Endpoint: hard — all-cause mortality, prospective general-population cohort (not a referral/angiography population).
In 5,469 PREVEND (Netherlands) participants, median 8.3-year follow-up, 322 deaths: highest vs. lowest TMAO quartile had 1.86-fold higher crude all-cause mortality. After adjusting for cardiovascular risk factors the association weakened (HR 1.36, 95% CI 0.97-1.91); further adjustment for eGFR and urinary albumin excretion abolished it. The TMAO-mortality association was significantly modified by eGFR (interaction P=0.002).
relevance_note: a primary demonstration, in an independent general-population cohort, that a TMAO-mortality association is substantially explained by renal function — directly supports the “TMAO may mark impaired renal clearance rather than independently cause CVD/mortality” skeptical position (thread iii).