Share this article on:

Cost Drivers in Patient-Controlled Epidural Analgesia for Postoperative Pain Management After Major Surgery

Schuster, Martin MD, MA; Gottschalk, André MD; Freitag, Marc MD; Standl, Thomas MD

Section Editor(s): Cousins, Michael J. Section Editor

doi: 10.1213/01.ANE.0000096040.02725.74
PAIN MEDICINE: Research Report

In this retrospective study, we determined efficiency, treatment length, and resource use for postoperative pain management with patient-controlled epidural analgesia (PCEA) in 350 consecutive patients undergoing major abdominal, thoracic, gynecological, or orthopedic surgery. Average pain scores on a visual analog scale were 16 ± 23 and 9 ± 16 (visual analog scale range, 0 to 100) on postoperative Days 1 and 3, respectively, and were similar among groups. The treatment length was 4.9 ± 2.2 days in general surgical, 5.2 ± 3.1 days in gynecological, and 4.5 ± 2.8 days in orthopedic patients. The total volumes of the mixture of local anesthetic and opioid received epidurally were 707 ± 507 mL, 770 ± 576 mL, and 593 ± 456 mL in the general surgical, gynecological, and orthopedic groups, respectively. The average total costs for all groups for the full treatment course with PCEA were €447 ± €218 per case (€1 equals approximately US$1). Fifty-one percent of these costs were staff costs, 20% were costs for the applied drugs, 15% were costs for PCEA pumps and pump material, and 13% were costs for the initial catheter insertion. In the light of these costs and the availability of less costly alternatives, measurements for cost containment by using PCEA are recommended. Because treatment length is the main cost driver both for drug and staff costs, close monitoring of treatment length and a predefined migration path to alternative techniques after PCEA should be considered.

IMPLICATIONS: Patient-controlled epidural analgesia is increasingly used as first-line treatment for postoperative pain management. In this study, costs and cost drivers are analyzed for the first time for this new technique, based on 350 cases of pain therapy after major surgery in a university hospital.

Department of Anesthesiology, University Hospital Hamburg-Eppendorf, Hamburg, Germany

August 26, 2003.

Address correspondence and reprint requests to Martin Schuster, MD, MA, Department of Anesthesiology, University Hospital Hamburg-Eppendorf, Martinistr. 52, 20246 Hamburg, Germany. Address e-mail to

The introduction of epidural analgesia (EA) with catheter techniques has considerably changed postoperative pain management over the last decades (1). EA has gained increasing popularity on the basis of its excellent ability to control postoperative pain. In several studies comparing EA with IV patient-controlled analgesia (PCA), it has been demonstrated that EA is superior in terms of pain relief in general surgical (2), orthopedic (3), and gynecological patients (4) when compared with PCA. Additional advantages of EA in comparison with IV opioids for postoperative pain control have been identified: reduction of pulmonary (5) and cardiac complications (6), improvement of myocardial oxygenation and tissue reperfusion (6,7), and reduction of postoperative paralytic ileus after abdominal surgery (2). In several studies, EA decreased the time to mobilization and hospital discharge both in orthopedic (3,8) and general surgical (9,10) patients. The advantage of EA has also been shown in elderly (11) and high-risk (12) patients. It has been suggested that postoperative EA is most effective when integrated in a multimodal approach to optimize the preoperative, intraoperative, and postoperative management (1).

Patient-controlled EA (PCEA) is the latest development in EA, in which in addition to a background infusion, the patients are able to self-administer predefined top-up boluses on demand. This can be used to optimize pain control, to reduce the amount of required drugs, or to decrease the side effects because of a decreased background infusion (13).

Data on costs of EA for postoperative pain control in clinical practice have not been published in detail previously. In this study, we have retrospectively analyzed the costs of 350 consecutive general surgical, orthopedic, and gynecological patients treated with PCEA for postoperative pain management in our hospital from July 1, 2001, until June 30, 2002. The cost drivers are identified, and possible measures for cost reduction are discussed.

