As neuroanesthetic induction and maintenance must balance the preservation of stable hemodynamics and perfusion of cerebral tissues with alterations to intracranial pressure that such perfusion may incur, it is pertinent to note the systemic effects of these various anesthetic agents in neurosurgical procedures ^[1][3]. This consideration not only ensures that the metabolic demands of the organs of the central nervous system are met, but also reduces injury and trauma to them. Most anesthetic agents, with the exception of ketamine, reduce the overall global oxygen metabolism of the brain, whereas the cerebral blood flow is increased by volatile anesthetics, nitrous oxide, and ketamine but reduced by other intravenous agents ^[1][3]. Such changes in physiological parameters impinge upon several endogenous mechanisms including cerebral autoregulation, vasomotor reactivity, and neurovascular coupling, of which anesthetic agents have varied effects ^[1][3].

Cerebral autoregulation entails adjustments to vascular tone that are sensitive to both long-term and short-term changes in systemic blood pressure ^[2][9]. These changes are hypothesized to be regulated myogenically or endothelially in response to stress on the vasculature, neurogenically by the autonomic nervous system and neurons, and to some extent metabolically ^[3][10]. Volatile anesthetic agents have been found to reduce cerebral autoregulation at higher doses, and propofol in particular has been found to reduce it at levels >200 mcg/kg/min ^[1][3]. In addition, synthetic opioids have been found to increase cerebral blood flow by vasodilation as a result of such cerebrovascular autoregulation ^[4][11]. The vasomotor reactivity of cerebral arterioles arises from pH changes to vasculature and surrounding smooth muscles, and as such, is largely dependent on the partial pressure of carbon dioxide: increases in PaCO₂ have been shown to have vasodilatory effects, increasing blood flow ^[1][3]. The vasomotor reactivity is reduced at higher concentrations of certain volatile agents and has been shown to be variably affected by propofol ^[1][3]. Neurovascular coupling can be thought of as a combination of phasic and tonic vasomodulation regulated by the production of substances and metabolites by neurons and astrocytes, respectively. The role of these various anesthetic agents on neurovascular coupling has yet to be fully determined but they have been shown to affect neural activity as well as subsequent vasoactive signal transmission, and vasoactive response as well ^[1][5][3,12]. Most notably, etomidate, propofol, and barbiturates have been found to decrease cerebral blood flow secondary to decreased cerebral metabolic demands ^[1][3].

Intracranial pressure, determined by the pressure within the skull exerted by its content in the form of blood volume, tissue volume, and cerebrospinal fluid volume, is functionally related to cerebral perfusion pressure, and affected by overall cerebral blood flow ^[6][13]. As such, alterations in cerebral blood flow following the use of anesthetic agents subsequently coincide with changes in intracranial pressure, as demonstrated through the Cushing reflex equation: where CPP is cerebral perfusion pressure, MAP is mean arterial pressure, and ICP is intracranial pressure ^[3][10].

The first mention of total intravenous anesthesia (TIVA) in the literature dates to 1872 in a report by Pierre-Cyprine describing the use of chloride hydrate ^[7][15]. Following the introduction of propofol in 1977, the archetypal TIVA, TIVA utilization rapidly increased ^[8][16]. Propofol increases GABA-mediated chloride channels in the brain resulting in inhibitory tone in the CNS. Furthermore, the drug increases the duration of the GABA’s effects by decreasing dissociation from its receptor, leading to the hyperpolarization of the cell membrane ^[9][17]. Its effects have a rapid onset and are short-acting, which allows for rapid recovery post-surgery and evaluation of neurological function ^[10][18]. Propofol also decreases intracranial pressure, cerebral blood flow, cerebral metabolism, and edema while supporting cerebral perfusion pressure and mean arterial pressure. This aggregate of effects is neuroprotective during cerebral ischemia ^[11][19]. Furthermore, the administration of propofol with an opioid provides hypnosis, amnesia, and minimizes response and movement to surgical stimulation, achieving all of the factors of a true anesthetic while also decreasing postoperative nausea and vomiting ^[8][10][16,18]. TIVA is now generally classified as the combined use of a hypnotic agent (e.g., propofol) and an opioid (e.g., fentanyl or remifentanil) without concurrent use of inhaled anesthetics ^[12][13][20,21]. Additionally, TIVA is conducive to new considerations that have emerged in neurosurgery anesthesia such as minimizing affected brain function and electrophysiological monitoring. Due to propofol’s rapid onset and short duration, infusion rates can be adjusted to allow for patient cooperation when necessary during a procedure ^[10][18]. Current common intravenous agents include thiopentone, propofol, etomidate, ketamine, benzodiazepines, and opioids ^[11][19].

