Ann Thorac Surg 2003;75:490-495
© 2003 The Society of Thoracic Surgeons
Original article: cardiovascular
Pain treatment with a COX-2 inhibitor after coronary artery bypass operation: a randomized trial
Franz F. Immer, MDa,
Alexsandra S. Immer-Bansi, MDb*,
Nathalie Trachsel, MDa,
Pascal A. Berdat, MDa,
Verena Eigenmann, MDa,
Michele Curatolo, MD, PhDb,
Thierry P. Carrel, MDa
a Department of Cardiovascular Surgery and Department of University Hospital, Berne, Switzerland
b Department of Anesthesiology, Division of Pain Therapy, University Hospital, Berne, Switzerland
Accepted for publication August 19, 2002.
* Address reprint requests to Dr Immer-Bansi, Institute of Anesthesiology, University Hospital, 3010 Berne, Switzerland.
e-mail: immerbansi{at}yahoo.de
 |
Abstract
|
|---|
BACKGROUND: Adequate analgesic medication is mandatory after cardiac operations. Cyclooxygenase-2 inhibitors represent a new therapeutic option, acting primarily on the response to inflammation.
METHODS: We compared a cyclooxygenase-2 inhibitor (etodolac) with two traditional drugs: a nonselective cyclooxygenase inhibitor (diclofenac) and a weak opioid (tramadol) on postoperative pain and renal function in patients undergoing coronary artery bypass operations. Sixty consecutive patients were randomized into three groups: (1) group A patients who received tramadol; (2) group B patients who received diclofenac; and (3) group C patients who received etodolac. For measurement of analgesic effect, the visual analogue scale was assessed up to postoperative day 4. Creatinine-clearance was determined before and at the end of study medication, and serum creatinine and urea were monitored daily for renal effects. Study medication was given on postoperative days 2 and 3. Side effects and additional pain medication were recorded.
RESULTS: The visual analogue scale was lower in group C (p < 0.05) from postoperative days 2 to 4 and in group B (p < 0.05) from postoperative days 3 to 4 compared with group A. Amount of additional pain medication and incidence of side effects were significantly less in group C compared with group A. We observed a short-lasting elevation of serum creatinine and urea in groups B and C compared with group A (p < 0.05).
CONCLUSIONS: At the doses analyzed, etodolac and diclofenac produced better postoperative pain relief with less side-effects than tramadol. A short-lasting impairment of renal function was found in patients treated with etodolac and diclofenac.
|
Introduction
|
|---|
Pain therapy is an important aspect in the postoperative care of patients undergoing surgical interventions. Median sternotomy, which is performed in the majority of patients requiring coronary artery bypass grafting (CABG) and harvesting of the internal mammary artery may be very painful [1–5]. Fortunately, postoperative pain after CABG is normally transient with a maximum up to postoperative day 3 [1, 5]. Pain is caused by the surgical trauma itself, which consists of musculoskeletal distortion of the thoracic cage, as well as surgical manipulation of the parietal pleura and costal and sternal periosteum. Because renal function may deteriorate after nonpulsatile cardiopulmonary bypass used for cardiac surgical procedures, nonsteroidal anti-inflammatory drugs (NSAIDs) should be used with caution and their use in the early postoperative period is controversial [6–8]. Only a few studies using NSAIDs in the early postoperative period after CABG operation have been published [2]. The analgesic mechanism of NSAIDs is inhibition of prostaglandin biosynthesis by inhibiting cyclooxygenase (COX). COX is known to exist as two isoforms: COX-1 and COX-2. COX-1 expression is ubiquitous and its activity predominates during normal physiologic conditions. COX-2 expression is restricted to a few specialized loci during basal conditions. In response to factors such as inflammation and synaptic activity, COX-2 expression is markedly enhanced by mediators, including growth factors, cytokines, and mitogens. It has been proposed that the therapeutic activity of NSAIDs in pain treatment is primarily the result of inhibition of COX-2, whereas the toxicity of these drugs concerning gastrointestinal function and coagulation results from inhibition of the constitutive COX-1 isoform [9]. Recently, selective COX-2-inhibitors have become available. Their effect in analgesic treatment after cardiac operation has not yet been investigated.
