For years I filed vitamin D under harmless. Take it, move on, it can only help; the bottle sits in my cabinet next to the fish oil and I have never once thought of it as something that could hurt an organ. Then I read a trial where a form of vitamin D drove kidney function down four times faster in exactly the patients it was meant to help, and I want to be precise about which form, because it matters. This is not the softgel you buy at the pharmacy. It is the prescription-strength hormone version, the one doctors reach for when they want to force a stubborn lab number into line.

The number they were chasing is parathyroid hormone, PTH, the signal that climbs as kidneys fail and starts leaching calcium out of bone. In a randomized trial in the American Journal of Nephrology, 78 adults with chronic kidney disease that had not yet reached dialysis were all already taking extended-release calcifediol, a slow-release prohormone that gently nudges the body toward more active vitamin D. Their kidneys were running at about a quarter of normal. Then the researchers split them: some stayed on calcifediol alone, and the rest had a second, much more aggressive drug bolted on, one of the older active vitamin D analogs (calcitriol, doxercalciferol, or paricalcitol) that hit PTH hard and fast.

On PTH, it worked, and worked beautifully. Adding the active vitamin D dropped the hormone by an extra 35 percent. Calcifediol alone barely moved it, about 2 percent. Stop reading the chart there and you would call the second drug a clean win and write the prescription.

I kept reading, because the kidney numbers underneath were telling the opposite story. Over just fourteen weeks, estimated kidney function fell 11.8 percent in the group that added active vitamin D, against 3.0 percent in the group that did not. Same short window, same starting point, and roughly four times the loss in the people who got the extra drug. The lab value everyone was optimizing looked its best in the arm that was losing kidney function fastest.

So why would a vitamin corrode filtration in the first place? Here is where the biology stops being abstract. Active vitamin D does its main job in the gut: it tells the intestine to pull in more calcium and more phosphorus from whatever you eat. You can watch that happen in the trial. Serum calcium rose, phosphorus rose, and then the body’s phosphate alarm went off. That alarm is a hormone called FGF23, and in the group taking the extra vitamin D it more than doubled, climbing 155 percent.

FGF23 is supposed to be the cleanup crew, dumping surplus phosphate back out through the kidney. But when it is chronically screaming, it stops looking protective. In failing kidneys, a persistently high FGF23 is one of the most consistent independent predictors of heart disease and death we have, and it has been implicated in thickening heart muscle and stiffening blood vessels. The arm whose phosphate alarm was screaming loudest is the arm whose kidneys slipped the most.

Read it that way and the whole thing clicks. Nobody poisoned anyone. The doctors chased a surrogate, PTH, with a tool strong enough to crush it, and in doing so they poured calcium and phosphorus into a system that had already lost the ability to clear them. The printout improved while the patient did worse. Treating the number as the goal instead of the person it is supposed to stand in for is one of the oldest traps in medicine, and this trial catches it on camera in fourteen weeks.

My skeptical reflex flagged something, and it is worth saying plainly. Eight of the study’s authors are or have been employees or consultants of OPKO Health, the company that makes the extended-release calcifediol every patient in the trial was already taking. The design sets that product, taken alone, against the same product plus an older competitor drug, and it is the competitor that comes out looking dangerous. A 78-person, 14-week study that lands this neatly on the sponsor’s commercial interest earns a raised eyebrow, and the authors themselves say the result needs a larger, longer trial to confirm.

But I do not just wave it away as a company guarding its franchise, and here is why. The biology does not care whose logo is on the paper. Forcing calcium and phosphorus up in a damaged kidney and watching FGF23 double in response is not a marketing artifact; it is the mechanism nephrologists have been circling for years. It is also why the major kidney guidelines already tell doctors not to use these active vitamin D analogs routinely in pre-dialysis patients, and to reserve them for the severe, progressive cases. The sponsor had a motive, and the trial still surfaced a mechanism that holds up on its own. Both are true at once.

None of this is an argument against the vitamin D in your cabinet. Ordinary D3, the nutritional kind, is a different molecule doing a gentler job, and fixing a genuine deficiency is not what went wrong here. What went wrong was the reflex to escalate to the hormone-strength version to win a number, in the population least able to absorb the collateral damage.

If it were my kidney, or my mother’s, I would treat that PTH number as a dial to nudge, not a score to max out, and I would ask the doctor point-blank what the added drug was doing to phosphorus and FGF23 before agreeing to a single dose. I would not trade a better-looking lab value for kidney function I cannot get back.

Sources

  1. American Journal of Nephrology – Ashfaq et al., adjunctive active vitamin D decreases kidney function during ERC treatment of secondary hyperparathyroidism in ND-CKD, randomized trial (2025)
  2. KDIGO 2017 Clinical Practice Guideline Update for CKD–Mineral and Bone Disorder (active vitamin D analogs not for routine use in non-dialysis CKD)
  3. Toxins – Vervloet, “FGF23 and Phosphate: Cardiovascular Toxins in CKD” (2019)
  4. FGF23, cardiovascular disease and mortality in chronic kidney disease (PubMed)
  5. OPKO Health – Rayaldee (extended-release calcifediol) product page