clinically significant, applying interpretive standards for the
latter described elsewhere.48 Results were similar at
6 months. There were insufficient data for metaanalysis beyond 6 months.
No individual studies demonstrated a reduction
in pain intensity at 3 months, and only one39 demonstrated a reduction in intensity in those receiving
ketamine at 6 months. In this study, those receiving
ketamine had a mean visual analog scale pain intensity of 2.8 1.1 vs. those receiving placebo 3.4 1.1
(P = 0.02), although the number of participants was
low in each study arm (n = 36 and n = 38 in the
ketamine and placebo groups, respectively).
In the single active comparator study,42 participants in the gabapentin group reported lower pain

scores at 3 and 6 months vs. ketamine and placebo,

although mean verbal rating scale scores were low
in all groups. The authors stated that those receiving
gabapentin suffered less impact of pain on their
daily activities at 3 months, but did not present supporting data.
Adverse events were almost exclusively monitored during the acute administration phase, either
for the duration of the ketamine infusion or shortly
thereafter. Adverse events were only mentioned at
long-term follow-up (3 months and onward) in relation to participants withdrawing from the
study.31,32,36,38,40,44,46 No long-term adverse events
were attributed to the interventions by the investigators, and they appeared to be evenly balanced
between ketamine and placebo groups. Mostly,
long-term adverse events included death or recurrence of presenting disease, at rates typical in such
populations. For adverse events monitored in the
acute phase (typically up to 72 h post-operatively)
and commonly associated with ketamine, there was
no statistical difference between ketamine and
placebo for hallucinations and nightmares/vivid
dreams, but there was an increase in visual disturbances (e.g., nystagmus, diplopia), with an RR of
3.13 (95% CI: 1.655.94), translating to a NNH of
around 9 (95% CI: 617). Similarly, there were no
differences between ketamine and placebo for sedation, nausea and vomiting, which are adverse effects
that might be expected to be less frequent in the
ketamine arms because of reduced opioid consumption attributed to ketamine administration. The
single study comparing ketamine with gabapentin
reported no difference in the incidence of common
side effects. There were insufficient data to compare
study withdrawal rates because of adverse events or
severity of adverse events. Lastly, there was no difference in the rate of serious adverse events, all of
which were attributed to underlying disease and not
to the interventions.

Subgroup analysis investigation of heterogeneity

Dose of ketamine. We were unable to perform subgroup analysis based on dose of ketamine as the
regimens varied widely, particularly in the dose and
duration of continuous infusions.
Follow-up period and surgical procedure. Results for
each time point (3 months, 6 months and 1 year or
later) and type of surgical procedure are presented
as subgroups within our analyses of the primary
outcome; however, participant numbers for individual procedures are small, and only arthroplasty

1205

E. D. McNicol et al.

I2

P
P

I2

P
P

P
P
2

I2
P

I2

Fig. 2. Forest plot of comparison: Incidence of persistent post-surgical pain: ketamine vs. placebo; outcome: Incidence at 3 months, all
studies. M-H, Mantel-Haenszel; CI, confidence interval.

1206

Ketamine for prevention of PPSP

at 6 months shows a statistically significant reduction in PPSP in those participants receiving