The aim of this study was to evaluate the efficacy of a relative selective COX-2 inhibitor (etodolac) on postoperative pain after coronary artery bypass operation and to compare it with two traditional drugs used in postoperative pain treatment (diclofenac, an unselective COX inhibitor and tramadol, a weak opioid). Renal function was assessed before and after pain treatment.
|
Patients and methods
|
|---|
Patients
Sixty-nine consecutive patients, scheduled for elective CABG operation were enrolled in this prospective trial between June 2000 and October 2000. None of the patients who fulfilled the inclusion criteria refused to participate in the study. Exclusion criteria were age more than 70 years, left ventricular ejection fraction less than 30%, previous history of peptic ulcer disease or gastrointestinal bleeding, hepatic or renal insufficiency, known allergy to tramadol or NSAIDs, and preoperative analgesic treatment. Postoperative period exclusion criteria were delayed transfer to the general ward (eg, intensive care treatment requiring for more than 24 hours), serum creatinine more than 150 mmol/L, and altered mental status. Patients were randomly assigned to either group A patients who received tramadol slow-release (150 mg every 12 hours orally) (Tramal Retard, Grünenthal, Glarus, Switzerland) or group B patients who received diclofenac (50 mg every 8 hours orally) (Voltaren, Novartis Pharma, Basel, Switzerland), or group C etodolac (300 mg every 8 hours orally) (Lodine, Sigma-Tau, Zofingen, Switzerland).
Methods
The study was approved by the local Ethical Committee and informed consent was obtained from all patients. Demographic data of the patients were recorded, as well as specific data like coronary angiography report for the extent of coronary heart disease; ejection fraction reflecting left ventricular function; previous history of myocardial infarction, as well as history of smoking, diabetes mellitus, and chronic obstructive pulmonary disease. During the procedure any fracture of the sternum or ribs, the number and type of grafts (internal mammary artery, radial artery, or saphenous vein, or any combination thereof) as well as the duration of the intervention were assessed. Conventional sternal and left internal mammary artery retractors were used, and harvesting of the left internal mammary artery was performed intrapleurally. Three thoracic tubes were inserted before closing of the sternum (one in the left pleura, one in the pericardium, and one retrosternally).
Blood samples were taken preoperatively to analyze serum creatinine and daily from the first up to the fourth postoperative day and before discharge. Creatinine-clearance was recorded on the first postoperative day (before starting study medication) and on postoperative day 4 (after receiving study medication). In patients with insufficient analgesic treatment (defined as visual analogue scale [VAS] more than 5), morphine (7.5 mg) was given subcutaneous with a minimal interval of 4 hours. Total morphine consumption and occurrence of nausea were recorded daily. A basic analgesic regimen with 1 g paracetamol (Dafalgan, UPSA, Switzerland) every 6 hours orally was started after extubation in every patient and maintained until discharge. Study medication was given on postoperative days 2 and 3. Treatment efficacy was monitored by the VAS, which is a well established and frequently used instrument to assess pain intensity. Postoperatively, one of the authors (AIB) who specializes in pain treatment, asked patients daily (in the late afternoon) to indicate pain intensity during coughing and in-bed mobilization, and maximal and average pain level within the past 24 hours. Furthermore, verbal rating score (VRS) was assessed similarly. Pain by VRS was quantified by the patient according to five categories: no, slight, moderate, severe and unbearable pain. Patients with regular preoperative pain medication were excluded from the study so as not to influence postoperative VAS and VRS scores by pain of nonsurgical origin. Both examiner and patient were blinded to the type of study medication. All patients were additionally treated with 100 mg acetylsalicylic acid daily to have a slight inhibition of cyclooxygenase 1 to prevent aggregation of thrombocytes. In this regimen acetylsalicylic acid was not assumed to be an analgesic treatment. According to the protocol, all patients randomized in the study received omeprazole 20 mg daily until discharge to prevent gastric ulcer, which is a well-known complication after medical treatment with diclofenac.
Data analysis
Demographic factors, clinical preoperative measurements, intraoperative data, and postoperative nonrepeated measurements were analyzed as follows: one-way analysis of variance for normally distributed numerical data, Kruskal-Wallis one-way analysis of variance on ranks for non-normally distributed numerical data, and 
2 test for categorical data. Repeated postoperative measurements were analyzed by two-way repeated measures of analysis of variance. The Tukey test was used for multiple comparison in the analysis of variance. The verbal rating score was reported for description and was not analyzed statistically. A p value less than 0.05 was considered as significant.
|
Results
|
|---|
Nine patients were excluded postoperatively; 8 patients were not transferred to the general ward on day 1, and 1 patient had an increase in serum creatinine to 150 µmol/L 12 hours postoperatively. No hepatic insufficiency was observed in the postoperative period. Sixty patients were randomized (20 to each group). The groups were similar in respect to age, gender, left ventricular function, cardiovascular risk factor profile, surgical procedure, and technique (Table 1).
