Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 
  • Users Online:236
  • Home
  • Print this page
  • Email this page

 Table of Contents  
Year : 2016  |  Volume : 30  |  Issue : 2  |  Page : 101-107

A 6-month audit of epidural analgesia in a teaching hospital

Department of Anaesthesia, Khoo Teck Puat Hospital, 90 Yishun Central, 768828, Singapore

Date of Web Publication18-Jul-2016

Correspondence Address:
Shahla Siddiqui
Department of Anaesthesia, Khoo Teck Puat Hospital, 90 Yishun Central, 768828
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0970-5333.186465

Rights and Permissions

Introduction: Major laparotomies generally herald high pain scores postoperatively and have high intraoperative hemodynamic shifts. Traditional analgesic regimens with intravenous opioids were supplanted with the superior epidural analgesia (EA) in the mid-1980s. This was based on the perception that EA provided highly effective postoperative analgesia for patients undergoing major abdominal surgery. However, recent literature points to a high failure rate. We aimed to retrospectively audit our EA performance in terms of success and complications and take an in-depth look at factors associated with failure. Methods: We retrospectively reviewed charts and our EA forms over 6 months. Results: Our results show a low rate of mortality and neurological morbidity. However we have a high rate of failure of 37% as judged by high pain scores, use of adjuvant analgesics and also an unacceptably high rate of hypotension from epidural infusions, requiring fluid boluses, inotropes and Intensive Care Unit admission. Registrar level and senior insertion did not show any difference in failure rate. Conclusion: Level of training per se does not equate to experiential skip and prior level of experience with administration of this type of anesthesia. We can conclude that EA in our setting is safe but not effective and requires further and frequent scrutiny in terms of procedures, technical skill, education and perhaps looking at its cost-effectiveness and need for standardization.

Keywords: Audit, complications, efficacy, epidural analgesia

How to cite this article:
Siddiqui S, Quek C, Prasad MM, Venkatesan K. A 6-month audit of epidural analgesia in a teaching hospital. Indian J Pain 2016;30:101-7

How to cite this URL:
Siddiqui S, Quek C, Prasad MM, Venkatesan K. A 6-month audit of epidural analgesia in a teaching hospital. Indian J Pain [serial online] 2016 [cited 2020 Dec 2];30:101-7. Available from: https://www.indianjpain.org/text.asp?2016/30/2/101/186465

  Introduction Top

Major laparotomies generally herald high pain scores postoperatively and have high intraoperative hemodynamic shifts. [1] Traditional analgesic regimens with intravenous (IV) opioids were supplanted with the superior epidural analgesia (EA) in the mid-1980s. [2] This was based on the perception that EA provided highly effective postoperative analgesia for patients undergoing major abdominal surgery. EA entails placing a catheter preoperatively at an appropriate level in the epidural space and infiltrating local anesthetic via a continuous infusion. [3] Side effects were perceived to be few and less serious, and the risk versus benefit ratio was thought to be in favor of the latter. However, there has been a spate of recent case reports of major complications occurring during EA to the extent that the practice is now at a crossroads and many clinicians are questioning whether more peripheral regional anesthetic techniques (such as transversus abdominis plane [TAP] blocks) are superior and whether the risks involved in placing and maintaining EA are worth embracing. [4] Although a successful epidural provides a patient with excellent pain scores, satisfaction, early mobilization and narcotic sparing effects, recently many societies such as the Association of Anaesthetists of Great Britain and Ireland have recommended frequent audits of EA practice to assess efficacy, complications and patient satisfaction. [5]

In a recent review, Hermanides et al. describe epidural failure more common than recognized in clinical practice. Technical factors, misplacement, choice of drug and dosage all seem to influence rates of failure. [6] One study conducted by Scherer et al. showed no permanent neurological sequelae in 4185 patients although unsuccessful catheter placement occurred in 1 in 93 patients and self-limiting peripheral nerve lesions in 1 in 174 patients. The incidence of dural tap was 1 in 140 patients but was less common at higher insertion sites. [7] In a similar audit to ours, Chilvers et al. audited 54 patients who received multimodal analgesia versus 59 previous patients who received EA. They found that satisfactory pain scores were obtained with both techniques, but that multimodal analgesia was associated with reduced anesthesia preparation time, reduced requirement for high dependency unit stay, reduced need for staff interventions, and reduced length of hospital stay. [8]

