Dear Editor:
I read with great interest the study by Chang et al1 which found an association between postdiagnosis sodium-glucose cotransporter 2 inhibitor (SGLT2i) use and reduced dialysis initiation for patients with autosomal-dominant polycystic kidney disease (ADPKD). The authors state that “all analyses were conducted using the TriNetX analytics platform”. However, the authors also state that they have used a pseudo-index event design for their control cohort. I have previously outlined that this is not possible to implement through the TriNetX analytics platform, as the platform only allows users to select inclusion criteria to act as index events.2 TriNetX employees replied to this study confirming that pseudo-index events cannot be conducted through the platform.3 Although the authors have stated that they conducted the study using the TriNetX analytics platform and that they used a pseudo-index event design for their control cohort, these 2 statements cannot both be true.
Index event settings majorly influence the outcomes of TriNetX-based studies.4 It is therefore important that the authors clarify how their index event for their control cohort was configured. For example, if the index event for the control cohort was selected as the date of ADPKD diagnosis, then the control cohort’s outcome analysis would count patients with underlying kidney failure only diagnosed in the days following their ADPKD diagnosis.5 In contrast, the treatment cohort would only be followed from their postdiagnosis SGLT2i prescription, not counting patients who presented with underlying kidney failure and required dialysis in the days after ADPKD diagnosis. This would heavily bias the study in favor of the treatment cohort. Evidence of this bias can be seen in the shape of the Kaplan-Meier curve of the original study—there is a drastic drop in kidney failure-free survival for the control cohort in the days after the index event.
An appropriate design for this study would be to set the index event as the first ADPKD diagnosis for both cohorts, only include patients with an SGLT2i prescription within a finite period of time after ADPKD diagnosis, and measure outcomes starting after the prescription time window. For example, the treatment cohort could be set to include only patients who received an SGLT2i prescription in the 12 months after ADPKD diagnosis. Then, the outcome period could be set to begin 366 days following the index event for both cohorts. This would ensure that both cohorts are assessed during the same post-ADPKD time window and eliminate any immortal time bias.6 I would encourage the authors to perform this re-analysis so that their important research question can be answered robustly.
Potential Competing Interests
The author reports no competing interests.
References
- 1.Chang C.H., Kuo Y.C., Kuan Y.H., et al. Real-world outcomes of sodium-glucose cotransporter 2 inhibitor therapy in nondiabetic autosomal-dominant polycystic kidney disease: effects on progression to dialysis in a TrinetX cohort. Mayo Clin Proc Innov Qual Outcomes. 2026;10(2) doi: 10.1016/j.mayocpiqo.2026.100709. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lin Y.J., Tsai K.W., Lu K.C., Wang J. On the reported methodology in published TriNetX-based studies: an analysis of impossible index event designs. Eur J Epidemiol. Published online February 21, 2026 doi: 10.1007/s10654-025-01342-6. [DOI] [PubMed] [Google Scholar]
- 3.Williford S.E., Arnold Chan K., Brown, Re J.S., Liu, et al. “On the reported methodology in published TriNetX-based studies: an analysis of impossible index event designs.”. Eur J Epidemiol. Published online February 21, 2026 doi: 10.1007/s10654-026-01374-6. [DOI] [PubMed] [Google Scholar]
- 4.Tsai K.W., Chong K.H., Wang J. Research design concerns in a TriNetX study of tirzepatide and heart failure prevention in obese patients. Eur J Preventive Cardiol. 2025 doi: 10.1093/eurjpc/zwaf721. [DOI] [PubMed] [Google Scholar]
- 5.Harrison T.N., Chen Q., Lee M.Y., et al. Health disparities in kidney failure among patients with autosomal dominant polycystic kidney disease: a cross-sectional study. Kidney Medicine. 2022;5(2) doi: 10.1016/j.xkme.2022.100577. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Li S.Y., Nag A., Ben-Israel D., et al. Serotonin reuptake inhibition and intracerebral hemorrhage risk after ischemic stroke: a multicenter retrospective study. Clin Neurol Neurosurg. 2026;262 doi: 10.1016/j.clineuro.2026.109309. [DOI] [PubMed] [Google Scholar]