3. Volatile Anesthetics

Diethyl ether was identified centuries ago and may have been originally compounded by an 8th-century Arabian philosopher. However, it was not until 1842 that the first use of ether for surgical anesthesia was documented ^[14][22]. More than 150 years after this discovery of a form of general anesthesia, volatile agents are still in clinical use today ^[15][23]. While the precise mechanism of action of volatile anesthetics remains largely unknown, studies have demonstrated that they are active in the central nervous system, augmenting GABA receptors and stimulating potassium channels ^[16][8]. Additionally, these agents have a role in depressing excitatory pathways including acetylcholine receptors, glutamate receptors, serotonin receptors, muscarinic receptors, and nicotinic receptors ^[16][8]. They are most often administered through a face mask, laryngeal mask, or endotracheal tube. While induction with volatile agents is often preferred in infants and young children because it allows for a needleless experience while awake, adult patients typically undergo an intravenous induction to reduce risks inherent to a full-grown patient. In a randomized double-blind comparison of 8% sevoflurane and propofol as anesthetic agents, Thwaites et al. found that induction with sevoflurane was associated with a lower rate of apnea, decreased time to establish spontaneous ventilation, and a smoother transition to maintenance. Furthermore, emergence was found to be earlier with sevoflurane ^[17][24]. While volatile agents have few absolute contraindications beyond gene variants for malignant hyperthermia, in the field of neurosurgery, these agents have concerns as they have been shown to decrease cerebral perfusion and increase intracerebral pressure ^[13][18][21,25]. Some of the most common volatile anesthetics in use today include halothane, isoflurane, desflurane, and sevoflurane.

4. Advantages and Disadvantages of Anesthetic Options

When administering anesthesia to neurosurgical patients, the onus is on the anesthetist to gauge whether inhaled or intravenous anesthetics would be more advantageous. Three of the most significant considerations regarding a neurosurgical anesthetic are the effects on hemodynamics, intracranial pressure, and postoperative conditions ^[11][19]. Hemodynamic stability is needed to maintain cerebral autoregulation ^[19][26]. Furthermore, the intrinsic autoregulation of factors (e.g., partial CO₂ pressure and mean arterial pressure) that influence hemodynamic stability may be lost via the circumstances precipitating neurosurgical management ^[20][21][27,28]. Therefore, hemodynamic consideration is especially important within the context of neurosurgical anesthesia. In a randomized study, Strebel et al. demonstrated that volatile agents can impair autoregulation while propofol is helpful in preserving it ^[22][29]. Furthermore, Van Hemelrijck et al. demonstrated that the use of a propofol-loading infusion did not change the measure of blood pressure or heart rate, demonstrating the benefits of TIVA within the context of hemodynamics ^[23][30]. Low intracranial pressure is a significant factor allowing for optimal operating conditions and can also be neuroprotective. In a randomized prospective study, Petersen et al. found that subdural intracranial pressure was lower and cerebral perfusion pressure was higher in patients anesthetized by propofol compared to sevoflurane- or isoflurane-anesthetized patients ^[24][31]. This finding suggests that TIVA is beneficial in neurological surgery by minimizing local hypoperfusion and cerebral ischemia. Following surgery, rapid recovery from anesthesia is important to allow for neurological examination. Regarding the duration of effect, both TIVA with propofol-remifentanil and volatile agents have short half-lives, allowing for a rapid recovery after surgery ^[25][32]. This is supported in the literature as demonstrated by a randomized sample of patients undergoing craniotomy. Here, Talke et al. demonstrate that there was no difference between early post-operative recovery variables such as recovery time and early cognition ^[26][33]. An appraisal of the present literature of randomized-controlled trials comparing anesthetic agents head-to-head across various peri and postoperative parameters are included in Table 1 and Table 2 ^{[23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64]}[30,31,32,33,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72].