Back to Top | Article Outline


In our hospital we use EA as the standard of care for postoperative pain management after major general, orthopedic, and gynecological surgery. The epidural catheter (B. Braun, Melsungen, Germany) is placed before surgery. For abdominal or thoracic surgery, thoracic placement of the epidural is used, whereas patients with lower-limb surgery receive lumbar epidural catheters. During the operation, the epidural catheters are used in addition to a balanced anesthesia for continuous epidural infusion with 0.125% bupivacaine plus sufentanil 0.5 μg/mL in saline at a rate of 6–8 mL/h. After tracheal extubation, the PCEA infusion pump (Pegasus Light; Logomed, Windhagen, Germany) is connected to the epidural catheter in the postanesthesia care unit (PACU). The Pegasus Light pump is an electronically controlled, peristaltic infusion pump, similar to an IV PCA pump. It has a microdesign with only a 180-g pump weight. Because of its very small design, the pumps allow patients to ambulate with their PCEA device. All pump functions (infusion rate, top-up bolus volume, lockout time, hourly total, and so on) can be set by the physician at bedside. The physician can recall information on a display on given and refused boluses, total volume, and estimated time until the drug container will be empty. For postoperative pain control, a mixture of 0.2% ropivacaine and 0.5 μg/mL sufentanil in saline is delivered by the PCEA pump at 6–8 mL/h to the epidural space. Patients can release additional boluses (top-up doses), typically of 2 mL, with a lockout time of 15 min. Sufentanil is not approved in Germany for epidural use in patients older than 70 yr because of a supposed increased risk of respiratory depression in this population. These patients received ropivacaine 0.2% without sufentanil during the study period. Because we have shown (14) recently that larger concentrations of epidural ropivacaine are safe and provide excellent pain control without opioids, we are now using 0.375% ropivacaine for epidural use in postoperative patients who cannot receive opioids. The acute pain service (APS) is staffed by an anesthesia fellow and a nurse and is supervised by a staff anesthesiologist with subspecialty training in pain medicine. All patients are visited by the APS staff at least once a day, including weekends, and more often if pain control is insufficient or technical problems are present. Nonsteroidal antiinflammatory drugs and metamizole are frequently added to the medication scheme. Written protocols for pain management are kept with the patient record, and pain scores (at rest, during movement, and while coughing), heart rate, respiratory rate, blood pressure, motor function (Bromage scale) (15), level of analgesia, sedation score, and side effects are documented during every visit. Besides the APS, the nurses of the surgical wards routinely check the vital signs and monitor the patients for major side effects, such as over-sedation or motor block, at least three times a day. During the night, an anesthesiologist on call with expertise in pain management is available in-house for all difficulties with pain management.

After IRB approval, the charts of all patients older than 18 yr who were treated with epidural catheters between July 1, 2001, and June 30, 2002, were analyzed retrospectively. The number of contacts with the APS, duration of treatment, total analgesic volume required from the PCEA pumps, and daily pain scores were extracted. Three different types of costs were included: cost for insertion of the epidural catheter, material costs (local anesthetic, opioid, disposable pump set, and the share of pump costs), and staff cost for the APS. Costs for insertion of the epidural catheter were calculated on the basis of the material cost of the catheter set and the average time for placement of an epidural catheter by the anesthesiologist in the operating room. The epidural drug mixture is prepared by the hospital pharmacy in standardized sets, including a 300-mL reservoir. The sets are connected to the handheld pumps in the PACU and on the wards. Material costs were calculated on the basis of the actual number of sets used per patient. The costs for the pumps are included in the cost for the set material, based on the contract with the pump manufacturer. The drug costs reflect the acquisition cost for the hospital pharmacy. Costs for preparing the sets are not charged by the pharmacy in our institution. The calculation of the personnel costs is based on average annual costs used in our institution for anesthesia fellows (€65,400) and nurses (€41,400) and includes health insurance and social security costs. The total yearly personnel costs were divided by the total number of patient visits per year to calculate the cost for a single patient visit. This accounts for the effect that the personnel cost of the APS is independent of the actual number of patients treated at any given time. The costs for the regular weekend and additional night consultations were added on an hourly basis. General overhead costs (hospital administration costs, supervision costs, insurance, and so on) are not included in the calculation. Statistical analysis was performed with the statistical software SPSS 9.0. Results are given as mean ± sd.

Back to Top | Article Outline


Patient demographics and the most common diagnoses are shown in Table 1. A total of 350 patients were treated with PCEA: 176 after abdominal or thoracic surgery, 51 after gynecological operations (mainly gynecological tumor operations; no obstetric patients), and 123 after orthopedic surgery. The age of the patients was similar in the three groups. A few operations accounted for the majority of patients in each group: pancreas and colorectal surgery in the general surgical population, debulking operations for ovarian cancer in the gynecological patients, and hip and knee replacements in the orthopedic surgery group. Pain scores among the three patient groups were similar at rest on postoperative day (POD) 1 and POD 3 (Fig. 1). The treatment length tended to be shorter in the orthopedic surgery group (4.5 ± 2.8 days) versus in the general surgical (4.9 ± 2.2 days) and gynecological (5.2 ± 3.1 days) patients (Table 2). As a consequence, the total epidural volume received by the patients was smaller in the orthopedic population (593 ± 456 mL versus 707 ± 507 mL in the general surgical group and 770 ± 576 mL in the gynecological surgery group). The volume infused per day did not differ among groups. The average total cost for the full treatment course was €447 ± €218 (Table 3). The costs were highest in the gynecological surgery group (€471 ± €244) and lowest in the orthopedic surgery group (€419 ± €221). More than 50% of the costs were staff costs of the APS. Approximately one-third were material costs, with 20% of the costs for drugs and 15% for the PCEA pump and pump material. The insertion of the catheter accounted for only 13% of all costs.

