Continuous monitoring methods of cerebral compliance and compensatory reserve: a scoping review of human literature

Objective. Continuous monitoring of cerebrospinal compliance (CC)/ cerebrospinal compensatory reserve (CCR) is crucial for timely interventions and preventing more substantial deterioration in the context of acute neural injury, as it enables the early detection of abnormalities in intracranial pressure (ICP). However, to date, the literature on continuous CC/CCR monitoring is scattered and occasionally challenging to consolidate. Approach. We subsequently conducted a systematic scoping review of the human literature to highlight the available continuous CC/CCR monitoring methods. Main results. This systematic review incorporated a total number of 76 studies, covering diverse patient types and focusing on three primary continuous CC or CCR monitoring metrics and methods—Moving Pearson’s correlation between ICP pulse amplitude waveform and ICP, referred to as RAP, the Spiegelberg Compliance Monitor, changes in cerebral blood flow velocity with respect to the alternation of ICP measured through transcranial doppler (TCD), changes in centroid metric, high frequency centroid (HFC) or higher harmonics centroid (HHC), and the P2/P1 ratio which are the distinct peaks of ICP pulse wave. The majority of the studies in this review encompassed RAP metric analysis (n = 43), followed by Spiegelberg Compliance Monitor (n = 11), TCD studies (n = 9), studies on the HFC/HHC (n = 5), and studies on the P2/P1 ratio studies (n = 6). These studies predominantly involved acute traumatic neural injury (i.e. Traumatic Brain Injury) patients and those with hydrocephalus. RAP is the most extensively studied of the five focused methods and exhibits diverse applications. However, most papers lack clarification on its clinical applicability, a circumstance that is similarly observed for the other methods. Significance. Future directions involve exploring RAP patterns and identifying characteristics and artifacts, investigating neuroimaging correlations with continuous CC/CCR and integrating machine learning, holding promise for simplifying CC/CCR determination. These approaches should aim to enhance the precision and accuracy of the metric, making it applicable in clinical practice.


Introduction
Cerebrospinal compliance (CC)/cerebrospinal compensatory reserve (CCR) is a key factor in preserving healthy intracranial pressure (ICP).It reflects the ability of the cerebrospinal system to accommodate or buffer the changes in intracranial volume without a significant shift in ICP (Portella et al 2005).In cases of acute traumatic neural injury (i.e.traumatic brain injury (TBI)), hydrocephalus or hemorrhages, ICP can increase in a potentially fatal manner (Donnelly et al 2020).Continuous monitoring of CC/CCR potentially allows for the identification of abnormalities in ICP at an early stage, which is critical for timely interventions and preventing more significant deterioration (Kiening et al 2003, Ng et al 2005).
In recent years, research has been conducted on the different approaches of continuous monitoring of CC/CCR aimed at objective quantification of patient state to get a more robust understanding, potentially leading to personalized routes of care.Initially, compliance measurements relied on assessing ICP responses to changes in volume induced invasively (such as through bolus injection or infusion of fluid), a method impractical for continuous monitoring.Consequently, alternative approaches were necessary.As time has advanced, the importance of monitoring compliance in a continuous manner for critically head-injured patients has become evident.Recent advancements making continuous monitoring of CC/CCR possible are the analysis of ICP pulse waveform, sophisticated tools like transcranial doppler (TCD), Spiegelberg Compliance Monitor, and data acquisition software platforms for live time signal analytics at the bedside (Abdullah et al 2005, Ziółkowski et al 2021).
RAP is the most popular metric for measuring continuous CC/CCR, defined as the moving Pearson's correlation between ICP pulse amplitude waveform (AMP) and ICP (Zeiler et al 2018a).RAP value ranges from −1 to +1.The lack of correlation in AMP and ICP (i.e., RAP is close to 0) suggests a favourable state of CC/CCR.On the contrary, the RAP value close to +1 indicates the increase of ICP with the rise of AMP, in other words, decreased CC/CCR (Zhu et al 2023).However, when ICP continues to rise beyond the critical ICP level, the cerebral autoregulation will collapse, and AMP and ICP show a negative correlation.That means a negative RAP value signifies an exhausted CC/CCR (Jin et al 2019).It is noteworthy to mention that the RAP index does not estimate compliance directly but assesses the compensatory reserve based on its position on the pressure-volume (PV) curve.Conversely, by placing an intraventricular or intraparenchymal catheter within the patient's brain, the Spiegelberg Compliance Monitor provides real-time data on compliance (Piper et al 1999).CC/CCR (∆V/∆P) is calculated from the small ICP perturbations (∆P) caused by a sequence of up to 200 pulses of added volume (∆V).After achieving a stable average, a minute-by-minute measurement of CC/CCR can be obtained from the Spiegelberg Compliance Monitor (Yau et al 2002).Furthermore, examining the relation between cerebral blood flow velocity (CBFV) and compliance (more specifically, cerebral arterial compliance (Ca) and cerebrospinal space compliance (Ci)), continuous compliance can also be assessed using TCD (Carrera et al 2011, Kim et al 2012).From CBFV, Cerebral arterial Blood Volume (CaBV) can be calculated.Ca is the ratio of the pulsatile amplitude of CaBV (AMP_CaBV) and the pulsatile amplitude of ABP (AMP_ABP), whereas Ci is calculated as the ratio between the pulsatile amplitude of CaBV (AMP_CaBV) and the pulsatile amplitude of ICP (AMP_ICP) (Kim et al 2012).Another approach for estimating CC/CCR continuously is the high frequency centroid (HFC), determined as the average power-weighted frequency within the 4-15 Hz frequency range of the ICP power density spectrum (Robertson et al 1989, Contant et al 1995).Additionally, the utilization of another centroid metric, higher harmonics centroid (HHC), defined as the centre of mass of the ICP pulse waveform harmonics from the 2nd to the 10th, was also notable (Zakrzewska et al 2021, Uryga et al 2022).Finally, distinct peaks from the ICP pulse wave (ICPW), namely P1 (percussion wave) and P2 (tidal wave), were observed to be utilized for assessing continuous CC/CCR, as alterations in the P2/P1 ratio can indirectly indicate changes in CC/CCR (Lee et al 2016, Brasil et al 2021).Aside from these more commonly described continuous CC/CCR measures, researchers use a few additional techniques and metrics in this aspect, which are parabolic regression models derived from the PV curve and the ICP-PCO2 compliance index (PCI) (Lai et al 2016, Wolf et al 2021).
In summary, while the RAP index is extensively researched, the Spiegelberg Compliance Monitor comes closest to directly measuring compliance.This method resembles fully invasive intermittent approaches but operates continuously (Kiening et al 2005, Carrera et al 2011).Conversely, RAP and TCD-based measurements do not directly estimate compliance but operate under the assumption that the ICP pulse waveform is the response to changes in CaBV within a single cardiac cycle (Steiner et al 2005, Carrera et al 2011a, Pineda et al 2015, Kazimierska et al 2021).However, TCD-based measurements uniquely distinguish between cerebrospinal and arterial components of total cerebral compliance (Ci and Ca), a differentiation absent in RAP (Carrera et al 2011a, Kazimierska et al 2021).Additionally, the P2/P1 ratio and centroid metrics (HFC and HHC) also indirectly estimate compliance by analysing characteristics of the ICP pulse shape (i.e. the height of the peaks and frequency of the ICP pulse, respectively) (Robertson et al 1989, Brasil et al 2021, 2023, Zakrzewska et al 2021, Galdino et al 2022, Uryga et al 2022).Combining these five principal approaches will present a persuasive depiction of the continuous measurement of CC/CCR.Despite these previously described methods, the literature on continuous CC/CCR methods in humans remains difficult to navigate and often scattered.This systematic review aims to provide an insightful and comprehensive overview of existing literature for continuous CC/CCR measurement by highlighting the methodologies employed, significant findings, areas requiring further investigation, and notable trends or developments in the field.Our objective is to illuminate the developing field of continuous CC/CCR monitoring, sub-categorizing by type of techniques and patients, aiming to enhance comprehension of compliance and improve patient care within the neurocritical care domain.