The higher incidence of diabetes in group B was not significant in comparison with the other two groups. All patients suffering from diabetes were on oral antidiabetic medication. The incidence of fractures of the sternum or ribs, or both, was not different in the three groups (Table 1). We observed no significant increase of pain in patients with fractures of the sternum or ribs, or both. The maximum pain intensity, assessed by the VAS is presented in Figure 1.
On postoperative day 1 on which the study drugs were not administered, no statistically significant difference was found among the groups (VAS: group A, 6.3 ± 1.5; group B, 6.8 ± 1.8; and group C, 6.4 ± 1.7). The test for multiple comparison revealed significantly lower VAS levels in group C from postoperative days 2 to 4 compared with group A (p < 0.05). Group B displayed lower maximum VAS than group A only at postoperative days 3 and 4 (p < 0.05). All other pain-related comparisons were not statistically significant between the groups. Although we found no significant difference between groups B and C (p = not significant), group C patients tended to have lower VAS and VRS scores after application of study medication at postoperative day 2. We found an overall decrease of maximal VAS from postoperative day 1 up to 4 of 64% in group C, 47% in group B, and 11% in group A. Similar results were obtained with the VRS (Fig 2).
On postoperative day 1 the total dose of morphine administered was similar in all three groups. On postoperative day 4, group C patients required significantly less morphine compared with group A (p < 0.05) (Table 2).
From postoperative days 2 to 4, group B and C patients required significantly less anti-emetic medication compared with those of group A (p < 0.05) (Table 2). Creatinine clearance was similar in all groups. A reduction of creatinine clearance from postoperative days 1 to 4 of 9.7% in group B, 10.7% in group C, and 14.3% in group A was found, which did not reach statistical significance (Table 1).
View larger version (18K):
[in this window]
[in a new window]
|
Fig 1. Visual analogue scale (maximum) for groups A, B, and C on postoperative days 1 to 4. Results are displayed as mean values. Study medication was given on postoperative days 2 and 3.
|
|
View larger version (51K):
[in this window]
[in a new window]
|
Fig 2. Verbal rating scale (maximum) for patients treated with tramadol (group A), diclofenac (group B), and etodolac (group C) on postoperative days 1 to 4. Number of patients with no, mild, moderate, severe, and unbearable pain (in %).
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Additional Anti-Emetic Therapy (Average No. of Applications Daily) and Pain Treatment (Average Doses of Opiate in mg Daily) for Each Group
|
|
Serum creatinine and serum urea were significantly higher at postoperative days 3 and 4 in groups B and C compared with group A (p < 0.05) (Figs 3, 4).
However, at discharge no significant difference was found between the three groups (Table 3).
View larger version (19K):
[in this window]
[in a new window]
|
Fig 3. Serum creatinine (µmol/L) value for groups A, B, and C on postoperative days 1 to 4 and at discharge. Results are displayed as mean values. Upper limit of norm 110 µmol/L.
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig 4. Serum urea value (mmol/L) for groups A, B, and C on postoperative days 1 to 4 and at discharge. Results are displayed as mean values. Upper limit of norm 7.3 mmol/L.
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Visual Analogue Scale (Maximum), Serum Creatinine (µmol/L), and Serum Urea (mmol/L) for Groups A (n = 20), B (n = 20), and C (n = 20)
|
|
A borderline significance for shorter hospitalization was found in group C (7.3 ± 1.6 days) compared with group A (8.6 ± 1.7 days) (p = 0.57) (Table 1).
|
Comment
|
|---|
This study demonstrated that diclofenac and etodolac combined with paracetamol reduced postoperative pain and the need for additional opioid and anti-emetic medications after CABG. Similar observation had already been reported by others in patients treated with NSAIDs compared with opioid treatment after hysterectomy [10], as well as after abdominal operation [11, 12]. This opioid-sparing effect allows reduction of several of the well-known side effects of morphine medication, such as drowsiness and anti-emetic therapy [13, 14], which is consistent with our observations. As tramadol is known to be a weak opioid, it may be assumed that pain reduction would have been better by using a stronger opioid. However, no significant increase in additional pain medication was observed in the subset of patients treated with tramadol on postoperative days 2 and 3. Furthermore, more pronounced side effects may have been expected with the use of a stronger opioid.