Authors and Year	Surgical Procedure	Comparison	Findings
Aken et al., 1990 [27][35]	Unspecified cranial procedure	Balanced anesthesia (loading thiopental and fentanyl + maintenance fentanyl, droperidol, thiopental, and isoflurane in nitrous oxide, n = 20) vs. TIVA (loading propofol + alfentanil infusion, n = 20)	During induction, TIVA had a significantly greater hemodynamic stability. Balance anesthesia was associated with a significantly longer emergence time than TIVA.
Hemelrijck et al., 1991 [23][30]	Craniotomy for resection of brain tumor	Propofol (n = 20) vs. thiopental (n = 20)	Postoperative return to orientation time was shorter in the propofol group (7 +/− 5 min vs. 27 +/− 23 min).
Ornstein et al., 1993 [28][36]	Craniotomy for resection of supratentorial lesion	Anesthetic maintenance via desflurane (n = 12) vs. isoflurane (n = 12)	CBF values were non-significantly different as measured at 1 MAC and 1.5 MAC concentrations for both desflurane and isoflurane (p > 0.05), as well as at 1.25 MAC as measured in n = 15 patients (p > 0.05).
Talke et al., 1996 [29][37]	Hypophysectomy	Propofol (n = 10) vs. loading propofol + maintenance desflurane (n = 10) vs. loading propofol + maintenance isoflurane (n = 10)	Minimum CPP was significantly lower in desflurane (p < 0.05) and isoflurane (p < 0.05) groups compared to propofol-only control. Minimum SBP was significantly lower in desflurane (p < 0.05) and isoflurane (p < 0.05) compared to propofol-only control.
Artru et al., 1997 [30][38]	Unspecified cranial procedure	Anesthetic maintenance via sevoflurane (n = 8) and isoflurane (n = 6) following induction via mannitol	Neither sevoflurane or isoflurane significantly altered ICP, and both decreased middle cerebral artery flow velocity (Vmca). Notably, decreased Vmca with sevoflurane was related to decreased CPP at 0.5 MAC (p < 0.05), and increased CVRe at 1.0 and 1.5 MAC (p < 0.05). The CPP decreased from baseline at 0.5, 1.0, and 1.5 MACs of isoflurane (p < 0.05).
Hoffman et al., 1998 [31][39]	Craniotomy for unspecified pathology	Thiopental induction (n = 10) vs. desflurane (n = 10)	Neither thiopental nor desflurane changed tissue gases or pH, but desflurane increased PO2 70% (p < 0.05), whereas thiopental decreased PO2 30% during temporary brain artery occlusion.
Talke et al., 1999 [32][40]	Transsphenoidal Hypophysectomy	Anesthetic maintenance via propofol (n = 10) vs. sevoflurane (n = 20)	Sevoflurane increased lumbar CSF pressure and decreased CPP and systolic blood pressure following infusion while propofol did not affect lumbar CSF pressure, CPP, nor systolic blood pressure.
Talke et al., 2002 [26][33]	Craniotomy for resection of supratentorial lesion	Propofol (n = 20) vs. isoflurane (n = 20)	Emergence time to eyes opening was not different between anesthetic agents (p > 0.05). There was no difference in occurrence of hypertension (p > 0.05).
Iwata et al., 2003 [33][41]	Unspecified intracranial surgery	Propofol (n = 13) vs. sevoflurane (n = 13)	There was no difference in the rate of temperature decrease and recovery in induced hypothermia (p < 0.05).
Fraga et al., 2003 [34][42]	Craniotomy for resection of supratentorial lesion	Inhalation of isoflurane (n = 30) vs. desflurane (n = 30) following induction via fentanyl, thiopental, and vecuronium maintained with 60% nitrous oxide in oxygen	There were no significant differences between MAP, ICP, and CPP between use of desflurane and isoflurane, but notable decreases (p < 0.05) in both groups from baseline values with regard to MAP and CPP. The ratio between the cerebral metabolic oxygen requirement and cerebral blood flow decreased significantly for both groups as well.
