The Effect of IM Ketorolac Tromethamine on
Bleeding Time: A Prospective, Interventional,
ADAM J. SINGER, MD, CHRISTOPHER J. MYNSTER, MD, AND BRIAN J. MCMAHON, MD
Opiates, although effective analgesics, have significant adverse side
effects. Ketorolac, the only parental nonsteroidal antiinflammatory drug
available for use in the United States does not cause significant respi-
ratory depression or hypotension, but it is a reversible inhibitor of plate-
let aggregation with a theoretical increased bleeding risk, which limits its
use. The objective of this study was to determine the effect of a single
intramuscular dose of 60 mg ketorolac on 4-hour bleeding times in
healthy volunteers. This was a prospective, paired, unblinded, before-
and-after interventional study performed in a suburban university-based
EM residency training program. Subjects were 20 healthy volunteer EM
residents. Standard Ivy bleeding times were measured before and 4
hours after intramuscular administration of 60 mg ketorolac. Before-and-
after bleeding times were compared using a paired t-test. The study had
90% power to detect an effect size of 0.5. The subjects’ mean age was
31.6 and 7 (35%) were females. Bleeding time was increased from a mean
baseline time of 3 minutes 34 seconds (ⴞ 1 min 20 sec) to a mean 4-hour
postinjection time of 5 minutes 20 seconds (ⴞ 3 min 8 sec). The mean
prolongation of bleeding time was 1 minute 46 seconds (50% increase
with 95% confidence interval, 25%-75%). There were no adverse events.
A standard intramuscular dose of 60 mg ketorolac resulted in prolonga-
tion of the bleeding time in healthy volunteers. The clinical significance
of this prolongation in patients is unclear. (Am J Emerg Med 2003;21:
441-443. © 2003 Elsevier Inc. All rights reserved.)
The opioids have long provided the gold standard for
analgesia in many circumstances, including trauma, postop-
erative pain, painful crises of sickle cell anemia, nephroli-
thiasis, and biliary colic. However, the parenteral opiates are
associated with multiple adverse effects such as sedation,
hypotension, respiratory depression, emesis, and paralytic
ileus. This has led to a search for safer alternative analgesic
agents such as the nonsteroidal antiinflammatory agents.1-3
Marketed in 1990, ketorolac is the only parenteral nonste-
roidal antiinflammatory drug (NSAID) currently available
in the United States.
The use of ketorolac is frequently limited by concern over
the reversible inhibition of platelet aggregation, which the-
oretically predisposes patients to abnormal bleeding.4 This
concern has led many surgeons to avoid its use before any
From the Department of Emergency Medicine, Stony Brook Uni-
versity Hospital, Stony Brook, New York.
Manuscript received January 5, 2003; accepted January 5, 2003.
Presented at the Annual Meeting of the Society for Academic
Emergency Medicine, May 2001, Atlanta, GA.
Address reprint requests to Adam J. Singer, MD, Department of
Emergency Medicine, Stony Brook University Hospital, UH-L4-515,
Stony Brook, NY 11794-7400. Email: adam.singer@sunysb.edu
Key Words: Ketorolac, bleeding time, nonsteroidal antiinflamma-
tory agents.
© 2003 Elsevier Inc. All rights reserved.
0735-6757/03/2105-0007$30.00/0
doi:10.1016/S0735-6757(03)00100-1
Controlled Study
anticipated surgery in which operative and postoperative
bleeding might pose a risk to the patient. This concern has
been further supported by studies performed in the operat-
ing room demonstrating increased operative and postoper-
ative bleeding in patients receiving ketoralac.5,6 A prior
small study limited to 10 females suggested that a single
dose of intramuscular ketorolac prolongs bleeding time 3
hours after administration.7
The purpose of the current study was to evaluate the
effect of a single intramuscular injection of Ketorolac on
bleeding times, an accepted measure of platelet function, in
healthy adult male and female subjects. Our null hypothesis
was that ketorolac would not significantly alter bleeding
time.
METHODS
Study Design
A prospective paired, unblinded interventional trial study
design in which each test subject served as his or her own
control was used. The project was approved by the institu-
tional review board. Written informed consent was obtained
from all study participants.
