Anaesthesia, 1990, Volume 45, pages 9 16-9 19
Platelet function after intramuscular diclofenac
I. POWER, W. A . CHAMBERS, 1. A. GREER, D. RAMAGE
AND
E. SIMON
Summary
A randomised double-blind controlled study was performed to examine the effect of diclofenac on skin bleeding time and in vitro whole blood platelet aggregation. Twenty thoracotomy patients were studied; I0 were given diclofenac 75 mg intramuscularly at induction of anaesthesia, and I0 formed a control group. Skin bleeding times and platelet aggregation tests were performed the day before and repeated one hour after induction of anaesthesia. Diclofenac prolonged skin bleeding time and reduced platelet aggregation. There were no significant changes in the control group. Key words
Analgesics; diclofenac. Blood; coagulation.
The use of nonsteroidal anti-inflammatory drugs (NSAIDs) to provide postoperative analgesia is becoming accepted practice. Intramuscular diclofenac was shown to be an effective analgesic,' but there is concern over the perioperative use of the NSAIDs since they have potential side effects related to their mechanism of action, inhibition of the enzyme cyclo-oxygenase.2 This enzyme is essential in platelets for the production of thromboxane A, which is an important mediator of platelet aggregation and vasoconstriction. These are processes that constitute the primary haemostatic response to vessel injury. Peri-operative administration of a NSAID could affect platelet function and inhibit haemostasis. It is clear that NSAIDs and aspirin, which has a similar inhibitory effect on cyclo-oxygenase, inhibit aggregation and prolong bleeding time in v ~ l u n t e e r s ,but ~ , ~there is little information on the peri-operative situation where the haemostatic response may be altered by the stress of surgery. It has recently become possible to study platelet aggregation in whole b l ~ o d . This ~ . ~ may be more physiological than traditional turbidometric methods using platelet-rich plasma, since the platelets are left in their natural milieu surrounded by red and white cells that can influence the aggregatory r e s p o n ~ e .To ~ . ~our knowledge there is no previous study which uses the whole blood aggregation technique to investigate the effect of NSAIDs on platelet aggregation in the peri-operative situation. The aim of this study
was to investigate the effect of intramuscular diclofenac on skin bleeding time and whole blood platelet aggregation in a randomised, double-blind, controlled, peri-operative study. Methods
Twenty ASA 1 and 2 patients aged 27 to 79 years undergoing thoracotomy were studied. The investigation was approved by the regional ethics committee and each subject provided written informed consent before enrolment. Exclusion criteria consisted of a history of peptic ulcer disease, bleeding tendency, asthma, allergies, recent aspirin or NSAID ingestion, and alcohol or narcotic abuse. The general anaesthetic technique was standardised (premedication, papaveretum and hyoscine; induction, thiopentone, suxamethonium; maintenance, nitrous oxide, oxygen, enflurane, fentanyl, alcuronium). Subcutaneous heparin (2500 IU subcutaneously, one hour before operation) was given as prophylaxis against venous thrombosis. All patients had intercostal nerve blocks before surgery (bupivacaine 0.5%). The subjects were randomly allocated so that 10 of them were given diclofenac 75 mg intramuscularly at induction of anaesthesia. The clinician who administered the injection was not otherwise involved in the study. Skin bleeding times (all subjects) and platelet function
I. Power, FCAnaes, Lecturer, D. Ramage, FCAnaes, E. Simon, MB, ChB, Registrars, University Department of Anaesthetics, Royal Infirmary, Edinburgh, W.A. Chambers, MD, FCAnaes, Consultant, The City Hospital, Edinburgh, I.A. Greer, MD, MRCP, MRCOG, Lecturer, University Department of Obstetrics and Gynaecology, Centre for Reproductive Biology, Edinburgh EH3 9EW. Accepted 29 March 1990. 0003-2409/90/ 1 10916 + 04 $03.00/0
Q 1990 The Association of Anaesthetists of Gt Britain and Ireland
916
917
Platelet function after intramuscular diclofenac Table 1. Patient demography and blood loss. Values are shown as mean (SEM).