In our center, roughly 50 emergency or elective abdominal surgeries occur every month. In 6 months from July to December 2014, we had 53 epidurals placed for intra and postoperative analgesia. That accounts for 26% of all abdominal surgeries (200). We aimed to retrospectively audit our EA performance in terms of success and complications and take an in-depth look at factors associated with failure. We have trainees under supervision as well as consultant level anesthetists performing the epidurals.

  Methods Top


We are a 500 bedded regional tertiary care hospital serving adult patients. In our setting, we have in place an acute pain service with a protocolized EA management closely liaised with pharmacy and nursing. Epidural insertion level and drug selection are the choice of the primary anesthetist. We usually use anatomical landmark technique for level confirmation. Postoperatively patients can go to specialized general wards if no hemodynamic instability is present. A dedicated pain management nurse and team follow-up these patients with daily rounds and notes input from the primary teams. Ward nurses are also in-serviced in trouble shooting the infusion pumps and the required patient monitoring. In general, the decision to place an epidural is the primary anesthetist's in conjunction with the surgeon. Patient refusal is an absolute contraindication of course. Most open hepatobiliary surgeries and laparotomies are offered EA. These are placed preoperatively in the Operation theatre itself. We are a teaching hospital, and we have trainees of differing experience. Ropivacaine is the usual preference and is available in two concentrations, 0.1 and 0.2%, and the choice of concentration is at the discretion of the anesthesiologist. Usually, fentanyl is added as an additive at 1-2 μg/ml.

Ethics and inclusion criteria

After it had been determined by the Central Institutional Review Board that an ethics review was not required for this retrospective audit, we used our EA forms used to record all patients who receive EA and also looked through the patient records. Data over 6 months from July till December 2014 were included. All adult patients receiving epidurals for perioperative analgesia were included in the audit.

Aims and standards

Our aim was to assess efficacy and complications as well as outcome. Our standards were taken from literature which points to a pain score of >3 and use of adjunct analgesia as an unsuccessful EA, also any complication is counted as unacceptable. We counted hypotension with a mean arterial blood pressure <65 mmHg requiring treatment as one of the unacceptable side effects of an epidural. Literature reviews have reported 13-47% failure rate in heterogeneous surgical populations and 0-75% incidence of hypotension following an epidural. [9] In our center, it is usually up to the primary physician caring for the patient to treat hypotension as deemed fit. The initial treatment is usually fluid boluses as the situation allows as well as lowering or stopping the epidural infusion. There does not exist a protocol for such treatment.

Data collection

By means of a data collection form and an excel spreadsheet, we tabulated all 53 patients and their characteristics, including type of surgery, site of surgery, emergency or elective status, pain score >3, other adjuncts used, level of insertion, person inserting the epidural, postoperative Intensive Care Unit admission, length of stay, outcome, day of mobilization, day of catheter removal, patient satisfaction, and complications.


Apart from looking at success of EA, outcome and complications we specifically looked for hypotension and delay in mobilization as a potential drawback of EA after the initial postoperative period (more than 24 h later). We cross-tabulated some variables to study the effect of seniority of person inserting the epidural, the status of the surgery, and appropriateness of the level of insertion with success. A successful epidural was defined as one where the patient had "pain (visual analogue score or [VAS]) scores of <3, no adjuvant mode of analgesia and early mobilization with no motor block or prolonged sensory deficits." Early mobilization usually means within 24 h of the surgery. We do not have fast track surgery protocols in place for all laparotomies. However, elective colorectal surgery patients are often placed on the enhanced recovery after surgery (ERAS) (enhanced recovery) protocol. [10] VASs are recorded by the nurses in charge every shift (usually twice a day) as well as the pain team (once a day). Patient satisfaction scores are assessed by means of a simple yes or no answer at the end of the hospital stay.