ketamine, with an RR of 0.43 (95% CI: 0.190.99).
Route of administration. While subgroup analysis by
route of administration was possible for the primary
outcome, only three studies used the epidural
route,33,36,46 therefore sub-analyses are presented for
intravenous data only (including the study by Suppa
et al.,43 which administered an intramuscular bolus
before the intravenous infusion). Unlike the metaanalysis of combined routes, analysis of exclusively
intravenous studies at 3 months demonstrated a statistically significant RR of 0.75 (P = 0.01, 95% CI:
0.600.93); that is, a 25% reduction in risk of developing PPSP. This translates to an NNT of 12; that is,
on average, 12 patients would need to be treated
with ketamine for one less patient to develop PPSP
than would be the case if they received placebo
(Fig. 3). Similarly, at 6 months, removal of epidural
data resulted in meta-analysis demonstrating a
statistically significant RR of 0.70 (P = 0.04,
95% CI: 0.500.98); that is, a 30% reduction in risk of
developing PPSP (NNT of 14, Fig. 4). At 12 months
or later, overall risk reduction remained
nonstatistically significant for the intravenous data
sub-analysis.
Definition of PPSP. We scrutinized the prevalence
of PPSP in the placebo group of each study. If each
study was measuring the same outcome in similar
populations and employing similar methodologies,
PPSP rates should also be similar among studies of
similar surgeries. In our analysis, where a study presented data that would allow more than one definition of PPSP prevalence (e.g., number of patients
with pain intensity > 1 on 010 numeric rating
scale), we chose the one which most closely aligned
with other studies in an attempt to homogenize
overall data (Table 1). Consequently, prevalence of
PPSP in placebo groups in the included studies was
generally consistent (and within ranges reported for
various surgeries in the literature), but with some
outliers. For example, placebo rates of PPSP in the
included thoracotomy studies were 69%, 63%, 43%
and 64% at 3 months, and 39% and 50% at 6 months.
Removing studies with outlying rates did not significantly affect our analysis.

Sensitivity analysis
We re-analyzed our results using random effects
models. All meta-analyses that had previously not
demonstrated a statistically significant difference

remained so. In addition, our sub-analyses of

studies that only employed the intravenous or intramuscular route were no longer statistically significant; that is, at 3 and 6 months, although best point
estimates were similar. For secondary outcomes,
changing model of analysis had no effect on statistical significance of any of the analyses (i.e., they
remained nonsignificant), with the exception of
vision disorders, which, when analyzed using a
random effects model no longer demonstrated statistical significance.

Discussion
Summary of main results
The results for our primary outcome demonstrate
that perioperative ketamine reduces the risk of
developing PPSP at 3 and 6 months, but only after
removal of epidural studies from our metaanalyses. This result is consistent with a prior
study. De Kock et al.36 compared epidural and
intravenous routes and, perhaps surprisingly, given
experimental work demonstrating the importance
of spinal NMDA receptors in nociceptive input
amplification, showed that intravenous, but not
epidural, administration was successful in reducing analgesic requirements at 6 months after
surgery. Several reasons may be considered for this
apparently contradictory finding. The involvement
of spinal NMDA receptors demonstrated in animal
studies of peripheral nerve injury may be less
important clinically in post-operative pain than
previously thought.24 If spinal NMDA receptors do
not serve as the primary target of ketamine, the
lower plasma concentrations achieved with epidural administration when compared with intravenous administration49 likely translates to a lower
systemic efficacy. On the other hand, ketamine
given epidurally might be rapidly absorbed into
the bloodstream via epidural veins, and the epidural ketamine dose might be lower than the IV
doses employed, accounting for these results.
Additionally, superior analgesia and opioidsparing effects associated with ketamine administration during the acute post-operative phase may
be mediated primarily by attenuation of opioidinduced hyperalgesia, which has a substantial
supraspinal component.
Best point estimates at 6 months and at 1 year or
later show a trend toward greater risk reductions in
ketamine groups than at 3 months, suggesting that
lack of statistical significance may be due to the
small sample size at these later time points.

1207

E. D. McNicol et al.

P
P

I2

P
P

I2

P
P
2

I2
P

I2

Fig. 3. Forest plot of comparison: Incidence of persistent post-surgical pain: ketamine vs. placebo; outcome: incidence at 3 months,
intravenous studies only. M-H, Mantel-Haenszel; CI, confidence interval.