Diclofenac and etodolac are NSAIDs that inhibit the COX enzyme with consecutive decrease in prostaglandin synthesis. These drugs effectively provide analgesia and suppress inflammation. However, COX inhibition may lead to undesired side effects by inhibiting synthesis of prostaglandins that serve important functions in other organs. This may promote complications, including upper gastrointestinal ulcer formation and hemorrhage, renal insufficiency, and platelet dysfunction, resulting in increased postoperative blood loss [9, 15]. Since the discovery of the COX-2 isoforms, a more selective approach became feasible. COX-1 is expressed normally in the gastrointestinal tract, the kidneys, and the platelets; in these organs and cells its inhibition is not desired. Etodoloac is a relative selective COX-2 inhibitor and therefore should mainly act on inhibition of COX-2 activity, which is primarily up-regulated in response to inflammation [9]. In particular, side effects like gastrointestinal events are significantly lower as Bombardier and colleagues [16] reported in a large randomized trial of 8,076 patients with rheumatoid arthritis treated with rofecoxib (a selective COX-2 inhibitor) compared with patients treated with naproxen (a nonselective inhibitor).
Comparing the maximal VAS score in groups B and C, we found a trend toward a faster decrease of VAS score on postoperative day 2 in group C without reaching statistical significance. An onset of analgesic effect appears earlier in patients treated with etodolac (group C) compared with patients treated with diclofenac (group B), which already had been reported [17, 18]. Looking at the side effects, we found no hemorrhage or increased blood loss in our patients (Table 1). On the other hand, a slight but reversible elevation of serum creatinine and serum urea, reflecting smoothly impaired renal function, was found in groups B and C. During and shortly after medication, creatinine clearance remained basically unchanged. Nevertheless, renal function can be impaired by the use of NSAIDs and COX-2 inhibitors, which is in accordance with two recently published studies of elderly patients and of healthy patients [19, 20]. Inhibition of COX is associated with adverse renal effects caused by reduction in synthesis of renal prostaglandins. Both isoforms of COX, COX-1 and COX-2, are expressed in the kidney in constitutive and inducible forms. In the renal cortex, COX-2 expression increases in high renin states, and selective COX-2 inhibitors significantly decrease plasma renin levels. In the medullary region of the kidney, expression of COX-2 increases in response to a high-salt diet and water deprivation. The most important prostanoids in the kidney are prostaglandin I2, or prostacyclin, and prostaglandin E2. Prostaglandin I2 is a potent vasodilator and helps maintain renal perfusion during conditions of decreased effective circulating blood volume; its inhibition in hypovolemic patients may result in acute renal failure. It is therefore assumed that the COX-2 selective inhibitors may have an effect similar to that of nonselective NSAIDs [21, 22]. As the cause of renal abnormalities after cardiopulmonary bypass is multifactorial, future studies should focus on the effect of COX inhibitors on renal function, especially in patients undergoing beating heart operations, which are assumed to have a less deleterious effect on renal function than on-pump operations [23].
In conclusion, in patients properly screened for renal insufficiency and history of peptic ulcer, short-term analgesic treatment with diclofenac and etodolac is superior to tramadol in quality of pain treatment and incidence of side effects after cardiac operation procedures and may be associated with shorter hospital stays. The clinical relevance of the slight deterioration in renal function should be investigated by outcome studies on large patient populations.
|
References
|
|---|
- Walther T., Falk V., Metz S. Pain and quality of life after minimally invasive versus conventional cardiac surgery. Ann Thorac Surg 1999;67:1643-1647.[Abstract/Free Full Text]
- Rapanos T., Murphy P., Szalai J.P., Burlacoff L., Lam Mc Culloch J., Kay J. Rectal indomethacin reduces postoperative pain and morphine use after cardiac surgery. Can J Anesth 1999;46(8):725-730.[Medline]
- Holmer Petterson P., Anjou Lindskog E., Öwall A. Patient-controlled versus nurse-controlled pain treatment after coronary artery bypass surgery. Acta Anaesthesiol Scand 2000;44:43-47.[Medline]
- Müller X.M., Tinguely F., Tevaearai H.T., Revelly J.P., Chioléro R., von Segesser L.K. Pain location, distribution, and intensity after cardiac surgery. Chest 2000;118:391-396.[Abstract/Free Full Text]
- Müller X.M., Tinguely F., Tevaerai H.A.T., Revelly J.P., Chiolero R., von Segesser L.K. Pain pattern and left internal mammary artery grafting. Ann Thorac Surg 2000;70(6):2045-2049.[Abstract/Free Full Text]
- Souter A.J., Fredmann B., White P.F. Controversies in the perioperative use of nonsteroidal antiinflammatory drugs. Anesth Analg 1994;79:1178-1190.[Free Full Text]
- Ralley F.E., Day F.J., Cheng D.C. Pro: nonsteroidal anti-inflammatory drugs should be routinely administered for postoperative analgesia after cardiac surgery. J Cardiothorac Vasc Anesth 2000;14(6):731-734.[Medline]
- Griffin M. Con: nonsteroidal anti-inflammatory drugs should not be routinely administered for postoperative analgesia after cardiac surgery. J Cardiothorac Vasc Anesth 2000;14(6):735-738.[Medline]
- Peterson W.L., Cryer B. COX-1-Sparing NSAIDs—is the Enthusiasm Justified?. JAMA 1999;282(20):1961-1963.[Free Full Text]
- Engel C., Lund B., Kristensen S.S., Axel C., Nielsen J.B. Indomethacin as an analgesic after hysterectomy. Acta Anaesth Scand 1989;33:498-501.[Medline]
- Reasbeck P.G., Rice M.L., Reasbeck J.C. Double-blind controlled trial of indomethacin as an adjunct to narcotic analgesia after major abdominal surgery. Lancet 1982;2:115-118.[Medline]
- Sims C., Johnson C.M., Bergesio R., Delfos S.J., Avraamides E.A. Rectal indomethacin for analgesia after appendectomy in children. Anaesth Intensive Care 1994;22:272-275.[Medline]
- Reuben S., Connelly N.R., Lurie S., Klatt M., Gibson C.S. Dose-response of ketorolac as an adjunct to patient-controlled analgesia morphine after spinal surgery. Anesth Analg 1994;79:1178-1190.