Back to Top | Article Outline


There are still areas of uncertainty regarding the relative benefit of EA for some indications, and some recent studies could not reproduce the benefit of EA in decreasing the postoperative morbidity (16,17). Also, the specific value of PCA versus continuous EA has to be studied in more detail. However, as has been underscored in a recent overview on the effect of postoperative analgesia on surgical outcome, there is strong evidence for the superiority of EA compared with PCA regarding pain control and postoperative complications after major surgery (1). Accordingly, the current ASA practice guidelines for postoperative pain management state that EA should be made available by any postoperative pain service (18). A recent German survey showed that EA for postoperative pain management is available in almost all hospitals with dedicated anesthesiologists for the APS. More than 60% of these APSs used EA as the preferred method for postoperative pain management, both after major abdominal surgery and limb amputations (19). Again, our data stress the excellent pain control that can be achieved in normal clinical practice with PCEA with local anesthetics and opioids after major surgery. The visual analog scale (VAS) pain mea-surements recorded at rest in our pain service are comparable with the VAS values measured in most clinical studies of EA with a similar surgical population (3,20,21).

However, given the increasing cost-containment efforts, the benefits of sophisticated pain management tools must be weighed against their costs. In our study, we found that the average APS treatment with PCEA caused direct costs of approximately €450 per patient, with few differences among the three groups. This amount was much more than we expected. Despite extensive work on postoperative pain management, only few cost data regarding postoperative pain management have been published (22–24). The data published by Rawal and Berggren (22) summarize APS costs of all patients in a hospital but do not distinguish between different types of postoperative pain management regimens. The focus of the study of Brodner et al. (23) was financial savings due to PCEA by avoiding intensive care unit (ICU) stay. However, it is unclear how the APS costs were distributed among cases, and APS costs are reported only as a yearly total, not by case. The data on EA published by Rockemann et al. (24) were sampled during a study protocol, thereby not necessarily reflecting actual clinical practice. In addition, the staff costs were calculated on the basis of hours actually spent with the patient. However, in clinical practice the APS has to be staffed continuously and sufficiently to meet high demand times. In consequence, staff costs are fixed to a large extent. Similar to emergency room staffing, this causes inefficiency but has to be regarded as the trade-off for professional full-service postoperative pain management that is available 24 hours per day. Moreover, a comparison with total hospital costs shows that even with €450 per case, postoperative pain management is only a small portion of total hospital costs. PCEA is used in our institution mainly for postoperative pain control after major operations, in which total hospital costs per patient typically range from €8,000 to €10,000. In these cases, pain management causes only 5% of all costs, which seems to be reasonable, especially given the importance of efficient pain therapy for patients (25).