Age; years Height; cm Weight; kg Smoker; Y/N Sex; M/F Blood loss; ml
Diclofenac
Control
P
61.2 (4.4) 163.1 (1.4) 62.5 (3.2) 416 416 352 (125)
61.1 (4.1) 164.9 (1.7) 67.7 (3.1) 218 416 300 (165)
0.99 0.43 0.25 0.63* 0.8
p. unpaired 1-test; *Chi-squared test. tests (10 subjects) were performed the day before surgery, and repeated one hour after induction of anaesthesia. Surgical blood loss was estimated by swab weight at operation. Skin bleeding times were performed in duplicate on the forearm by the modified method of Ivy','' using a Simplate I1 bleeding-time device to standardise the length and depth of incision. Bleeding time was taken for the two incisions and the mean time recorded. The normal bleeding time is up to 10 minutes with this technique.1° Platelet aggregation was studied in v i m in five subjects in each group using a Clay-Adams Ultra-Flo whole blood platelet counter."," Collagen (2 pg/ml, Semmelweiss) was used as an aggregating agent. Whole venous blood was anticoagulated with 3.8% trisodium citrate. A red cell count was first determined from each sample and dialled into the whole blood platelet counter. A 10-pl aliquot was withdrawn and the baseline platelet count determined. The aggregating agent was then added and the sample stirred at a constant 1000 rpm and 37°C. Further aliquots were taken at 1, 3, and 5 minutes and the platelet count determined. Platelet aggregation was recorded as the percentage decrease in the platelet count from baseline to 1, 3, and 5 minutes. Twenty patients entered and completed the trial. No adverse events were encountered. The randomisation code was broken after the end of the study and the results obtained. Values are shown as mean (SEM). Subject demography and surgical blood loss were compared between the diclofenac and control groups using unpaired t-tests and Chi-squared tests. Before and after dose bleeding times and platelet aggregation were compared using paired t-tests. A p value of 0.05 or less was accepted as indicative of a significant difference, and all the tests were two-tailed. Ninety-five percent confidence intervals were calculated where appropriate (95% CI).
Surgical procedures were lobectomy (five diclofenac; four control), pneumonectomy (two diclofenac; two control), open lung biopsy (two in each group), pleurectomy (one diclofenac), and thoracotomy with no resection (two control). Blood loss did not differ between the groups (p = 0.80). There was a wide range of blood loss within each group (diclofenac 20 to 1200 ml; control 20 to 1700 ml) . Skin bleeding time (Table 2) was significantly prolonged one hour after diclofenac (p < 0.0001, 95% CI 47 to 115 seconds). There was no change in the control group (p = 0.66, 95% CI -24 to 16 seconds). The range of bleeding
1 0
I
I
I
. 2
I
I
I
3
4
5
Time (minutes)
Fig. 1. Diclofenac group; five subjects. Collagen-induced platelet and one hour after (- -0--) induction aggregation before (-o-) of anaesthesia. Mean (SEM).
'"OK 90
Results There was no difference between the groups in age, height, weight, smoking history, or sex distribution (Table 1). Table 2. Skin bleeding time. Bleeding times in seconds, mean
(SEM), are shown for each group of 10 subjects before and one hour after induction of anaesthesia. Before induction Diclofenac Control p, paired t-test.
250 (16.9) 226 (18.7)
After induction 330 (19.7) 222 (15.5)
P 0.0001 0.66
0
I
2
3
4
5
Time (minutes)
Fig. 2. Control group; five subjects. Collagen-induced platelet
aggregation before (-o-) and one hour after (- -0--) induction of anaesthesia. Mean (SEM).
918
I . Power et al.
Table 3. Diclofenac group platelet aggregation in five subjects. Values shown are the percentage decreases from baseline, mean (SEM), in platelet counts over 5 minutes in the presence of a collagen aggregator.
