Author + information
- Yaling Han, MD, PhD∗ ()
- ↵∗Department of Cardiology, Shenyang Northern Hospital, 83 Wenhua Road, Shenyang, China
We thank Dr. Cecere and colleagues for their interest in and comments on our study (1). They raise a very important question of whether rosuvastatin was consistently effective among various subgroups, especially the subgroup of patients with or without intravenous hydration.
We agree that intravenous hydration is an important factor that should be well controlled in a study in either the design or analysis phase. Unfortunately, no special consideration (e.g., stratified randomization) was given in the design phase back in 2008 when this preventing strategy was not yet well established in the guidelines (2). Therefore, hydration was administered at the discretion of individual investigators according to their daily practice.
Is hydration a confounding factor to our conclusion? Looking at our data, first in Table 1 (1), we observed that hydration therapy was well balanced across 2 arms, probably as a result of our large sample size (44.9% and 42.8% in the rosuvastatin and control groups, respectively), which reassured us a bit that hydration therapy is unlikely to confound our overall conclusion that rosuvastatin can significantly reduce the risk of contrast-induced acute kidney injury (CI-AKI) in patients with diabetes mellitus (DM) and chronic kidney disease (CKD) undergoing arterial contrast medium injection. This was further examined and confirmed in a multivariate analysis as shown in Table 4 (2) that hydration was not an independent predictor of CI-AKI (odds ratio: 0.83, 95% confidence interval: 0.53 to 1.31, p = 0.43), and most importantly, in a test of treatment by hydration interaction (p > 0.05, data not shown).
Dr. Cecere and colleagues also point out that no effect (p = 0.89) was observed in approximately 43% of patients given intravenous hydration. That is probably not valid, and is too early a conclusion, because subgroup analysis was usually underpowered. A nonsignificant result does not mean no effect. We think a more appropriate way is to have an interaction test, and as mentioned in the preceding text, the test assured us that rosuvastatin was consistently effective among patients with or without hydration.
We admit that the point estimate of the rosuvastatin effect observed in patients with hydration was less significant. Whether rosuvastatin can provide an add-on effect in pure hydration patients needs further investigation. We also admit that hydration remains the cornerstone of the prevention of CI-AKI based on accumulated evidence in recent years.
In summary, we think our conclusion of “rosuvastatin significantly reduced the risk of CI-AKI in patients with DM and CKD undergoing arterial contrast medium injection” remains valid in a population with hydration or without hydration as a whole.
- American College of Cardiology Foundation
- Han Y.,
- Zhu G.,
- Han L.,
- et al.
- King S.B. 3rd.,
- Smith S.C. Jr..,
- Hirshfeld J.W. Jr..,
- et al.