An Analysis of Postoperative Epidural Analgesia Failure

by Computed Tomography Epidurography

Downloaded from http://journals.lww.com/anesthesia-analgesia by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsI

Cyrus Motamed, MD* In this prospective study involving 125 patients, we analyzed epidural analgesia
Ho4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 07/23/2023

Fayezi Farhat, MD*
Francis Rémérand, MD*
Jean Stéphanazzi, MD*
Agnès Laplanche, MD†
Christian Jayr, MD*
failure after major abdominal surgery using computed tomography (CT) epidu-
rographies to compare the incidence of dislodgement of epidural catheters and
leakage of solution from the epidural space between two groups of patients:
patients with successful or failed epidural analgesia. Our hypothesis was that the
incidence of dislodgement and leakage should be low when epidural analgesia is
successful. A thoracic epidural catheter was inserted before general anesthesia and
secured by subcutaneous tunneling. Bupivacaine (0.25%) was administered during
surgery followed by continuous epidural analgesia with 0.125% bupivacaine (10
mL/h) and morphine (0.25 mg/h) for 48 h. Failure was defined as a visual analog
scale pain score at rest more than 30 mm and/or interruption of epidural analgesia
before 48 h for any reason. When failure was not due to unintentionally withdrawn,
kinked catheters or adverse events (n ⫽ 11), a CT scan with contrast injection was
performed. Control CT scans were also performed in patients with adequate
analgesia (i.e., the success group). The incidence of failure was 24.8% (n ⫽ 31). CT
scans in the failure group (n ⫽ 20) showed seven patients with catheters outside the
epidural space, nine with normal distribution, one with unilateral spread, and three
with leakage of solution outside the epidural space. In the success group, CT scans
(n ⫽ 19) showed 11 patients with normal distribution, five with unilateral spread,
and three with leakage. We conclude that the major cause of epidural analgesia
failure was dislodgment of the catheter. CT scans were mostly useful for detecting
leakage of injectate, which may be the early phase of dislodgment.
(Anesth Analg 2006;103:1026 –32)