Statistical analysis

As this was a retrospective audit we did not calculate sample size and hence cannot do prospective analysis. However, we did find frequencies and cross tabulations hinting at associations. All data were entered into Excel® and analyzed.

  Results Top

Fifty-three epidurals were placed in our center during the audit period for a total of 200 abdominal surgeries. As already mentioned, our rate of epidural placement for major abdominal surgeries is 26% over 6 months. The majority of these were laparotomies 15 (28%) followed by low anterior resections 10 (18%) and open sigmoid colectomies 9 (17%). Overall, 34 (65%) of the surgeries were lower abdominal incisions while 19 (35%) were higher abdominal incisions such as hepatectomies or Whipple's procedure. 42 (79%) of the cases were elective surgeries while the rest, 11 (21%) were emergencies. A VAS score of >3 was found in 36 (32%) of the patients who received an epidural. Adjuvant modes of analgesia were used in 10 (19%) of the patients. Of these, IV patient-controlled analgesia (IV PCA) was used in eight patients or 80% of the time, while others included IV multimodal analgesia in two patients or (20%), TAP block, one patient (10%) and repeating the epidural in one patient (10%).

Ropivacaine was used in all cases with fentanyl as an additive in 51 patients (96%) of the infusions. The concentration of ropivacaine was 0.2% in majority of the cases 35 (66%) while 18 (33%) had 0.1% ropivacaine used. Postoperatively, there was a sensory band seen on examination in only 31 (58%) of the patients postoperatively denoting a good analgesic effect, while two patients (3%) developed a motor block, of which one patient had bilateral motor weakness. This motor weakness in both patients lasted for 1 week each. In both cases, an magnetic resonance imaging scan ruled out an epidural hematoma. There were no residual weaknesses noted on follow-up. Another complication which we considered as undesirable was the incidence of hypotension with a mean arterial pressure of <65 mmHg requiring treatment. This was in the range of 22% or 12 patients. Of the treatments used 10 (83%) patients were given fluids while 3 (25%) received vasopressor support. One patient (8%) had their concentration changed from 0.2% to 0.1%.

[Table 1] [Table 2] [Table 3] describe the type of surgery, level of the block, and characteristics of the EA among our cohort. [Table 4] and [Table 5] show the outcomes and complications as well as patient satisfaction score. [Table 6] displays the cross tabulations looking for associations if any. Due to the variety of surgical techniques as well as myriad levels chosen, it was difficult to tabulate all outcomes in a single table. Also, our numbers being so small precluded statistical significance analysis.
Table 1: Types of surgery

Click here to view
Table 2: Level of blocks given

Click here to view
Table 3: Epidural Analgesia characteristics

Click here to view
Table 4: Epidural analgesia Outcomes

Click here to view
Table 5: Satisfaction and complications

Click here to view

Our level of placement of EA varied from T5-6 to L3-4, however, the most common level was T8-9 in 25 patients (47%) followed by T7-8 in 11 (20%) and T9-10 in seven patents (13%). Taking the recommended ideal level of T6-8 for high abdominal incisions and T8-12 for lower abdominal incisions we identified 43 or 81% of the epidurals to be placed appropriately for the level of incision. Appropriateness of level also corresponded well to success of the EA [Table 2]. Our patients went to the general ward 54% of the time (29 patients) while to the high dependency unit in 45% (24) instances. The epidurals were performed by "seniors" (consultants) 81% of the time (43 cases) whereas the rest were performed by trainees (10 cases or 19%). However, the failure rates were equal (30%) in both reflecting a lack of association between level of anesthetist and epidural success. Our average rate of infusion was 5 ml/h and did not seem to influence the success or failure of the epidural. There was no bowel or urinary incontinence noted.

