Do you give a Flux about Flux?

There is a spirited debate in our hospital system about the use and availability of high flux membranes. Part of the debate centered around middle molecule clearance and its relationship to uremia. I fell back to my old standard the HEMO trial.

A study as long and as expensive as HEMO will not be let to rot after its first paper, and the research parasites extracted a number of tantalizing findings. In 2003 JASN published an analysis that looked at flux when patients were divided by vintage. Dialysis patients who have been on dialysis for more than 3.7 years had improved outcomes with high flux dialyzers:

In the subgroup that had been on dialysis for >3.7 yr, randomization to high-flux dialysis was associated with lower risks of all cause mortolity (RR, 0.68; 95% CI, 0.53 to 0.86; P = 0.001), and cardiac deaths (RR, 0.63; 95% CI, 0.43 to 0.92; P = 0.016), compared with low-flux dialysis.

Though the data is not so compelling when the vintage is divided by quintiles:

Even more damning ,was the fact that the longer the patients were randomized, the smaller the effect of high flux membranes:

Longer years of follow-up should show more protection from high-flux membranes (lower RR in the table) but the opposite is actually seen.

Additional post-hoc analysis was done here, showing decreased cerebrovascular disease for longer vintage patients:

I then came across the membrane permeability outcome study, MPO, published in 2009 in JASN. This European study was designed to answer the following question:

This prospective, randomized Membrane Permeability Outcome (MPO) study was designed to compare the impact of membrane permeability on survival in incident HD patients who had either low ( 4 g/dl) or normal albumin ( 4 g/dl) and were treated with a minimum dialysis dose (single-pool Kt/V [spKt/V]) of 1.2. 

Use of incident patients will eliminate the vintage advantage seen in the two post-hoc HEMO studies discussed above. Separating patients based on albumin at baseline seems a bit wonky. Lots of patients have low albumin at the start of dialysis. And I doubt there is an inherent biological reason for this. Turns out, the two tiers were due to pokey enrollment after 11 months and they changed enrollment rules.  They also changed the rules by extending the time patients could take to get to a spKt/V from 1 month to 3 months. The things we do for enrollment!

The top-line results showed no difference in survival by flux:

But when the authors looked at the pre-specified sub-group analysis for patients with a baseline albumin less than 4, the data showed protection with high-flux membranes:

This isn't just any pre-specified sub-group, this was supposed to be the entire cohort, and only became a sub-group after they amended the protocol.

They also found an interaction between flux and diabetes. This was post-hoc analysis:

So, in the end this seems like a tale told by an idiot (your truly), full of sound and fury signifying nothing because we have all moved on to high flux dialyzers, except in the hospital where F16 low flux dialyzers are available and cheaper than their high flux brethren.