1 minute 3 minutes 5 minutes
Before induction
After induction
P
66.4 (11.5) 82.6 (6.1) 82.6 (5.2)
51.8 (14.4) 73.8 (6.0) 64.0 (8.6)
0.077 0.024 0.038
p, paired i-test. Table 4. Control group platelet aggregation in five subjects. Values shown are the percentage decreases from baseline, mean (SEM), in platelet counts over 5 minutes in the presence of a collagen aggregator.
1 minute 3 minutes 5 minutes
Before induction
After induction
P
70.2 (15.8) 85.8 (3.2) 83.8 (3.8)
60.2 (11) 81.6 (4.4) 78 (4.3)
0.38 0.1 0.09
p, paired t-test
times after diclofenac was from 247 to 480 seconds. The longest prolongation of bleeding time in a subject was from 350 seconds before diclofenac t o 480 seconds after. Figures 1 and 2 show the platelet counts in the I-, 3-, and 5-minute sample for the diclofenac and control groups. Each figure shows the decrease in platelet count before and after injection of diclofenac (or control). Tables 3 and 4 show the change in platelet aggregation after diclofenac and the control. Diclofenac significantly reduced in vitro platelet aggregation at the 3- and 5-minute points, but not at one minute (Table 3). The 95% CI for the reduction in platelet aggregation with diclofenac at 5 minutes was from -35.5 to - 1.7%. The control group showed a tendency towards reduced aggregation one hour after induction of anaesthesia (Fig. 2 , Table 4), but this was not statistically significant (e.g. at five minutes p = 0.09, 95% CI - 13 t o 1.4%).
seconds). The magnitude of this effect is similar to that seen with aspirin and may not be of clinical significance,' since none of the bleeding times after diclofenac were above the normal upper limit of 600 seconds. Therefore, although diclofenac does affect platelets, it may not produce a n ' abnormal haemostatic state in previously normal individuals. However, it was shown that certain members of the population are more sensitive to the haemostatic effects of aspirin than othersI0 and so some patients could have a greater effect from diclofenac than seen in this study. The tendency towards reduced platelet aggregation in the control group was statistically insignificant, but since it could reflect the effects of anaesthesia and surgery on platelet function it may merit further study. The effect of intramuscular diclofenac 75 mg on surgical blood loss is unclear. This investigation was not designed to address that problem. Previous studies have found that pre-operative administration of lower doses of diclofenac do not increase blood loss after gynaecological laparotomy (33.6 mg intravenously)I2 o r transurethral prostatectomy (50 mg orally).I3 Experience with other NSAlDs suggests that the dose used is important in determining whether blood loss is increased or not. For example, in gynaecological laparotomy pre-operative intravenous indomethacin 0.8 mg/kg increased intra-operative blood loss,14but 25 mg did not.I2 The clinically used intramuscular dose of diclofenac is 75 mg. Further studies are required to ascertain whether the platelet dysfunction produced by diclofenac 75 mg given intra-operatively is associated with increased blood loss. In conclusion, diclofenac 75 mg intramuscularly does impair platelet function within one hour. This should be considered if blood loss seems excessive when diclofenac has been given intra-operatively. In addition it may be prudent to avoid the use of diclofenac or another NSAID in the presence of other defects of haemostasis o r coagulation. Indeed, the administration of a NSAID could reveal a subclinical haemostatic problem, and if this was suspected the skin bleeding time would be a useful clinical test to ~erf0rm.l~
References Discussion This study has demonstrated that one hour after an intramuscular dose of diclofenac 75 mg skin bleeding time is prolonged and whole blood platelet aggregation is inhibited. Skin bleeding time is a useful clinical test of platelet function, and collagen-induced aggregation depends largely on thromboxane production by platelet cyclo-oxygenase. Previous work found no change in bleeding time in patients 24 hours after an intramuscular dose of diclofenac 33.6 mg followed by an intravenous infusion of 6.7 mg/ hour." This was a surprising observation in view of the known effect of cyclo-oxygenase inhibitors o n thromboxane production. Our work has shown that the clinically used dose of diclofenac 75 mg does affect skin bleeding time within one hour of intramuscular administration in patients having surgery. The increase of bleeding time seen with diclofenac was statistically significant but modest (95% CI 47 to 115
1. BUCHANAN JM, BALDASERA J, PWLE PH, HALSHAWJ,
2. 3.
