Anaesthesia, 2011, 66, pages 913–918 doi:10.1111/j.1365-2044.2011.06820.x
.....................................................................................................................................................................................................................
ORIGINAL ARTICLE
Appropriate length of epidural catheter in the epidural
space for postoperative analgesia: evaluation by
epidurography
G. Afshan,1 U. Chohan,2 F. A. Khan,2 N. Chaudhry,3 Z. E. Khan4 and A. A. Khan5
1 Associate Professor, 2 Professor, 5 Instructor, Department of Anaesthesia, 3 Associate Professor, 4 Assistant Professor,
Department of Obstetrics and Gynaecology, Aga Khan University, Karachi, Pakistan
Summary
In current practice, the length of epidural catheter that should be left in the epidural space is not
standardised for effective postoperative analgesia. This prospective, randomised, double-blinded
study aimed to determine the most appropriate length of epidural catheter that should be inserted
into the epidural space for postoperative analgesia. We recruited 102 women and assigned them
into three study groups (3, 5 and 7 cm insertion). An epidural catheter was inserted and epidur-
ography was performed. Postoperatively, mean pain scores, motor and sensory levels, and any
complications associated with the epidural catheter were recorded. No statistically significant di-
fference for mean postoperative pain score was found at all study timings. Motor and sensory
blockade was also statistically insignificant. Unilateral sensory analgesia developed in one patient in
the 7 cm group and epidural catheter dislodgement was observed in four patients in the 3 cm
group. In order to minimise catheter-related complications for postoperative analgesia, the most
appropriate length that an epidural catheter should be left in the epidural space is 5 cm.
. ......................................................................................................
Correspondence to: Dr G. Afshan
Email: gauhar.afshan@aku.edu
Accepted: 15 May 2011
You can respond to this article at http://www.anaesthesiacorrespondence.com
Continuous epidural analgesia is considered one of migration of the epidural catheter out of the epidural
the most effective techniques for postoperative anal- space. However, if a longer length of epidural catheter
gesia [1]. The position of an epidural catheter and the is left in the epidural space, this may increase the
distribution of local anaesthetic drugs in the epidural likelihood of a unilateral block or intravenous cann-
space are two of the most important determining ulation [3]. Clinical trials have shown that even if the
factors for successful epidural analgesia. The incidence epidural catheter was left 2–5 cm in the epidural space,
of failed or inadequate epidural analgesia has been this was found to provide satisfactory analgesia.
reported as 4.2–6.3% in one study, because of However, one trial has shown that if more than 3 cm
suboptimal placement of the epidural catheter within of the epidural catheter is left in the epidural space, this
the epidural space [2]. In our institution, regular audit increases the risk of transforaminal escape [4].
of the acute pain service also showed that catheter- The length of the epidural catheter that should be
related problems are one of the main reasons for failed left in the epidural space has not been standardised.
or incomplete epidural analgesia. The main aim of this study was to determine the most
In current practice, controversy exists regarding the appropriate length of epidural catheter that should be
length of epidural catheter that should be left within inserted into the epidural space for postoperative
the epidural space for successful pain management. analgesia. The secondary aim was to determine the
Inserting a limited length of the epidural catheter, i.e. complications associated with different lengths of
3–4 cm, may result in an increased incidence of epidural catheter.
Ó 2011 The Authors
Anaesthesia Ó 2011 The Association of Anaesthetists of Great Britain and Ireland 913
G. Afshan et al. Æ Length of epidural catheter in the epidural space Anaesthesia, 2011, 66, pages 913–918
. ....................................................................................................................................................................................................................
16-G Tuohy needle (Portex; Smith Medical ASD,
Methods
Weston, MD, USA), with the bevel of needle facing
After ethical approval from the ethics committee of cephalad, and using loss of resistance to air in the lateral
Aga Khan University hospital, we conducted this decubitus position. A median approach was used in all
prospective, randomised, double-blinded trial. In- patients. A multi-orifice epidural catheter was inserted
formed written consent was obtained from all patients. into the space according to the group allocation.