A 25% relative risk reduction in PPSP prevalence

at 3 months in the intravenous meta-analysis is
small, as reflected in the NNT of 12. Indeed, when
sensitivity analysis was performed using a random
effects model, the result was no longer statistically

1208

significant, reflecting the slim margin of statistical

significance in our original analysis. However, even
minor reductions in the incidence of PPSP may have
a considerable public health impact. A 1996 estimate
of cost savings based on preventing the transition

Ketamine for prevention of PPSP

P
P

I2

P
P

P
P
2

I2

I2
P

I2

Fig. 4. Forest plot of comparison: Incidence of persistent post-surgical pain: ketamine vs. placebo; outcome: incidence at 6 months,
intravenous studies only. M-H, Mantel-Haenszel; CI, confidence interval.

1209

E. D. McNicol et al.

from acute to PPSP in a 30-year-old patient, suggested that $1 million may be saved during the
patients lifetime.50
At 3 months, both the Chi2 and I2 statistics imply
that the data are homogeneous, apparently justifying the use of a fixed effect model; however, the
variability in types of surgery, definitions of PPSP
and dosing of ketamine suggest that heterogeneity
does exist. The overall NNT of 12 may not be indicative of reductions in risk for specific surgeries where
the prevalence of PPSP differs. In surgeries with a
lower PPSP prevalence, reductions in risk of developing PPSP are likely to be of a lesser margin; that is,
more patients would need to receive ketamine in
order to achieve a statistically significant reduction
in the incidence of PPSP, resulting in a higher NNT.
The dosage regimens employed between studies
were heterogeneous and prevented a subgroup
analysis. However, in studies that assigned more
than one ketamine regimen, higher doses of
ketamine do not appear to consistently improve the
persistent pain outcomes.3537,39 Therefore, we
cannot draw any conclusions regarding the optimal
dose or timing (pre-emptive vs. preventative) of
ketamine in reducing PPSP. It has been suggested
that the inflammatory response may peak around
48 h post-surgery and therefore, ketamine should be
administered for at least this duration postoperatively;33 but those studies32,34,38,40,44,46 that did
administer ketamine for 48 h or longer did not demonstrate increased efficacy vs. those that did not.
Similarly, the variability between studies in
anesthetic and analgesic regimens employed in
addition to ketamine precluded us from performing
sensitivity analyses based on general vs. regional
anesthesia or multimodal vs. conventional postoperative analgesia.
Reductions in pain intensity in participants
receiving ketamine were neither statistically nor
clinically significant at either 3 or 6 months.
However, the mean pain intensity was generally low
at either time point in both ketamine and placebo
groups.
When subanesthetic doses of ketamine were
administered, rates of ketamine-associated (including neuro-psychiatric) side effects did not differ
between ketamine and placebo for the majority of
comparisons, confirming a widely accepted clinical
perception. The single statistically significant difference was observed for the rate of visual disturbances. However, the data were heterogeneous and
may have reflected differences in definitions of, or
methods of, reporting this adverse event. It appears

1210

that any reductions in opioid use in those treated

with ketamine were not accompanied by reductions
in opioid-induced side effects. Long-term safety
data were scarce and what were available were not
attributed to ketamine administration. Direct
neurotoxicity of ketamine, particularly with intrathecal administration at high doses, has been reported
in animal models.20,51,52 None of the included studies
reported signs of clinical neurotoxicity, either
acutely or in those patients completing long-term
follow-up.

Overall completeness and applicability of

evidence
The included studies covered a wide variety of surgeries, but did not include many others associated
with developing PPSP, such as nephrectomy, cholecystectomy or hernia repair. Additionally, adverse
events were rarely monitored beyond the acute
post-operative stage. The clinical relevance of the
number of patients with any pain at study endpoints is unclear, but the evidence would undoubtedly be more applicable if more of the studies had
measured pain that was clinically meaningful to a
patient. Lastly, it was unclear in many studies
whether patients had pre-existing pain; that is,
chronic pain before surgery. Therefore, the efficacy
of ketamine in such a population has not been separately investigated.