E pidural analgesia provides superior postoperative

analgesia compared with parenteral opioids (1– 4).
Unfortunately, epidural analgesia is often associated
with failures that may be difficult to resolve when the
patient is in a surgical ward or during the night (5–9).
Epidural analgesia failures may result from technical
difficulties (10,11), insufficiencies or overdosing of
local anesthetics (12,13), epidural septum or midline
adhesions (14), placement of the epidural catheter
through an intervertebral foramen or into the anterior
epidural space (15) and problems related to the cath-
eter itself (13,16 –18). Mechanical problems with the
catheter include the formation of kinks and knots,
breakage, dislodgement, leakage, and disconnections
(19 –22). Dislodgement and leakage are common but
From the *Department of Anesthesia; and †Department of
Biostatistics; Institut Gustave Roussy, 39 rue Camille Desmoulins,
94805 Villejuif, Cedex, France.
Accepted for publication June 23, 2006.
Address correspondence and reprint requests to Cyrus Mota-
med, Service d’Anesthésie, Institut Gustave Roussy, 39 rue Camille
Desmoulins, 94805 Villejuif, Cedex, France. Address e-mail to
motamed@igr.fr.
Copyright © 2006 International Anesthesia Research Society
DOI: 10.1213/01.ane.0000237291.30499.32
1026
are probably under-estimated, especially at the begin-
ning of catheter displacement outside the epidural
space (13,23).
We, therefore, conducted a prospective study to
analyze epidural analgesia failure by CT epidurogra-
phy after major abdominal surgery for cancer. Our
hypothesis was that the incidence of dislodgement
and leakage should be low when epidural analgesia is
successful. Therefore, the main objective of our study
was to compare the incidence of dislodgement of
epidural catheters and leakage of liquid infusion out
of the epidural space between two groups of patients:
patients with successful or failed epidural analgesia.
METHODS
This prospective study was performed after approval
of the IRB. Written informed consent was obtained from
each patient. Inclusion criteria were patients undergoing
major elective abdominal surgery for cancer via a mid-
line, subcostal, or bisubcostal incision, 18 –75 yr of age,
ASA I–III and weight 50 –100 kg. Exclusion criteria were
patient refusal, contraindications to epidural analgesia
(e.g., preoperative coagulopathy and localized or uncon-
trolled systemic infection) and technically impossible
epidural catheter placement. Patients were premedi-
cated with 0.5 mg alprazolam 1 h before surgery.
Vol. 103, No. 4, October 2006
 Epidural Catheter Placement and Management
of Anesthesia
Before general anesthesia, an epidural catheter was
placed by an attending anesthesiologist in a patient who
was seated upright. After local anesthesia of the skin and
the intervertebral tissue, the epidural space (T9 –11) was
identified using an 18-gauge Tuohy needle via a midline
approach, and the loss-of-resistance technique with sa-
line solution. An open-end, single-hole catheter (Portex
Downloaded from http://journals.lww.com/anesthesia-analgesia by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsI
Ho4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 07/23/2023
Ltd威) was then advanced 4 cm into the epidural space
and 5-cm subcutaneous tunneling was performed. The
catheter was sutured and dressed with Tegaderm威. A
5-mL test dose of 2% lidocaine with epinephrine (5
␮g/mL) was injected, and 0.25% bupivacaine (8 –10 mL)
with epinephrine (5 ␮g/mL) and fentanyl (0.1 mg) was
injected. Bupivacaine was reinjected until the sensory
block attained the T4 level. The extent of analgesia was
determined by bilateral loss of sensation to ice. When the
T4 level was obtained, general anesthesia was induced
with etomidate, fentanyl, and neuromuscular blocking
drugs, and was maintained with isoflurane and nitrous
oxide in oxygen. Intraoperative analgesia was obtained
by intermittent epidural injections of 0.25% bupivacaine
approximately every 45 min. If hypotension occurred, a
bolus of ephedrine (3– 6 mg) was injected. Adequate
intraoperative analgesia was evaluated according to
hemodynamic stability (heart rate and arterial blood
pressure within 25% of baseline) and fentanyl was
administered IV when bupivacaine failed to provide
adequate intraoperative analgesia. After completion of
surgery, tracheal extubation was performed after rever-
sal of the neuromuscular block.
Postoperative Pain Management
In the recovery room, a mixture of 0.125% bupiva-
caine (10 mL/h) and morphine (0.25 mg/h) was
infused continuously into the epidural space. The
infusion rate was reduced if motor block was above 0
according to a modified Bromage scale (0 ⫽ no motor
block; 1 ⫽ inability to raise extended legs; 2 ⫽ inability
to flex knees; 3 ⫽ inability to flex ankle joints) (2– 4).
Paracetamol (1 g/4 h) was injected IV on request when
pain relief exceeded 30 mm at rest or when the patient
complained of pain other than from abdominal origin
(e.g., sore throat).
Abdominal pain was assessed at rest and while
coughing using a 100-mm visual analog scale (VAS) (0
mm ⫽ no pain; 10 mm ⫽ worst pain imaginable) every
15 min in the recovery room and every 4 h thereafter.
The spread of sensory block was determined by loss
and return of cold sensation to ice. Patients were
discharged from the recovery room when the pain
score at rest was ⬍30 mm.
On the surgical ward, the following variables were
recorded every 4 h: level of sedation, VAS at rest and
while coughing, extent of sensory blockade, presence
of motor block as well as vital signs.
Vol. 103, No. 4, October 2006
The following criteria signaled failure of analgesia:
• Interruption of epidural analgesia before 48 h for
any reason.
• A VAS score that exceeded 30 mm at rest and
persisted for 45 min after a rescue 5 mL epidural
0.125% bupivacaine injection and 1 g paracetamol
IV were administered.
If the patient did not meet the failure criteria after
the rescue dose, the infusion rate was then increased.
The absence of a sensory level with a VAS not
exceeding 30 mm was considered successful analgesia.
If an epidural analgesia failure occurred, a vertebral
axial CT scan was acquired after injection of contrast
medium via the catheter into the epidural space (5 mL
iopamidol, 400 mg/mL). Image slices (2.5 and 5 mm
thick) were acquired; two vertebrae above and below the
site where the epidural catheter had been inserted. A
tiny lead ball taped onto the skin marked the vertebral
level where the catheter had been inserted for CT scan
detection.
On the third postoperative day, before the removal
of the epidural catheter, a CT scan was also performed
in 19 patients in the success group (control group).
These patients were selected as follows: after CT-
epidurography in a patient in the failure group, CT-
epidurography was performed in the next patient
with successful epidural analgesia.
Both a radiologist and anesthesiologist who were
blinded to the efficacy of epidural analgesia interpreted
the CT scans. Three categories of images were defined: if
contrast medium was present in the epidural space on at
least one slice, it was “IN,” (Figs. 1a and 2), if contrast
medium was not found in the epidural space, it was
“OUT” (Fig. 1b). When the contrast medium was “IN”
and “OUT” on the same image, it was considered to be
“LEAKING” out of the epidural space (Fig. 2).
Figure 1. (a) Left: Lower thoracic axial computed tomogra-
phy scan after contrast injection through a catheter in a
patient with a normally functioning catheter. (b) Right:
Computed tomography scan after contrast injection through
a catheter in a patient in whom epidural analgesia failed.
The contrast medium accumulated in the subcutaneous
tissue outside the epidural space.
© 2006 International Anesthesia Research Society 1027

