06.11.2019 | Letter to the Editor
A missed protective drug–drug interaction of DPP-4 inhibitors and statins on myopathy risk
Erschienen in: Acta Diabetologica | Ausgabe 1/2020
Einloggen, um Zugang zu erhaltenExcerpt
With great interest, we read the recent paper of Antonazzo et al. [1] on the risk of myopathy related to DPP-4 inhibitors and statin use. Antonazzo et al. used 142,888 myopathy cases and 858,627 non-cases from FAERS, the FDA adverse event reporting system [1]. The authors were especially interested in potential interactions between the two drugs and therefore refer to and use the approach of Thakrar et al. [2] which entails a multiplicative and an additive model to assess drug interactions. Antonazzo et al. computed both models, but did not give confidence intervals for the estimated interaction terms. When using the raw data from Table 2 of Antonazzo et al. and the GENMOD procedure in SAS (SAS Institute Inc., Cary, NC, USA, see Appendix for the code), as recommended by Thakrar et al., we observed statistically significant and clinically relevant negative interactions from the multiplicative as well as from the additive model (Table 1). Therefore, we disagree with the conclusion of Antonazzo et al., that a “drug interaction is very unlikely.”
Parameter
|
Estimate
|
95% CI
|
p value
|
---|---|---|---|
Additive model
|
|||
DPP4
|
0.017
|
[0.013; 0.020]
|
< 0.0001
|
Statin
|
0.121
|
[0.120; 0.123]
|
< 0.0001
|
Interaction
|
− 0.033
|
[− 0.040; − 0.026]
|
< 0.0001
|
Multiplicative model
|
|||
Exp (DPP4)
|
1.173
|
[1.140; 1.207]
|
< 0.0001
|
Exp (Statin)
|
2.265
|
[2.238; 2.292]
|
< 0.0001
|
Exp (Interaction)
|
0.789
|
[0.757; 0.824]
|
< 0.0001
|