Methods
The Cochrane Handbook for Systematic Reviews (Cochrane Handbook) was used as a guide for this systematic review.We followed the guidelines outlined in the preferred reporting items for systematic reviews and meta-analysis (PRISMA) (Page et al 2020) and the PRISMA Extension for Scoping Review in our reporting (Tricco et al 2018).The methodology and search resembled those utilized in previous systematic reviews conducted by the research team (Gomez et al 2022, Sainbhi et al 2023).The collaborative efforts of the primary authors (AI and LF) and senior author (FAZ) formulated the review objectives and developed the search strategy.The PRISMA checklist can be found in supporting information A.

Search questions, population, and inclusion/ exclusion criteria
The question addressed in this systematic review is as follows: What techniques and metrics have been used for the continuous measurement of CC/CCR in humans?
This systematic review encompasses all human studies that involve continuous quantified measurement of CC/CCR regardless of whether they were prospective or retrospective in nature.There are no limitations on sample size, patient characteristics, age, or data sampling method.There were studies where CC/CCR measurement was not the primary focus but rather a factor contributing to the main objective, which were also considered and incorporated in this review (Calviello et al 2018, Zeiler et al 2018a, Froese et al 2020).
The exclusion criteria posed in this study were as follows: non-English language studies, animal studies, theoretical studies, non-continuous measurement of compliance, qualitative measurement of compliance, non-original studies, and abstract-only studies.Non-original and abstract-only studies were excluded to prioritize significant and original research contributions in our analysis.Recognizing the critical significance of continuous measurement of CC/CCR, the primary emphasis of this particular scoping review was real-time data, and therefore excluded any non-continuous and qualitative studies.The animal studies were excluded because we aimed to comprehensively review the continuous CC/CCR measurement techniques directly applied to human subjects.

Search strategy
Searches were conducted in major databases, including PubMed, Embase, Scopus, BIOSIS, and Cochrane Library, which covered records dating from the inception of each database up to mid-June 2023.Individualized search strategies were developed for each of these databases and can be seen in supporting information B. Additionally, a meticulous search in the reference lists of the finally selected studies was conducted to ensure that no studies were overlooked.

Study selection
Two reviewers, AI and LF, performed a thorough two-step review of all the articles retrieved through the individual search strategies for each database.In the first step, reviewers independently screened all the retrieved articles without access to each other's review progress.In this phase, the inclusion or exclusion decision was made based on the title and abstract content.Following the initial screening, the second phase of study selection involved a thorough assessment of full texts.Studies that deviated from our primary focus, which is identifying continuous measurement techniques quantifying CC/CCR, were excluded.Similar to the first one, this phase was also conducted independently, and any disagreements between the two reviewers were resolved by a third party (FAZ).To enhance the comprehensiveness of our review, we meticulously examined the reference lists of the reviewed articles, focusing on the continuous measurement of CC/CCR.

Data collection
The data field encompasses patient characteristics such as age, population, male/female quantity or percentage, and Glasgow Coma Scale (GCS) score.In addition, it includes measurement and processing methods of CC/CCR, CC/CCR quantified values and relation with other factors, primary and secondary outcomes, and lastly, limitations and conclusions, particularly regarding continuous CC/CCR metrics/methods.

Bias assessment
As the aim of this review was to offer a comprehensive and wide-ranging overview of the existing literature, a formal bias assessment was not carried out.

Statistical analysis
Since the objective of this review was to provide a scoping overview of the existing literature, a meta-analysis was not conducted.This decision was also influenced by the existence of extensive heterogeneity in the study designs and data.

Results
Using a PRISMA flow diagram, the search and filtration method of this systematic review has been summarized in figure 1.A total of 15 682 papers were identified from the search strategies applied from the five databases.Among them, 8725 studies were identified as duplicates and removed, resulting in 6957 studies.These 6957 studies were screened through their title and abstracts.6497 studies were excluded for being irrelevant studies (n = 5891), non-continuous studies (n = 344), abstract-only studies (n = 80), review studies (n = 16), animal studies (n = 133) or non-English studies (n = 33).Then the 460 retrieved studies were studied through full-text examination, and 387 studies were excluded based on these criteria-non-continuous studies (n = 363), review studies (n = 8), animal studies (n = 13), and non-English studies (n = 3).Later, three studies were identified from the reference lists of the included papers, and finally, 76 papers were incorporated into this systematic review.