Potential biases from the studies and the review

process
The 17 included studies generally had low or
unclear risk of bias for the five assessed domains,
with high risk occurring only in four studies and
only for attrition and reporting bias. More likely,
sources of bias included the low number of participants contributing data and the heterogeneity of
reporting of PPSP. For the former, there were a
mean number of 60 participants (median 42)
assigned to receive ketamine in each study and a
mean number of 46 (median 33) assigned to receive
placebo, with the number of participants reporting
data dropping substantially at later time points. As
mentioned in the Methods section, we included
studies where enrolled participants did have, or may
have had, pre-existing pain of any kind. If we had
included only those studies where the authors
explicitly reported that they excluded patients with
pre-existing pain, only five studies would have met
the selection criteria.3133,42,43
Our search strategies were concise and did not
contain terms for individual surgeries; therefore, it

Ketamine for prevention of PPSP

is possible that we may have missed relevant

studies. However, all included studies found in reference sections of other papers (original reports or
reviews) were also found in at least one of our
search strategies, suggesting that our searches were
sufficiently sensitive. By searching for nonpublished
data, we believe that our review is not subject to
publication bias, but we did not contact study
authors or manufacturers for unpublished studies.

Comparison with other reviews

Several qualitative and quantitative reviews have
studied the acute effects (7 days or less postoperatively) of perioperative ketamine. Findings
were inconsistent both within and between reviews
(based on different inclusion criteria for the latter) in
the measured outcomes, which included pain intensity, opioid use and opioid-related and ketaminerelated adverse events.18,5356 In general, any
differences in the outcomes studied between
ketamine- and placebo-treated patients were minor.
One review of various agents for prevention of
chronic pain after surgery employed different inclusion criteria to ours and did not perform an analysis
of mean pain severity or of adverse events.57 Despite
these differences, the review reported similar minor
reductions in the number of patients developing
chronic pain when comparing ketamine with
placebo.
As with our analysis, the above reviews discussed
the heterogeneity of included studies and were
unable to recommend an optimal timing or route of
administration or dose regimen of ketamine. Direct
extrapolations of short-term results to long-term
effectiveness are challenging in that acute studies
generally assess pain severity or opioid consumption, whereas long-term studies generally assess
prevalence of pain as their primary outcome.
However, many studies included in our review
demonstrated short-term efficacy of ketamine that
did not translate to long-term reductions in PPSP.
This suggests that while increased acute pain
appears to be a risk factor for developing PPSP,58
reduction of acute pain intensity may not be directly
linked to reduced prevalence of developing persistent pain. Short-term studies may not predict longterm efficacy, but the reduced numbers of
participants at later time points also impairs the
ability to demonstrate a statistically significant
result. In the future, larger studies presenting acute
data for the subset of participants who also reported
long-term data may clarify this relationship.

Conclusions
In this meta-analysis of all eligible studies, combining intravenous and epidural administration,
ketamine did not provide a significant reduction of
PPSP at 3 and 6 months. In common with systematic
reviews of short-term effects of ketamine on acute
pain outcomes, the study data from our review of
outcomes at 3 months or later are heterogeneous
and suggest efficacy of intravenous ketamine only in
comparison with placebo in preventing PPSP at 3
and 6 months. There was no evidence to support
epidural ketamine administration for PPSP prevention. This may be important as concerns about direct
ketamine neurotoxicity exist.
Highly variable timing and dosing of ketamine
in these studies suggest that no unifying effective
regimen has emerged despite the long history of
ketamine in clinical practice. Based on the variability of anesthetic and analgesic regimens employed,
there may be increased efficacy of ketamine when
used as part of a multimodal regimen. Given
the multiplicity of neurotransmitters, growth
factors and pathways involved in the transition
from acute to chronic pain, it would seem logical
that a combination of mechanisms should be targeted via a variety of approaches in order to
increase the likelihood of inhibiting central
sensitization. It is also unclear whether ketamine
may be more effective in certain surgeries, or in
specific populations, for example, in those who are
opioid tolerant and/or who have pre-existing pain.
However, these data suggest that perioperative
ketamine is safe; therefore, it may be appropriate in
patients undergoing painful surgeries or who are
expected to require large doses of opioids postoperatively. We found insufficient data to make
comparisons of ketamine with other interventions
used to prevent PPSP.
Future research should focus on clinically relevant outcomes; that is, pain that affects functioning
or health-related quality of life. Several large studies
recently addressed the risk factors for PPSP development,59,60 and future studies should attempt to
stratify populations at high risk for PPSP, to identify
which patients may benefit the most from specific
interventions. It may be necessary to identify
patients with pre-operative pain and possible
central sensitization prior to enrollment. The dose
and the duration of perioperative ketamine administration should be adequate to cover the critical
period of susceptibility to central sensitization and
neural plasticity. Lastly, sufficiently large numbers