Downloaded from http://journals.lww.com/anesthesia-analgesia by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsI
Ho4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 07/23/2023
Figure 2. Lower thoracic axial computed tomography scan
depicting a posterior leak after contrast injection through a
catheter in a patient in whom postoperative epidural analgesia
failed.
The presence or absence of symmetrically sur-
rounding contrast medium was recorded as well as its
unilaterality or bilaterality.
In the failure group, all catheters that were displaced
from the epidural space were leaking, exhibited total
unilateral distribution, or were in the foramen were
removed, and epidural analgesia was switched to par-
enteral morphine, but those with normal distribution
inside the epidural space were maintained and the
infusion rate was readjusted so as to provide efficient
analgesia.
Statistical Analysis
The main objective of our study was to compare the
incidence of epidural catheter dislodgement and leak-
age of liquid infusion out of the epidural space be-
tween two groups of patients: patients with successful
versus failed epidural analgesia. As dislodgment and
leakage may be responsible for 66% of epidural anal-
gesia failure (13) and the incidence of dislodgement
and leakage should be low when epidural analgesia is
successful, we estimated that 40 patients would be
required to demonstrate a difference of 50% in dis-
lodgment and leakage on CT scan between groups
with a 5% type I error and a 10% type II error using a
two-tail test.
Preoperative characteristics in the two groups were
compared using ␹2 analysis for category variables and
the Student’s t-test for continuous variables. Analysis
was performed using Statistical Analysis System (SAS,
Cary, NC).
Univariate and multivariate analyses were per-
formed to search for factors associated with epidural
analgesia failure, such as demographic characteristics,
type of surgery and incision, the physician performing
the procedure, distance from the skin to the epidural
space, extent of sensory block, presence of motor
blockade, low arterial blood pressure, and asymmetri-
cal sensory blockade.
Postoperative category variables were compared by
␹2 analysis or Fisher’s exact test when necessary.
1028 CT Scan in Epidural Failure
Data were expressed as mean ⫾ sd, or percent-
ages. A P-value of ⬍0.05 was considered statistically
significant.
RESULTS
One-hundred-twenty-five patients were included
in this study. No patients were excluded because of
technical difficulties during catheter insertion.
Thirty-one patients (24.8%) were classified as fail-
ures while the others were classified as successes.
Twenty CT scans were performed in the failure group.
For the other 11 patients, CT scanning was not re-
quested because the cause of analgesia failure was
obvious: adverse events, unintentional catheter with-
drawal, and difficult venous access in a patient who
was able to ingest fluids. Nineteen CT scans were
performed in the success group. Different CT images
are shown in Figures 1 and 2.
Demographic characteristics and sensory block in-
formation are detailed in Table 1 and Figure 3. There
were no differences in the body mass index, the
duration and type of surgery, or in the amount of
intraoperative fentanyl administered between groups.
Intraoperative ephedrine was administered more of-
ten in the success group (Table 1, P ⬍ 0.05).
After multivariate analysis, only intraoperative
ephedrine requirements were associated with success-
ful epidural analgesia.
The Failure Group: (n ⴝ 31)
Interruption of Epidural Analgesia Despite Good
Pain Relief (n ⴝ 2)
In one patient, morphine was removed from the
infusion because of intense pruritus. Despite the pres-
ence of an adequate symmetrical sensory block, pain
scores increased to 45 mm at 48 h. Epidural analgesia
was therefore interrupted and subsequently replaced
by parenteral morphine.
In another patient, epidural analgesia was stopped
after 36 h because of a total, unilateral (right) block
and concomitant prolonged hypotension, which was
treated with ephedrine and intravascular volume re-
placement. Pain scores remained less than 10 mm on
the VAS. However, a CT scan was performed before
removal of the catheter. The epidural space was
bilaterally opaque without any leakage, but right
asymmetry (more opacity on the right side) was
noted. No contrast medium was detected in the sub-
dural or intradural spaces.
Interruption Unrelated to Epidural Analgesia (n ⴝ 2)
Two catheters were removed at 24 and 36 h. It was
not possible to achieve peripheral venous access in
one patient. However, oral fluid and per os medication
were allowed because he had undergone a nephrec-
tomy. The second patient expressed major anxiety
concerning the epidural analgesia. Pain scores, how-
ever, were ⬍30 mm in both cases.
ANESTHESIA & ANALGESIA
 Table 1. Demographic and Operative Characteristics
Success Failure
(n ⫽ 94) (n ⫽ 31)
Age (years) 52 ⫾ 14 50 ⫾ 12
Sex (F/M) 29/62 11/23
Weight (kg) 62 ⫾ 13 61 ⫾ 12
Height (cm) 164 ⫾ 9 164 ⫾ 8
Duration of surgery (min) 168 ⫾ 64 189 ⫾ 70
Intraoperative bupivacaine 0.25% (mL) 18 ⫾ 4 19 ⫾ 4
Fentanyl (␮g) 420 ⫾ 70 450 ⫾ 120
Downloaded from http://journals.lww.com/anesthesia-analgesia by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsI
Ho4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 07/23/2023