Petersen et al., 2003 [24][31]	Craniotomy for resection of supratentorial tumor	Propofol (n = 41) vs. isoflurane (n = 38) vs. sevoflurane (n = 38)	No differences in ICP or CPP between anesthetic agents (p > 0.05).
Günes et al., 2005 [35][43]	Unspecified intracranial procedure	Anesthetic maintenance via propofol (n = 39) vs. dexmedetomidine (n = 39)	Systolic blood pressure and MAP were not different between the two agents. Extubation time was shorter for propofol (p < 0.05). Analgesic requirements were higher for propofol (p = 0.013).
Magni et al., 2005 [25][32]	Supratentorial craniotomy for unspecified pathology	Propofol (n = 64) vs. sevoflurane (n = 64)	Emergence time was not different between anesthetic agents. Occurrence of hypertension was higher in propofol than sevoflurane use (p = 0.0046), and hypotension was higher in propofol than sevoflurane (p = 0.02).
Sekimoto et al., 2006 [36][44]	Craniotomy for resection of brain tumor	Anesthetic maintenance via halothane vs. isoflurane vs. sevoflurane after induction via propofol/fentanyl/nitrous oxide	Halothane, isoflurane, and sevoflurane were all found to reduce systolic blood pressure, but only sevoflurane and isoflurane decreased train-of-four ratios significantly at 1.0 MAC (p < 0.001). Amplitudes of transcranial motor-evoked potentials were reduced by isoflurane and sevoflurane at 0.5 MACs, but not halothane, reflecting the reduced extent of the neuromuscular blockade initiated by halothane.
Djian et al., 2006 [37][45]	Unspecified intracranial procedure	Remifentanil vs. sufentanil in combination with propofol for maintenance of anesthesia	Remifentanil was associated with the need for less adjustments with regard to hemodynamic stability (p = 0.037), greater use of morphine (p = 0.01), and higher intraoperative opioid costs. However, there was no significant differences in extubation times between groups.
Bhagat et al., 2008 [38][46]	Craniotomy for unspecified pathology	Anesthetic maintenance via propofol (n = 50) vs. isoflurane (n = 50)	Hypertension occurrence and MAP change were not different between the two agents. Emergence time was higher for propofol (p = 0.008).
Bonhomme et al., 2009 [39][47]	Unspecified intracranial procedure	Propofol (n = 30) vs. sevoflurane (n = 31)	Propofol was associated with higher occurrence of intraoperative hypertension (p < 0.001) and sevoflurane was associated with higher occurrence of intraoperative hypotension (p = 0.015).
Ali et al., 2009 [40][48]	Resection of pituitary tumor	Propofol (n = 30), isoflurane (n = 30), sevoflurane (n = 30)	Emergence time was significantly longer with use of isoflurane (p < 0.001). Hypertension occurrence was higher in isoflurane than in propofol or sevoflurane, and higher in sevoflurane than propofol (p < 0.001). Hypotension was not difference between anesthetics (p = 0.36).
Bilotta et al., 2009 [41][49]	Craniotomy for resection of supratentorial lesion	Sevoflurane (n = 28) vs. desflurane (n = 28)	Significant delays in cognitive “awakening” for obese and overweight patients receiving sevoflurane-based anesthesia as compared to those receiving desflurane-based anesthesia as measured by post-operative short orientation memory concentration test scores at 15 and 30 min (p < 0.005, p < 0.005) as well as with the Rancho Los Amigos scale (p < 0.005)
Güneş et al., 2009 [63][71]	Craniotomy for resection of supratentorial lesion	Anesthetic maintenance with dexmedetomidine in addition to sevoflurane (n = 30), desflurane (n = 30), and isoflurane (n = 30)	MAP was elevated following intubation for all groups. Rates of eyes opening and responsiveness following verbal commands were lower in desflurane–dexmedetomidine than in other groups (p = 0.001).
