Cavernous Carotid Aneurysms: To Do or Not To Do?

Unruptured asymptomatic aneurysms

These aneurysms are known to have an overall low risk of rupture and life-altering complications. Being enclosed in safe venous pouch, the cavernous sinus, these aneurysms tend to grow from large to giant size before they manifest clinically. In the ISUIA study, the 5-year cumulative rupture rate of unruptured CCAs was 0% for aneurysms with size ≤12 mm, 3% for aneurysms of size 13–24 mm and 6.4% for aneurysms >5 mm.[7] Similarly, risk prediction for bleed based on the PHASES score remains high for these aneurysms as the majority of these aneurysms tend to be giant.[8] However, for some reason, the common predictors of bleed for an incidental aneurysm such as hypertension, age, sex, and previous bleeds, do not apply strictly for CCAs There are several published series on successful conservative management of cavernous segment aneurysms the largest being by Stiebel-Kalish et al.[9] In their series of 132 aneurysms followed up for 4 years, 39 patients improved 21 remained unchanged, eight patients worsened and two died, suggesting a rather benign course compared to other aneurysms. The need for any prophylactic intervention purely based on the size of the aneurysm, thus, remains debatable.

Intervention for an asymptomatic aneurysm is justified in three situations. Aneurysms with an intradural component carry a risk of subarachnoid hemorrhage and need to be promptly excluded from the circulation. Such intradural components are often missed if not looked for specifically. Appropriate imaging of the dural rings is, therefore, mandatory for all CCAs. Similarly, an expanding aneurysm can result in sphenoid bone erosion, which can eventually result in fatal epistaxis. The presence of sphenoid bone erosion on bone window scans is the second indication for prophylactic intervention in an asymptomatic aneurysm. The third instance for preemptive treatment for a CCA would be an aneurysm seen to have increased in size on serial angiograms, especially in a young individual.

Symptomatic aneurysms

It is reasonable to assume that all symptomatic CCAs should be treated. Symptomatic aneurysms which have bled, which produce thromboembolism, which cause intolerable pain and deteriorating vision definitely merit treatment. Cranial nerve palsies other than vision, association with previously ruptured aneurysms, etc., are gray areas where treatment needs to be individualized. However, treatment is justified only if a successful cure can be offered without major complications. Both surgical and endovascular modalities of treatment essentially attempt to exclude the aneurysm from the circulation and initiate thrombosis. Except for direct clipping of the aneurysm none of the other modalities attempt to decompress the aneurysm and thereby reduce mass effect. Pain is one symptom which responds moderately well to treatment. Successful intervention certainly reduces the risk of further bleed and ischemic stroke. Intervention helps in arresting the progression of cranial nerve deficits and complete recovery seldom happens. The second cranial nerve is an exception and visual recovery following decompression has been reported in few cases.[12] Like other causes of cranial nerve deficits, the duration of the deficits before treatment can have an impact on the resolution of the deficit.

Unfortunately, all modalities carry a considerable risk of complications. Both endovascular and surgical options carry significant risks, and the overall mortality and morbidity varies from 3.23%–22.6% to 9.2%–14.8%, respectively.[1] The decision to intervene thus needs to be made judiciously and the type of treatment ought to be justified adequately.

Treatment modalities

Exclusion of the aneurysm from the circulation can be done either by surgery or by endovascular techniques. Treatment can be either occlusive or reconstructive. Occlusive strategies include parent artery ligation surgically or by endovascular techniques. Reconstructive strategies include direct microsurgical clip application, coil embolization with or without the use of a vascular reconstruction device, flow-diverting devices, or the use of liquid embolic agents. With the advent of flow diverters, the results have improved with an obliteration rate of 60%–80% with morbidity ranging from 10.44% (9.9%–15.2%) and mortality of 6.86% (2.3%–9.2%).[1] However, in developing countries like India, cost is a major limitation for endovascular techniques, and surgical option remains the mainstay of treatment.

Surgery

Literature provides reports of <150 cases of CCA which have undergone direct clipping so far. Direct clipping probably provides the chance of near 100% obliteration maintaining parent vessel patency at the same time. Of these Dolenc[1011] with 115 cases have the largest series and report a morbidity of 9.5% and morbidity 2.6% while Diaz et al.[12] with 15 cases report a 20% morbidity and 6.6% risk of mortality. However, the technique is complex and carries a high complication rate. Other authors have not been able to reproduce the same results and report a high risk of morbidity (17.45%) and mortality (31.27%).[1] Direct clipping is thus seldom preferred.

Occlusive surgical options include proximal carotid ligation with or without trapping the aneurysmal carotid segment by placing a clip proximal to the ophthalmic artery. Proximal carotid ligation carries a risk of ischemia to the ipsilateral hemisphere. It also carries a risk of new aneurysm formation on the contralateral side or increase in the size of any contralateral aneurysm if any. Risk of ischemia can be predicted by various techniques such as rate of venous filling, balloon test occlusion (BTO), single-photon emission computed tomography (SPECT), and positron emission tomography (PET). The risk of infarct is 32%–60% if carotid ligation is carried out without any preoperative assessment of cardiovascular reserve.[13141516] Tests of cerebrovascular reserve like BTO, SPECT, etc., are at times difficult to perform. A simple alternative is to look for a delay of more than 0.5 s in venous filling in the vascular territory of the occluded vessel which indicates nontolerance. This simple test described by Müller-Forell and Valavanis in 1995 and later validated by van Rooij et al. is a simple safe and effective in predicting tolerance to carotid occlusion and has become the current standard of practice in many centers.[1718] The risk of infarction comes down to 22% in patients who successfully complete all preoperative assessment tests.[1920] The risk, however, does not completely disappear. With an additional bypass, this risk comes down to 14.6% (1.8–29.4). A bypass procedure before parent artery occlusion is, therefore, preferable to reduce the risks of postocclusion stroke even in patients who tolerate BTO successfully (universal bypass).

Bypass could be an augmentative Superficial temporal to Middle cerebral artery (low flow STMC) bypass for patients with moderate cerebrovascular reserve or a replacement bypass (high flow EC-IC) bypass for those with poor cerebrovascular reserve. The choice between an arterial radial artery graft and a venous saphenous graft depends on the surgeon's preference. Radial artery graft lasts longer but has potential risk of spasm. Venous graft is easier to harvest but is prone to kinking and occlusion. Similarly, the choice between parent artery occlusion alone or trapping is difficult as it has been shown that there is no difference in complications and outcome between trapping and carotid occlusion.[13] Trapping supposedly prevents backflow into the aneurysm due to cross circulation and also prevents dislodged thrombi from reaching the IC circulation. Trapping is preferred to mere carotid occlusion for aneurysms with significant intradural extension and patients who demonstrate significant retrograde flow during the BTO.[21] The next controversy pertains to the choice of the proximal site for anastomosis. Common carotid artery, ICA and external carotid artery all have comparable results and the decision essentially depends on the surgeon's preference. We prefer the ICA or the external carotid artery to avoid manipulation of the carotid bulb situated close to the bifurcation.

Our patient presented with poor vision and intolerable pain justifying treatment. We did not perform a BTO, SPECT, or PET study for our patient as the cross compression study revealed poor cross circulation. Saphenous vein graft and ICA for proximal anastomosis was chosen due to surgeon's preference and past experience.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.