% Patients receiving ephedrine 74 55*

Ephedrine received (mg) 19 ⫾ 13 19 ⫾ 12
Incision type: median (mid-abdominal)/subcostal/bi-subcostal, n (%) 83/9/2 (88/10/2) 26/4/1 (84/13/3)
Type of surgery: n (%)
Colorectal 33 (35) 12 (39)
Gastrectomy 7 (7) 2 (6)
Hepatectomy 2 (2) 1 (3)
Hysterectomy with pelvic and lumbo-aortic lymphadenectomy 25 (27) 7 (23)
Cystectomy with enteroplasty 6 (7) 1 (3)
Nephrectomy 9 (10) 4 (13)
Lumbo-aortic lymphadenectomy 7 (7) 3 (10)
Laparotomy 5 (5) 1 (3)
Values are mean ⫾ SD.
Only the percentage of patients receiving ephedrine was statistically different between the two groups.
*P ⬍ 0.05.

Obvious Failure of Epidural Analgesia (n ⫽ 8) postoperative hours). In one case, catheter kinking
Seven catheters were unintentionally completely blocked the infusion on the first postoperative day and
withdrawn (1 in. the postanesthesia care unit (PACU), subsequent reinfusion was not possible. CT scans
4 during the first 24 h, and 2 between 24 and 48 were not performed.

Figure 3. Upper and lower left and right sensory blocks observed on both sides of patients assessed after insertion (test dose),
in the postanesthesia care unit, at days 1 and 2. When sensory blockade was present, no differences were noted between
groups. Patients with no sensory block were not included in the figure.

Vol. 103, No. 4, October 2006 © 2006 International Anesthesia Research Society 1029
 Table 2. Results of Epidurography by Computed Tomography (CT Scan) in the Failure and Success Groups
Success Failure
CT scan performed (n ⫽ 19) (n ⫽ 20) P value
CT scan images
Contrast medium “OUT” of the epidural space 0 7 ⬍0.01
Contrast medium “IN” the epidural space: 16 10 0.05
Normal (symmetrical) spread of contrast medium in 11 9 0.07
the epidural space
Asymmetrical spread of contrast medium 5 1 0.09
Downloaded from http://journals.lww.com/anesthesia-analgesia by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsI

Contrast medium “leaking” out of the epidural space 3 3 0.9

Ho4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 07/23/2023

Sensory spread/CT scan injectate spread

Clinical asymmetrical sensory blocks confirmed by CT 4/5 1/1 NA
scan
Clinical symmetrical sensory blocks confirmed by 10/11 8/9 NA
normal spread on CT scan
No. of clinical asymmetrical sensory blocks out of all 14/94 (15%) 5/31 (16%) 0.9
the patients in each group
Day when epidural failure occurred Day 0 (postanesthesia
(n ⫽ number of patients) care unit): 7
Day 1: 16
Day 2: 8
Epidural catheters outside epidural space according to Day 0: (5/7) 71% NS
the day when failure occurred: 14 catheters were Day 1: (6/16) 37% Day 0
considered outside the epidural space (with seven Day 2: (3/8) 37% Versus Day 1 and 2
CT scans performed) among 31 epidural failures
(n) and %
Five milliliters of contrast medium was injected into the catheter. In the success group, 19 CT scans were performed for 94 patients; in the failure group, 20 CT scans were performed for 31
patients. Therefore, 11 patients in the failure group did not have a CT scan for the following reasons: (a) epidural analgesia was interrupted despite efficient analgesia due to intense pruritus,
difficult venous access, and major anxiety (n ⫽ 3); catheters were unintentionally withdrawn (n ⫽ 7); kinked catheter and injection was not possible (n ⫽ 1). NA, not applicable.

