RCT
Citation: Home, P. D. Et al (2007). Rosiglitazone Evaluated for Cardiovascular Outcomes — An Interim Analysis, NEJM, 357 (1):28-38. http://content.nejm.org/cgi/content/full/NEJMoa073394
THERAPY WORKSHEET: page 1 of 2
Are the results of this single preventive or therapeutic trial valid?
Was the assignment of patients to treatments randomized? Yes, subjects were randomly assigned to the case group or the control group. Patients were initially receiving sulfonylurea were given additionally metformin (control group) or rosiglitazone (Avandia). People who were initially receiving Metformin were given additional sulfonylurea or Avandia. The randomization occurred at the 4 weeks of treatment with the initial drug. The random allocation occurred over the phone. Stratification of the random permuted blocks was done according to the previous medications.
-and was the randomization list concealed?
The random allocation occurred over the phone. Stratification of the random permuted blocks was done according to the previous medications.
Were all patients who entered the trial accounted for at its conclusion? -and were they analyzed in the groups to which they were randomized?
In the trial about 7428 patients were initially screened for diabetes and consumption of sulfonylurea or metformin. About 4458 patients were selected for the study. About 6 patients in the case group and 5 in the control group were not given the study medication. Initially about 2222 patients were receiving metformin monotherapy, out of which 1117 were assigned to the case group and 1105 to the control group. On the other hand, 2225 patients who were receiving sulfonylurea, 1103 were assigned to the case group and 1122 were assigned to the control group. Hence, 2220 patients were in the case group and 2227 patients were in the control group. In the both groups, about 10 % of the patients were lost during the follow-up. About 74 patients in the case group and 80 patients in the control group were died due to various causes. 47 patients in the case group and 22 patients from the control group were hospitalized for heart failure. 267 patient in the case group and 243 patients in the control group had primary end points (hospitalization or death from CVD)
Were patients and clinicians kept “blind” to which treatment was being received?
The study was not kept blinded to the patient, physician or the evaluator because the patient would be consuming a vast number of preparations with specific doses. This was a long-term study and blinding was not possible. Besides, in some patients insulin therapy may have been required, which would be difficult to enable in case the patients or the physician did not know which drug was being consumed. However, the adjudication of events was blinded for various potential cardiovascular end points.
Aside from the experimental treatment, were the groups treated equally?
Yes, both the groups were treated equally. This was evident from the fact that 218 patients in the case group and 223 patients in the control group left the trial and did not report for follow-up. Both the groups contained equal number of patients. The patients were randomly allocated to either group. If any of the patients developed an HbA1c level greater than 8.5 %, they were given an additional anti-diabetic drug if they were in the case group or insulin if they were in the control group.
Were the groups similar at the start of the trial?
Yes, both the groups were similar at the start of the trial. Equal number of patients who were receiving metformin therapy and sulfonylurea therapy were assigned to the case and control groups. The patients were basically type 2 diabetics with poor glycaemic group with either metformin or sulfonylurea. Their treatment required the use of an additional drug.
Are the valid results of this randomized trial important?
SAMPLE CALCULATIONS:
Occurrence of heart failure – 47 in case group; 22 in control group (morbidity); 267 patients in the case group and 243 patients in the control group had primary end points (hospitalization or death from CVD)
CER – 47/267 = 17.6 %
EER – 22/243 = 9.05 %
RRR = CER-EER/CER
17.6-9.05/17.6 = 48.75%
APR à CER-EER
17.6-48 = -30.4 (Absolute Risk Reduction is 30.4 higher)
NNT = 1/APR = 1/30.4 = 3.33
95% Confidence Interval (CI) on an NNT = 1 / (limits on the CI of its ARR) =
95% CI, 1.30 to 3.57
THERAPY WORKSHEET: page 2 of 2
Can you apply this valid, important evidence about a treatment in caring for your patient?
Do these results apply to your patient? Is your patient so different from those in the trial that its results can’t help you?
Yes, this can be applied to the patient suggesting a greater chance of side-effects with heart failure following administration of Avandia.
How great would the potential benefit of therapy actually be for your individual patient?
Patients with already higher risk for heart failure or CVD should not be administered Avandia, as they are not going to benefit. Instead drugs used in the control groups such as metformin and sulfonylurea can be utilized.
Method I: f Risk of the outcome in your patient, relative to patients in the trial. Expressed as a decimal: 0.5; NNT/F = 3.33/0.5 = 6.66
(NNT for patients like yours)
Method II: 1 / (PEER x RRR) Your patient’s expected event rate if they received the control treatment:
PEER:______
1 / (PEER x RRR) = 1/________ =
_______
(NNT for patients like yours)
Are your patient’s values and preferences satisfied by the regimen and its consequences?
Do your patient and you have a clear assessment of their values and preferences?
The risk of heart failure in diabetics is higher when Avandia is administered compared to sulfonylurea or metformin. The risk for heart failure is 3.33 (NNT).
Are they met by this regimen and its consequences?
The evidence in the literature does support the issue that Avandia can increase the risk for heart failure. However, it does not support the issue that Avandia can increase the risk for other heart diseases such as myocardial infarction and other CVD.