1211

E. D. McNicol et al.

of patients should be enrolled to account for inevitable loss to follow-up in long-term studies.
Conflicts of interest: No conflicts of interest
declared.
Funding: The review was funded in part by the
Richard Saltonstall Charitable Foundation, USA. All
other funding was departmental.

References
1. Macrae WA. Chronic post-surgical pain: 10 years on. Br J
Anaesth 2008; 101: 7786.
2. Haroutiunian S, Nikolajsen L, Finnerup NB, Jensen TS. The
neuropathic component in persistent postsurgical pain: a
systematic literature review. Pain 2013; 154: 95102.
3. Macrae WA. Chronic pain after surgery. Br J Anaesth 2001;
87: 8898.
4. Perkins FM, Kehlet H. Chronic pain as an outcome of
surgery: a review of predictive factors. Anesthesiology 2000;
93: 112333.
5. Macrae WA, Davies HTO. Chronic postsurgical pain, In:
Crombie IK, Linton S, Croft P, Von Korff M, LeResche L eds.
Epidemiology of pain. Seattle: International Association for
the Study of Pain, 1999: 12542.
6. Johansen A, Romundstad L, Nielsen CS, Schirmer H,
Stubhaug A. Persistent postsurgical pain in a general population: prevalence and predictors in the Troms study. Pain
2012; 153: 13906.
7. Macintyre PE, Schug SA, Scott DA, Visser EJ, Walker SM,
APM:SE Working Group of the Australian and New Zealand
College of Anaesthetists and Faculty of Pain Medicine eds.
Acute pain management: scientific evidence, 3rd. Melbourne: ANZCA & FPM, 2010.
8. Woolf CJ. Central sensitization: implications for the diagnosis and treatment of pain. Pain 2011; 152: S215.
9. Woolf CJ, Salter MW. Neuronal plasticity: increasing the
gain in pain. Science 2000; 288: 17658.
10. Colvin LA, Fallon MT. Opioid-induced hyperalgesia: a clinical challenge. Br J Anaesth 2010; 104: 1257.
11. Sandkuhler J. Models and mechanisms of hyperalgesia and
allodynia. Physiol Rev 2009; 89: 70758.
12. Du J, Zhou S, Coggeshall RE, Carlton SM. N-methylD-aspartate-induced excitation and sensitization of normal
and inflamed nociceptors. Neuroscience 2003; 118: 547
62.
13. Kaneko M, Kaneko T, Kaneko R, Chokechanachaisakul U,
Kawamura J, Sunakawa M, Okiji T, Suda H. The role of
N-methyl-d-aspartate receptor subunits in the rat thalamic
mediodorsal nucleus during central sensitization. Brain Res
2011; 1371: 1622.
14. Richebe P, Rivat C, Laulin JP, Maurette P, Simonnet G.
Ketamine improves the management of exaggerated postoperative pain observed in perioperative fentanyl-treated rats.
Anesthesiology 2005; 102: 4218.
15. Davies SN, Lodge D. Evidence for involvement of
N-methylaspartate receptors in wind-up of class 2 neurones in the dorsal horn of the rat. Brain Res 1987; 424: 4026.
16. Kovcs G, Kocsis P, Tarnawa I, Horvth C, Szombathelyi Z,
Farkas S. NR2B containing NMDA receptor dependent
windup of single spinal neurons. Neuropharmacology 2004;
46: 2330.
17. Weinbroum AA. Non-opioid IV adjuvants in the
perioperative period: pharmacological and clinical aspects
of ketamine and gabapentinoids. Pharmacol Res 2012; 65:
41129.