Analgesia Failure Without Any Obvious Cause epidural space between patients with failed epidural
(n ⴝ 19) analgesia and those with efficient epidural analgesia
In these patients, the VAS score at rest exceeded 30 in a large homogenous population of patients after
mm (see Methods section). The results of clinical major abdominal surgery for cancer.
assessments and CT scan findings are detailed in The results of this observational study revealed a
Table 2. relatively frequent incidence (24.8%) of suboptimal
epidural analgesia using bupivacaine and morphine.
CT Scan Results
The major cause of failure of epidural analgesia was
The Failure Group (n ⴝ 20 of 31) dislodgment of the epidural catheter (45%, P ⬍ 0.01).
In seven patients, no contrast medium was seen in No other differences between the two groups were
the epidural space (OUT) (Fig. 1b). This difference seen on CT scan. Eight percent (n ⫽ 10) of the total
with the control group was statistically significant
patient population had inadequate pain relief despite
(P ⬍ 0.01).
catheter placement in the epidural space. The inci-
Contrast medium was strictly seen in the epidural
dence of perioperative hypotensive episodes was
space in 10 patients (IN) (Fig. 1a and Table 2) but there
more frequent in the success group; however, it is
was no statistical difference between groups.
difficult to compare the two groups as they were
Contrast medium leakage was observed in three
unequal in size (Table 1). Interestingly, 5/6 (83%)
patients; in one case from the intervertebral foramen
and in the other two from the posterior wall (Fig. 2 dislodgment events in the PACU did not result in
and Table 2). perioperative hypotension, suggesting that the cath-
eter had not functioned properly during surgery.
The Success Group (n ⴝ 19 of 94) Intraoperative hypotension may, therefore, be predic-
In none of the success group patients was the tive of epidural analgesia efficacy.
catheter outside the epidural space (Table 2, P ⬍ 0.05). We found that the spread of sensory block was often
Three patients had leakage of contrast medium (in one confirmed by the spread of contrast medium in the
case from the intervertebral foramen, in two cases epidural space (Table 2), which is consistent with what
from the posterior wall). has been reported (24). CT scans were therefore mostly
useful for detecting leakage of contrast medium.
DISCUSSION Other published reports cite dislodgment as a com-
The main objective of this prospective study was to mon cause of technical failure, with incidence rates
compare the incidence of dislodgement of epidural comparable to those in our study. Studies on large
catheters and leakage of liquid infusion out of the patient populations (i.e., 1,014 –5,628 patients) reported
1030 CT Scan in Epidural Failure ANESTHESIA & ANALGESIA
 dislodgment rates of between 10% and 13% of all popu-
lation (5,13,23). However, the proportion of dislodge-
ment and leaking responsible for inadequate epidural
analgesia may be as high as 66% (13) and 45% in our
study. By contrast, less dislodgment (2%–3%) was re-
ported in other studies (7,25,26), but the reason for these
discrepancies among studies is unclear.
In our experience, dislodgment occurred early in
the PACU or late in the surgical ward. Nonetheless,
Downloaded from http://journals.lww.com/anesthesia-analgesia by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsI
some studies have reported an increase in the failure rate
Ho4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 07/23/2023
according to the duration of epidural catheterization
(27,28).
Lumbar epidural catheter tip position and the dis-
tribution of injectate assessed by CT scanography have
been reported exclusively in patients receiving suc-
cessful epidural anesthesia (29). In that study, CT was
performed within 4 h of surgery and with an addi-
tional 10 mL injection of contrast medium. The author
found that nonuniform distribution of injectate was
common (at least 57% of cases) and compatible with
uniform anesthesia (29) which is similar to our results
since 42% of CT epidurography did not show uniform
spread of contrast medium in our success group. Only
5 mL of contrast medium was injected in our study,
which differs from that mentioned above (29). We
used a volume mostly associated with postoperative
analgesia conditions, larger volumes would simulate
anesthesia conditions.
Interestingly, failures in the PACU were mainly
due to dislodgment (Table 2, 6/7 failures, 86%). One
explanation could be that when the catheter is prop-
erly placed in the epidural space, failed analgesia does
not occur during the early postoperative phase be-
cause large amounts of local anesthetics are injected
during and at the end of surgery. The second reason
for analgesia failure was the dosing of the epidural
medication with unilateral analgesia that may have
been solved in some cases by increasing the infusion
rate if the catheter is correctly placed in the epidural
space (15,29). Epidurography has demonstrated that
adequate analgesia with bilateral sensory blockade
does not necessarily mean bilateral and symmetrical
diffusion of contrast medium into the epidural space.
Satisfactory analgesia can sometimes be maintained
despite leakage outside the epidural space (Table 2). In
the Hogan study (29), spread of injectate through the
intervertebral foramina was seen in all patients.
Scott et al. (23) reported that catheter migration
occurs when fluid accumulates under the clear adhe-
sive dressing. Their group was able to minimize such
migration and reduced dislodgment rates from 15% to
8% by placing a porous, nonclear dressing over the
clear dressing. In our study, leakage was demon-
strated in 15% of all analgesia failures on CT scans
(Table 2, 6/39). We speculate that leakage from the
posterior wall even though the catheter is lodged in
the epidural space may be the sign of the early phase
of complete dislodgment.
Vol. 103, No. 4, October 2006
Burstal et al. (30) tried a subcutaneous tunneling
technique that resulted in a significant reduction of
backward and frontward movement of the epidural
catheter, whereas Ballantyne et al. (13) tried various