Magni et al., 2009 [64][72]	Craniotomy for resection of supratentorial lesion	Anesthetic maintenance via sevoflurane (n = 60) vs. desflurane (n = 60)	Mean emergence was similar between the two groups, but extubation and recovery time were lower (p < 0.001) in the desflurane group. Hemodynamic stability differences were non-significant between the two groups.
Lauta et al., 2010 [42][50]	Craniotomy for resection of supratentorial lesion	Anesthetic maintenance via propofol (n = 153) vs. sevoflurane (n = 149)	Propofol was associated with a significantly longer emergence time to eyes opening (p < 0.014. Sevoflurane was associated with higher occurrence of hypotension (p < 0.0167).
Yildiz et al., 2011 [43][51]	Craniotomy for resection of supratentorial lesion	Anesthetic maintenance via desflurane (n = 35) vs. isoflurane (n = 35)	Heart rate was not different between the two agents. MAP was higher for desflurane (p < 0.05). Extubation time and eyes opening time was shorter for desflurane (p < 0.05).
Ghoneim et al., 2015 [44][52]	Craniotomy for resection of supratentorial tumors	Anesthetic maintenance via isoflurane (n = 20) vs. sevoflurane (n = 20) vs. desflurane (n = 20)	Emergence times were significantly shorter for desflurane or sevoflurane than with isoflurane in pediatric patients following a craniotomy for supratentorial tumors.
Hernandez et al., 2015 [45][53]	Craniotomy for hematoma	Anesthetic maintenance via propofol (n = 20) or sevoflurane (n = 20)	SSEPs amplitudes and latencies were not different between the two agents. TceMEPs amplitudes were higher for propofol (p < 0.05). Latencies were shorter in the propofol group (p < 0.05).
Goettel et al., 2016 [46][54]	Awake craniotomy for unspecified pathology	Dexmedetomidine (n = 25) vs. propofol (n = 25)	There were no differences in level of sedation (OAA) (p = 0.13). There were no differences in intraoperative hypertension (p = 0.60), hypotension (p = 0.50), or complications (p = 0.99). There was no difference in postoperative complications (p > 0.05).
Gokcek et al., 2016 [47][55]	Unspecified intracranial procedure	Anesthetic maintenance via sevoflurane (n = 25) vs. desflurane (n = 25)	Emergence time and time to eyes opening were higher with sevoflurane (p < 0.001).
Lin et al., 2016 [48][56]	Resection of supratentorial lesion	Anesthetic maintenance via propofol (n = 31) vs. dexmedetomidine (n = 31)	NIHSS-positive change was higher in propofol than dexmedetomidine (p < 0.001). Focal neurologic deficits were higher in propofol than dexmedetomidine (p < 0.05).
Rajan et al., 2016 [49][57]	Craniotomy or transsphenoidal approach for resection of brain tumor	Dexmedetomidine (n = 68) vs. remifentanil (n = 71)	Dexmedetomidine was associated with significantly lower postoperative MAP (p < 0.001). Dexmedetomidine was associated with significantly longer emergence time to open eyes (p < 0.001).
Thongrong et al., 2017 [50][58]	Craniotomy for unspecified pathology	Anesthetic maintenance via fentanyl (n = 30) vs. dexmedetomidine (n = 30) after propofol induction	Dexmedetomidine infusions reduced adverse effects, with signs of effectively controlled systolic blood pressure one minute prior to skull pin insertion (p < 0.05), as well as during skull pin insertion (p < 0.01) in comparison to fentanyl. Similarly, dexmedetomidine infusions were related to reduced adverse hypertensive and hypotensive responses in patients.
Bhardwaj et al., 2018 [51][59]	Surgical clipping for aneurysmal subarachnoid hemorrhage	Propofol (n = 35) vs. desflurane (n = 35)	There was no difference in blood loss (p < 0.05), hypotension (p < 0.05), hypertension (p < 0.05), or emergence time for eyes opening (p < 0.05).
Gracia et al., 2018 [52][60]	Unspecified intracranial procedure	Anesthetic induction via propofol (n = 20) vs. thiopental (n = 20)	There was no difference in heart rate (p > 0.05). MAP was significantly higher in thiopental groups (p < 0.05). Systolic and diastolic blood pressure was significantly lower in thiopental groups (p < 0.05).