It has been argued that optimal pain control with EA also has a direct economic benefit based on a reduced or avoided stay in the ICU (23). It is a limitation of our study that we did not include these avoided direct costs in our analyses. However, because our study was retrospective in nature and focused on PCEA only, we did not have two similar patient populations with different pain regimens to calculate this effect. Reliable data in this respect can be extracted only from randomized controlled trials with clearly defined criteria, e.g., for discharge from the ICU or the PACU. Kehlet and Holte (1) have reviewed the literature regarding differences of hospital stay after PCEA versus PCA pain management regimens and found the data to be inconclusive because of the large numbers of confounding factors. Cost calculations of medical procedures have inherent weaknesses. Personnel and material costs differ largely between hospitals and especially between different health systems. The scope of analysis can include different sets of costs, and large amounts of costs can be hidden in overhead costs. Finally, a specific cost can be viewed very differently depending on the value of a specific good to the respective individual. Nevertheless, given the tremendous pressure to reduce hospital costs, any newly or additionally occurring cost is under tight observation both from the side of health insurance and of the hospital administration. This brought us to consider possibilities for reducing this cost without jeopardizing patient comfort and outcome. The main cost driver behind both personnel and drug costs is the length of treatment with PCEA. Some patients use PCEA for an extended time and seem to be prime candidates for reduction in treatment length (Fig. 2). However, in our experience, this is a misleading assumption. First, these complex cases benefit extensively from invasive pain management techniques and close monitoring by the APS. Second, the effect on overall costs would be small. To limit the maximum treatment length with PCEA to seven days would reduce total APS costs for all patients by only 5%. If cost reduction is necessary, the real challenge is to reduce the treatment length and APS involvement in most cases. To achieve this, we are now beginning to monitor treatment length more closely. After a predefined time (e.g., three days after total hip replacement or four days after pancreas surgery), we consider putting the PCEA pump on hold and switching to a fixed schedule of IV or oral analgesics if the patient can mobilize and cough (especially after abdominal surgery) without experiencing severe pain (VAS ≤30). The alternative regimen is started before stopping the PCEA, and the epidural catheter is kept in place. Only if the alternative regimen results in equal pain control do we remove the catheter, usually on the same day, and follow-up of the APS is discontinued. Otherwise, we continue the PCEA and try to omit PCEA on the next day. There are no data in the literature regarding the optimal time to change the regimen of pain therapy. Therefore, the time frame we suggest to start such a progression is based only on our clinical experience. The key to success is a concerted approach with the surgical colleagues and integration into a multimodal approach with consensual clinical pathways for postoperative therapy. Our aim is to reduce the average treatment length by at least one day, which would in turn result in a 20% reduction of APS drug and personnel costs.

Given the expense of PCEA, it is tempting to reduce the number of patients treated with PCEA and to use presumably less-costly techniques, such as PCA or oral analgesics, as the initial postoperative pain management regimen. We doubt that this approach would be successful. First, insufficient pain management is not only unethical, but it is also costly, because excessive time of the ward staff is needed to deal with a patient in pain. Second, it is important to keep in mind that more than half of the costs are the personnel costs of the APS. State-of-the-art postoperative pain management that fulfills standard quality criteria, such as regular assessment and documentation of pain scores and side effects, is time consuming, independent of the technique used to control pain. Decreasing the standard of care for the immediate postoperative phase should not be considered. In our opinion, changing from PCEA to IV PCA would not significantly reduce the time the APS spent for each case, even though we cannot prove this in our patient population, because patients who are treated with PCEA and those who are treated with IV PCA are quite different groups in our hospital. We treat patients scheduled for major surgery—such as pancreas resection, extensive tumor resections of the bowel, hip and knee surgery, or amputation—with PCEA. Patients with less extensive surgery and no significant comorbidity receive treatment with IV PCA exclusively. The treatment with IV PCA is started in the PACU, but after discharge from the PACU the treatment management is taken over by the nurses and the surgical staff of the ward.

Because of the decreased expense of bupivacaine compared with ropivacaine in Germany, the replacement of 0.2% ropivacaine with 0.175% bupivacaine would theoretically be a cost-saving measure. Brodner et al. (9) found similar pain scores and similar epidural drug volumes when comparing these two drugs for PCEA after major abdominal surgery. However, in light of the superior toxicity profile of ropivacaine and the reduced incidence of motor block (9), we prefer ropivacaine for EA on general wards because sophisticated electrocardiogram monitoring to detect early cardiac side effects of bupivacaine is not routinely available on general wards.

A detailed cost comparison of PCEA and IV PCA has not yet been published and might be very difficult to perform in a randomized, controlled trial. However, any comparison not only should include direct costs, such as material cost and staff cost, in which PCEA has disadvantages, but should also account for indirect costs, such as duration of PACU and ICU stay and costs of complications, in which PCEA might have decisive advantages.

We want to thank the anesthesia fellows of the Department of Anesthesiology and Andrea Domke, RN, for their outstanding continuous support of the APS at the University Hospital Hamburg-Eppendorf.