means of securing catheters. None of these methods
improved the standard fixation method that involves
securing the catheter with a clear, adhesive dressing,
surrounded by paper or plastic tape. In another study,
the Lockit威 device was compared to the standard
dressing with a clear adhesive and was found to be
superior (31). The standard tunneling technique and
the suture technique were compared in one study, but
no differences were observed for dislodgment (32).
Secondary dislodgement might not be due entirely to
the fixation device; other mechanisms such as skin
movement could also cause dislodgement.
In summary, the major cause of epidural analgesia
failure in our study was catheter displacement. The
second reason for analgesia failure was the dosing of
the epidural medication that may partly be solved by
a more adaptable infusion rate and/or a higher bolus
dose in a patient-controlled epidural analgesia device.
ACKNOWLEDGMENTS
The authors thank Lorna Saint Ange for her patience and
her editing.
REFERENCES
1. Kehlet H. Postoperative opioid sparing to hasten recovery: what
are the issues? Anesthesiology 2005;102:1083–5.
2. Jayr C, Beaussier M, Gustafsson U, et al. Continuous epidural
infusion of ropivacaine for postoperative analgesia after major
abdominal surgery: comparative study with i.v. PCA morphine.
Br J Anaesth 1998;81:887–92.
3. Jayr C, Thomas H, Rey A, et al. Postoperative pulmonary
complications. Epidural analgesia using bupivacaine and opi-
oids versus parenteral opioids. Anesthesiology 1993;78:666 –76.
4. Motamed C, Spencer A, Farhat F, et al. Postoperative hy-
poxaemia: continuous extradural infusion of bupivacaine and
morphine vs. patient-controlled analgesia with intravenous
morphine. Br J Anaesth 1998;80:742–7.
5. Burstal R, Wegener F, Hayes C, Lantry G. Epidural analgesia:
prospective audit of 1062 patients. Anaesth Intensive Care
1998;26:165–72.
6. Andersen G, Rasmussen H, Rosenstock C, et al. Postoperative
pain control by epidural analgesia after transabdominal sur-
gery. Efficacy and problems encountered in daily routine. Acta
Anaesthesiol Scand 2000;44:296 –301.
7. McLeod G, Davies H, Munnoch N, et al. Postoperative pain
relief using thoracic epidural analgesia: outstanding success and
disappointing failures. Anaesthesia 2001;56:75– 81.
8. Ready LB. Acute pain: lessons learned from 25,000 patients. Reg
Anesth Pain Med 1999;24:499 –505.
9. Dolin SJ, Cashman JN, Bland JM. Effectiveness of acute postop-
erative pain management I. Evidence from published data. Br J
Anaesth 2002;89:409 –23.
10. Lang SA, Korzeniewski P, Buie D, et al. Repeated failure of
epidural analgesia: an association with epidural fat? Reg Anesth
Pain Med 2002;27:494 –500.
11. Tolksdorf W, Thurigen W, Lutz H. [The influence of abnormal
vertebral column on failure in epidural anaesthesia]. Prakt
Anaesth 1977;12:413– 8.
12. Russell AW. Inadvertent epidural overdose. Anaesth Intensive
Care 1994;22:501–2.
13. Ballantyne JC, McKenna J, Ryder E. Epidural analgesia –
experience of 5628 patients in a large teaching hospital derived
through audit. Acute Pain 2003;4:89 –97.
14. Gallart L, Blanco D, Samso E, Vidal F. Clinical and radiologic
evidence of the epidural plica mediana dorsalis. Anesth Analg
1990;71:698 –701.
© 2006 International Anesthesia Research Society 1031
 15. Beilin Y, Zahn J, Bernstein HH, et al. Treatment of incomplete
analgesia after placement of an epidural catheter and adminis-
tration of local anesthetic for women in labor. Anesthesiology
1998;88:1502– 6.
16. Valentine SJ. Defects in epidural catheters. Anaesthesia
1988;43:1062 (letter).
17. Smith S. Faulty markings on an epidural catheter. Anaesth
Intensive Care 1998;26:594 –5.
18. Yau G, Ewart MC, Oh TE. Duplicate markings on an epidural
catheter. Anaesthesia 1991;46:795– 6.
19. Stuart AL, McDavid AJ. Knotted epidural catheters. Reg Anesth
1996;21:606 (letter).
Downloaded from http://journals.lww.com/anesthesia-analgesia by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsI
20. Willens JS. Disconnected epidural catheter. Nursing 1991;21:33
Ho4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 07/23/2023
(letter).
21. Asai T, Yamamoto K, Hirose T, et al. Breakage of epidural
catheters: a comparison of an arrow reinforced catheter and
other nonreinforced catheters. Anesth Analg 2001;92:246 – 8.
22. Macfarlane J, Paech MJ. Another knotted epidural catheter.
Anaesth Intensive Care 2002;30:240 –3.
23. Scott DA, Beilby DS, McClymont C. Postoperative analgesia using
epidural infusions of fentanyl with bupivacaine. A prospective
analysis of 1,014 patients. Anesthesiology 1995;83:727–37.
24. Yokoyama M, Hanazaki M, Fujii H, et al. Correlation between
the distribution of contrast medium and the extent of blockade
during epidural anesthesia. Anesthesiology 2004;100:1504 –10.
1032 CT Scan in Epidural Failure
25. Ready LB, Loper KA, Nessly M, Wild L. Postoperative epidural
morphine is safe on surgical wards. Anesthesiology 1991;75:452– 6.
26. de Leon-Casasola OA, Parker B, et al. Postoperative epidural
bupivacaine-morphine therapy. Experience with 4,227 surgical
cancer patients. Anesthesiology 1994;81:368 –75.
27. Scott AM, Starling JR, Ruscher AE, et al. Thoracic versus lumbar
epidural anesthesia’s effect on pain control and ileus resolution
after restorative proctocolectomy. Surgery 1996;120:688 –95.
28. Tsui BC, Gupta S, Finucane B. Confirmation of epidural catheter
placement using nerve stimulation. Can J Anaesth 1998;45:640 – 4.
29. Hogan Q. Epidural catheter tip position and distribution of
injectate evaluated by computed tomography. Anesthesiology
1999;90:964 –70.
30. Burstal R, Wegener F, Hayes C, Lantry G. Subcutaneous tun-
nelling of epidural catheters for postoperative analgesia to
prevent accidental dislodgement: a randomized controlled trial.
Anaesth Intensive Care 1998;26:147–51.
31. Clark MX, O’Hare K, Gorringe J, Oh T. The effect of the Lockit
epidural catheter clamp on epidural migration: a controlled
trial. Anaesthesia 2001;56:865–70.
32. Chadwick VL, Jones M, Poulton B, Fleming BG. Epidural catheter
migration: a comparison of tunnelling against a new technique of
catheter fixation. Anaesth Intensive Care 2003;31:518 –22.
ANESTHESIA & ANALGESIA