Studies with TBI patients
This systematic review identified 26 studies that used the RAP metric to assess CC/CCR and included TBI patients as subjects (Czosnyka et al 1988, Steiner et al 2005, Shahsavari et al 2008, 2011, Smielewski et al 2008, Timofeev et al 2008a, 2008b, Budohoski et al 2012, Howells et al 2012, Haubrich et al 2013, 2015, 2016b, Pineda et   In studies that reported the age of the patients, in most cases, the mean age was between 30-50 years range, except for three studies (Lalou et al 2020, Levrini et al 2021, Zhu et al 2023), where the mean age was 59.6, 55.7, and 53.2 years, respectively.The majority of these studies experimented and reported the association of RAP with patient outcome, other physiology, and brain imaging.

Association with other physiologic parameters
Three studies (Calviello et al 2018, Pineda et al 2018, Zeiler et al 2018a) demonstrated the relationship between ICP and RAP.Evident sign changes and significant variance were shown with the change of ICP.Moreover, when ICP was unchanged, RAP was found to be constant.This proves the future potential of RAP for predicting ICP events and patient outcome.The effect of surgery, medical treatment and agents on RAP was investigated in five studies (Timofeev et al 2008a, 2008b, Howells et al 2012, Froese et al 2020, 2021).The association of RAP with decompressive craniectomy was observed in two studies (Timofeev et al 2008a, Howells et al 2012).Both of the studies were aligned with the findings of decreasing RAP after decompressive craniectomy.In addition, Howells and collaborators studied RAP's association with the Glasgow Motor Score (GCS-M), motor component of GCS, age and thiopental treatment.One of the major findings was that RAP had a less significant correlation with GCS-M and age compared to that of AMP, whereas in the case of thiopental treatment, a significant or marginally significant downward shift of RAP was noticed, which indicates improved intracranial compliance.Another medical procedure, ventriculostomy, demonstrated improving compensatory reserve, i.e. reducing RAP (Timofeev et al 2008b).Two studies examined the impact of vasopressor and sedative agents on RAP (Froese et al 2020(Froese et al , 2021)).In both cases, their findings revealed that, overall, there was no significant change in RAP due to these medications.This was experimented with mean hourly dose and incremental dose change protocols.Moreover, sedation depth was examined, and a negative linear relation was revealed between the bispectral index (BIS) and RAP (Froese et al 2021).
The relation between respiratory oscillations (R waves) and RAP was researched by Haubrich and colleagues, and they found that an increase in RAP is associated with a decrease in GainFv (the transfer function gains of R waves between arterial blood pressure (ABP) and Doppler flow velocity (FV)), i.e. reduction in R waves.This suggests an exhausted cerebral compensatory reserve (Haubrich et al 2015).
One study conducted research on the association of RAP with the transfer function of ABP to ICP at the fundamental cardiac frequency.This transfer function was denoted as MTF, and the result showed that RAP is positively correlated with MTF and negatively correlated in their variances.This indicates an increased value of MTF during poor compensatory reserve (Shahsavari et al 2008).Shahsavari and colleagues also explored the status of RAP in various conditions of cerebral autoregulation and corresponding regimes within the cerebrovascular system in TBI patients.In this experiment, the population was divided into three groups based on pressure reactivity as follows-a.the intact pressure reactivity group, b. impaired pressure reactivity case I group, and c. impaired pressure reactivity case II group.In the case of the intact pressure reactivity group and the impaired pressure reactivity case I group, the result showed that the average RAP is close to +1.However, in the impaired pressure reactivity case II group, RAP was close to +1 before the plateau wave.During the plateau wave, RAP decreased significantly (Shahsavari et al 2011).In another experiment, after exploring the association of RAP with hyperventilation in TBI patients, the result showed that RAP was reduced due to hyperventilation and showed a significant correlation with hyperventilation-induced reduced ICP (Steiner et al 2005).Baseline ICP and RAP were both significant predictors of ICP reduction, with RAP being the more powerful parameter (Steiner et al 2005).A different study showed that RAP decreased due to spindle waves, a distinctive pattern observed in EEG during the non-rapid eye movement (NREM) sleep stage.This was evident in both scenarios-scenario 1, a comparison between the control and spindle wave group, and scenario 2, a comparison (within the spindle wave group) between during spindle wave occurrence and before and after the period of that occurrence (Zhu et al 2023).
Three studies reported their subject as head injured patients (Smielewski et al 2008, Haubrich et al 2013, 2016b).One of the studies analysed the impact of increasing ICP on R-waves based on RAP value.RAP exceeding 0.85 reflects a substantial impact on R-waves gain with increasing ICP.Conversely, while RAP was below 0.85, it resulted in a lower impact on R-waves gain with rising ICP.Haubrich and colleagues also studied the effect of RAP in response to mild hypocapnia.Their findings indicated a significant level of decrease below baseline in RAP, and hypocapnic ICP showed a positive correlation with baseline RAP.Additionally, the hypocapnic vasomotor response (VMR) was linearly and positively correlated with RAP (Haubrich et al 2013).The remaining study centred on the ICM+ software and illustrated its usefulness in measuring the RAP metrics (Smielewski et al 2008).
The relations of RAP with two newly derived indices were observed in three studies (Calviello et al 2018, Zeiler et al 2018a, 2019).A new physiological index RAC, defined as the moving correlation coefficient between AMP and cerebral perfusion pressure (CPP), exhibited a significant negative correlation with RAP.The study also delved into the association between RAP and CPP.With the increase of CPP, RAP was reduced to zero.However, a steep increase was evident in RAP when CPP decreased towards the lower limit of autoregulation (Zeiler et al 2018a).The compensatory reserve weighted variable, weighted ICP (wICP), is another derived variable obtained from ICP and RAP.In the cases of good compensatory reserve (i.e.low RAP) and exhausted compensatory reserve (i.e.RAP close to +1), if the ICP remained low, then wICP stayed low as well.On the other hand, when ICP crossed its critical threshold, RAP led toward negative, resulting in an abrupt increase in wICP (Calviello et al 2018, Zeiler et al 2019).
Two studies conducted a comparative analysis of the RAP metric between TBI and idiopathic normal pressure hydrocephalus (iNPH) patients (Lalou et al 2020, Levrini et al 2021).RAP baseline was insignificantly different between the two groups, with the TBI group exhibiting a slightly higher RAP (Lalou et al 2020).Additionally, when comparing shunted post-traumatic hydrocephalus (PTH) vs non-shunted PTH, the group of shunted PTH showed a slightly higher RAP, but this difference was not significant (Lalou et al 2020).However, following continuous infusion tests, a notable increase in RAP of the PTH group was noticed, which is higher than the atrophy group (Levrini et al 2021).