We recruited female patients between 40 and 60 years Patients were randomly assigned to have the epidural
of age, with ASA physical status 1–3 scheduled for total catheter inserted either 3, 5, or 7 cm in the epidural
abdominal hysterectomy. Exclusion criteria included space. Patients were placed in the supine position after
patients with any contraindication to regional withdrawing the Tuohy needle and securing the
anaesthesia and with a known history of allergy to catheter.
radio-opaque dye. All patients had an epidural catheter Figure 1 gives an overview of the study design. Any
inserted and were randomly assigned into three study epidural catheter that was found on test dose to be
groups: 3, 5 and 7 cm inside the epidural space. either intrathecal or intravascular was removed before
Randomisation was performed using a computerised the induction of general anaesthesia. For epidurogra-
table of random numbers. phy, 2–3 ml non-ionic, iso-osmolar contrast medium
All patients were premedicated with 7.5 mg oral Omnipaque (GE-Health, Ireland) was injected via the
midazolm, 1 h before surgery. Patients were preloaded epidural catheter, to ascertain the level of entry into the
with 7–8 ml.kg)1 crystalloid solution. Baseline heart epidural space and to ascertain the location of the
rate, blood pressure and oxygen saturation were epidural catheter. This was followed by injection of
recorded. Using a full aseptic technique, the epidural 12–15 ml contrast medium to demonstrate the spread
space was localised at the L2-3 or L3-4 level with a and number of segments covered above or below the
Insertion of 2–3 ml non-ionic
Epidural test
epidural catheter iso-osmolar contrast
dose of 3 ml
lidocaine 2% medium epidurally
Fluoroscopy
AP/lateral view for 8–10 ml iso- Epidurography
level of entry and osmolar contrast AP/lateral view of
medium spread of contrast
catheter location
medium
Epidural bolus of Desired level of
T8-T10 (2 Failed epidural if
5 ml bupivacaine
additional desired level was not
0.25% and 50 µg
boluses of 5 ml achieved in 30 min
fentanyl (2
bupivacaine
increments)
only)
Continuous End of surgery :
General induced epidural infusion transfer to recovery
for intra-operative room + pain score
pain relief monitoring
Transfer to ward and
continued monitoring
for pain score and any
complication at
4, 8, 12, 16, 20, 24, 48 h Figure 1 Overview of the study
design.
Ó 2011 The Authors
914 Anaesthesia Ó 2011 The Association of Anaesthetists of Great Britain and Ireland
Anaesthesia, 2011, 66, pages 913–918 G. Afshan et al. Æ Length of epidural catheter in the epidural space
. ....................................................................................................................................................................................................................
point of entry. Anterior and lateral images were taken for 48 h. A pain score of more than 3 was treated with
under fluoroscopy (Figs 2 and 3). All these findings of additional boluses of 5–7 ml bupivacaine 0.25%. If pain
epidurography were interpreted and recorded in the was not relieved, then intravenous rescue analgesia was
data collection form. given as per departmental guidelines. Complications
After epidurography, the patient was handed over to associated with the catheter including dislodgment,
the primary anaesthetist, who was blinded to the unilateral analgesia and nausea and vomiting were also
allocation of the study group. The epidural block was monitored.
then established with a total of 12–15 ml bupivacaine For the study, we defined an intravenous catheter as
0.25% with 50 lg fentanyl over a period of 10 min, to a catheter from which blood could be aspirated or that
achieve a block to T8–T10. Two additional 5-ml was associated with central neurological symptoms after
boluses of 0.25% bupivacaine were also given if the the administration of local anaesthetic. An intrathecal
block level was not achieved in 15–20 min. The catheter was defined as a catheter from which
epidural was considered failed if the sensory level was cerebrospinal fluid could be aspirated or that was
not achieved in 30 min, and the patient was then given associated with motor block after administration of a
rescue analgesia in the form of intravenous morphine test dose. Unilateral sensory analgesia was defined as
or pethidine. During establishment of the epidural any epidural catheter associated with more than 2
block, if systolic blood pressure dropped more than dermatomes, sensory disparities. Dislodgement of the
25% from the baseline or less than 90 mm Hg, 50- epidural catheter was defined as any catheter that
100-lg boluses of phenylephrine were given intrave- functioned well and subsequently ceased to function
nously. A heart rate of less than 50 beats.min)1 was despite additional boluses. A failed epidural in our
treated with 0.5–1 mg atropine. After establishment of study was one that failed to achieve the desired level of
epidural analgesia, anaesthesia was induced with prop- sensory block at the time of establishment of epidural
ofol 1.5–2 mg.kg)1 and fentanyl 2 lg.kg)1. Atracuri- analgesia.
um 0.5–0.6 mg.kg)1 was given to facilitate tracheal At the end of 48 h, all patients were asked to
intubation. Isoflurane was used for maintenace of rate the overall pain management as excellent ⁄ good ⁄
anaesthesia. Intra-operative analgesia was maintained satisfactory ⁄ unsatisfactory.
with an epidural infusion of 12–15 ml bupivacaine A total of 30 subjects per group achieved 80%
0.0625% with fentanyl 2 lg.ml)1. Morphine power, with a minimum 20% difference in average
0.1 mg.kg)1 or pethidine 1 mg.kg)1 was used as rescue pain score between any of the two groups, a maximum
analgesia and was administered at the discretion of of 30% variability in pain scores and 5% level of
the primary anaesthetist. Postoperatively, pain relief significance [3]. An interim analysis supervised by an
was maintained with an infusion of 12–15 independent statistician led to an adjustment of the
ml.h)1 bupivacaine 0.0625% with fentanyl 2 lg.ml)1 total number of patients to 102, to increase the power
for 48 h. Postoperative pain scores at rest (visual of the study. Data were entered twice in epidata
analogue scale 1–10), sensory level (pinprick method) software by two different data operators. Data entry
and motor power of lower limbs (0–3 on Bromage was verified by re-examining at 5% the data collection
scale) were assessed by the acute pain service, every 4 h form manually. For analysis purpose, data were
Figure 2 Lateral and anteroposterior
and lateral images after epidurography
in a patient who had an epidural
inserted for a total abdominal hyster-
ectomy (after 3 ml contrast medium).
The arrow indicates the catheter tip
in the epidural space.
Ó 2011 The Authors
Anaesthesia Ó 2011 The Association of Anaesthetists of Great Britain and Ireland 915
G. Afshan et al. Æ Length of epidural catheter in the epidural space Anaesthesia, 2011, 66, pages 913–918
. ....................................................................................................................................................................................................................
among the three groups for the total number of
segments covered above or below the level of catheter
insertion or to the right and left of the midline.
Sensory blockade at T10 was successfully achieved in
all groups before the induction of general anaesthesia
and the intra-operative course was unremarkable in all
cases. No statistically significant difference was found
for mean postoperative pain scores among the groups at
all study timings (Fig. 4). Motor and sensory blockade
was also statistically insignificant among all three
groups.
Unilateral sensory analgesia developed in one patient
in the 7 cm group in the first postoperative period.
This patient was successfully managed according to our
routine departmental guidelines, which are to with-
draw the catheter by 1–2 cm and re-administer the
local anaesthetic dose accordingly. Epidural catheter
dislodgement was observed in four patients in the 3 cm
Figure 3 Lateral and anteroposterior images after epiduro- group (p 2 cm groups and
Performing epidurography and its interpretation in they recommended inserting 6 cm, when prolonged
terms of catheter tip position and contrast medium labour was expected or caesarean section was likely
spread was found to be a reliable process, as two of the [12]. In our study, we chose a minimum length of
investigators in this study were familiar with these 3 cm considering previous literature supporting dis-
procedures, due to their routine interventional pain lodgment with 2 cm. Despite this, we found dislodg-
practice. In this study, the epidural catheter tip was ment with the 3 cm group resulting in an increased
found in the epidural space in all patients except one in demand for rescue analgesia. Thus, a 3-cm length of
the 3 cm group. Successful sensory level was also epidural catheter may be appropriate for short-term use
achieved in all confirmed catheter positions. This (e.g. rapidly progressing labour) but for a longer period
showed that fluoroscopy reliably indicates the catheter of time, e.g. postoperative analgesia, the length of the
position, inside the epidural space, resulting in epidural catheter should be more than 3 cm. This
successful analgesia. finding is consistent with the result of the study by
Ó 2011 The Authors
Anaesthesia Ó 2011 The Association of Anaesthetists of Great Britain and Ireland 917
G. Afshan et al. Æ Length of epidural catheter in the epidural space Anaesthesia, 2011, 66, pages 913–918
. ....................................................................................................................................................................................................................
Bhiston et al. who found that the epidural catheter has
References
a tendency to migrate both further into and out of the
epidural space after it has been secured. They found 1 Dolin SJ, Cashman JN, Bland JM. Effectiveness of acute
that 22% of the catheters migrated more than 1 cm out pain management: evidence from published data. British
of the epidural space [13]. Journal of Anaesthesia 2002; 89: 409–23.
