Will Starting Dialysis “Fix” Heart Dysfunction?

A patient with CKD nearing dialysis asked the following (edited) question:

“ … my eGFR is between 12 and 14%, but my nephrologist feels it is not accurate. I've read that Cystatin C is a better indicator of kidney function, especially in women. Is this so?

In addition, the toxins are beginning to affect my heart muscle as I have Stage1 diastolic dysfunction. Would starting dialysis slow this process?”

Unfortunately, this issue is not at all a straightforward one. Firstly, diastolic dysfunction is not specific to CKD. It can occur as an echocardiographic finding in a vast array of conditions, which include, but are not restricted to:

• Longstanding hypertension
• Any cause of aortic outflow tract obstruction eg: aortic valve narrowing (stenosis)
• Other outflow tract abnormality
• Left ventricular hypertrophy of any cause
• Myocardial fibrosis
• Diabetes…

The list goes on. It can also occur—without significant symptoms—in an otherwise healthy elderly person.

In diastolic dysfunction, the basic problem is an increase in ventricular wall stiffness. This can affect either ventricle, though it is most commonly seen (or echo-reported) with the left ventricle, and simply reflects a loss of normal ventricular wall relaxation as the ventricle re-fills after the previous systolic contraction. Pretty much anything that can cause the wall of the heart to thicken or stiffen will produce this finding.

It is seen in CKD, of course, as many (if not most) CKD patients also have longstanding hypertension and/or a developing anaemia…both of which are major contributors to left ventricular hypertrophy (LVH; thickening). But, it is this change in architecture of the ventricular wall that alters the normal relaxation sequence…and this is by no means specific to CKD.

I also wouldn't want to say, as the questioner has implied, that “the toxins” of CKD5 are the primary cause. Rather, the alterations in wall elasticity and thickness that translate into diastolic dysfunction are slow mechanical changes that have developed over some considerable time in response to a range of factors. These factors are many, with the accumulating “toxins” of CKD being relatively unimportant.

For example, the syndrome we call “uraemic cardiomyopathy”—a rather more advanced cause of diastolic dysfunction—is more a product of prolonged hypertension +/- anaemia +/- salt and water overload +/- lipids and calcium, hyper-homocysteinaemia, and a range of other impacts on vascular wall health, than due to urea (or toxins) per se… even though, to be sure, toxins may play some role.

Will starting dialysis slow this process? Unfortunately starting early dialysis is not the cure for diastolic dysfunction the patient seeks…and an eGFR of 12-14 would still be regarded, for most patients, as an early start.

While effective dialysis, especially extended hour and higher frequency dialysis, will help to control the volume factors that contribute to LVH has been seen to regress with intensive dialysis, close attention to salt and water balance, BP control and the optimal management of pre-dialysis CKD5 anaemia are much more likely to be of benefit than any early start to dialysis. Early start dialysis is something that I would strongly argue against—and the data confirm that an early start in most patients is an incorrect approach.

As for cystatin C …while there are certainly proponents of this technique as a more accurate measure of eGFR, especially at low levels of residual renal function, this is also not as straightforward as it seems.

Cystatin C, while less dependent on age, gender, and muscle mass than creatinine, has not reliably been shown to be significant superior to a CKD epi-GFR. Cystatin C generation can also be influenced by:

• Body composition
• Thyroid status
• Corticosteroid use
• Some cancers
• Smoking
• A range of other factors

It is still, largely, an investigatory rather than a clinically integrated tool. At present, any (small) advantage is likely more than counterbalanced by cost, by clinical uncertainty of its place and value, and by a broader unfamiliarity with exactly what factors influence, and by how much, the interpretation of cystatin C levels.

On a practical level, the CKD-epi GFR is a satisfactory (if imperfect) marker. But, any method of CKD5 eGFR measurement is only relevant if taken into account with other clinical indicators. eGFR alone should neverbe used as a sole indication to start dialysis, rather, it should be used to help guide decision-making in CKD5…when to initiate pre-dialysis education and, through lifestyle impact, modality preferencing (eGFR 20), when to initiate vascular access creation (eGFR 18-20), and when to ensure all ducks are in a row for a smooth transition to the chosen 1^st preference dialysis (eGFR 12-14).

Most patients will not need to start (nor should they start) until the eGFR—by whatever means it is measured—drops into the 7-10 range and, even then, the decision to start dialysis should more commonly be made on clinical grounds, and not in response to an imperfect and unreliable laboratory concoction, be it eGFR, CKD-epi GFR, or cystatin C.