Secondary Logo

Journal Logo

Original Article

Does ibuprofen increase perioperative blood loss during hip arthroplasty?

Slappendel, R.; Weber, E. W. G.; Benraad, B.; Dirksen, R.; Bugter, M. L. T.

Author Information
European Journal of Anaesthesiology (EJA): November 2002 - Volume 19 - Issue 11 - p 829-831
  • Free



Non-steroidal anti-inflammatory drugs (NSAIDs) are used in the perioperative period for analgesia and reduction of oedema in the surgical field. Beside these benefits, there are some unwanted side-effects: a reduction of renal blood flow, gastric complaints and an increase of blood loss during surgery by influencing the coagulation cascade. NSAIDs are widely used in orthopaedic surgery, and ibuprofen is a very commonly used NSAID in The Netherlands. There is concern over the perioperative use of NSAIDs since they have the potential to increase perioperative blood loss related to their mechanism of action [1-4].

We decided to assess the effect of ibuprofen on perioperative blood loss in routine practice in patients undergoing hip arthroplasty by means of a randomized double-blinded placebo-controlled study.


The Ethics Committee of our hospital approved the study, and written informed consent was obtained from all patients. To exclude as many external factors as possible, we only included patients who were to undergo their first elective total hip replacement for coxarthrosis during spinal (intrathecal) anaesthesia. Patients were allocated and randomized to two groups in a double-blind manner 2 weeks before surgery. All patients were pretreated for 2 weeks before surgery: one group with placebo drug, the other group with ibuprofen 600 mg. Both placebo and ibuprofen 600 mg (total 1800 mg) were given orally three times a day. The pharmacist who prepared the identical tablets was the only person aware of the type of treatment. The exclusion factors were: any patients receiving NSAIDs, aspirin or anticoagulants before starting the trial, and any patients with a history of peptic ulcer, renal dysfunction or allergy to any NSAID.

On the day of surgery, all patients were premedicated with midazolam 7.5 mg orally 1 h before spinal anaesthesia. Spinal anaesthesia was performed in each patient by administering bupivacaine 20 mg plus morphine 0.1 mg contained in 4 mL solution. Glucose saline solution (2.5% glucose and NaCl 0.9%) 500 mL was given intravenously immediately before starting surgery. A continuous infusion of the same solution was administered during surgery 250 mL h−1 and after surgery 100 mL h−1. A colloid solution (Gelofusine®; B Braun, Melsungen AG, Melsungen, Germany) was also given to match the volume of blood lost to a maximum of 2.5 L in 24 h.

Adequate sedation was provided by the patient's request during the procedure: the anaesthesiologist administered midazolam 1 mg at a minimum interval of 5 min until the patient indicated that the desired level of sedation had been reached. Non-invasive blood pressure, heart rate (from the electrocardiograph), transcutaneous oxygen saturation and respiratory rate were continuously monitored during anaesthesia and in the intensive care unit during the first 24 h after surgery.

Perioperative blood loss

All operations were performed by the same orthopaedic surgeon. Prophylaxis against thromboembolism was started in all patients on the evening before surgery with acenocoumarol 3 mg. On the day of surgery, acenocoumarol 2 mg was given 24 h after the initial dose. The third dose of acenocoumarol was given after the trial was completed. Nurses in the operating room measured perioperative blood loss. They were unaware of the NSAIDs given. Total blood loss was calculated by taking into account the amount in the suction bottles, the weight of the surgical sponges and the irrigation fluid used.

After operation, all patients stayed in the intensive care unit according to our normal policy. The volume of blood collected in the high-vacuum wound drainage containers was measured for 24 h after surgery. The transfusion trigger for homologous packed cells was a haemoglobin concentration <8 g L−1 in the whole postoperative period.

Statistical analysis

The t-test tested for differences between the groups. P < 0.05 was considered as significant.


Patient characteristics data are given in Table 1. The two groups did not differ for age, height, weight or gender. Likewise other variables, e.g. preoperative use of β-adrenoceptor receptor blocking drugs, patients who received sedation during surgery, the use of cement and a decline in blood pressure (>25% decrease in mean arterial pressure after cementation), showed no differences between groups. There was no difference in the duration of surgery in either group. Eight patients in the ibuprofen group and six in the placebo group terminated their participation in the trial because of adverse effects or severe pain (Table 2).

Table 1
Table 1:
Patient characteristics data.
Table 2
Table 2:
Reasons to stop the trial.

Perioperative blood loss

The volume of blood loss was significantly higher in patients pretreated with ibuprofen than with placebo. The volume of perioperative blood loss was 40% greater in the ibuprofen group (mean 700 mL) compared with the placebo group (mean 416 mL) (P < 0.01). The measured blood loss in the first 24 h after surgery also showed a 17% higher blood loss in the ibuprofen group compared with the placebo. This was not statistically different. The overall blood loss, i.e. the perioperative blood loss plus the blood loss in the first 24 h after surgery, showed an increase of 45% (P < 0.05) (Table 3). The study had an 86% power to demonstrate a 45% difference in expected blood loss at a P = 0.05 level of significance. The number of homologous blood transfusions was nine in the ibuprofen group and six in the placebo group (not significant) during the whole period the patients remained in the hospital.