1212

18. Bell RF, Dahl JB, Moore RA, Kalso EA. Perioperative
ketamine for acute postoperative pain. Cochrane Database
Syst Rev 2006; (1): CD004603, doi: 10.1002/14651858.
CD004603.pub2.
19. Katz J, Clarke H. Preventive analgesia and beyond: current
status, evidence, and future directions, In: Macintyre PE,
Walker SM, Rowbotham DJ eds. Clinical pain management:
acute pain, 2nd edn. London: Hodder Arnold, 2008: 15498.
20. Schmid RL, Sandler AN, Katz J. Use and efficacy of low-dose
ketamine in the management of acute postoperative pain: a
review of current techniques and outcomes. Pain 1999; 82:
11125.
21. Bell RF, Eccleston C, Kalso EA. Ketamine as an adjuvant to
opioids for cancer pain. Cochrane Database Syst Rev 2012;
(11): CD003351, doi: 10.1002/14651858.CD003351.pub2.
22. Hocking G, Cousins MJ. Ketamine in chronic pain management: an evidence-based review. Anesth Analg 2003; 97:
17309.
23. White PF, Way WL, Trevor AJ. Ketamine its pharmacology
and therapeutic uses. Anesthesiology 1982; 56: 11936.
24. De Kock MF, Lavandhomme PM. The clinical role of NMDA
receptor antagonists for the treatment of postoperative pain.
Best Pract Res Clin Anaesthesiol 2007; 21: 8598.
25. Oye I, Paulsen O, Maurset A. Effects of ketamine on
sensory perception: evidence for a role of N-methyl-Daspartate receptors. J Pharmacol Exp Ther 1992; 260: 1209
13.
26. Moore RA, Gavaghan D, Tramr MR, Collins SL, McQuay
HJ. Size is everything large amounts of information are
needed to overcome random effects in estimating direction
and magnitude of treatment effects. Pain 1998; 7: 20916.
27. Review Manager (RevMan) [Computer program]. Version
5.2. Copenhagen: The Nordic Cochrane Centre, 2012; The
Cochrane Collaboration.
28. Deeks JJ. Issues in the selection of a summary statistic for
meta-analysis of clinical trials with binary outcomes. Stat
Med 2002; 21: 1575600.
29. Deeks JJ, Higgins JPT, Altman DG. Chapter 9: analysing data
and undertaking meta-analyses, In: Higgins JPT, Green S
eds. Cochrane handbook for systematic reviews of interventions. 5.0. Chichester: John Wiley & Sons, 2008: 24396.
30. Crousier M, Cognet V, Khaled M, Gueugniaud P-Y, Piriou
V. Effect of ketamine on prevention of postmastectomy
chronic pain. A pilot study. Ann Fr Anesth Reanim 2008; 27:
98793.
31. Duale C, Sibaud F, Guastella V, Vallet L, Gimbert Y-A, Taheri
H, Filaire M, Schoeffler P, Dubray C. Perioperative ketamine
does not prevent chronic pain after thoracotomy. Eur J Pain
2009; 13: 497505.
32. Joseph C, Gaillat F, Duponq R, Lieven R, Baumstarck K,
Thomas P, Penot-Ragon C, Kerbaul F. Is there any benefit to
adding intravenous ketamine to patient-controlled epidural
analgesia after thoracic surgery? A randomized doubleblind study. Eur J Cardiothorac Surg 2012; 42: e5865.
33. Ryu HG, Lee CJ, Kim YT, Bahk JH. Preemptive low-dose
epidural ketamine for preventing chronic postthoracotomy
pain: a prospective, double-blinded, randomized, clinical
trial. Clin J Pain 2011; 27: 3048.
34. Spreng AJ, Dahl V, Raeder J. Effects of perioperative S (+)
ketamine infusion added to multimodal analgesia inpatients
undergoing ambulatory haemorrhoidectomy. Scand J Pain
2010; 1: 1005.
35. Bilgen S, Kner O, Tre H, Menda F, Fiicioglu C, Ayka B.
Effect of three different doses of ketamine prior to general
anaesthesia on postoperative pain following caesarean delivery: a prospective randomized study. Minerva Anestesiol
2012; 78: 4429.

