Should patients with autosomal dominant polycystic kidney disease be screened for cerebral aneurysms?
This is the title of a review paper presented in the January edition of the American Journal or Neuroradiology. It is a review of the current literature from a Chicago team, led by M Rozenfeld, triggered by the absence of any screening protocols for intracranial aneurysms in ADPKD patients.
Like the majority of people with ADPKD, I have not had any scans to look for cerebral aneurysms. When I was diagnosed (early 1980s), the association of PKD with intracranial aneurysms (IA) had certainly been described but the prevailing view was that risks associated with angiograms outweighed any benefit.
[As an aside here, the first cerebral angiograms in living humans were performed in 1920s by a neurologist, Egas Moniz, in Lisbon. He published over 60 academic papers on the topic. He was awarded a Nobel prize, but not for the work on angiography, rather for establishing frontal lobotomy as a treatment for mental illness.]
The first cerebral angiograms entailed injections into the carotid artery. Then in the 1970s a safer technique was perfected, digital subtraction angiography (DSA), which enabled intravenous injections of contrast to be used. The field advanced again in the next two decades and by 1990s Computed Tomography (CT) became the standard investigative procedure.
One of the problems with CT angiography of the head is the dose of radiation it entails – up to 1.9mSv for each procedure, which is equivalent to about 16 chest X-rays, or even more if no contrast medium is used. Though to put this into a different perspective, some people live in areas where the background radiation is more than this each year.
Recent advances in magnetic resonance imaging (MRI) has brought the issue of screening to a head (pun intended).
If you are interested, the physics of MRI angiography is described in intricate detail in this guideline paper. Much of it went over my head (OK, no more jokes) but it seems that it is now possible to obtain precise images of cerebral blood vessels using a process called “3T time-of-flight MRA” that will detect aneurysms as small as 3mm with around 79% sensitivity.
That means that for 100 aneurysms of 3mm diameter, 79 of them would be found by using a 3T TOF MRA. For larger aneurysms, above 5mm diameter, 95 out of 100 would be seen.
This scan does not require the use of contrast media.
Many people with PKD will have been warned away from scans with contrast – some very scary pictures and statistics on a side effect called “nephrogenic systemic sclerosis”. It appears to be triggered by certain forms of gadolinium contrast, used in MRI, but it is exceedingly rare if the GFR is over 30. (American College of Radiography, 2012)
The CT angiography scans use iodine based contrast media which is also potentially nephrotoxic. The rule of thumb there is that if GFR is greater than 60 the contrast is not likely to adversely affect the kidneys.
So we have reached a point where the scans can be performed without contrast and can detect pretty small aneurysms. What about the cost?
For CT cerebral angiography the costs are in the region of $6200
This includes the physician interpretation. An MRA can cost anything up to $10,000.
The cost effectiveness is harder to determine and depends on the proportion of screens that result in treatment and finding a measure for the reduction in morbidity and mortality this would bring about. The authors of the paper have considered this and conclude that costs compare favourably to other national screening programmes such as mammography and colonoscopy. They even suggest that in the terms of mortality it is as effective as use of seatbelt in cars.
How many people are we talking about?
Some background statistics might help here:
- ADPKD affects around 1 in 1000 people.
- Intracranial aneurysms affect between 4% and 20% of these people. (The data is highly variable ranging from a study from 1983 quoting 41% to a more recent one in 2002 quoting 14%)
- Autopsy-derived data on the general population gives a prevalence of 2-3% for comparison.
- 6% of all ADPKD patients will die from sub arachnoid haemorrhage – that is the bleeding that occurs when an intracranial aneurysm ruptures.
Some neuroanatomy might help too:
Where patients with ADPKD found to have cerebral aneurysms, nearly all occur in the anterior circulation of the brain.
In a 2011 study of 355 patients with ADPKD, 12.4% were shown to have aneurysms, and 1 in 5 of those had more than one found.
The only risk factor above PKD was a positive family history of a cerebral aneurysm, in which case nearly 22% of that group were affected. There was no relationship with blood pressure, renal function, age, duration of disease or presence of liver cysts. In other words the surest way to detect an aneurysm was by MRA scan and there were no additional pointers to its presence.
What does it mean to have an intracerebral aneurysm?
Obviously the most worrying thing about cerebral aneurysms is that they can rupture. The larger the aneurysm the greater the risk of rupture. Clinically this will present as a “subarachnoid haemorrhage”. There are 3 layers of tissue covering the brain – meninges. The arachnoid membrane is the middle one. A sub-arachnoid haemorrhage will be under the arachnoid membrane, pressing onto the very thin and delicate inner pia mater. A bleed like this carries up to 60% mortality and for the survivors some 40% will be permanently disabled. You may have some warning headaches or a stiff neck but usually it comes on out of the blue as a sudden “thunderclap” pain in the head.
Can this be prevented if an aneurysm is found by a scan?
The conservative approach we can all do whether we know about aneurysms or not: BP control, stop smoking, minimise other risks such as alcohol and oestrogen-contraceptive pills, treat atherosclerosis and high cholesterol. But the real benefit of knowing you have an aneurysm is surgery to clip, coil or embolise it. I am not going into the details but it is generally recommended to treat aneurysms over 7mm in diameter and new techniques are being established every year.
In this review Rozenfeld et al. have demonstrated three findings:
- Intracerebral aneurysms in ADPKD patients are prevalent and carry a significant morbidity
- The tests used to diagnose them are safe and efficient
- There is effective treatment which can extend life and reduce morbidity
Those are the factors necessary for an ideal screening programme.
- Screen ALL patients with ADPKD using a non-contrast 3T TOF MRA at the time of initial diagnosis
- Further scans should be done at intervals of 2-10 years, based on other clinical risks factors
- Coil embolisation should be undertaken for all over 7mm diameter, and some with specific features
- Aneurysms less than 7mm should be reviewed with regular repeat MRA
The paper was published in January 2014 so I searched to see if this recommendation had found its way into any guidance documents – the answer, sadly, is not yet!