As we are learning more and more about targets of hemoglobin, a recent publication by Dr Coyne in Kidney International stuns us even more. The Normal hematocrit trial was the largest trial to date on hemoglobin targets and trying to show that a goal of 13-15g/dl was needed for better quality of life. This trial was terminated early due to increased deaths and cardiac events in the high target groups. The trial did show that it had improved quality of life significantly. The new KI article compares the FDA 1996 clinical trial filed report to the actual publication given in 1998 in NEJM.
Few summary points
1. Higher quality of life was not achieved and was almost reaching statistically significance
2. Higher target hemoglobin had increased risk of death, non access thrombotic events and hospitalization rates( statistical significance) - not noted in the actual paper
3. Renal community guidelines to treat anemia were based on this original paper
Swinging of the pendulum as some call it is what is happening. More data and more data supporting not to target higher doses of hemoglobin. The real question to ask is could the renal community guidelines be different and the early 2000s be different if this was known earlier to us as physicians.
Some of us feel disgusted by this corrupt data presentations and really applaud this author for showcasing what should have been published much earlier.
Just because a major journal publishes an article, it doesn't make the validity of the publication accurate.
take a look at the KI publication and the original trial as well
http://www.ncbi.nlm.nih.gov/pubmed/9718377 ( original 1998 publication)
http://www.nejm.org/doi/full/10.1056/NEJMc076523 ( a follow up 2008 publication on this matter)
This comment has been removed by a blog administrator.
ReplyDeleteMinor correction on your post: There is no improvement in quality of life in any domain and none were approaching statistical significance. The 1998 NEJM paper suggested "physical function" domain improved at 12 months, The results showed that mean +SD score was 35+- 27.3 in the high target arm and 35+-27.5 in the low arm, p=0.97. The difference in area under the curve for physical function over 18 months, which was a predefined secondary end point, similarly was not different between arms:p=0.88.
ReplyDeleteFrom
Daniel Coyne, MD