Molina et al., 2018 [53][61]	Craniotomy for resection of tumor	Propofol–remifentanil (n = 105) for asleep sedation vs. conscious sedation with dexmedetomidine (n = 75)	Patients sedated with dexmedetomidine used less opiates, antihypertensive drugs, and had a lower postoperative duration and length of stay (all p < 0.001).
Xinyan et al., 2018 [54][62]	Awake craniotomy for unspecified pathology	Dexmedetomidine (n = 20), propofol (n = 20), etomidate (n = 20)	There was no significant difference in perioperative wake up duration (p > 0.05) and postoperative emergence time (p > 0.05). The rate of adverse events was lower in dexmedetomidine than propofol and etomidate (p < 0.05). The rate of adverse events was lower in propofol than etomidate (p < 0.05).
Khallaf et al., 2019 [55][63]	Craniotomy for hematoma	Anesthetic maintenance via propofol (n = 20) vs. dexmedetomidine (n = 20)	Tachycardia, bradycardia, and hypertension occurrences were not different between the two agents. IPP and CPP changes were not different between the two agents. Hypotension occurrences were higher in the propofol group (p = 0.024).
Preethi et al., 2021 [56][64]	Craniotomy for hematoma	Anesthetic maintenance via propofol (n = 45) vs. isoflurane	Change in heart rate, systolic blood pressure, diastolic blood pressure, and MAP were not different between the two agents. Brain relaxation was higher for propofol (p < 0.05). ICP was higher for isoflurane (p = 0.01).
Balasubramanian et al., 2021 [57][65]	Surgical clipping/endovascular coiling for aneurysmal subarachnoid hemorrhage	Propofol (n = 8) vs. isoflurane (n = 8) vs. sevoflurane (n = 8), vs. desflurane (n = 8)	There was no significant difference found between anesthetic on levels of CSF caspase-3 levels.

Authors and Year	Surgical Procedure	Comparison	Findings
Laureau et al., 1999 [58][66]	Posterior instrumentation for treatment of idiopathic scoliosis	Induction via intravenous propofol (n = 15) vs. midazolam (n = 15)	Cortical somatosensory-evoked potentials did not deteriorate in either the propofol or the midazolam induction groups.
Inoue et al., 2005 [59][67]	Cervical spine surgery for unspecified pathology	Anesthetic maintenance via fentanyl and propofol (n = 25) vs. fentanyl and <1% sevoflurane (n = 25) vs. sevoflurane (n = 25)	Perception of pain and bucking scores following emergence- were greater for patients exposed to sevoflurane versus propofol and fentanyl and fentanyl and sevoflurane in combination.
Kurt et al., 2005 [60][68]	Unspecified spinal procedure	Anesthetic maintenance via isoflurane (n = 12) vs. sevoflurane (n = 10) vs. desflurane (n = 10)	Sevoflurane and isoflurane administered via volatile anesthetics were able to achieve controlled hypotension in comparison to desflurane with systolic blood pressures outside the target range of 32% and 26% for isoflurane and sevoflurane, respectively, and 44% with desflurane.
Albertin et al., 2008 [61][69]	Lumbar spine surgery for unspecified pathology	Induction via sevoflurane (n = 14) or propofol (n = 14) as main anesthetic agents	Peripheral blood flow was greater in the propofol group before and during the hypotensive period, but had reduced blood loss and intra-operative bleeding as compared to the sevoflurane group (p < 0.005).
Turgut et al., 2008 [62][70]	Lumbar laminectomy	Pre-operative bolus and anesthetic maintenance via dexmedetomidine (n = 25) vs. fentanyl (n = 25) following induction via propofol as well as maintenance	Extubation and discharge times were similar between dexmedetomidine and fentanyl, but MAP values after intubation for those exposed to dexmedetomidine were higher for those exposed to fentanyl before and after extubation. Supplemental analgesia was required earlier for fentanyl group patients (34.8 +/− 1.35 min vs. 60.4 +/− 1.04 min).