Population and Setting
This study was conducted in the ED at the State Univer-
sity of New York at Stony Brook. Healthy adult volunteers
were recruited from among EM residents. Pregnant sub-
jects; those with a history of bleeding disorders; recent
NSAID, aspirin, alcohol, or anticoagulant use; and those
allergic to NSAIDs were excluded.
Study Protocol
Subjects had their baseline bleeding time measured ac-
cording to the method originally described by Ivy.8 The
volar aspect of both forearms was prepped with isopropyl
alcohol 70% and allowed to dry. A standard small superfi-
cial laceration (1 ⫻ 5 mm) was made on one of the forearms
distal to the antecubital fossa (the side was randomly se-
lected) using a special spring-operated device (Surgicutt,
International Technidyne Corp, Edison, NJ). The bleeding
wound was gently dabbed with filter (Bleeding Time Blot-
ting Paper, International Technidyne Corp) every 15 to 30
seconds until hemostasis was obtained. The bleeding time
was measured starting from creation of the laceration and
ending when hemostasis was obtained. Subjects then re-
ceived an intramuscular injection of 60 mg ketorolac. Four
hours later, their bleeding times were again determined with
this method using the contralateral forearm for the wound
site.
441
442 AMERICAN JOURNAL OF EMERGENCY MEDICINE ■ Volume 21, Number 5 ■ September 2003
FIGURE 1. Individual baseline and post-Ketorolac bleeding
times.
Outcomes
The primary outcome in this study was the absolute and
relative increase in the Ivy bleeding time 4 hours after
intramuscular ketorolac administration. A secondary out-
come was the presence of adverse events such as allergic
reactions and hematoma formation.
Data Analysis
Data was entered into SPSS for Windows 10.1 (SPSS,
Inc, Chicago, IL) for statistical analysis. Continuous data
are presented as means with 95% confidence intervals (CIs).
Comparisons of bleeding times before and after ketorolac
administration were performed using a paired t-test using
subjects as their own controls. Age and gender were ad-
justed for using analysis of covariance (ANCOVA). This
study had 90% power to detect an effect size of 0.5 on
bleeding time.9
RESULTS
Twenty subjects were enrolled in the study. Their mean
age was 31.6 ⫾ 6.6; 7 subjects (35%) were females. The
mean interval between bleeding time measurements was 4
hours 8 minutes (range, 225-275 min). The mean baseline
bleeding time was 3 minutes 34 seconds (95% CI, 177-252
sec). The mean bleeding time after ketorolac administration
was 5 minutes 20 seconds (95% CI, 236-405 sec).
The mean prolongation of bleeding time after Ketorolac
administration was 1 minute 46 seconds (95% CI, 47-165
sec). This corresponds to a relative increase of 50% (95%
CI, 25%-75%) in the bleeding time. There was no difference
in bleeding time prolongation between males and females.
None of the subjects developed any adverse events. Indi-
vidual values for bleeding times before and after adminis-
tration of ketorolac are presented in Figure 1.
DISCUSSION
Although ketorolac has a favorable analgesic profile,
inhibition of cyclooxygenase activity by ketorolac with
concomitant reduction in thromboxane A2 levels causes
reversible inhibition of platelet aggregation,2 which could
predispose patients to an increased risk of bleeding, partic-
ularly in the setting of trauma and in postoperative patients.
Ketorolac has also been shown to increase the rate of
perioperative hemorrhage in the tonsillectomy population.
A chart review of 169 patients undergoing tonsillectomy
who were administered ketorolac revealed a postoperative
hemorrhage rate of 10.1% in comparison with a rate of 2.2%
in those given opioids.6 This complication has not been
reported after other surgical procedures, including dental
surgery. Additionally, a large postmarketing surveillance
study by Strom et al. concluded that there was only a slight
increase in overall operative-site bleeding limited to elderly
patients and those taking high doses of ketorolac.10 Further-
more, in this study there was no increases in the risk of
clinically significant operative-site bleeding episodes asso-
ciated with use of ketorolac.