Ketamine for prevention of PPSP

36. De Kock M, Lavandhomme P, Waterloos H. Balanced analgesia in the perioperative period: is there a place for
ketamine? Pain 2001; 92: 37380.
37. Dullenkopf A, Muller R, Dillmann F, Wiedemeier P, Hegi
TR, Gautschi S. An intraoperative pre-incision single dose of
intravenous ketamine does not have an effect on postoperative analgesic requirements under clinical conditions.
Anaesth Intensive Care 2009; 37: 7537.
38. Hayes C, Armstrong-Brown A, Burstal R. Perioperative
intravenous ketamine infusion for the prevention of persistent post-amputation pain: a randomized, controlled trial.
Anaesth Intens Care 2004; 32: 3308.
39. Katz J, Schmid R, Snijdelaar DG, Coderre TJ, McCartney CJ,
Wowk A. Pre-emptive analgesia using intravenous fentanyl
plus low-dose ketamine for radical prostatectomy under
general anesthesia does not produce short-term or long-term
reductions in pain or analgesic use. Pain 2004; 110: 70718.
40. Mendola C, Cammarota G, Netto R, Cecci G, Pisterna A,
Ferrante D, Casadio C, Della Corte F. S+ -ketamine for
control of perioperative pain and prevention of post thoracotomy pain syndrome: a randomized, double-blind study.
Minerva Anestesiol 2012; 78: 75766.
41. Remerand F, Le Tendre C, Baud A, Couvret C, Pourrat X,
Favard L, Laffon M, Fusciardi J. The early and delayed analgesic effects of ketamine after total hip arthroplasty: a prospective, randomized, controlled, double-blind study.
Anesth Analg 2009; 109: 196371.
42. Sen H, Sizlan A, Yanarates O, Emirkadi H, Ozkan S, Dagli G,
Turan A. The effects of gabapentin on acute and chronic pain
after inguinal herniorrhaphy. Eur J Anaesthesiol 2009; 26:
7726.
43. Suppa E, Valente A, Catarci S, Zanfini BA, Draisci G. A study
of low-dose S-ketamine infusion as preventive pain treatment for cesarean section with spinal anesthesia: benefits
and side effects. Minerva Anestesiol 2012; 78: 77481.
44. Suzuki M, Haraguti S, Sugimoto K, Kikutani T, Shimada Y,
Sakamoto A. Low-dose intravenous ketamine potentiates
epidural analgesia after thoracotomy. Anesthesiology 2006;
105: 1119.
45. Sveticic G, Farzanegan F, Zmoos P, Zmoos S, Eichenberger
U, Curatolo M. Is the combination of morphine with
ketamine better than morphine alone for postoperative
intravenous patient-controlled analgesia? Anesth Analg
2008; 106: 28793.
46. Wilson JA, Nimmo AF, Fleetwood-Walker SM, Colvin LA. A
randomised double blind trial of the effect of pre-emptive
epidural ketamine on persistent pain after lower limb amputation. Pain 2008; 135: 10818.
47. Arendt-Nielsen L, Nielsen J, Petersen-Felix S, Schnider TW,
Zbinden AM. Effect of racemic mixture and the (S+) isomer
of ketamine on temporal and spatial summation of pain. Br J
Anaesth 1996; 77: 62531.
48. Dworkin RH, Turk DC, Wyrwich KW, Beaton D, Cleeland
CS, Farrar JT, Haythornthwaite JA, Jensen MP, Kerns RD,
Ader DN, Brandenburg N, Burke LB, Cella D, Chandler J,
Cowan P, Dimitrova R, Dionne R, Hertz S, Jadad AR, Katz
NP, Kehlet H, Kramer LD, Manning DC, McCormick C,
McDermott MP, McQuay HJ, Patel S, Porter L, Quessy S,

49.