4.
5.
6.
7.
DALLARDJK. Postoperative pain relief; a new approach: narcotics compared with non-steroidal anti-inflammatory drugs. Annals of the Royal College of Surgeons of England 1988; 7 0 332-5. VANE JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature 1971; 231: 232-5. WEISSHJ, ALEDORT LM, KOCHWAS. The effect of salicylates on the hemostatic properties of platelets in man. Journal of Clinical Investigation 1968; 47: 2169-80. RORARIUS M, MIRALLES J, BAERGA, PALOMAKI E. Diclofenac versus indomethacin given as intravenous infusions: their effect on haemodynamics and bleeding time, and side-effects in healthy subjects. Annals of Clinical Research 1985; 17: 306-9. LUMLEYP, HUMPHREY PPA. A method for quantitating platelet aggregation and analyzing drug-receptor interactions on platelets in whole blood in vitro. Journal of Pharmacological Methods 1981; 6: 153-66. Fox SC, BURGESS-WILSON M, HEPTINSTALL S, MITCHELL JRA. Platelet aggregation in whole blood determined using the Ultra-Flo 100 Platelet Counter. Thrombosis and Haernosiasis 1982; 48: 327-9. SANIABADI AR, LOWEGDO, BARBENEL JC, FORBESCD. A
Platelet function after intramuscular diclofenac comparison of spontaneous platelet aggregation in whole blood with platelet rich plasma: additional evidence for the role of ADP. Thrombosis and Haemostasis 1984; 51: 115-8. 8. BLACKWELL GJ, FLOWER RJ, RUSSELL-SMITH N, SALMON JA, THOROGOOD PB, VANEJR. Prostacyclin is produced in whole blood. British Journal of Pharmacology 1978; 64: 436P. G. The standardization of certain 9. IVYAC, NELSOND, BUCHER factors in the cutaneous venostasis bleeding time technique. Journal of Laboratory and Clinical Medicine 1941; 2 6 1812-22. 10. MIELKECH, KANESHIRO MM, MAHERIA, WEINERJM, SI. The standardized normal Ivy bleeding time and RAPAPORT its prolongation by aspirin. Blood 1969; 3 4 20415. JJ. CALDER AA, FORBES CD. Inhibition of 11. GREERIA, WALKER platelet aggregation in whole blood by adrenoreceptor antagonists. Thrombosis Research 1985; 4 0 63 1 4 3 . 12. RORARIUS MGF, BAERGA, METSA-KETELA T, MIRALLES J,
919
PAMMAKIE, VAPAATALO H. Effects of peri-operatively administered diclofenac and indomethacin on blood loss, bleeding time and plasma prostanoids in man. European Journal of Anaesthesiology 1989; 6 33542. 13. BRICKERSRW, SAVAGEME, HANNINCCD. Peri-operative blood loss and non-steroidal anti-inflammatory drugs: an investigation using diclofenac in patients undergoing transurethral resection of the prostate. European Journal of Anaesthesiology 1987; 4 429-34. 14. ENGELC, LUND B, KRISTENSEN SS, AXELC, NIELSENJB. Indomethacin as an analgesic after hysterectomy. Acta Anaesthesiologica Scandinavica 1989; 3 3 498-50 I . 15. STUARTMJ, MILLERML, DAVEYFR, WOLKJA. The postaspirin bleeding time: a screening test for evaluating haemostatic disorders. British Journal of Haematology 1979; 4 3 649-59.
Platelet function after intramuscular diclofenac
I. POWER, W. A . CHAMBERS, 1. A. GREER, D. RAMAGE
AND
E. SIMON
Summary
A randomised double-blind controlled study was performed to examine the effect of diclofenac on skin bleeding time and in vitro whole blood platelet aggregation. Twenty thoracotomy patients were studied; I0 were given diclofenac 75 mg intramuscularly at induction of anaesthesia, and I0 formed a control group. Skin bleeding times and platelet aggregation tests were performed the day before and repeated one hour after induction of anaesthesia. Diclofenac prolonged skin bleeding time and reduced platelet aggregation. There were no significant changes in the control group. Key words
Analgesics; diclofenac. Blood; coagulation.
The use of nonsteroidal anti-inflammatory drugs (NSAIDs) to provide postoperative analgesia is becoming accepted practice. Intramuscular diclofenac was shown to be an effective analgesic,' but there is concern over the perioperative use of the NSAIDs since they have potential side effects related to their mechanism of action, inhibition of the enzyme cyclo-oxygenase.2 This enzyme is essential in platelets for the production of thromboxane A, which is an important mediator of platelet aggregation and vasoconstriction. These are processes that constitute the primary haemostatic response to vessel injury. Peri-operative administration of a NSAID could affect platelet function and inhibit haemostasis. It is clear that NSAIDs and aspirin, which has a similar inhibitory effect on cyclo-oxygenase, inhibit aggregation and prolong bleeding time in v ~ l u n t e e r s ,but ~ , ~there is little information on the peri-operative situation where the haemostatic response may be altered by the stress of surgery. It has recently become possible to study platelet aggregation in whole b l ~ o d . This ~ . ~ may be more physiological than traditional turbidometric methods using platelet-rich plasma, since the platelets are left in their natural milieu surrounded by red and white cells that can influence the aggregatory r e s p o n ~ e .To ~ . ~our knowledge there is no previous study which uses the whole blood aggregation technique to investigate the effect of NSAIDs on platelet aggregation in the peri-operative situation. The aim of this study
was to investigate the effect of intramuscular diclofenac on skin bleeding time and whole blood platelet aggregation in a randomised, double-blind, controlled, peri-operative study. Methods
Twenty ASA 1 and 2 patients aged 27 to 79 years undergoing thoracotomy were studied. The investigation was approved by the regional ethics committee and each subject provided written informed consent before enrolment. Exclusion criteria consisted of a history of peptic ulcer disease, bleeding tendency, asthma, allergies, recent aspirin or NSAID ingestion, and alcohol or narcotic abuse. The general anaesthetic technique was standardised (premedication, papaveretum and hyoscine; induction, thiopentone, suxamethonium; maintenance, nitrous oxide, oxygen, enflurane, fentanyl, alcuronium). Subcutaneous heparin (2500 IU subcutaneously, one hour before operation) was given as prophylaxis against venous thrombosis. All patients had intercostal nerve blocks before surgery (bupivacaine 0.5%). The subjects were randomly allocated so that 10 of them were given diclofenac 75 mg intramuscularly at induction of anaesthesia. The clinician who administered the injection was not otherwise involved in the study. Skin bleeding times (all subjects) and platelet function
I. Power, FCAnaes, Lecturer, D. Ramage, FCAnaes, E. Simon, MB, ChB, Registrars, University Department of Anaesthetics, Royal Infirmary, Edinburgh, W.A. Chambers, MD, FCAnaes, Consultant, The City Hospital, Edinburgh, I.A. Greer, MD, MRCP, MRCOG, Lecturer, University Department of Obstetrics and Gynaecology, Centre for Reproductive Biology, Edinburgh EH3 9EW. Accepted 29 March 1990. 0003-2409/90/ 1 10916 + 04 $03.00/0
Q 1990 The Association of Anaesthetists of Gt Britain and Ireland
916
917
Platelet function after intramuscular diclofenac Table 1. Patient demography and blood loss. Values are shown as mean (SEM).
