Thank you for asking me to comment on this case, which I think is difficult rather than controversial – clearly the options are to go for surgery or not and there are arguments in favour of either. Also, we need to remember the limitations of what we can learn from the outcome in a single case. If whichever path is finally chosen results in a ‘successful’ outcome, that is not the same as saying that the correct option was taken. Any option has a chance of success and a chance of failure, which we cannot now (or probably ever) calculate accurately. It is quite possible to miscalculate, opt for the low return option and get lucky.

A case such as this is particularly topical for two reasons. Firstly, it highlights the enormous limitations of evidence-based medicine. EBM has nothing to offer us here – don’t bother to consult the Cochrane collobaration or the latest issue of. Clinical Evidence Secondly, everyone is now very interested in surgical outcomes. The physicians can probably do what they want in this case without concern, but the surgeons have their mortality rates to consider. In a rational environment, it would be obvious that such infrequent operations (in the UK) as this cannot be meaningfully audited for their outcomes. It is not self-evident that consideration of surgical outcomes currently is always rational.

What we can do here is examine the case in great detail to see what we really know about this man. We also need to carefully formulate the questions that are making the decision about intervention difficult. The nature of such a case means there is always some data missing in the truncated summary. I’ll mention where this might have helped decision-making.

Analysis of the case

At first glance this looked like just another case of atheromatous renal artery stenosis (ARAS) but it turns out to have some quite interesting and unusual features. The failure of magnetic resonance angiography will be mentioned only to console those of us currently without access to this technique. (Another explanation is that the artery occluded between MRA and definitive angiography. If this was important, I’d expect to have seen a rise in serum creatinine, of which there is no mention in the case history). The split GFR results are fascinating – I would not have guessed that the small kidney with one occluded artery would have the same GFR as the big kidney with two arteries, one recently occluded but one unstenosed. I’m assuming that each of these arteries supplies about half the kidney. It is obvious therefore that all of the left kidney function and perhaps some of the right function depends on renal blood flow through collateral vessels. One thing we would like to know is whether or not restoring blood supply through the main arteries would improve renal function. There are some clues in the history as given, and probably more clues not given.

The adverse but reversible effect of ACE inhibition (ACEi) does not necessarily mean that surgical revascularisation will improve GFR. It could be that the GFR is angiotensin II dependent because of poor renal blood flow through collaterals and that revascularisation would improve this. It could also be that the roughly 30% of surviving nephrons are hyperfiltering, and the cessation of this following ACEi caused the rise in creatinine. Other explanations not pointing to reversible ischaemia include a coincident degree of volume depletion, or simply too much ACEi.

Whenever the contribution of ARAS to renal failure is considered, we also have to think of the other common problem in this setting which impairs GFR. This is an ill-defined but increasingly recognised nephropathy due to a combination of hypertension and severe atheroma – probably to be called athero-embolic nephropathy [1]. This causes progressive renal failure and is not going to be helped by revascularisation. A kidney with ARAS is just as likely to have this as the cause of poor function [2]. It is vitally important to distinguish between renal impairment due to low renal blood flow, which is therefore reversible by intervention, and renal impairment due to nephron loss because of other factors, which is not reversible. A recent study of split renal function in cases of unilateral stenosis due to atheroma or fibromuscular disease shows an average of 18 ml/min less GFR in the non-stenotic kidney when atheroma was the cause [3].

So what do I think is going on here? If the renal impairment was solely due to athero-embolic nephropathy, the creatinine should have been rising steadily to the current level. The patient has an extensive medical history and probably has had lots of blood tests over the years – a reciprocal creatinine plot would be interesting. A steady rise in creatinine would favour this diagnosis. Against the diagnosis is the asymmetry in kidney size.

If the renal impairment was solely due to ARAS, I’d expect the left kidney to have a lower GFR, and the right kidney GFR to be closer to 25% of normal total – about 25 to 30 ml/min. I’d also expect the creatinine to have risen after the unsuccessful angioplasty which occluded the upper right renal artery.

Perhaps the likeliest scenario is a bit of both! The left kidney had a longstanding ARAS, which encouraged collateral growth. When the artery occluded, the kidney remained viable due to the collaterals but many nephrons were lost causing low GFR and shrinkage. The reason that the better blood supply to the right kidney is not associated with better GFR may be that there was insufficient collateral growth and most of the upper pole died after occlusion. The kidney size was estimated before occlusion. A touch of athero-embolic disease might explain the less than perfect function in the lower half of the kidney.

This picture would suggest that revascularisation will not change right upper pole function, but might improve left kidney function. I don’t think a left renal biopsy would help – there is bound to be a lot of nephrosclerosis but the kidney is clearly viable. Further clues to help decide about the right would include a renogram or Doppler ultrasound to look at regional blood flow in that kidney. A lower pole biopsy wouldn’t help (that bit can’t be improved). It is of course the case that the smaller the kidney or the worse its function, the less likely benefit will be seen with revascularisation [4,5]. I am choosing to not be over influenced by this because I think there is a big difference between a small kidney beyond a stenosis of say 50 to 90%, and a small kidney beyond an occlusion with a documented GFR of 18 ml/min. In the former circumstance, significant loss of nephrons in the absence of a very tight stenosis is probably a sign of athero-embolic disease. This is the common scenario in many reports and it is not surprising that these kidneys don’t stabilise after intervention. The current case is very different.