Association with neuroimaging features
The association of RAP with imaging was observed in one study (Zeiler et al 2018b).The relation of RAP with admission computer tomography (CT) injury characteristics was examined, and the integrated area under the curve was found to be statistically associated with CT markers of diffuse TBI across different RAP thresholds (Zeiler et al 2018b).

Association with patient outcomes
Research was conducted to find out the association of RAP with patient outcome in four studies (Budohoski et al 2012, Calviello et al 2018, Pineda et al 2018, Donnelly et al 2020).Significant distinction in RAP was found between the survivor and dead group, though non-invasive AMP (nAmp) showed the most significant difference (Budohoski et al 2012).Whereas Calviello and colleagues found RAP to be significantly distinctive between individuals with severe disability and those with a favourable outcome.RAP performed better in distinguishing between severe disability and favourable outcome than differentiating between survival and fatal outcome (Calviello et al 2018).In a different study, RAP illustrated a significantly lower value in dead patients or patients in the vegetative state and RAP declined and oscillated towards 0 and negative value (Czosnyka et al 1996).Similarly, low RAP was identified in patients with severe ICH as well (Donnelly et al 2020).Another study investigated the potential of RAP in predicting cerebral hemodynamic instability and found RAP can identify unstable periods if it exceeded 0.6, providing a better distinguishing factor than ICP or AMP (Pineda et al 2018).However, the study conducted by Kazimierska and colleagues did not find a significant correlation between RAP and patient outcome (Uryga et al 2022).Nevertheless, they observed a significant positive association of RAP with the dominant ICP pulse type.This refers to the pulse type occurring most frequently throughout the entire recording, excluding pulses classified as artifacts from the analysis (Uryga et al 2022).

Studies with hydrocephalus patients
The evaluation of the RAP metric of patients affected with hydrocephalus was discussed in 11 studies (Czosnyka et al 1988, Petrella et al 2008, Schuhmann et al 2008, Kim et al 2009a, 2015, Weerakkody et al 2011, Speil et al 2012, Varsos et al 2015, Haubrich et al 2016a, Green et al 2021, Ziółkowski et al 2021).The majority of these subjects were elderly patients, more than 50 years old.However, two studies included children as their subjects (Czosnyka et al 1988, Schuhmann et al 2008).

Association between RAP, infusion testing and shunting
The effect of cerebrospinal fluid (CSF) infusion on RAP was explored in 6 studies (Czosnyka et al 1988, Kim et al 2009a, Weerakkody et al 2011, Varsos et al 2015, Haubrich et al 2016a, Ziółkowski et al 2021).Czosnyka and colleagues divided subjects into two groups considering RAP trends; group 1: RAP was low before and after infusion, whereas during infusion, RAP increased to a high value; group 2: before infusion, RAP was high, close to 1, whereas during infusion RAP typically decreased.Before and during recovery after an infusion test, RAP can be used as an indicator of the equilibrium state (Czosnyka et al 1988).Increased RAP during infusion of CSF was observed in four studies (Kim et al 2009a, Weerakkody et al 2011, Haubrich et al 2016a, Ziółkowski et al 2021).In the case of non-shunted patients with ventriculomegaly, during infusion, RAP stayed high, usually exceeding 0.6 (Weerakkody et al 2011).However, the majority of the atrophic profile exhibited low RAP during infusion.These atrophic profiles were caused by cerebrovascular or neurodegenerative disease and typically demonstrated low baseline pressure, resistance to CSF outflow (RCSF) and AMP (Weerakkody et al 2011).
Experiments were conducted on NPH patients to assess the impact of shunting on RAP (Petrella et al 2008, Schuhmann et al 2008, Kim et al 2009a, Weerakkody et al 2011).Among patients with ventriculomegaly, RAP significantly improved (i.e.RAP decreased) in both RAP baseline and RAP plateau (Petrella et al 2008, Kim et al 2009a).A functioning shunt had an important impact on CSF circulation and pressure-volume compensation, and RAP.On the other hand, Schuhmann and collaborators evaluated the effect of a shunt malfunction on RAP, which led to the classification of 3 groups: normal, questionable and pathological group, with the latter indicating shunt malfunction.RAP was significantly different between these groups and notably elevated in the pathological group compared to the normal group (Schuhmann et al 2008).In a comparison between properly functioning shunt, overdrainage and underdrainage states, RAP was noted to be low for the first two cases, while exceeding 0.6 for the latter (Weerakkody et al 2011).

Association with other physiologic measures
Furthermore, several studies investigated the association of RAP with other parameters and cases (Kim et al 2009a, 2015, Speil et al 2012, Varsos et al 2015, Green et al 2021, Ziółkowski et al 2021).For instance, exploring the relationship between RAP and phase shift angle between fundamental harmonics of CBFV and ICP illustrated a negative correlation (Kim et al 2015).In a study comparing mean RAP values, both iNPH and late-onset idiopathic aqueductal stenosis (LIAS) patients exhibited impaired values, with no significant difference noted (Green et al 2021).In another study, a strong positive correlation was observed between RAP and the RCSF, while a weaker correlation was observed with ventriculomegaly (Kim et al 2009a).Speil and collaborators conducted a study comparing RAP with both elastance (E) and pressure-volume index (PVI).In this study, E and PVI were measured through a lumbar infusion test.However, no significant correlation existed between them (Speil et al 2012).
In a different experiment, the association between cerebral hemodynamic indices (critical closing pressure (CrCP), wall tension (WT), closing margin (CM)) and CSF compensatory parameter (i.e.RAP) was observed.No association between RAP and CrCP, WT or CM was found.However, it was suggested that CM at baseline pressure could be utilized as an indicator of RAP since it was negatively correlated with cerebrospinal elasticity (Varsos et al 2015).Additionally, analysis was conducted to determine the relation between ICP pulse shape changes, CBFV, and intracranial compliance (Ziółkowski et al 2021).A new index called the ratio of pulse slopes (RPS) was introduced for this purpose, based on the inclinations of the ascending parts of the ICP and CBFV pulse waveforms.The findings included a strong positive correlation of baseline RAP with intracranial elasticity and negative correlations with the height ratio of the first and second peaks of the ICP pulse (P1/P2) and RPS (Ziółkowski et al 2021).

