To start, let's look a pamphlet produced by the AAEM "tPA for Stroke - Potential Benefit, Risk and Alternatives." It represents the National Institute of Neurological Disorders and Stroke tPA Stroke Study Group (NINDS) data in the following way:
This diagram shows that without treatment 6/18 patients recover by three months without significant disability. With tPA treatment, this increases by 2 to 8/18 while unfortunately causing one intracranial hemorrhage. Stated in this way, the benefit to harm ratio is not great.
However, this over looks several subtle caveats which have been written about extensively in the neurology literature. First, while tPA did cause one intracranial bleed per eighteen treated, this DOES NOT mean that had this patient not been treated, that his or her three month outcome would have been much better (ie the patient could have ended up with a severe debilitating stroke instead; either way, functional outcome would have been similar). Second, while tPA will increase the number of patients living without significant disability by two, it will increase the number of patients living with a lesser disability (ie perhaps the patient won't be walking but he'll be wheelchair bound with a lesser degree of left sided weakness). Taking these two items into consideration and doing some statistical crunch and munch and munch, the risk and benefit of tPA as appreciated by neurologists is that for every one hundred patients with acute ischemic stroke treated with tPA, thirty-two will benefit and three will be harmed. In this light, tPA looks much more favorable.
Bottom line, it's all the same original data but the framing of how it is presented will make a significant difference on whether physicians encourage and patients subsequently opt for tPA therapy.
Source
Follow links to original source information.