Age; years Height; cm Weight; kg Smoker; Y/N Sex; M/F Blood loss; ml
Diclofenac
Control
P
61.2 (4.4) 163.1 (1.4) 62.5 (3.2) 416 416 352 (125)
61.1 (4.1) 164.9 (1.7) 67.7 (3.1) 218 416 300 (165)
0.99 0.43 0.25 0.63* 0.8
p. unpaired 1-test; *Chi-squared test. tests (10 subjects) were performed the day before surgery, and repeated one hour after induction of anaesthesia. Surgical blood loss was estimated by swab weight at operation. Skin bleeding times were performed in duplicate on the forearm by the modified method of Ivy','' using a Simplate I1 bleeding-time device to standardise the length and depth of incision. Bleeding time was taken for the two incisions and the mean time recorded. The normal bleeding time is up to 10 minutes with this technique.1° Platelet aggregation was studied in v i m in five subjects in each group using a Clay-Adams Ultra-Flo whole blood platelet counter."," Collagen (2 pg/ml, Semmelweiss) was used as an aggregating agent. Whole venous blood was anticoagulated with 3.8% trisodium citrate. A red cell count was first determined from each sample and dialled into the whole blood platelet counter. A 10-pl aliquot was withdrawn and the baseline platelet count determined. The aggregating agent was then added and the sample stirred at a constant 1000 rpm and 37°C. Further aliquots were taken at 1, 3, and 5 minutes and the platelet count determined. Platelet aggregation was recorded as the percentage decrease in the platelet count from baseline to 1, 3, and 5 minutes. Twenty patients entered and completed the trial. No adverse events were encountered. The randomisation code was broken after the end of the study and the results obtained. Values are shown as mean (SEM). Subject demography and surgical blood loss were compared between the diclofenac and control groups using unpaired t-tests and Chi-squared tests. Before and after dose bleeding times and platelet aggregation were compared using paired t-tests. A p value of 0.05 or less was accepted as indicative of a significant difference, and all the tests were two-tailed. Ninety-five percent confidence intervals were calculated where appropriate (95% CI).
Surgical procedures were lobectomy (five diclofenac; four control), pneumonectomy (two diclofenac; two control), open lung biopsy (two in each group), pleurectomy (one diclofenac), and thoracotomy with no resection (two control). Blood loss did not differ between the groups (p = 0.80). There was a wide range of blood loss within each group (diclofenac 20 to 1200 ml; control 20 to 1700 ml) . Skin bleeding time (Table 2) was significantly prolonged one hour after diclofenac (p < 0.0001, 95% CI 47 to 115 seconds). There was no change in the control group (p = 0.66, 95% CI -24 to 16 seconds). The range of bleeding
1 0
I
I
I
. 2
I
I
I
3
4
5
Time (minutes)
Fig. 1. Diclofenac group; five subjects. Collagen-induced platelet and one hour after (- -0--) induction aggregation before (-o-) of anaesthesia. Mean (SEM).
'"OK 90
Results There was no difference between the groups in age, height, weight, smoking history, or sex distribution (Table 1). Table 2. Skin bleeding time. Bleeding times in seconds, mean
(SEM), are shown for each group of 10 subjects before and one hour after induction of anaesthesia. Before induction Diclofenac Control p, paired t-test.
250 (16.9) 226 (18.7)
After induction 330 (19.7) 222 (15.5)
P 0.0001 0.66
0
I
2
3
4
5
Time (minutes)
Fig. 2. Control group; five subjects. Collagen-induced platelet
aggregation before (-o-) and one hour after (- -0--) induction of anaesthesia. Mean (SEM).
918
I . Power et al.
Table 3. Diclofenac group platelet aggregation in five subjects. Values shown are the percentage decreases from baseline, mean (SEM), in platelet counts over 5 minutes in the presence of a collagen aggregator.