Studies with other patient populations
Aside from TBI and NPH patients, there are an additional six studies that include diverse patient population investigating RAP metric (Czosnyka et al 1994, Balestreri et al 2004, Eide and Sorteberg 2013, Moyse et al 2016, Jin et al 2019, Sekhon et al 2019).The mean age across these studies varies from each other, since they explored different aspects within various types of patients.
Analysis of RAP in patients with ICH was conducted in two studies (Czosnyka et al 1994, Balestreri et al 2004).In the study of the comparison between patients with favourable and fatal outcomes, a progressive increase of RAP was observed over time, along with a sudden decrease when ICP started to increase in cases of the patients who died.In contrast, patients with a favourable outcome persistently showed a higher RAP (Balestreri et al 2004).Czosnyka and colleagues, while studying cerebrospinal dynamics, classified patients into group 1 with low RAP (good compensatory reserve), group 2: RAP is high and increasing with decreasing CPP, and group 3: RAP is close to +1 (nearly exhausted RAP).Furthermore, RAP became negative when CPP dropped below 30 mmHg, indicating a critical disturbance in cerebral circulation (Czosnyka et al 1994).
Two studies concentrated on patients with subarachnoid hemorrhage and intracerebral hemorrhage (Eide andSorteberg 2013, Jin et al 2019).Combining RAP with ICP monitoring proved to be an effective marker in an experiment of subarachnoid hemorrhage patients (Jin et al 2019).The total population was split into two groups: group 1: receiving ICP monitoring only (n = 5), and group 2: receiving ICP monitoring with RAP index (n = 2).Out of the 5 patients 3 died from group 1, whereas none from the group 2 faced fatal outcome.The monitoring of RAP index allowed an opportunity for implementing additional interventions to reduce ICP before clinical worsening, thus minimizing irreversible damage to brain tissue (Jin et al 2019).In another study, variations in RAP value were noted within the same patient when measured using two different sensors.The second sensor was categorized into four types-category A: a solid sensor; category B: a fluid sensor; category C: an air-pouch sensor; and category D: a fibreoptic sensor.A difference in RAP ⩾0.4 was found while comparing the measurements from the two sensors, among which 4%, 44%, 20%, and 28% belonged to category A, category B, category C, category D respectively.Additionally, when applying a threshold RAP of 0.6, the two sensors showed a difference in measurements in more than 20% of scores (Eide and Sorteberg 2013).
Moyse and collaborators examined states of RAP in the supratentorial and infratentorial compartment of a 70 year old patient with acute cerebellar infarction and infratentorial mass effect.The correlation of RAP in the two compartment was strong and their values did not align precisely but were close (Moyse et al 2016).Observation of RAP in HIBI patients showed impaired RAP along with normal ICP in most of the cases (Sekhon et al 2019).

Spiegelberg Compliance Monitor-based continuous measurement
Recently, Spiegelberg Compliance Monitor has become a prevalent tool for assessing continuous CC/CCR and in this systematic review 11 studies utilized this for their analysis (Piper et al 1999, Raabe et al 1999, Kiening et al 2002, 2003, 2005, Yau et (Kiening et al 2002, 2003, 2005, Abdullah et al 2005, Portella et al 2005, Salci et al 2006), four studies concentrated on hydrocephalus patients (Piper et al 1999, Raabe et al 1999, Yau et al 2002, 2005) one study explored patients with abnormal hematoma volume (Ng et al 2005).The median age of the population of these studies lies between 45-60 years, with the exception of one study (Abdullah et al 2005) that included younger subjects, aged between 18-27 years.

Studies with TBI patients
Out of the six studies (Kiening et al 2002, 2003, 2005, Abdullah et al 2005, Portella et al 2005, Salci et al 2006) focusing on TBI patients, Abdullah and colleagues explored the relationship between continuous CC/CCR and decompressive craniectomy.Their findings revealed that, after receiving decompressive craniectomy, compliance improved substantially in all alive patients (Abdullah et al 2005).Analysis was conducted on the potential of continuous CC/CCR to predict increased ICP (Kiening et al 2003(Kiening et al , 2005)).CC/CCR correctly predicted 37 increased ICP episodes from 225 episodes.Furthermore, continuous CC/CCR and ICP had a scattered correlation (Kiening et al 2003), which reduced significantly with a rise in ICP (Kiening et al 2005).Kiening and team also researched the association of age with continuous CC/CCR and observed that it declined with the increase of age in patients with high ICP (Kiening et al 2005).The association of continuous CC/CCR with CPP was explored with a designated threshold of CPP = 60 mmHg.CC/CCR was significantly lower when CPP was below 60 mmHg.Though CPP exceeding 60 mmHg resulted in a higher CC/CCR, it dropped dramatically when CPP reached 100 mmHg (Portella et al 2005).The relationship between continuous CC/CCR and the lactate/pyruvate (L/P) ratio was observed by Salci and team.As compliance decreased, the L/P ratio increased with or without high temperature.This relationship was differed by coma treatment.Furthermore, when the temperature is elevated, there is a greater increase in the L/P ratio.Furthermore, the analysis revealed that individuals with lower CC/CCR were more susceptible to secondary hyperthermia (Salci et al 2006).In a different research study, a relation between CC/CCR and age was observed; as age increased, CC/CCR increased as well (Kiening et al 2002).