Although it has been suggested that bleeding time is the
most reliable indicator of abnormal bleeding in patients
receiving antiplatelet agents, there is little evidence to sug-
gest that increased bleeding time can predict hemostatic
compromise; for example, studies have failed to show a
consistent correlation between aspirin-induced prolongation
of bleeding time and surgical blood loss.11 For example, the
“postaspirin” bleeding time is not a reliable indicator of
normal platelet function.12
The results of the current study confirm that a single
intramuscular 60-mg dose of ketorolac significantly pro-
longs bleeding time in healthy subjects. This is evidenced
by a nearly 2 minute or 50% prolongation in the standard
bleeding time. This prolongation of bleeding time is slightly
shorter yet similar to the 3-minute 12-second prolongation
reported by Dordoni et al.7 However, it is unclear whether
these results can be generalized to sick patients. More
importantly, it is not clear what the clinical relevance of a
2-minute prolongation in bleeding time would be. Although
increased bleeding has been noted in some patients with
prolonged bleeding times,13 bleeding time has not consis-
tently been found to predict the risk of perioperative bleed-
ing.14
Study Limitations and Future Questions
Our study subjects were healthy adult volunteers. It is not
clear whether the effects of ketorolac on bleeding times
would be similar in other age groups or in patients with
various illnesses. Second, our study looked at bleeding
times. As mentioned earlier, it is unclear how this outcome
correlates with operative bleeding or other bleeding com-
plications. Finally, we only assessed bleeding times 4 hours
after ketorolac administration. We chose this time interval
because many of our ED patients with surgical conditions
requiring emergent surgery are operated on within this time
period. It is unclear what the bleeding times would be for
other shorter or longer time intervals.
CONCLUSIONS
Our study suggests that a single intramuscular injection
of 60 mg ketorolac prolongs Ivy bleeding times in healthy
subjects. This finding needs to be validated in a larger
SINGER ET AL ■ KETOROLAC AND BLEEDING TIMES
population that includes ill patients. The clinical signifi-
cance of this finding remains to be determined.
REFERENCES
1. Lieh-Lai MW, Kauffman RE, Uy HG, et al: A randomized com-
parison of Ketorolac tromethamine and morphine for postoperative
analgesia in critically ill children. Crit Care Med 1999;27:2786-2791
2. Buckley MMT, Brogden RN: Ketorolac: a review of its phar-
macodynamic and pharmacokinetic properties, and therapeutic po-
tential. Drugs 1990;39:86-109
3. Mcardle P: Intravenous analgesia. Crit Care Clin 1999;15:89-104
4. Greer IA: Effects of Ketorolac tromethamine on hemostasis.
Pharmacotherapy 1990;10:71S-76S
5. Reinhart DI: Minimizing the adverse effects of Ketorolac. Drug
Safety 2000;22:487-497
6. Gallagher JE, Blauth J, Fornadley JA: Perioperative Ketorolac
tromethamine and postoperative hemorrhage in cases of tonsillec-
tomy and adenoidectomy. Laryngoscope 1995;105:606-609
443
7. Dordoni P, Ventura MD, Stefanelli A, et al: Effect of Ketorolac,
ketoprofen and nefopam on platelet function. Aneasthesia 1994;49:
1046-1049
8. Ivy AC, Nelson D, Buchet G: The standardization of certain
factors in the cutaneous “venostasis” bleeding time technique.
J Lab Clin Med 1941;26:1812-1822
9. Cohen J: Statistical Power Analysis for the Behavioral Sci-
ences. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988
10. Strom BL, Berlin JA, Kinman PW, et al: A postmarketing
surveillance study. Parenteral Ketorolac and risk of gastrointes-
tinal bleeding and operative site bleeding. JAMA 1996;275:376-
382
11. Rodgers RPC, Levin J: A critical reappraisal of the bleeding
time. Semin Thromb Hemost 1990;16:1-20
12. Hindman BJ, Koka BV: Usefulness of the post-aspirin bleed-
ing time. Anesthesiology 1986;64:368-370
13. Steiner RW, Coggins C, Carvalho ACA: Bleeding time in
uremia: a useful test to assess clinical bleeding. J Hematol 1979;7:
107-117
14. Lind SE: The bleeding time does not predict surgical bleed-
ing. Blood 1991;77:2547-2552