1 minute 3 minutes 5 minutes
Before induction
After induction
P
66.4 (11.5) 82.6 (6.1) 82.6 (5.2)
51.8 (14.4) 73.8 (6.0) 64.0 (8.6)
0.077 0.024 0.038
p, paired i-test. Table 4. Control group platelet aggregation in five subjects. Values shown are the percentage decreases from baseline, mean (SEM), in platelet counts over 5 minutes in the presence of a collagen aggregator.
1 minute 3 minutes 5 minutes
Before induction
After induction
P
70.2 (15.8) 85.8 (3.2) 83.8 (3.8)
60.2 (11) 81.6 (4.4) 78 (4.3)
0.38 0.1 0.09
p, paired t-test
times after diclofenac was from 247 to 480 seconds. The longest prolongation of bleeding time in a subject was from 350 seconds before diclofenac t o 480 seconds after. Figures 1 and 2 show the platelet counts in the I-, 3-, and 5-minute sample for the diclofenac and control groups. Each figure shows the decrease in platelet count before and after injection of diclofenac (or control). Tables 3 and 4 show the change in platelet aggregation after diclofenac and the control. Diclofenac significantly reduced in vitro platelet aggregation at the 3- and 5-minute points, but not at one minute (Table 3). The 95% CI for the reduction in platelet aggregation with diclofenac at 5 minutes was from -35.5 to - 1.7%. The control group showed a tendency towards reduced aggregation one hour after induction of anaesthesia (Fig. 2 , Table 4), but this was not statistically significant (e.g. at five minutes p = 0.09, 95% CI - 13 t o 1.4%).
seconds). The magnitude of this effect is similar to that seen with aspirin and may not be of clinical significance,' since none of the bleeding times after diclofenac were above the normal upper limit of 600 seconds. Therefore, although diclofenac does affect platelets, it may not produce a n ' abnormal haemostatic state in previously normal individuals. However, it was shown that certain members of the population are more sensitive to the haemostatic effects of aspirin than othersI0 and so some patients could have a greater effect from diclofenac than seen in this study. The tendency towards reduced platelet aggregation in the control group was statistically insignificant, but since it could reflect the effects of anaesthesia and surgery on platelet function it may merit further study. The effect of intramuscular diclofenac 75 mg on surgical blood loss is unclear. This investigation was not designed to address that problem. Previous studies have found that pre-operative administration of lower doses of diclofenac do not increase blood loss after gynaecological laparotomy (33.6 mg intravenously)I2 o r transurethral prostatectomy (50 mg orally).I3 Experience with other NSAlDs suggests that the dose used is important in determining whether blood loss is increased or not. For example, in gynaecological laparotomy pre-operative intravenous indomethacin 0.8 mg/kg increased intra-operative blood loss,14but 25 mg did not.I2 The clinically used intramuscular dose of diclofenac is 75 mg. Further studies are required to ascertain whether the platelet dysfunction produced by diclofenac 75 mg given intra-operatively is associated with increased blood loss. In conclusion, diclofenac 75 mg intramuscularly does impair platelet function within one hour. This should be considered if blood loss seems excessive when diclofenac has been given intra-operatively. In addition it may be prudent to avoid the use of diclofenac or another NSAID in the presence of other defects of haemostasis o r coagulation. Indeed, the administration of a NSAID could reveal a subclinical haemostatic problem, and if this was suspected the skin bleeding time would be a useful clinical test to ~erf0rm.l~
References Discussion This study has demonstrated that one hour after an intramuscular dose of diclofenac 75 mg skin bleeding time is prolonged and whole blood platelet aggregation is inhibited. Skin bleeding time is a useful clinical test of platelet function, and collagen-induced aggregation depends largely on thromboxane production by platelet cyclo-oxygenase. Previous work found no change in bleeding time in patients 24 hours after an intramuscular dose of diclofenac 33.6 mg followed by an intravenous infusion of 6.7 mg/ hour." This was a surprising observation in view of the known effect of cyclo-oxygenase inhibitors o n thromboxane production. Our work has shown that the clinically used dose of diclofenac 75 mg does affect skin bleeding time within one hour of intramuscular administration in patients having surgery. The increase of bleeding time seen with diclofenac was statistically significant but modest (95% CI 47 to 115
1. BUCHANAN JM, BALDASERA J, PWLE PH, HALSHAWJ,
2. 3.