Studies with hydrocephalus patients
The application of the Spiegelberg Compliance Monitor to measure continuous CC/CCR in hydrocephalus patients was researched for different aspects.The performance and result of this continuous measurement method were compared with a manual volume injection /withdrawal method, and a strong positive correlation was found between them.Their average measurements were quite similar, with the continuous method showing a slightly higher value (Piper et al 1999).In another study, continuous CC/CCR remained unchanged across different body positions (Raabe et al 1999).Analyzing the contrast between continuous CC/CCR and cerebral elastance, the latter exhibited fewer outliers and a better correlation with ICP.Moreover, the exponentially-weighted moving average (ewma) method was introduced in this analysis, and the result showed it was more strongly correlated with ICP than the conventional Spiegel 'rolling' method in terms of calculating continuous CC/CCR (Yau et al 2005).The functional and compliance improvement following CSF shunting was also investigated with the Spiegelberg Compliance Monitor measurement.Except for one, all the other patients had noteworthy improvement (Yau et al 2002).

Studies with other patient populations
Other than research on TBI and hydrocephalus patients, Spiegelberg Compliance Monitor was also used for other aspects to determine continuous CC/CCR.One of the studies researched continuous CC/CCR as a bedside monitoring technique.Ng and collaborators observed the impact of applying ultrasound-guided aspiration to patients with abnormal hematoma volume.It was found that this procedure led to a sustained improvement in compliance accompanied by a gradual reduction in ICP (Ng et al 2005).

TCD-based continuous measurement
Nine studies of this comprehensive systematic review included research that encompassed TCD-based continuous measurement of CC/CCR (Kim et al 2009b, 2010, 2012, Carrera et al 2011, 2011a, 2011b, Capel et al 2014, Kazimierska et al 2021, Moir et al 2021).Out of them, three focused on TBI patients (Kim et al 2009b, 2012, Carrera et al 2011), three researched patients with hydrocephalus (Kim et al 2010, Capel et al 2014, Kazimierska et al 2021), and two studied normal subjects (Carrera et al 2011a(Carrera et al , 2011b) and the remaining one involved patients with internal carotid artery disease (Moir et al 2021).

Studies with TBI patients
Among the TBI studies, Carrera and team investigated the changes of Ci and Ca induced by initial hyperventilation and observed an increase in Ci and a decrease in Ca.ICP was also reduced in this phase and correlated with Ca reduction.The period of sustained hyperventilation was also observed, and though ICP was elevated and Ci was reduced, no noteworthy change in Ca was observed (Carrera et al 2011).The state of Ca and Ci was observed in a study regarding continuous monitoring of the Monro-Kellie doctrine (Kim et al 2012).An index of CC/CCR (ICC), defined as a moving correlation coefficient between Ca and Ci, was also introduced.Three phases were monitored-a.during arterial hypertension, b. during plateau waves, and c. during ICH.Ca and Ci showed a negative correlation during arterial hypertension, leading to a negative ICC.During plateau waves, both Ca and Ci increased.Meanwhile, Ca and Ci were reduced, and a consistent and positive ICC was exhibited at the ICH phase.Regarding outcome, the study also depicted that higher ICC was associated with a higher mortality rate (Kim et al 2012).In addition, this defined ICC demonstrated a significant and positive correlation with pressure reactivity index (PRx) but none with ICP, mean ABP, or CPP (Kim et al 2012).Alongside these findings, in another study, Kim and collaborators observed the compartmental compliance changes, i.e. the changes of Ca and Ci, focusing on plateau waves.In the case of Ci, it reduced significantly during the plateau wave, followed by an increase.The opposite case was seen in terms of Ca since it increased during plateau waves and decreased afterwards (Kim et al 2009b).

Studies with hydrocephalus patients
CSF infusion tests were performed in all investigations involving hydrocephalus patients (Kim et al 2010, Capel et al 2014, Kazimierska et al 2021).All of the tests demonstrated poorer compliance during the plateau phase of infusion, characterized by an elevated value of Ca. whereas a reduction in Ci and other discussed CC indices (Kim et al 2010, Capel et al 2014, Kazimierska et al 2021).In one of the studies, after a total of 50 infusion tests, a strong negative correlation was found between Ca and cerebrovascular response (CVR) as well as between Ca and CPP.However, no correlation was noted between Ca and any CSF compensatory parameters-Rcsf, elasticity or ICP baseline (Capel et al 2014).In a comparative study evaluating three compliance measurement methods, all of them correlated strongly with each other.These three methods-CCSF, CCaBV, and CP1/P2 represented the CC and derived from Marmarou's model of CSF dynamics (CCSF), changes in CaBV (CCaBV), and the amplitudes of peaks P1 and P2 of ICP pulse waveform (CP1/P2) respectively.Comparing their values between the baseline and plateau phase of the infusion test revealed that all of them reduced, with CCaBV exhibiting the largest change, whereas CP1/P2 had the smallest (Kazimierska et al 2021).In another infusion test study, brain elasticity was positively correlated with Ci, which was evident when comparing values from the baseline to the plateau phase and during the plateau phase itself (Kim et al 2010).

Studies with other patient patients
Regarding predicting ischemic events, Carrera and collaborators researched patients with internal carotid artery disease (Carrera et al 2011b).Their findings encompassed a lower value of Ca for the diseased group during baseline, hyperventilation and 5% CO2 inhalation, as well as Ca and the degree of stenosis illustrated a reciprocal characteristic at baseline (Carrera et al 2011b).In addition, a comparative analysis was conducted on normal subjects, examining the differences between normo-, hypo-and hypercapnia.While comparing with normocapnia, Ca was eminently lower in hypocapnia; however, in terms of hypercapnia, there was no difference between them.Furthermore, Ca was associated with CVR in an inversely changing manner, evident in both hypo-and hypercapnia (Carrera et al 2011a).Another study was conducted on healthy subjects, experimenting with the effect of sublingual sodium nitroglycerin (SNG) protocol and hypercapnia protocol.In both cases, cerebrovascular compliance decreased, with hypercapnia protocol associated with a comparatively higher reduction (Moir et al 2021).