So, in short, there is a possibility that successful revascularisation might improve GFR from 35 ml/min to maybe 50 ml/min. So what?

Possible benefits of successful revascularisation

Revascularisation is not an end in itself. Would it help the patient? A modest increase in GFR in itself is unlikely to make a big difference to the patient or his prognosis. It might reduce the risk of renal anaemia, which of course could be treated with EPO, at some expense, BP permitting.

Would revascularisation reduce the risk of eventual end-stage renal disease? Quite possibly not in this case – the kidneys are surviving with collateral supply and I haven’t seen any analysis of what happens to such kidneys in the medium to long term.

BP control might be helped, although it should be pointed out that the tablet burden in this case could be considerably eased by simply changing to once-a-day drugs.
Will the patient live longer? Possibly. I get the impression that high levels of angiotensin II are a bad thing. Successful revascularisation addresses this problem both directly, by switching off the stimulus, and indirectly, by allowing use of ACEi or AIIR blockers. (In the interests of fairness, it needs to be pointed out that the use of these might precipitate anaemia of a degree sufficient to require EPO treatment).
On balance, therefore, there is a sufficient possibility that revascularisation would help for us to consider the downside of surgery.

Possible risks of surgery

The information in the case summary doesn’t help us decide. There are two factors to consider – the patient, and the surgical team.

The concern with the patient relates to anaesthesia and possible sudden fluctuations in BP in either direction in someone who could have critical stenoses of carotid, coronary or mesenteric arteries. I would imagine that carotid dopplers and coronary angiography or some form of testing for reversible ischaemia would be sensible. The mesenteric vessels may have been visualised on renal angiography. As the main renal arteries are already occluded, the risks of precipitating renal failure may be less than when trying to improve on a tight stenosis, and principally related to episodes of hypotension.

The surgical team will need to provide not only surgical expertise for this type of operation, but aggressive pre-, peri- and post-operative management of BP and fluid state. This operation is rarely carried out in the UK. There are very few surgeons who could tell prospective patients meaningful results in similar previous cases.

Decision time

Ultimately, my job here would be to tell the patient what he needs to know to allow him to decide. If testing suggested he was reasonably fit for surgery, and if a good surgical team was available, I would summarise our hopes for intervention as possibly improving his kidney function (which may or may not make him feel any different), but more importantly allowing us to use the best drugs in this setting to control BP and prevent future related problems, i.e. ACEi or AIIRa. Unfortunately, the surgery would carry a small but not insignificant risk of major complication including death, stroke, heart attack, renal failure and peripheral or mesenteric ischaemia. Other than the risks, the strongest reason not to proceed would simply be that surgery might well make no difference to how he felt or his future health. Clearly, in such a complex case, there will be a large variation between patients in how they interpret what the various possibilities mean to them. A significant proportion will ask me to do whatever I think is for the best. Such a decision would be based on many factors which can never be addressed in this sort of exercise. If pre-operative assessment was favourable (and I suspect it might well not be), if the patient was keen on intervention, and if a suitably skilled surgical team were available, then this is case where surgery might have something to offer. In this particular case, because the benefits of surgery are not clearly defined in advance, it would not take much in the way of extra anaesthetic/operative risk to dissuade me from recommending that approach.

Whether or not surgery is performed, this patient has a high risk of coronary artery disease and should take aspirin and almost certainly an HMG CoA reductase inhibitor.

I await the surgeon’s opinion with interest.

Conclusion

The left kidney has an occluded renal artery and a GFR of 18ml/min – presumably filtering blood from a collateral supply. I don’t know of any studies that tell us if revascularisation in this setting improves GFR, prevents eventual end-stage renal failure, and/or allows ACE inhibitor use without an acute decline in GFR. Given that these are at least reasonable theoretical possibilities, intervention should be considered if the surgical expertise is available and significant vascular disease in other vital organs is absent.

The function in the right kidney is similarly low despite adequate perfusion of the lower half. It seems less likely that there is much to salvage in the upper half – a repeat ultrasound scan and possibly renogram six weeks after the occlusion might confirm that.

References

1. Suresh M, Laboi P, Mamtora H, Kalra PA. Relationship of renal function to proximal renal arterial disease severity in atherosclerotic renovascular disease. Nephrol Dial Transplant 2000; 15: 631-6.

2. Farmer CKT, Reidy J, Kalra PA, Cook GJR, Scoble J. Individual kidney function in atherosclerotic nephropathy is not related to the presence of renal artery stenosis. Lancet 1998; 352: 288-9.

3. La Batide-Alanore A, Azizi M, Froissart M, Raynaud A and Plouin P-F. Split renal function outcome after renal angioplasty in patients with unilateral renal artery stenosis. J Am Soc Nephrol 2001; 12: 1235-41.

4. Beutler JJ, Ampting JM, ven PJ et al. Long term effects of arterial stenting on kidney function for patients with ostial atherosclerotic renal artery stenosis and renal insufficiency. J Am Soc Nephrol 2001; 12: 1475-81.

5. Harden P, MacLeod MJ, Rodger RSC et al. Effect of renal artery stenting on progression of renovascular renal failure. Lancet 1997; 349: 1133-36.