My colleagues and I read with interest your Cochrane Review, "Once versus twice daily low molecular weight heparin for the initial treatment of venous thromboembolism." We would like to address a few points from this review.
First, we would like to acknowledge that an important outcome presented in the review is mortality. We noted that the Holmoström 19921 trial did not report on mortality, and was therefore not included in the mortality analysis. Because this outcome takes precedence over all others in determining if one treatment is better than another, we believe that it would be important to attain as much data on it as possible. This can be achieved by contacting the authors of the original article. Furthermore, as the odds ratio confidence interval was so wide for this outcome [(OR 1.14, 0.62 to 2.08)], it is not possible to determine whether a clinically significant difference in mortality is seen when comparing treatments. Therefore, the statement in the review that "there was a statistically non-significant difference in mortality in favour of people who are treated with LMWH twice daily compared with people treated with LMWH once daily" can be reworded to "there was a statistically non-significant difference in mortality, but a clinically significant increase or decrease cannot be ruled out." This would allow readers to put the available data into clinical context. In our opinion, the wide confidence interval for mortality limits the ability to conclude with certainty that "once daily treatment with LMWH is as effective and safe as twice daily treatment with LMWH."
We would also like to address the protocol of the review regarding the types of studies that were to be included. The protocol was to include only those that were randomized with an intention-to-treat analysis, and that had adequate allocation concealment and blinded outcome measurement. We do not believe that trials should be excluded from a systematic review due to any form of bias. Rather, biases should be outlined in the bias reports, and sensitivity analyses should be done as necessary. Regardless, based on the table of excluded studies, all reasons for exclusion appear to be reasonable. However, it should be clarified in the review that biases did not serve as exclusion criteria, which would aid in strengthening the validity of the review.
We performed a risk of bias assessment for all the included trials, and noted that missing data was present in the Merli 20012 trial, and that Holmström 19921 was not an intention-to-treat trial. Missing data is important to address, as analyses using the missing data may yield different results from those presented in the review. For example, in the Merli 20012 trial, 34 patients in the once daily group and 37 in the twice daily group were lost to follow-up, and no sensitivity analyses were done. Sensitivity analyses assuming hemorrhagic event worst case scenarios for patients lost to follow-up either in the once or twice daily group would yield statistically significant different results: the odds ratio would increase to 2.39 (1.45, 3.93) if assuming all lost to follow-up patients in the once daily group had hemorrhagic events, and decrease to 0.28 (0.16, 0.48) if the same is done for the twice daily group.
Thank you for conducting this review; we believe that this is an important issue, as it affects convenience, patient comfort, and cost. We hope that you consider our suggestions, and we appreciate your time.
Hilary Wu (BSc.Pharm), Mark Ho (BSc.Pharm), Karen Hong (BSc.Pharm), and Yin Gong (BSc.Pharm)
1. Holmström M, Berglund MC, Granquist S, Bratt G, Törnebohm E, Lockner D. Fragmin once or twice daily subcutaneously in the treatment of deep venous thrombosis of the leg. Thromb Res. 1992 Jul 1;67(1):49–55.
2. Merli G, Spiro TE, Olsson CG, Abildgaard U, Davidson BL, Eldor A, et al. Subcutaneous enoxaparin once or twice daily compared with intravenous unfractionated heparin for treatment of venous thromboembolic disease. Ann Intern Med. 2001 Feb 6;134(3):191–202.
A reply from the review authors is awaited.
Feedback: Hilary Wu, Clinical Pharmacist, Providence Health Care, Canada
I certify that I have no affiliations with or involvement in any organization or entity with a financial interest in the subject matter of my feedback.