HFC/HHC-based continuous measurement
Five studies demonstrated the measurement method of continuous CC/CCR using HFC or HHC (Robertson et al 1989, Contant et al 1995, Lang et al 2003, Zakrzewska et al 2021, Uryga et al 2022).All of the studies concentrated on TBI patients.In the study conducted by Contant and colleagues, though a direct association of HFC with CC was not mentioned, it was found that HFC decreased during the refractory ICH changes (Contant et al 1995).In contrast, it increased during the transient changes.Since pulse amplitude of ICP (referred to as PAICP in the study) showed an inverse relation with CC/CCR and demonstrated an increased value in refractory ICH (not significant), it can be said that, probably, HFC and CC/CCR had a positive association, i.e. with the increase of HFC, CC/CCR increased (Contant et al 1995).However, the association of HFC with compliance is not straightforward since two other studies found that HFC was significantly greater in the group that met the fatal outcome, which was associated with poor CC/CCR (Robertson et al 1989, Uryga et al 2022).Regarding HHC, a negative correlation was reported with PRx, indicating a negative correlation between cerebral autoregulation and CC (Zakrzewska et al 2021).Additionally, Lang and the team found out from their experiment that a noninvasive measured HFC result was similar to the invasively measured HFC, concluding the possibility of measuring CC/CCR by HFC noninvasively (Lang et al 2003).

P2/P1 ratio-based continuous measurement
The measurement of continuous CC using the P2/P1 ratio was examined in six studies (Kuramoto et al 1986, Lee et al 2016, Nucci et al 2016, Brasil et al 2021, 2023, Galdino et al 2022, Uryga et al 2022).Among the studies, three specifically targeted TBI patients (Lee et al 2016, Brasil et al 2021, 2023), two focused on patients with hydrocephalus (Kuramoto et al 1986, Nucci et al 2016), and one examined patients with type 2 diabetes (Galdino et al 2022).In an approach to assess CC/CCR utilizing ICPW, an increase in P2/P1 ratio was observed, which was associated with impaired CC/CCR in the patients with craniectomy (Brasil et al 2021).Another parameter derived from the P2/P1 ratio, known as the brain compliance index, demonstrated a higher value (i.e.impaired compliance) in the early death (ED) group (Brasil et al 2023).Furthermore, P2/P1 was higher in patients with ICH (Brasil et al 2023).Conversely, in the group with type 2 diabetes mellitus (T2DM), no significant change in the P2/P1 ratio (i.e.no substantial change in CC/CCR) was observed compared to the control group (Galdino et al 2022).However, a significant decrease in the P2/P1 ratio was noticed due to active postural change only in the T2DM group (Galdino et al 2022).A different study by Kuramoto and colleagues illustrated the influence of P2 in bulk compliance (Kuramoto et al 1986).They observed that after the infusion test, P1 was smaller than P2 in the case of subdural hematoma cases, whereas P1 and P2 were either constant or hard to identify in hydrocephalus cases.Additionally, they noted an increment of P2 compared to P1 during the REM period (Kuramoto et al 1986).

Special cases
The importance of the P2/P1 ratio in assessing the CC in different circumstances was evident in four studies (Kuramoto et al 1986, Brasil et al 2021, 2023, Galdino et al 2022).In conjunction, two studies focused on identifying real-time P1, P2 and P3 peaks from ICPW (Nucci et al 2016, Uryga et al 2022).Using a clustering method, 95.3%, 87.8%, and 87.5% accuracy was obtained for P1, P2 and P3, respectively (Nucci et al 2016).Meanwhile, Nucci and the team used ANN to classify CSFPPW into different classes and found a match of 88.3% of cases between their result and an expert examiner's observation (Uryga et al 2022).These classes ranged from class I to class IV.A progressive increase in average elasticity index (EI) values was noted with the advancement of morphological class (Uryga et al 2022).

Other approaches for continuous measurement
Aside from RAP, Spiegelberg Compliance Monitor, TCD, HFC/HHC, and P2/P1 ratio, few other methods have been utilized to measure CC/CCR continuously.Two such studies (Lai et al 2016, Wolf et al 2021) are included in this systematic review focusing on patients with TBI.
The potential of representing continuous CC/CCR from a PV curve was experimented with.An indicator 'a' was derived from a parabolic regression model, expressed initially from the PV curve but without a conclusive statement (Lai et al 2016).PCI is another index defined as the moment-to-moment correlation between ICP change and end-tidal CO2 (ETCO2) change.Throughout most of the observation period, PCI remained within the range of 0.1-0.2,where PCI ≈ 0 suggests good compliance and PCI ≈ 0.2 suggests poor compliance (Wolf et al 2021).

