 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc.
CorrespondenceTO THE EDITOR: I read the article "Effect of Ketorolac Tromethamine on Bleeding and on Requirements for Analgesia after Total Knee Arthroplasty" (77-A: 998–1002, July 1995), by Fragen et al., with great interest. I find it hopeful that excellent analgesia can be obtained with a non-opiate substance in order to limit side effects in an often elderly population at risk for complications. I am still concerned, however, about bleeding and wound complications in the total joint population. These very important complications, which may prove catastrophic, were not addressed in this study. As we know from the literature regarding prophylactic therapy with low-dose warfarin and low-molecular-weight heparin, substantial bleeding episodes occur in as many as 5 per cent of treated patients2,4. In these reports, a major bleeding episode was defined as a decrease in hemoglobin of 0.002 gram per deciliter or more, the need for a transfusion of two units of blood or more after twenty-four hours, bleeding into the retroperitoneum, intracranial hemorrhage, or bleeding into the site of a major prosthetic joint2. None of the fifty-nine patients reported on by Fragen et al. had a substantial bleeding episode. I therefore question if enough patients were included for the authors to determine whether bleeding complications were affected or if the definition of a substantial bleeding complication was similar to that in previous studies of warfarin and low-molecular-weight heparin. Also, speaking anecdotally from clinical observations of our patients with a joint replacement, persistent serous drainage seems to occur with the use of anticoagulants in general. This drainage was more substantial when our patients were managed with ketorolac tromethamine and especially when it was used simultaneously with low-molecular-weight heparin. I would be interested to learn if these observations were made in the reported population and also if there was a difference in complications with respect to the use of warfarin and low-molecular-weight heparin. Hypothetically, given the rapid onset of both ketorolac and low-molecular-weight heparin, it would not be surprising to see more bleeding complications with this combination of agents. Since low-molecular-weight heparin interferes with factor Xa at the level of the common clotting pathway, there is a dependence on platelet function to form a loose platelet aggregate. The reversible and immediate cyclooxygenase-mediated platelet inhibition of ketorolac would then theoretically predispose the patient to bleeding because of a lack of formation of this so-called plug during primary hemostasis, which occurs within seconds after an injury of a vessel1. Conversely, as the half-life of warfarin is approximately 7.5 hours, a steady state is not reached for approximately thirty-six to seventy-two hours. The use of ketorolac in patients managed with warfarin allows formation of a clot with compaction to an organized and irreversible plug in the presence of fibrinogen, which is not immediately inhibited. Thus, fewer bleeding episodes may result. I hope that this study continues to include more patients, with specific observations regarding the cited parameters so that these critical issues may be addressed before the widespread simultaneous administration of these potentially interacting agents.
Douglas Linville, M.D.: Department of Orthopedics, State University of New York at Stony Brook, Stony Brook, New York 11794
Dr. Fragen, Dr. Stulberg, Dr. Wixson, Dr. Glisson, and Ms. Librojo reply: Dr. Linville correctly expresses concern regarding our article. Total knee arthroplasty was chosen for the study only because it is one of the few operations for which most of the blood loss is measured rather than estimated. The results should be applicable to other types of operations of the same or lesser severity, but we realize that they cannot be applied to all operations, such as, possibly, total hip arthroplasties. In our study, we found no wound infections or bleeding other than at the operative site. According to the definition of a major bleeding episode cited by Dr. Linville, twenty-five of the twenty-nine patients who received ketorolac and twenty-three of the thirty patients who received the placebo in our study had a decrease in hemoglobin of 0.002 gram per deciliter or more. However, these numbers are meaningless because many of the preoperative laboratory values were obtained before the patients donated blood for autologous transfusion. Furthermore, these numbers appear similar for the two types of treatment. A better variable for comparison is the number of patients who received transfusion of more than one unit of blood (some of the transfused blood came from the Hemovac drainage system). Twelve of the twenty-nine patients who received ketorolac received more than one unit of blood. Six of these twelve were given warfarin and six, low-molecular-weight heparin. Thirteen of the thirty patients who received the placebo received more than one unit of blood. Four of these thirteen were given warfarin and nine, low-molecular-weight heparin. Furthermore, with regard to the patients who had the greatest blood loss, seven who received the placebo and five who received ketorolac lost 750 to 1000 milliliters; two patients who received the placebo and one who received ketorolac lost more than 1000 milliliters. Of these fifteen patients, seven were given warfarin and eight were given low-molecular-weight heparin. It appears from these data that ketorolac was not associated with greater blood loss compared with the placebo regardless of whether warfarin or low-molecular-weight heparin was used for anticoagulation to prevent deep-vein thrombosis. We doubt if a larger number of participants would change the results, and we have not expanded the study. Despite our results, we reiterate part of our summary paragraph: "It is still possible that a bleeding problem could occur postoperatively with the combination of ketorolac and anticoagulants." This may be a special concern when preoperative platelet function may be abnormal from other causes, but ketorolac was recently shown not to worsen platelet function when platelets are healthy3.
Robert J. Fragen, M.D.; S. David Stulberg, M.D.; Richard Wixson, M.D.; Silas Glisson, Ph.D.; Eleanor Librojo, R.N.: Northwestern University Medical School, 360 Passavant Pavilion, 303 East Superior Street, Chicago, Illinois 60611
References
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |