Screening for Intracranial Aneurysms in Patients with Autosomal Dominant Polycystic Kidney Disease

Patient Presentation

A 40-year-old patient asks for a second opinion on the management of her autosomal dominant polycystic kidney disease (ADPKD). Her first cousin, who also has ADPKD, had a subarachnoid hemorrhage a few months earlier, leading her previous nephrologist to request a cerebral magnetic resonance angiography (MRA) that reveals a 3 mm aneurysm of the left anterior cerebral artery. Your colleague had previously discouraged the patient to perform screening in the absence of known familial history.

Prevalence of Intracranial Aneurysms, Risk of Rupture, and Consequences

The prevalence of intracranial aneurysm has been estimated between 9% and 12% of patients with ADPKD (1–3). In comparison, intracranial aneurysm prevalence in the general population is between 2% and 3%. The only identified risk factors for intracranial aneurysm in ADPKD are a personal or a family history of unruptured or ruptured intracranial aneurysm. In patients with a familial risk of intracranial aneurysm, the prevalence of intracranial aneurysm has been reported to be around 21%–22%, compared with 6%–8% in the absence of such family history (1,2).

The rupture of an intracranial aneurysm is the most serious complication of ADPKD, with a mortality rate of 10%–20% and a morbidity rate of 50%, related to the neurologic sequelae of the subarachnoid hemorrhage.

The rupture rate has rarely been reported in the literature. In our prospective experience, the global annual intracranial aneurysm rupture rate was 1.9 (95% confidence interval, 0.84–4.5) per 1000 patient-years, much higher than the risk of subarachnoid hemorrhage in the general population (0.06–0.10 per 1000 person-years), but also higher than previously reported in a large, retrospective study of patients with ADPKD (0.5 per 1000 patient-years) conducted by the Mayo Clinic between 1945 and 1984 (4).

Is there a Specific Natural History of Intracranial Aneurysms in ADPKD?

There is an abundance of literature on natural history detected in the general population, on the basis of large, prospective series. However, within these series the percentage of patients with ADPKD was very low. Identifying ADPKD intracranial aneurysms from what is known about intracranial aneurysms in the general population is possibly misleading. Indeed, the natural history of intracranial aneurysms in ADPKD is singular in several ways: (1) aneurysm rupture occurs at an earlier median age than in the general population (41 versus 51 years of age); (2) small aneurysms can rupture (20% of ruptured intracranial aneurysms measured <5 mm and 33% between 5 and 9 mm, whereas intracranial aneurysm rupture is rare in the general population for aneurysms <7 mm); and (3) the majority of the reported ruptures concerned the anterior brain territory.The results from our study were consistent with previous reports in the literature, including results from a recent Finnish study that specifically focused on patients with ADPKD (5).

Targeted Screening versus Universal Screening

Divergent opinions on intracranial aneurysm screening strategy have been published in the literature. A survey of practices conducted among French-speaking nephrologists in Europe shows that about one third of participants were in favor of universal screening for intracranial aneurysms in patients with ADPKD, whereas two thirds were in favor of targeted screening (6).

In 2014, a Kidney Disease Improving Global Outcomes (KDIGO) expert panel recommended targeted intracranial aneurysm screening only in cases of family history of intracranial aneurysm or subarachnoid hemorrhage, previous intracranial aneurysm rupture, high-risk professions (e.g., airline pilots), and patient anxiety despite adequate information (7).

Collecting information about the possible presence of aneurysm in polycystic relatives is complex. What about a family case of sudden death or a case of stroke without further details? Should we collect this information at each consultation, and should we be notified when there is new familial information? What about small or poorly informative families? What about patients without other known cases of polycystic disease?

Despite these uncertainties, restricting screening to “at-risk” families results in the detection of intracranial aneurysms in a greater proportion of cases because, as mentioned above, there is indisputable family aggregation of intracranial aneurysms.

However, no study has ever shown that the rate of rupture of a given aneurysm is different depending on whether one belongs to an at-risk family. Indeed, it has been consistently reported in the literature that approximately 80% of intracranial aneurysm rupture events in patients with ADPDK occurred in the absence of known family history of aneurysm (3,5,8).

Decisions regarding screening policy should consider the costs associated with intracranial aneurysm screening and preventive treatments, as well as the iatrogenic risks associated with them, and compare them with the cost reduction and clinical benefits obtained by avoiding the occurrence of intracranial aneurism rupture. To do so, we performed a cost–utility analysis, using a Markov model with probabilistic sensitivity analyses and recent medical data, showing that universal screening provides a gain of 1.29 quality-adjusted life years (QALYs) compared with no screening and 0.68 QALYs compared with targeted screening. For a willingness to pay of €50,000/QALY, universal screening was more effective than targeted screening in 89% of simulations (3).

On the basis of the above evidence, we recently decided to change our policy and now propose intracranial aneurysm screening to all patients with ADPKD. Universal screening for intracranial aneurysm in ADPKD remains debated. The KDIGO conference, published before our cost–utility analysis, did not recommend universal screening because it identifies mostly small aneurysms said to be of a low rupture risk, and because of the risks associated with aneurysm repair. We acknowledge that the cost–utility analysis we performed was on the basis of the results obtained from French patients referred to our tertiary referral center, and that additional cost–utility analyses should be performed in other clinical settings to strengthen and generalize our findings. It must be noted that cost estimates, neuroradiological expertise, and preventive treatment complication rates may vary between countries. Interestingly, a recent cost–utility analysis performed in the United States also showed that universal screening for aneurysms in ADPKD was cost-effective compared with no screening (targeted screening strategy was not studied in this analysis) (9).

What to Say to Patients?

It is possible to tell patients, without going into all the numerical details mentioned above, that there is a slightly higher incidence of cerebral aneurysm in ADPKD than in the general population, their rupture is a very rare event, and if an MRA screening is decided, there are two possible results: (1) most likely, no vascular abnormality is found, or (2) in rarer cases, an aneurysm has been detected, and it will be decided whether it should be simply monitored radiologically or treated preventively, most likely by radiologic intervention.

Some nephrologists also stress the need for urgent medical evaluation in case of signs suggestive of intracranial aneurysm rupture (i.e., the sudden occurrence of intense, nonmigraine headache) (6).

Intracranial Aneurysm Screening Modalities

Noninvasive techniques (computed tomography angiography requiring iodinated contrast injection, and 3T time-of-flight MRA without gadolinium-based contrast agent injection) have similar sensitivity and specificity for the detection of small (2–3 mm) intracranial aneurysms (7). The absence of contrast injection justifies the use of MRA instead of computed tomography angiography in this population of patients with compromised kidney function. Screening should be performed in an experienced neuroradiology center. The prescription should mention the context of ADPKD and require exploration of all cerebral arteries, from the occipital foramen to the corpus callosum.

A retrospective study (8) showed that 10% of aneurysm ruptures occurred before 21 years of age in ADPKD, and the youngest patient with aneurysm rupture in this study was 15 years old. The cost-efficiency of universal screening was maximal at 20 years, decreases with age, and was unproven after 60 years. MRA screening should be performed as early as possible once the diagnosis of ADPKD is made, ideally around 20 years of age. Because de novo aneurysms may develop in adults, MRA should be repeated every 5–10 years up to 60 years of age, to maintain screening efficiency (3,9).

Treatment of Unruptured Intracranial Aneurysms

The management of unruptured intracranial aneurysms should be discussed by a multidisciplinary team (neurologist, neuroradiologist, neurosurgeon) in an expert center. Some nonspecific measures are recommended to all patients to decrease the risk of intracranial aneurysm growth and rupture: smoking cessation, reduced alcohol intake, and optimal BP control. Aneurysms considered at low risk of rupture should be reassessed by serial imaging, with initial 6-month intervals and further spacing in case of stability. Aneurysms at high risk of rupture will benefit from endovascular or neurosurgical treatment.

Endovascular techniques (coil embolization of the aneurysm, sometimes with additional stent placement) have gradually improved over years. Their indications have extended whereas those of classic neurosurgery have progressively decreased. Neurosurgery is still recommended for very large aneurysms and those for which embolization is impossible. The complication rate of the preventive procedures has been specifically evaluated in patients with ADPKD. A recent registry study found a higher rate of iatrogenic events for both neurosurgical treatment (11.8% versus 6.4%) and endovascular treatment (9.4% versus 3.0%) in polycystic patients, although for patients with ADPKD, no procedure-associated mortality was detected in this study, and both hospital length-of-stay and hospital costs were lower in patients with ADPKD (10). These mitigated results of iatrogenic morbidity were taken into account in our cost-effectiveness analysis, which nonetheless indicates that widespread detection of aneurysms may reduce morbidity and mortality related to intracranial aneurysm rupture in patients with ADPKD (3).

Disclosures

Dr. Joly reports personal fees and nonfinancial support from Otsuka, and nonfinancial support from IPSEN. Dr. Flahault has nothing to disclose.