Research Article: Is topical or intravenous tranexamic acid preferred in total hip arthroplasty? A randomized, controlled, noninferiority clinical trial

Date Published: October 2, 2018

Publisher: Public Library of Science

Author(s): Kai-di Zhou, Hong-yi Wang, Yi Wang, Zhi-hong Liu, Chuan He, Jian-min Feng, Kirill Gromov.


The present study aimed to confirm the efficacy and safety of topical and intravenous tranexamic acid (TXA) compared with that of topical placebo and to assess the noninferiority between the two application methods of TXA in patients undergoing unilateral primary total hip arthroplasty.

Our randomized controlled trial investigated 170 patients with 1:1:1 allocation to two doses of 10-mg/kg intravenous TXA, 3-g topical TXA, and topical placebo of 60-ml physiological saline groups. The primary outcome, total blood loss, was calculated with Nadler and Gross formula. The secondary outcomes included allogeneic blood transfusion requirement, drain blood loss, decreased hemoglobin level. Noninferiority would be established when the upper limit 95% CI is lower than 250 ml of the noninferiority margin for the mean difference of total blood loss between topical and intravenous TXA. Thromboembolic complication incidence was considered as a safety outcome.

The total blood loss of patients administered intravenous (mean±standard deviation, 1125±514 ml) and topical TXA (1211±425 ml) was significantly reduced compared with that of those administered topical placebo (1464±556 ml) (p = 0.0012). Drain blood loss and hemoglobin level reduction in patients administered with TXA were also significantly lower than those in patients administered topical placebo. The mean difference of total blood loss between topical and intravenous TXA is 86 ml (95% CI, −88 to 260 ml). The complications were comparable between patients managed with TXA and patients with topical placebo.

The noninferiority of topical TXA to intravenous TXA can not be concluded. Considering no significant difference was found in all efficacy outcomes between the two administration methods. Any of the two TXA administration methods can be adopted for blood loss prevention in total hip arthroplasty.

Partial Text

Hemorrhage and subsequent anemia are highly prevalent in patients undergoing total hip arthroplasty (THA) [1, 2].

The present study was registered in the public registry (NCT02312440) after approval was granted by ethics board of Ruijin Hospital, Shanghai Jiao Tong University School of Medicine. The complete date range for patient recruitment and follow-up was also approved by ethics board of Ruijin Hospital. In addition, the study protocol has also been published [20]. The authors confirm that all ongoing and related trials for this intervention are registered. However, the trial was registered after patient recruitment began because the translation of the protocol delays its releasement. Written informed consent was obtained from the patients or their legal representatives.

Computer-generated randomization table with block sizes of 9 was performed to randomize patients a day before the surgery based on their admission sequence. Independent pharmacists prepared the study medication and topical placebo after an investigator provided the intraoperative prescription based on the patient’s assignments. In the operation room, a scrub nurse provided a 60-ml solution with or without TXA to surgeons, whereas ivTXA solution was provided to anesthesiologists to administer based on the prescription. Patients, surgeons, and nurses participating in treatment and evaluation were blinded to the group allocation throughout the study period.

From July 2014 to May 2015, 218 patients were scheduled for primary unilateral THA at our department (Fig 1). Based on the inclusion/exclusion criteria, 44 patients were excluded, and 174 were randomized to three cohorts. Of the 174 randomized patients, 170 completed the study and entered in statistical analysis. Fig 1 shows the reasons for excluding four randomized patients. No patient was lost or excluded during the follow-up.

We conducted a randomized controlled trial comparing two application routes of TXA with topical placebo. The first hypothesis that either tTXA or ivTXA would significantly reduce blood loss in patients undergoing THA compared with topical placebo has been confirmed. The second hypothesis that tTXA would be noninferior to ivTXA in conserving blood in patients undergoing THA cannot be established. Considering the lesser blood loss and easier to administer before and after the surgery compared with tTXA, we support administering two doses of 10-mg/kg ivTXA in unilateral primary THA as a standardized protocol for blood loss prevention. Considering the upper limit 95% CI of mean difference of total blood loss is only slightly higher than noninferiority margin and no significant differences were found for other efficacy outcomes when tTXA compared with ivTXA. In addition, tTXA can reduce 252 ml total blood loss when compared with topical placebo.

IvTXA and tTXA reduce total blood loss by 327 ml and 252 ml compared with topical placebo, respectively, in patients undergoing unilateral primary THA. Contrary to expectations, this study cannot conclude the noninferiority of tTXA relative to ivTXA in reducing total blood loss in THA, indicating loss of the effect of tTXA has not been ruled out. However, the upper limit 95% CI of mean difference of total blood loss is only slightly higher than noninferiority margin and we found no significant difference in other efficacy outcomes between ivTXA and tTXA. In addition, basing on the results of the present study, we speculate that the intravenous route and topical route of TXA has similar efficacy for blood loss prevention. Finally, no significant difference was found between patients managed by TXA and patients managed by placebo for thromboembolic complication incidence.




0 0 vote
Article Rating
Notify of
Inline Feedbacks
View all comments