| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Englewood Hospital and Medical Center, Englewood, NJ St. Agnes Healthcare, Baltimore, MD
Correspondence to: Tanuja Rijhwani, MBBS, MPH, Director, Clinical Research, Department of Anesthesiology, Englewood Hospital and Medical Center, Englewood, NJ 07631; e-mail: tanuja.rijhwani{at}ehmc.com
To the Editor:
The recent article by Avorn et al (October 2003)1 purports to show an association between the use of hydroxyethyl starch (HES) and excessive postoperative bleeding after coronary artery bypass surgery (CABG). This article resurrects the unresolved controversy regarding the use of HES in cardiac surgery and the emerging concern of clinically significant bleeding. Several prospective randomized trials,23456789 observational studies,91011 and meta-analyses,1213 have investigated the suspected association between HES use and bleeding after CABG.
Most randomized studies on HES and bleeding have failed to show any clinically significant bleeding differences. The published retrospective studies showing an increased incidence of blood loss have received the most press but are inherently limited due to study design. Cope et al11 retrospectively reviewed the use of hetastarch infusion based on perioperative exposure to HES and transfusion requirements during the first 24 h postoperatively. The selection bias of this study favored those with hemodynamic compromise or those with greater severity of illness. The meta-analysis by Wilkes et al14 shows that the difference in pooled mean blood loss in the albumin group was 693 ± 350 mL compared with 789 ± 487 mL in the HES group, a difference of 96 mL only.
We believe that the article by Avorn et al1 fails to show an association between HES and postoperative bleeding, as cited in the majority of related studies. The title leads the reader to believe that hetastarch increases the risk of bleeding, but the authors did not report any single measure of bleeding. They did not account for the number of bleeding episodes nor did they use quantifiable measures, ie, chest tube drainage volume. Some of the study design limitations were that no clear definitions of nonsurgical bleeding or proper criteria for correction of microvascular bleeding were used. Measurement of hematocrit from the drainage fluid and collection of blood samples for baseline laboratory parameters indicative of blood loss would have been more reliable estimates of perioperative blood loss. These indicators should be considered as one of the primary outcome variables. Although transfusions may reflect bleeding, previous data1516 suggest that transfusing blood in cardiac surgery is behavioral rather than a response to blood loss. Re-operation (exploration for bleeding) in coronary bypass surgery cannot be used as a measure of drug-induced coagulopathy unless one completely excludes other causes for bleeding, ie, heparin-protamine activity, large-vessel bleeding, and overall patient coagulation status. Simple changes in preoperative, intraoperative, and postoperative hemoglobin and hematocrit values would help clarify their claims, although even these are confounded by volume changes.
The patients selected in the study by Avorn et al1 were those who had received 
3 U packed RBCs within 72 h after undergoing CABG procedure; all other CABG patients were control subjects. The higher comorbidity in the cases (HES group) may act as a confounder leading to higher transfusion rates as a result of lower transfusion thresholds for more severely ill patients.
The editorial accompanying this study recommends a change in practice without convincing evidence. Lack of sufficient data should not be replaced with personal interpretation of study results. This invariably can be misleading to the medical community.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