Our eventual outcome showed that 2 (3.7%) of the patients expired, one patient (1%) was transferred to another facility and 50 (94%) patients were discharged home. Our patients with epidurals were mobilized most often on postoperative day 2 (15 patients or 28%) however, four patients (7%) were not mobilized till after day 5. This can be attributed to surgical causes, however, defeats the purpose of an EA. On average, the catheter was removed on day 3 (in 19 patients or 35%) but this varied between day 0 and day 5. Two catheters (3%) slipped out on days 2 and 3 unintentionally. The length of stay varied between <1 week to more than a month. The majority of the patients remained in the hospital between 1 and 2 weeks (26 or 49%). We did not have any reported "wet taps." 6 patients (10%) had "other complications" such as chronic pain or neuropraxia for more than 3 weeks, persistent motor weakness for more than 1 week, and hemodynamic instability. Interestingly, the rate of failed epidural was higher in the stronger concentration group (0.2% ropivacaine) at 70% or 37 patients.

When we carried out cross tabulations, we found that 90% of the "failed epidurals" or 15 patients (with pain scores >3, or other analgesia used) were elective cases. Both seniors (associate consultant and above) and juniors (registrars and below) had an approximately 30% rate of failed epidurals; however, more failures were seen by seniors as they performed the majority of the epidurals. The composite rate of failure (adding pain score >3 and use of other analgesia) was 37% or 20 patients. 80% of the epidurals placed at an inappropriate level had high pain scores (8 out of 10 patients). There was no association of rate of infusion or of the additive to success or failure of the epidural. There was a trend for greater hypotension in epidurals placed above T8. Average fluid bolused was 2 L to stabilize the blood pressure. There was no correlation between concentration of ropivacaine and complications; however, contrary to our expectation, the patients who received 0.2% infusion had more pain and were less satisfied. There was no correlation with hypotension, and the effect needs to be explored in a larger sample size prospectively.

  Discussion Top

EA is a safe and effective means of providing postoperative pain relief for major abdominal surgery. [11] However, international literature points to an overall high rate of failure. Despite this statistic, it is considered a gold standard in perioperative care of major abdominal surgery. The aim of thoracic postoperative analgesia is to provide optimal analgesia and minimal side effects for several days without the need for rescue medication. Freedom from pain to allow deep breathing and coughing without restriction helps attenuate the incidence of hypoxemic episodes and pulmonary complications experienced after surgery. Apart from their analgesic effects epidurals seem to bestow pulmonary, gastrointestinal and vascular beneficial effects as well and are a crucial part of the ERAS protocol. [12] However, recent trends point towards morbidity from its invasiveness as well as a high cost associated with using EA. Many regard analgesia alternatives to EA safer and less invasive and hence question the efficacy of EA. [13] Placing successful epidurals also require technical training and experience which requires constant practice. [14] In centers where ultrasound techniques were used to locate the correct intervertebral space and where patient controlled epidural infusions were used, success rates were higher. [15] It is recommended by many authorities to conduct detailed audits on practice of EA to assess performance and outcomes regularly. [16] With this aim in mind, we undertook this 6 months audit in our center.

Our results show nil mortality and a low rate of neurological morbidity (two patients or 3% with motor weakness for 1 week and one patient or 1% with sensory neuropraxia for 1 month). [17] However, we have a high rate of failure of 37% or twenty patients. This is an added score (composite score) as judged by high pain scores and use of adjuvant analgesics. We also have a high rate of hypotension requiring fluid boluses or vasopressors. Often an unsuitable spinal level was chosen for placement resulting in poor results. The seniority of the procedurist did not seem to affect the rate of failures as both failed around 30% of the time. This interesting finding needs further validation by larger prospective audits. Our results showed that level of training per se does not equate to experiential skip and prior level of experience with administration of this type of anesthesia.

Patients were mobilized late and epidurals were removed late in many cases. Our overall rate of placement of epidurals in all abdominal surgeries at 26% is low, which points to either an unfamiliarity in the technique or reliance on alternative modes of pain management. It is not possible to comment on surgeon or patient preference as these were not addressed in this retrospective audit. We had one each of persistent motor and sensory weakness which is deemed unacceptable in literature. Overall patient satisfaction was good in the majority of cases but the scores pertained to analgesia and not specifically to EA.