Table 3
Table 3:
Blood loss during and after operation.


The main finding is that pretreatment with ibuprofen before total hip replacement surgery was associated with an increase in blood loss both during operation and for the first 24 h afterwards. Besides the useful anti-inflammatory, analgesic and antipyretic action of the NSAIDs, the study demonstrated an undesirable effect, namely increased blood loss.

The anti-inflammatory, analgesic and antipyretic action of NSAIDs are mediated through inhibition of prostaglandin synthesis by inhibiting cyclo-oxygenase (COX). COX is the major enzyme in the biosynthesis of prostanoids. Following the discovery in the early 1990s of an inducible isoform of COX, it is now known that COX exists in at least two isoforms: COX-1 and COX-2 [4,5]. COX-1 exists in the stomach, intestine, kidneys and blood platelets. It synthesizes the prostaglandins that (a) regulate the normal physiological processes involved in protecting the gastrointestinal mucosa and (b) maintain the renal function and vascular homeostasis. This role of COX-1 has been referred as a 'housekeeping' function. In contrast, the inducible isoform COX-2, after expression induced by several cytokines or lipopolysaccharide, produces large amounts of prostanoids that mainly contribute to the pathophysiological process of inflammation.

The therapeutic effects of NSAIDs are largely the result of inhibition of the enzyme COX-2, whereas the toxic effects (disturbing platelets, the gut and the kidney) are primarily due to the inhibition of COX-1. This leads to a lack of thromboxane synthesis and impaired platelet aggregation. It has been suggested that NSAIDs that selectively inhibit COX-2 have fewer side-effects [6-8]. The relationship between platelet aggregation, thromboxane production and serum concentrations of the non-COX-2 selective drug ibuprofen has been examined. A single dose of ibuprofen (300-800 mg) blocked platelet aggregation 2 h after administration. However, the effect was lost within 24 h [9]. After ibuprofen, 200, 400 and 800 mg, had been given to healthy volunteers, platelet aggregation was inhibited for 6, 8 and 11 h, respectively [10]. In the light of the half-life of ibuprofen (2 h), these data suggest that oral ibuprofen should be stopped 24 h before surgery.

Although we tried to reduce as much possible the confounding factors in the study (one type of surgery performed one orthopaedic surgeon, blinded medicaments), the use of acenocoumarol for prophylaxis against thromboembolism could affect the outcome of the study. Ibuprofen, but not the placebo, increases the prothrombin time. Other weaknesses of the study are the technique of measuring blood loss and the relatively high dropout rate. The study was probably not powerful enough to show whether an increase in blood loss resulted in an increased transfusion requirement or perioperative morbidity or mortality. These are much more important outcome measures for the patient compared with the actual measured blood loss. However, they are much more difficult to measure and therefore were not primary end-points of the study. It is concluded that ceasing NSAIDs sufficiently long before major orthopaedic surgery reduces perioperative blood loss. NSAIDs should be replaced 3 days before surgery with other analgesics, e.g. paracetamol, or possibly COX-2 selective anti-inflammatory agents, which have a better safety profile concerning peroperative blood loss.


1. Schafer A. Effects of nonsteroidal anti-inflammatory drugs on platelet function and systemic hemostasis. J Clin Pharmacol 1995; 35: 209-219.
2. Campbell WI, Kendrick R, Patterson C. Intravenous diclofenac sodium. Does its administration before operation suppress postoperative pain? Anaesthesia 1990; 45: 763-766.
3. Rorarius MG, Baer GA, Metsa-Ketela T, et al. Effects of peri-operatively administered diclofenac and indomethacin on blood loss, bleeding time and plasma prostanoids in man. Eur J Anaesthesiol 1989; 6: 335-342.
4. Vane JR, Botting RM. New insights into the mode of action of anti-inflammatory drugs. Inflamm Res 1995; 44: 1-10.
5. McKenna F. COX-2: separating myth from reality. Scand J Rheumatol 1999; 109: 19-29.
6. Pardo A, Garcia-Losa M, Fernandez-Pavon, et al. A placebo-controlled study of interaction between nabumetone and acenocoumarol. Br J Clin Pharmacol 1999; 47: 441-444.
7. Cryer B, Feldman M. Cyclooxygenase-1 and cyclooxygenase-2 selectivity of widely used nonsteroidal anti-inflammatory drugs. Am J Med 1998; 104: 413-421.
8. Engelhardt G, Bögel R, Schnitzler C, Utzmann R. Meloxicam: influence on arachidonic acid metabolism. Biochem Pharmacol 1996; 51: 29-38.
9. McIntyre BA, Philp RB. Effect of ibuprofen on platelet function in normal subjects and hemophiliac patients. Clin Pharmacol Ther 1978; 24: 616-621.
10. Cox SR, VanderLugt JT, Gumbleton TJ, Smith RB. Relationship between thromboxane production, platelet aggregability and serum concentrations of ibuprofen and flurbiprofen. Clin Pharmacol Ther 1987; 41: 510-521.


© 2002 European Academy of Anaesthesiology