Back to Top | Article Outline


1. Kehlet H, Holte K. Effect of postoperative analgesia on surgical outcome. Br J Anaesth 2001; 87: 62–72.
2. Liu SS, Carpenter RL, Mackey DC, et al. Effects of perioperative analgesic technique on rate of recovery after colon surgery. Anesthesiology 1995; 83: 757–65.
3. Capdevila X, Barthelet Y, Biboulet P, et al. Effects of perioperative analgesic technique on the surgical outcome and duration of rehabilitation after major knee surgery. Anesthesiology 1999; 91: 8–15.
4. Wattwil M, Thoren T, Hennerdal S, et al. Epidural analgesia with bupivacaine reduces postoperative paralytic ileus after hysterectomy. Anesth Analg 1989; 68: 353–8.
5. Ballantyne JC, Carr DB, deFerranti S, 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.
6. Beattie WS, Buckley DN, Forrest JB. Epidural morphine reduces the risk of postoperative myocardial ischaemia in patients with cardiac risk factors. Can J Anaesth 1993; 40: 532–41.
7. Christopherson R, Beattie C, Frank SM, et al. Perioperative morbidity in patients randomized to epidural or general anesthesia for lower extremity vascular surgery: Perioperative Ischemia Randomized Anesthesia Trial Study Group. Anesthesiology 1993; 79: 422–34.
8. Williams-Russo P, Sharrock NE, Haas SB, et al. Randomized trial of epidural versus general anesthesia: outcomes after primary total knee replacement. Clin Orthop 1996; 331: 199–208.
9. Brodner G, Mertes N, Van Aken H, et al. Epidural analgesia with local anesthetics after abdominal surgery: earlier motor recovery with 0.2% ropivacaine than 0.175% bupivacaine. Anesth Analg 1999; 88: 128–33.
10. Carli F, Mayo N, Klubien K, et al. Epidural analgesia enhances functional exercise capacity and health-related quality of life after colonic surgery: results of a randomized trial. Anesthesiology 2002; 97: 540–9.
11. Mann C, Pouzeratte Y, Boccara G, et al. Comparison of intravenous or epidural patient-controlled analgesia in the elderly after major abdominal surgery. Anesthesiology 2000; 92: 433–41.
12. Yeager MP, Glass DD, Neff RK, et al. Epidural anesthesia and analgesia in high-risk surgical patients. Anesthesiology 1987; 66: 729–36.
13. Silvasti M, Pitkanen M. Patient-controlled epidural analgesia versus continuous epidural analgesia after total knee arthroplasty. Acta Anaesthesiol Scand 2001; 45: 471–6.
14. Gottschalk A, Freitag M, Burmeister MA, et al. Patient-controlled thoracic epidural infusion with ropivacaine 0.375% provides comparable pain relief as bupivacaine 0.125% plus sufentanil after major abdominal gynecologic tumor surgery. Reg Anesth Pain Med 2002; 27: 367–73.
15. Bromage PR. A comparison of the hydrochloride and carbon dioxide salts of lidocaine and prilocaine in epidural analgesia. Acta Anaesthesiol Scand Suppl 1965; 16: 55–69.
16. Norris EJ, Beattie C, Perler BA, et al. Double-masked randomized trial comparing alternate combinations of intraoperative anesthesia and postoperative analgesia in abdominal aortic surgery. Anesthesiology 2001; 95: 1054–67.
17. Rigg JR, Jamrozik K, Myles PS, et al. Epidural anaesthesia and analgesia and outcome of major surgery: a randomised trial. Lancet 2002; 359: 1276–82.
18. American Society of Anesthesiologists. Practice guidelines for acute pain management in the perioperative setting: a report by the American Society of Anesthesiologists Task Force on Pain Management, Acute Pain Section. Anesthesiology 1995; 82: 1071–81.
19. Stamer UM, Mpasios N, Stuber F, et al. A survey of acute pain services in Germany and a discussion of international survey data. Reg Anesth Pain Med 2002; 27: 125–31.
20. Brodner G, Mertes N, Van Aken H, et al. What concentration of sufentanil should be combined with ropivacaine 0.2% wt/vol for postoperative patient-controlled epidural analgesia? Anesth Analg 2000; 90: 649–57.
21. Hodgson PS, Liu SS. A comparison of ropivacaine with fentanyl to bupivacaine with fentanyl for postoperative patient-controlled epidural analgesia. Anesth Analg 2001; 92: 1024–8.
22. Rawal N, Berggren L. Organization of acute pain services: a low-cost model. Pain 1994; 57: 117–23.
23. Brodner G, Mertes N, Buerkle H. Acute pain management: analysis, implications and consequences after prospective experience with 6349 surgical patients. Eur J Anaesthesiol 2000; 17: 566–75.
24. Rockemann MG, Seeling W, Goertz AW, et al. Effectiveness, side effects and costs of postoperative pain therapy: intravenous and epidural patient-controlled analgesia (PCA). Anasthesiol Intensivmed Notfallmed Schmerzther 1997; 32: 414–9.
25. Strassels SA, Chen C, Carr DB. Postoperative analgesia: economics, resource use, and patient satisfaction in an urban teaching hospital. Anesth Analg 2002; 94: 130–7.
© 2004 International Anesthesia Research Society