4.
5.
6.
7.
DALLARDJK. Postoperative pain relief; a new approach: narcotics compared with non-steroidal anti-inflammatory drugs. Annals of the Royal College of Surgeons of England 1988; 7 0 332-5. VANE JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature 1971; 231: 232-5. WEISSHJ, ALEDORT LM, KOCHWAS. The effect of salicylates on the hemostatic properties of platelets in man. Journal of Clinical Investigation 1968; 47: 2169-80. RORARIUS M, MIRALLES J, BAERGA, PALOMAKI E. Diclofenac versus indomethacin given as intravenous infusions: their effect on haemodynamics and bleeding time, and side-effects in healthy subjects. Annals of Clinical Research 1985; 17: 306-9. LUMLEYP, HUMPHREY PPA. A method for quantitating platelet aggregation and analyzing drug-receptor interactions on platelets in whole blood in vitro. Journal of Pharmacological Methods 1981; 6: 153-66. Fox SC, BURGESS-WILSON M, HEPTINSTALL S, MITCHELL JRA. Platelet aggregation in whole blood determined using the Ultra-Flo 100 Platelet Counter. Thrombosis and Haernosiasis 1982; 48: 327-9. SANIABADI AR, LOWEGDO, BARBENEL JC, FORBESCD. A
Platelet function after intramuscular diclofenac comparison of spontaneous platelet aggregation in whole blood with platelet rich plasma: additional evidence for the role of ADP. Thrombosis and Haemostasis 1984; 51: 115-8. 8. BLACKWELL GJ, FLOWER RJ, RUSSELL-SMITH N, SALMON JA, THOROGOOD PB, VANEJR. Prostacyclin is produced in whole blood. British Journal of Pharmacology 1978; 64: 436P. G. The standardization of certain 9. IVYAC, NELSOND, BUCHER factors in the cutaneous venostasis bleeding time technique. Journal of Laboratory and Clinical Medicine 1941; 2 6 1812-22. 10. MIELKECH, KANESHIRO MM, MAHERIA, WEINERJM, SI. The standardized normal Ivy bleeding time and RAPAPORT its prolongation by aspirin. Blood 1969; 3 4 20415. JJ. CALDER AA, FORBES CD. Inhibition of 11. GREERIA, WALKER platelet aggregation in whole blood by adrenoreceptor antagonists. Thrombosis Research 1985; 4 0 63 1 4 3 . 12. RORARIUS MGF, BAERGA, METSA-KETELA T, MIRALLES J,
919
PAMMAKIE, VAPAATALO H. Effects of peri-operatively administered diclofenac and indomethacin on blood loss, bleeding time and plasma prostanoids in man. European Journal of Anaesthesiology 1989; 6 33542. 13. BRICKERSRW, SAVAGEME, HANNINCCD. Peri-operative blood loss and non-steroidal anti-inflammatory drugs: an investigation using diclofenac in patients undergoing transurethral resection of the prostate. European Journal of Anaesthesiology 1987; 4 429-34. 14. ENGELC, LUND B, KRISTENSEN SS, AXELC, NIELSENJB. Indomethacin as an analgesic after hysterectomy. Acta Anaesthesiologica Scandinavica 1989; 3 3 498-50 I . 15. STUARTMJ, MILLERML, DAVEYFR, WOLKJA. The postaspirin bleeding time: a screening test for evaluating haemostatic disorders. British Journal of Haematology 1979; 4 3 649-59.