Our findings are limited due to the retrospective nature of the audit. As in previous similar studies the utility of the audit is for quality improvement in the center, it was conducted. However, some of the implications are generalizable to a larger cohort, and we believe useful for helming future research. We also have a small fraction of laparotomies receiving EA. The possible causes of this have been speculated on prior in this article. Even though we hold ourselves to a very high standard as to the definition of failure we can argue that this is the recommended definition of a working EA and if a patient requires multimodal therapy including narcotics, is not pain-free, requires vasopressors and fluids and is not mobilized, we have failed that patient.

  Conclusion Top

We conclude that (1) EA in our setting is safe but, (2) not effective and requires further and frequent scrutiny in terms of procedures, technical skill, education and perhaps looking at its cost effectiveness as compared to other alternatives which are less labor intensive and technically challenging. Greater awareness and insight is needed in recognizing failed epidurals and their causes. [18] The greatest utility of an audit such as this is to assess quality of care delivered rather than establish a research answer. [19] We realize that initial hypotension may be due to other causes such as fluid depletion and sepsis or bleeding. We, therefore, deliberately looked for hypotension more than 24 h after operation.

In our practice, we note the lack of consensus on perfect analgesia for abdominal surgeries, and perhaps a need for a more stringent protocol for management of such choices. Placement seems to be important in success of the epidural and recognition of the correct space is crucial. We recommend use of techniques to correctly identify the epidural space by ultrasound guidance instead of looking for alternate methods of analgesia. Another suggestion is to use a uniform concentration of ropivacaine (0.1%) for thoracic epidural infusion with standard multimodal analgesics (paracetamol, nonsteroidal antiinflammatory drug) to be used in conjunction. As suggested previously, EA to enhance recovery of abdominal surgery patients require a heavily protocolized multidisciplinary effort. We would like to suggest the following:

  • Better technical expertise, hence educational endeavors in EA for physicians and nurses before credentialing and constant reaccreditation required
  • Explicit consideration and justification of choice of EA versus alternate means of analgesia such as other regional blocks and multimodal analgesia
  • An appropriate level, dose, contraindication consideration and infusion regimen confirmed at start of surgery with "time out" process as part of the surgical safety checklist
  • Possibility of spinal level confirmation by ultrasound explored
  • Certainly repeating this audit as well as further research in the area is required
  • Protocol for evaluation of postoperative hypotension and measures in place to exclude all other potential causes before establishing EA as the cause, as well as
  • Pathways for early mobilization and an algorithm based management of the EA by nurses.
Pain intensity differs with the site, extent of incision and area of surgery. Major hepatobiliary and colonic resections are associated with severe pain and based on current evidence EA is appropriate even with ERAS protocols. Here TAP blocks and IV PCA are not sufficient, especially for dynamic pain. Our findings may well be echoed in many centers across the world, especially in similar settings. Cohort studies such as ours can add to this ongoing debate on epidurals adding another dimension, that of expertise. Further prospective studies may be required to find possible statistical significance, plus perhaps a randomized controlled trial to compare EA with other modes (such as transversus abdominus plane block, and multimodal analgesia plus PCA) may provide a definitive answer to our question.


We are very grateful for the kind suggestions by our Department Audit Committee and Professor Narinder Rawal in the writing of this manuscript and Ms. Su Chang for data analysis check.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

Bauer M, George JE 3 rd , Seif J, Farag E. Recent advances in epidural analgesia. Anesthesiol Res Pract 2012;2012:309219.  Back to cited text no. 1
Faccenda KA, Finucane BT. Complications of regional anaesthesia incidence and prevention. Drug Saf 2001;24:413-42.  Back to cited text no. 2
Ballantyne JC, Carr DB, deFerranti S, Suarez T, Lau J, Chalmers TC, et al. The comparative effects of postoperative analgesic therapies on pulmonary outcome: Cumulative meta-analyses of randomized, controlled trials. Anesth Analg 1998;86:598-612.  Back to cited text no. 3
Rigg JR, Jamrozik K, Myles PS, Silbert BS, Peyton PJ, Parsons RW, et al. Epidural anaesthesia and analgesia and outcome of major surgery: A randomised trial. Lancet 2002;359:1276-82.  Back to cited text no. 4
Cook TM, Counsell D, Wildsmith JA; Royal College of Anaesthetists Third National Audit Project. Major complications of central neuraxial block: Report on the Third National Audit Project of the Royal College of Anaesthetists. Br J Anaesth 2009;102:179-90.  Back to cited text no. 5
Hermanides J, Hollmann MW, Stevens MF, Lirk P. Failed epidural: Causes and management. Br J Anaesth 2012;109:144-54.  Back to cited text no. 6
Scherer R, Schmutzler M, Giebler R, Erhard J, Stöcker L, Kox WJ. Complications related to thoracic epidural analgesia: A prospective study in 1071 surgical patients. Acta Anaesthesiol Scand 1993;37:370-4.  Back to cited text no. 7
Chilvers CR, Nguyen MH, Robertson IK. Changing from epidural to multimodal analgesia for colorectal laparotomy: An audit. Anaesth Intensive Care 2007;35:230-8.  Back to cited text no. 8
Ganapathi S, Roberts G, Mogford S, Bahlmann B, Ateleanu B, Kumar N. Epidural analgesia provides effective pain relief in patients undergoing open liver surgery. Br J Pain 2015;9:78-85.  Back to cited text no. 9
Rawal N. Epidural technique for postoperative pain: Gold standard no more? Reg Anesth Pain Med 2012-Jun; 37:310-7.  Back to cited text no. 10
Kitching AJ, O'Neill SS. Fast-track surgery and anaesthesia. Contin Educ Anaesth Crit Care Pain 2009;9:39-43.  Back to cited text no. 11
Melnyk M, Casey RG, Black P, Koupparis AJ. Enhanced recovery after surgery (ERAS) protocols: Time to change practice? Can Urol Assoc J 2011;5:342-8.  Back to cited text no. 12
Liu SS, Wu CL. The effect of analgesic technique on postoperative patient-reported outcomes including analgesia: A systematic review. Anesth Analg 2007;105:789-808.  Back to cited text no. 13
Joshi GP, Bonnet F, Kehlet H; PROSPECT collaboration. Evidence-based postoperative pain management after laparoscopic colorectal surgery. Colorectal Dis 2013;15:146-55.  Back to cited text no. 14
Lirk P, Messner H, Deibl M, Mitterschiffthaler G, Colvin J, Steger B, et al. Accuracy in estimating the correct intervertebral space level during lumbar, thoracic and cervical epidural anaesthesia. Acta Anaesthesiol Scand 2004;48:347-9.  Back to cited text no. 15
Kehlet H, Gray AW, Bonnet F, Camu F, Fischer HB, McCloy RF, et al. A procedure-specific systematic review and consensus recommendations for postoperative analgesia following laparoscopic cholecystectomy. Surg Endosc 2005;19:1396-415.  Back to cited text no. 16
Horlocker T, Kopp S. Epidural hematoma after epidural blockade in the United States: It's not just low molecular heparin following orthopedic surgery anymore. Anesth Analg 2013;116:1195-7.  Back to cited text no. 17
Kooij FO, Schlack WS, Preckel B, Hollmann MW. Does regional analgesia for major surgery improve outcome? Focus on epidural analgesia. Anesth Analg 2014;119:740-4.  Back to cited text no. 18
Leslie K, Myles P, Devereaux P, Williamson E, Rao-Melancini P, Forbes A, et al. Neuraxial block, death and serious cardiovascular morbidity in the POISE trial. Br J Anaesth 2013;111:382-90.  Back to cited text no. 19


  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Article Tables

 Article Access Statistics
    PDF Downloaded221    
    Comments [Add]    

Recommend this journal