Discussion
This scoping review comprehensively addressed the existing methods and metrics for calculating continuous compliance.Following a thorough review, the key insights encompass the performance of the metrics, a comparative analysis with conventional parameters, and the association of compliance with various parameters.The three major metrics and methods for assessing continuous CC/CCR are A. RAP, B. compliance measurement from Spiegel compliance monitor, C. TCD-based compliance-Ci and Ca.
Firstly, if attention is directed to the metric's performance, RAP was the most extensively researched continuous CC/CCR measurement metric (Czosnyka et al 1988, 1994, 1996, Balestreri et al 2004, Steiner et al 2005, Petrella et al 2008, Schuhmann et al 2008, Shahsavari et al 2008, 2011, Smielewski et al 2008, Timofeev et al 2008a, 2008b, Kim et al 2009a, 2015, Weerakkody et al 2011, Budohoski et al 2012, Howells et al 2012, Speil et al 2012, Eide and Sorteberg 2013, Haubrich et al 2013, 2015, 2016a, 2016b, Pineda et al 2015, 2018, Varsos et al 2015, Moyse et al 2016, Calviello et al 2018, Zeiler et al 2018a, 2018b, 2019, Jin et al 2019, Sekhon et al 2019, Donnelly et al 2020, Froese et al 2020, 2021, Lalou et al 2020, Green et al 2021, Levrini et al 2021, Ziółkowski et al 2021, Liu et al 2022, Zhu et al 2023).Nevertheless, in terms of predicting TBI patient outcome, AMP outperformed RAP (Budohoski et al 2012).The superiority of AMP over RAP was also evident in distinguishing between iNPH and PTH patients (Lalou et al 2020).In addition, it remained unclear if RAP could serve as a predictor for ICH, which led to the suggestion of further investigation (Pineda et al 2018).Nevertheless, RAP excelled as a compliance metric and proved to be a potential marker in various aspects.Its association with brain CT injury patterns further confirms its potential use as a measure of continuous CC/CCR in TBI (Zeiler et al 2018b).While RAP might not have shown superior performance in predicting mortality rate, it proved to be more effective in distinguishing between severe disability and favourable outcomes (Howells et al 2012).It was also proved to be beneficial for TBI patients in terms of controlling elevated ICP (Steiner et al 2005).Additionally, regarding indicating an equilibrium state after an infusion test, it also contributed significantly to patients affected with hydrocephalus (Czosnyka et al 1988).In the context of shunting applications, several studies utilized RAP as an indicator of improved compliance (Schuhmann et al 2008, Kim et al 2009a, Weerakkody et al 2011), and RAP provided the expected result, even in the case of the malfunctioning shunt (Schuhmann et al 2008, Weerakkody et al 2011).RAP emerged as a more effective indicator for HIBI patients compared to ICP, showing significant differences from normal cases (Sekhon et al 2019).Other than HIBI, for critical intracranial physiological parameters in poor-grade World Federation of Neurosurgical Societies patients, a combination of RAP and ICP was fruitful as a marker (Jin et al 2019).
While RAP is the most centric topic regarding the continuous measurement of CC/CCR, the Spiegelberg Compliance Monitor also contributed extensively to the field of research as a tool for assessing continuous compliance (Piper et al 1999, Raabe et al 1999, Kiening et al 2002, 2003, 2005, Yau et al 2002, 2005, Abdullah et al 2005, Ng et al 2005, Portella et al 2005, Salci et al 2006).Similar to RAP, it also has its limitations.Research conducted in 2003 concluded that the current hardware and software version of the Spiegelberg Compliance Monitor lacked satisfactory data quality, which might hamper predicting increased ICP (Kiening et al 2003).In another study, after analysing the effect of shunt application using the Spiegelberg Compliance Monitor, further research was suggested to clarify its clinical implications (Yau et al 2002).Nonetheless, comparing the measurement of the Spiegelberg Compliance Monitor with one of the conventional manual methods illustrated a significant correlation, implying its validity as a continuous compliance measurement method (Yau et al 2005).As an indicator, compliance measurement from the Spiegelberg Compliance Monitor was competent, indicating when to avoid hyperthermia in TBI patients (Salci et al 2006).
As mentioned, Ci and Ca are the two primary compliance metrics derived from TCD measurements.The performance of Ca was evaluated in one of the studies analysing the association of the metric with ETCO 2 (Carrera et al 2011).In addition, this metric was associated with CPP after infusion, similar to the other defined metrics (Capel et al 2014).Lastly, both Ca and Ci showed a strong correlation with the direct method of volumetric manipulation, affirming their validity as valuable parameters for continuous compliance measurement (Kazimierska et al 2021).
Additionally, two centroid metrics-HFC and HHC-were observed to assess continuous CC/CCR.While a direct correlation between CC/CCR and HFC/HHC was not evident, these metrics significantly contributed to predicting patient outcomes, which could be associated with either poor or improved CC/CCR (Robertson et al 1989, Zakrzewska et al 2021, Uryga et al 2022).In these scenarios, HFC/HHC were negatively correlated with the state of the CC/CCR (Robertson et al 1989, Zakrzewska et al 2021, Uryga et al 2022).Furthermore, the alternation in the distinct P1 and P2 peaks of ICPW was also used to assess continuous CC/CCR.P2 became greater than P1 (i.e.P2/P1 ratio > 1) in the impaired compliance cases in contrast to the P1 > P2 in normal cases.P2/P1 could also predict the patients' outcomes, with P2/P1 being higher in the ED group and ICH group patients (Brasil et al 2023).
Aside from RAP, Spiegelberg Compliance Monitor, TCD, HFC/HHC, and P2/P1 ratio-based compliance assessment, some other approaches were explored with the purpose of measuring compliance continuously.Originating from the PV curve, and modifying through a parabolic regression model, an indicator 'a' was derived.The absolute value of 'a' represented the compliance measurement in the experiment (Lai et al 2016).In another study, the newly derived metric PCI looked promising in the field of an indirect assessment of dynamic intracranial compliance measurement (Wolf et al 2021).
Secondly, in the realm of exploration and observation, a predominant number of the studies reported the association of continuous CC/CCR with one or multiple different parameters, conditions or factors (Czosnyka et al 1994, Abdullah et al 2005, Portella et al 2005, Steiner et al 2005, Yau et al 2005, Petrella et al 2008, Schuhmann et al 2008, Timofeev et al 2008b, Kim et al 2009a, 2010, 2012, Carrera et al 2011, Shahsavari et al 2011, Weerakkody et al 2011, Budohoski et al 2012, Howells et al 2012, Capel et al 2014, Haubrich et al 2015, Varsos et al 2015, Zeiler et al 2018a, Lalou et al 2020, Froese et al 2021, Green et al 2021, Kazimierska et al 2021, Levrini et al 2021).Several studies involving TBI patients found a strong correlation between compliance with decompressive craniectomy, that is undertaking this surgical procedure significantly improved compliance (Abdullah et al 2005, Timofeev et al 2008a, Howells et al 2012).Researchers were interested in studying the change of CC/CCR resulting from CSF infusion, which was applied to hydrocephalus patients.Consistent findings across studies indicate a notable decrease in compliance leading to a poorer state during infusion (Czosnyka et al 1988, Kim et al 2009a, Weerakkody et al 2011, Speil et al 2012, Varsos et al 2015, Haubrich et al 2016a, Levrini et al 2021, Ziółkowski et al 2021) and also during the plateau phase of the infusion, as some studies reported (Kim et al 2010, Capel et al 2014, Kazimierska et al 2021).Another focal area of investigation in the context of hydrocephalus was the effect of shunting (Yau et al 2005, Petrella et al 2008, