Postoperative imaging
Perioperative imaging can be useful to detect complications during and/or immediately after surgical clipping [12, 13]. In particular, intraoperative angiography has proven to be useful in recognizing residual aneurysm and vessel flow compromise resulting in a rapid treatment [12, 13]. The most recent SAH guidelines of the American Heart Association (AHA) recommend, after any aneurysm repair, an immediate cerebrovascular imaging to identify remnants or recurrence of the aneurysm that may require treatment (class I; level of evidence B). We agree with this recommendation and, in addition, we believe that, as soon as possible after cerebral aneurysm/s securing, the assessment of the neurological status is advisable in the absence of contraindications to sedation hold such as ICP instability, radiological signs of intracranial hypertension, and severe respiratory failure.
DCI-related vasospasm monitoring
DCI, especially if associated to vasospasm, is a major cause of death and disability after SAH [2, 7]. In this regard, a rapid diagnosis and treatment is of paramount importance to prevent cerebral infarction [2, 7]. Serial neurological examinations are fundamental to detect the occurrence of new ischemic insults — i.e., especially in awake patients or in those who may undergo a reliable neurological evaluation — but they are of limited sensitivity in patients with poor clinical grade [3, 5, 7]. Regarding neurological examination and according to available literature [14], we consider suggestive of DCI-related vasospasm the development of a new focal or a global neurologic deficit or a decrease of 2 or more points on the GCS score that lasts for at least 1 h and cannot be explained by another cause (i.e., seizures, hydrocephalus, etc.). Intermittent screening or continuous monitoring methods can be also useful for the detection and confirmation of DCI, especially in sedated or poor-grade SAH patients [5, 7, 14]. TCD/TCCD examination is commonly used as a noninvasive tool to monitor for the presence of cerebral vasospasm following acute SAH [3, 7]. TCD/TCCD has a good sensitivity and specificity to detect vasospasm in the proximal segments of the middle cerebral artery (MCA) and internal carotid artery (ICA) but is less reliable regarding anterior cerebral artery (ACA) branches and posterior circulation arteries [3, 7]. The Lindegaard ratio (the ratio of mean MCA flow velocity divided by mean ICA flow velocity) is generally utilized to diagnose vasospasm in the MCA when the ratio is greater than 3 [3, 7, 12]. However, the sensitivity and specificity of TCD for cerebral vasospasm detection is operator dependent and some patients do not have adequate temporal bone windows to allow the detection of TCD signals [3, 7, 12]. According to available literature [15], we consider suggestive of vasospasm an increase in mean FVMCA of more than 50 cm/s from basal over 24 h and/or a mean FVMCA of at least 120 cm/s (with a suggestive Lindegaard ratio). Considering the above, DSA is considered the gold standard for the detection of cerebral vasospasm [7, 16]. Also, CTA can be used in this setting [16]. We recommend the utilization of CTA and/or DSA to confirm the presence of cerebral vasospasm as the cause of the DCI in case of neurological signs or TCD/TCCD suggestive for vasospasm. Moreover, considering that perfusion imaging with CT or MRI can be useful to identify regions of potential brain ischemia [3, 5, 7, 12], we recommend their use early in SAH patients with DCI related to cerebral vasospasm. This evaluation could be very important to modulate the intensity of care according to the extent of brain damage. Despite their possible utility and likely important applications, we have not considered continuous monitoring methods (i.e., brain tissue oxygenation monitoring, continuous electroencephalogram, etc.) because at present they are not yet widely available.
DCI-related vasospasm prophylaxis
Nimodipine, a dihydropyridine calcium channel antagonist blocking the flux of extracellular calcium via voltage-gated calcium channels, is the only therapeutic agent with class I evidence for decreasing the risk of poor outcome in SAH [3, 7]. The beneficial effects of nimodipine are related to the reduction of delayed cerebral infarction occurrence but no effect on cerebral vasospasm was detected in large clinical trials. Thus, different potential mechanisms were proposed such as reduction of calcium-dependent excitotoxicity and reduced platelet aggregation [17]. It should be administered orally or by nasogastric tube at a dose of 60 mg every 4 h for 21 days [16]. In case of feeding intolerance, nimodipine should be applied intravenously [18]. The dose needs to be reduced or discontinued in case of arterial hypotension. According to above, we recommend:
-
Administration of oral nimodipine (60 mg every 4 h) after bleeding for DCI prevention (administration of intravenous nimodipine at 2 mg/h in case of feeding intolerance)
-
No administration of oral/intravenous nimodipine in hemodynamically unstable SAH patients (i.e., under inotropes and/or vasopressors therapy)
-
To withhold oral/intravenous nimodipine in case of a significant drop in arterial blood pressure
Regarding this last point, it must be considered that the optimal MAP target, in SAH patients without vasospasm and intracranial hypertension after cerebral aneurysm treatment, has yet to be established. An individualized approach for blood pressure management, especially in unconscious patients, could be performed with a multimodality neuromonitoring (brain tissue oxygenation monitoring, electroencephalography, invasive quantitative cerebral blood flow monitoring, cerebral microdialysis, and electrocorticography) [19]. Unfortunately, not all centers worldwide have this possibility. For this reason, we recommend to maintain SAH patients slightly hypertensive with the maintenance of a MAP between 80 and 100 mmHg (close to the lower limit for patients without a history of arterial hypertension and close to the upper limit for patients with a history of arterial hypertension).
Magnesium is a non-competitive calcium antagonist with several vascular and neuroprotective effects (i.e., vasodilation, reduction in glutamate release, etc.) [20]. At the moment, the induction of hypermagnesemia for the prevention of DCI-related vasospasm is not supported while hypomagnesemia should be avoided [7, 16]. In this regard, we recommend the maintenance of a magnesium blood level in the normal ranges (1.6–2.5 mg/dl to 0.65–1 mmol/L) in all SAH patients for 21 days after bleeding for vasospasm prevention.
DCI-related vasospasm management
Despite DCI has a complex multifactorial pathogenesis, ischemia from vasospasm is one of the most important potentially clinically reversible factors [3, 19]. Vasospasm typically occurs between 3 and 14 days post-bleeding, although it can occasionally persist up to 21 days [3, 7]. Angiographic vasospasm may or may not be clinically symptomatic [3]. Precisely, some degrees of vasospasm are visible angiographically in up to 70% of patients with SAH but only 30% of all patients develop clinical symptoms (symptomatic vasospasm) [3]. Symptomatic vasospasm, being associated with DCI and poor outcome following SAH, requires a prompt intervention [3, 19]. Generally, the first step in case of DCI-related vasospasm is to increase arterial blood pressure maintaining the patient euvolemic [5, 16, 19]. In this case, the co-administration of oral/intravenous nimodipine is not advisable [16]. We recommend a gradual increase of MAP (up to 120 mmHg), maintaining euvolemia, aimed to the resolution of clinical symptoms and/or radiological findings. This process should take into account the patient’s cardiovascular status to minimize the risks associated with MAP augmentation. We recommend that CT perfusion and/or advanced multimodal neuromonitoring (i.e., brain tissue oxygenation monitoring, electroencephalography, invasive quantitative cerebral blood flow monitoring, cerebral microdialysis, and electrocorticography), if available, should be utilized to guide (individualize) therapy for DCI associated with cerebral vasospasm in poor-grade SAH patients where neurological assessment is not possible. This strategy is in agreement with the available literature [5, 16, 19].
Generally, a variety of invasive intra-arterial procedures (angioplasty, vasodilators) are utilized in case of vasospasm refractory to MAP augmentation [5, 16, 19, 21]. Percutaneous transluminal balloon angioplasty (PTCA), based on mechanical stretching/dilation of vasospastic arteries, can be utilized only in case of proximal vessel vasospasm (i.e., internal carotid artery, M1 segments of the MCA) [19, 21]. This type of treatment, with respect to vasodilators, has a high success rate and is long-lasting [19, 21]. However, PTCA can present serious complications including embolism, thrombosis, dissection, and vessel rupture [19, 21]. Over the years, numerous intra-arterial vasodilators have been evaluated such as papaverine, nicardipine, verapamil, nimodipine, and milrinone [7]. Possible advantages of intra-arterial vasodilators with respect to PTCA are a distal and more diffuse effect and a better safety profile [7,8,9,10,11,12,13,14,15,16,17,18,19,20,21]. In some cases (distal + proximal vessels vasospasm), they can be used in conjunction with PTCA [21]. Possible disadvantages of vasodilators are recurrent vasospasm due to the short-lasting effect of these agents, intracranial hypertension due to vasodilation, and arterial hypotension due to systemic effects [21]. Data demonstrating the superiority of one method of treatment with respect to another are not available and more studies are necessary. Considering the above, we recommend, in case of symptomatic vasospasm refractory to MAP augmentation, the utilization of invasive intra-arterial procedures. In our opinion, the choice of intra-arterial procedure — i.e., vasodilators, angioplasty — should be individualized after discussion with the interventional neuroradiologist.
Increasing cardiac output with inotropes (i.e., dobutamine, milrinone) can improve brain perfusion after SAH [21]. A trial of inotropic therapy has been suggested if DCI-related vasospasm does not improve with blood pressure augmentation [16]. In this case, the utilization of an advanced hemodynamic monitoring (i.e., arterial pulse contour/waveform analysis, pulmonary artery catheter, transpulmonary thermodilution, ultrasound, etc.) was generally utilized [21]. We recommend the utilization of inotropes for refractory vasospasm with the utilization of an advanced hemodynamic monitoring. Milrinone, possessing a mechanism of action for the reversal of cerebral vasospasm as well as potentially anti-inflammatory effects, has been identified as a promising therapeutic agent for DCI [22]. Several recent preliminary studies showed a potential benefit of milrinone for the management of DCI-related vasospasm and encourage the conduction of confirmatory randomized trials [23,24,25]. Awaiting more data from well-powered studies, we are unable to provide any recommendation regarding the utilization of milrinone in case of DCI related to refractory vasospasm.
Anemia, reported in more than 50% of SAH patients, is associated with poor outcome [3, 5, 16, 21]. The appropriate target Hb concentration in SAH patients with and without DCI is unknown [3, 16]. Hb levels < 9g/dl are associated with brain tissue hypoxia and metabolic distress in poor-grade patients [26]. Packed red blood cell (RBC) transfusion increases brain tissue oxygen tension in poor-grade SAH patients with a baseline hemoglobin level of 8 g/dL [27]. RBC transfusion seems to be useful to optimize cerebral oxygen delivery in case of brain ischemia related to vasospasm. According to Neurocritical Care Society (NCS) guidelines, patients should receive packed RBC transfusions to maintain Hb concentration above 8–10 g/dl. Awaiting more data from well-powered studies, we recommend the maintenance of a Hb level > 8 gr/dl in poor-grade SAH patients without DCI-related vasospasm and > 9 gr/dl in case of DCI associated with cerebral vasospasm.
Fever is associated with a high rate of DCI and unfavorable outcomes after SAH [3, 28]. Temperature should be monitored frequently and infectious causes of fever should always be investigated and treated [16]. An aggressive control of fever, especially during the period of risk for DCI, is advisable [5, 16]. The intensity of fever control should be proportional to the risk of cerebral ischemia [16]. We recommend a continuous BCT monitoring in poor-grade SAH patients. Being fever (regardless of the cause that needs to be investigated) associated with poor outcome after SAH, we recommend the administration of antipyretics for a BCT > 37.5 in poor-grade SAH patients without DCI-related vasospasm and the maintenance of normothermia (BCT 36–37 °C) in case of DCI associated with cerebral vasospasm. Therapeutic hypothermia has been suggested as a possible rescue therapy for refractory symptomatic vasospasm [21], but considering their potential side effects [29] and the lack of robust data in this setting, we are unable to provide any recommendation regarding the utilization of therapeutic hypothermia in case of DCI related to refractory vasospasm. The panel expressed the same concerns for the utilization of metabolic suppression.
ICP monitoring and intracranial hypertension management
Intracranial hypertension after SAH is associated with poor outcomes [30,31,32]. Existing SAH international guidelines [5, 16, 18] do not provide specific recommendations regarding ICP/CPP monitoring and treatment despite there being several reasons for doing so [33]. Considering the above, we recommend:
-
ICP monitoring in all salvageable SAH patients in coma (GCS ≤ 8) with radiological signs of intracranial hypertension
-
The management of elevated ICP in all salvageable SAH patients (aneurysm/s secured) taking into account the underlying pathophysiological mechanism responsible of intracranial hypertension
-
The maintenance of a CPP ≧ 70 mmHg in patients with intracranial hypertension
Limitations
This paper, similarly to the first part [9], has several limitations. We deliberately did not base our statements on systematic literature reviews because of the lack of evidence (previously underlined) and in favor of simple, basic topics that are rarely subject to investigation. We decided to produce the clinical questions and recommendations on the basis of the panel’s expertise and using a pragmatic approach based on both literature and clinical experience to provide support.