50.
51.
52.
53.
54.
55.

56.

57.

58.
59.

60.

Rappaport BA, Rauschkolb C, Revicki DA, Rothman M,

Schmader KE, Stacey BR, Stauffer JW, von Stein T, White RE,
Witter J, Zavisic S. Interpreting the clinical importance of
treatment outcomes in chronic pain clinical trials: IMMPACT
recommendations. J Pain 2008; 9: 10521.
Xie H, Wang X, Liu G, Wang G. Analgesic effects and
pharmacokinetics of a low dose of ketamine preoperatively
administered epidurally or intravenously. Clin J Pain 2003;
19: 31722.
Cousins M, Smith G, Power I. 1996 Labat lecture: pain a
persistent problem. Reg Anesth Pain Med 2000; 25: 621.
de Lima J, Beggs S, Howard R. Neural toxicity of ketamine
and other NMDA antagonists. Pain 2000; 88: 3112.
Malinovsky JM, Lepage JY, Crozian A, Mussini JM, Pinaudt
M, Souron R. Is ketamine or its preservative responsible for
neurotoxicity in the rabbit? Anesthesiology 1993; 78: 10915.
Elia N, Tramer MR. Ketamine and postoperative pain a
quantitative systematic review of randomised trials. Pain
2005; 113: 6170.
Laskowski K, Stirling A, McKay WP, Lim HJ. A systematic
review of intravenous ketamine for postoperative analgesia.
Can J Anaesth 2011; 58: 91123.
Mathews TJ, Churchhouse AM, Housden T, Dunning J.
Does adding ketamine to morphine patient-controlled analgesia safely improve post-thoracotomy pain? Interact
Cardiovasc Thorac Surg 2012; 142: 1949.
McCartney CJ, Sinha A, Katz J, McCartney CJL, Sinha A,
Katz J. A qualitative systematic review of the role of
N-methyl-D-aspartate receptor antagonists in preventive
analgesia. Anesth Analg 2004; 98: 1385400.
Chaparro LE, Smith SA, Moore RA, Wiffen PJ, Gilron I.
Pharmacotherapy for the prevention of chronic pain after
surgery in adults. Cochrane Database Syst Rev 2013; (7):
CD008307, doi: 10.1002/14651858.CD008307.pub2.
Kehlet H, Jensen TS, Woolf CJ. Persistent postsurgical pain:
risk factors and prevention. Lancet 2006; 367: 161825.
Kaunisto MA, Jokela R, Tallgren M, Kambur O, Tikkanen E,
Tasmuth T, Sipila R, Palotie A, Estlander A-M, Leidenius M,
Ripatti S, Kalso EA. Pain in 1000 women treated for breast
cancer: a prospective study of pain sensitivity and postoperative pain. Anesthesiology 2013; 119: 141021.
Deumens R, Steyaert A, Forget P, Schubert M,
Lavandhomme P, Hermans E, De Kock M. Prevention of
chronic postoperative pain: cellular, molecular, and clinical
insights for mechanism-based treatment approaches. Prog
Neurobiol 2013; 104: 137.

Address:
Ewan McNicol
Departments of Anesthesiology and Pharmacy
Tufts Medical Center
800 Washington Street, Box 420
Boston, MA 02111
USA
e-mail: emcnicol@tuftsmedicalcenter.org

1213

Copyright of Acta Anaesthesiologica Scandinavica is the property of Wiley-Blackwell and its

content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.