- Parker R.K., Holtmann B., Smith I., White P.F. Use of ketorolac after lower abdominal surgery. Effect on analgesic requirement and surgical outcome. Anesthesiology 1994;80:6-12.[Medline]
- Jackson L.M., Hawkey C.J. COX-2 selective nonsteroidal anti-inflammatory drugs: do they really offer any advantages?. Drugs 2000;59(6):1207-1216.[Medline]
- Bombardier C., Laine L., Reicin A., Shapiro D., Burgos-Vargas R., Davis B. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med 2000;343:1520-1528.[Abstract/Free Full Text]
- Amadio P., Cummings D.M., Amadio P. Nonsteroidal antiinflammatory drugs, tailoring therapy to achieve results and avoid toxicity. Postgrad Med 1993;93(4):73-76.
- Bontoux D. Lombo-Radiculalgies: Efficacité, tolérance et bénéfice thérapeutique comparés de l’étodolac (600mg/j) du diclofénac (150mg/j) et du placebo. Rhumatologie 1990;42(7a):201-206.
- Svendsen K.B., Bech J.N., Sorensen T.B., Pedersen E.B. A comparison of the effects of etodolac and ibuprofen on renal haemodynamics, tubular function, renin, vasopressin, and urinary excretion of albumin and
-glutathione and S-transferase in healthy subjects: a placebo-controlled cross-over study. Eur J Clin Pharmacol 2000;56:383.[Medline]
- Swan S.K., Rudy D.W., Lasseter K.C. Effect of cyclooxygenase-2 inhibition on renal function in elderly persons receiving a low-salt diet. Ann Intern Med 2000;133:1-9.[Abstract/Free Full Text]
- Harris R.C., Jr Cyclooxygenase-2 inhibition and renal physiology. Am J Cardiol 2002;89(6A):10D-17D.[Medline]
- Schwartz JI, Vandormael K, Malice MP, et al. Clin Pharmacol Ther 2002;72(1):50–61
- Ascione R., Lloyd C.T., Underwood M.J., Gomes W.J., Angelini G.D. On-pump versus off-pump coronary revascularization: evaluation of renal function. Ann Thorac Surg 1999;68(2):493-498.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
M. R. Khajavi, S. B. M. Aghili, R. S. Moharari, A. Najafi, R. Mohtaram, P. Khashayar, and M. Mojtahedzade
Subcutaneous Tramadol Infiltration at the Wound Site Versus Intravenous Administration after Pyelolithotomy
Ann. Pharmacother.,
March 1, 2009;
43(3):
430 - 435.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. F. White
The Changing Role of Non-Opioid Analgesic Techniques in the Management of Postoperative Pain
Anesth. Analg.,
November 1, 2005;
101(5S_Suppl):
S5 - 22.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. SPINK, S. BAHN, and R. GLICKMAN
Clinical implications of cyclo-oxygenase-2 inhibitors for acute dental pain management: Benefits and risks
J Am Dent Assoc,
October 1, 2005;
136(10):
1439 - 1448.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Kulik, M. Ruel, M. E. Bourke, L. Sawyer, J. Penning, H. J. Nathan, T. G. Mesana, and P. Bedard
Postoperative naproxen after coronary artery bypass surgery: a double-blind randomized controlled trial
Eur. J. Cardiothorac. Surg.,
October 1, 2004;
26(4):
694 - 700.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