Current literature suggests that threading too much 2 Tanaka K, Watanabe R, Harada T, Dan K. Extensive
application of epidural anesthesia and analgesia in a
of the epidural catheter into the epidural space may
university hospital: incidence of complications related to
direct the catheter right or left, rather than into the
technique. Regional Anesthesia 1993; 18: 34–8.
middle of the space. In this study, only one case of 3 Beilin Y, Bernstein H, Zucker-Pinchoff B. The optimal
unilateral analgesia after 24 h was reported in the 7 cm distance that a multiorifice epidural catheter should be
group. This may be attributed to catheter migration or threaded into the epidural space. Anesthesia and Analgesia
prolonged positioning of the patient on one side, 1995; 81: 301–4.
causing the drug spread to spread unilaterally. 4 Kumar CM, Dennison B, Lawler PG. Excessive dose
Interestingly, mean pain score trends were similar in requirements of local anesthesia for epidural analgesia.
all groups at all study timings despite reported catheter- How far an epidural catheter should be inserted. Anaes-
related complications in the 3 and 7 cm groups. This thesia 1985; 40: 1100–2.
may be explained by provision of intravascular rescue 5 Schier R, Guerra D, Aguilar J, et al. Epidural space
analgesia and appropriate action of the departmental identification: a meta-analysis of complications after air
versus liquid as the medium for loss of resistance.
acute pain management guidelines in the study proto-
Anesthesia and Analgesia 2009; 109: 2012–21.
col. As a result of ethical consideration, these were 6 Johannes G, Tomi T, Markku T, Saana I, Rosenberg
strictly followed to ensure that patients were pain-free. PH. An evaluation of the epidural catheter position by
In summary, if there are no anatomical abnormalities epidural nerve stimulation in conjunction with
within the epidural space, all lengths of epidural continuous epidural analgesia in adult surgical patients.
catheter i.e. 3–7 cm may cause similar spread of local Anesthesia and Analgesia 2009; 108: 351–8.
anaesthetic within the epidural space. However, if the 7 Hogan Q. Epidural catheter tip position and distribution
catheter is as short as 3 cm or as long as 7 cm, then this of injectate evaluated by CT. Anesthesiology 1999; 90:
may lead to performance problems of the epidural 964–70.
catheter. Although epidurography is not a routine 8 SDe Medicis E, Tetrault JP, Martin R, Robichaud R,
procedure for routine epidural catheter placement, it is Laroche L. A prospective comparative study of two
indirect methods for confirming the localization of an
a definitive tool to confirm the position of the epidural
epidural catheter for postoperative analgesia. Anesthesia
catheter and may be used in difficult cases. Considering
and Analgesia 2005; 101: 1830–3.
the overall findings of this study and the supporting 9 Nagaro T, Yorozuya T, Kamei M, Kii N, Arai T, Abe S.
literature available, 5 cm of epidural catheter is the Fluoroscopically guided epidural block in the thoracic
most appropriate length for postoperative analgesia, and lumbar regions. Regional Anesthesia and Pain Medicine
with minimum catheter-related complications such as 2006; 31: 409–16.
dislodgment and unilateral analgesia. 10 Nakatsuka H, Takahashi T, Matsumi M, et al. Can
epidurography help to predict the extent of epidural
blockade? Anesthesiology 2005; 102: 479.
Acknowledgements 11 Magides AD, Sprigg A, Richmond MN. Lumbar
We would like to extend our thanks to: Mr. Iqbal epidurography with multi-orifice and single orifice
Azam, Department of Community Health Sciences; epidural catheters. Anaesthesia 1996; 51: 757–63.
Dr Khalid Munir, Instructor, Department of Radiology; 12 D’Angelo R, Berkebile BL, Gerancher JC. Prospective
Examination of Epidural Catheter insertion. Anesthesiol-
Dr. Sunita Dodhani, Assistant Professor, Department of
ogy 1996; 84: 88–93.
Family Medicine, Aga Khan University Hospital; and 13 Bishton M, Martin P H, Vernon J M. Factors influencing
Mrs. Zohra I Khan for their support for the completion epidural catheter migration. Anaesthesia 1992; 47: 610–2.
of our work.
Competing interests
No external funding or competing interests declared.
Ó 2011 The Authors
918 Anaesthesia Ó 2011 The Association of Anaesthetists of Great Britain and Ireland