CANCER RISK WITH AMIODARONE. FACT OR FALLACY?

In my journal review, I came across this biggish observational study from Taiwan, which is a little disturbing. Disturbing maybe because I do use a fair bit of Amiodarone for  atrial fibrillation and also in my CAD patients with large infarcts and low LVEF, in an attempt to prevent sudden cardiac death ( cheap AICD ).
Dr Vincent Yi-Fong Su and colleagues from the Taipei Veteran General Hospital, Taiwan, looksed into the Taiwan national Health Insurance database, to pick out those who were given Amiodarone from 1997-2008, with at least 2 years of follow-up. They identified 6,418 patients who had a mean follow-up of 2.8 years. From the database, they also calculated the "Standardised incidence ratio" for the cancers. This study is published in the April 8th edition of Cancer.
They found that after 1 year of follow-up, there was a 12% increase incidence of cancers, and that the higher the dose, the more likely. But the cancers reported was of a mixed variety, including solid tumours of the GIT, GU tract, Head and Neck, and also blood tumours.
Their conclusion was that the use of amiodarone was associated with a small increase risk of cancers , and that patients should be closely monitoered and encourage those of us using amiodarone to begin to be aware and report cancers when they appear in patients using amiodarone.
Well, I was a little troubled, as there are still a few of my patients on amiodarone, Thinking over the study, I had a few issues with the study and conclusion.
Firstly, although >6,000 is a reasonable size study, the average years of follow-up is 2.6 years, which is relatively short, especially for cancer tracking.
Secondly, using standardised incidence ratio is full of errors and not so accurate, especially with a short duration.
Thirdly, I could hardly believe that cancer can be caused within 1 year, unless the patient already had them.
Fourthly, the cancers were an assortment. usually cancer producing drugs trigger only one type of cancer predominantly. Not an assortment.

It is good that we should monitor our patients on Amiodarone and report cancers should they occur, but I think that we obviously need more data, before we can remove amiodarone and say that there is a link between the use of amiodarone and cancers.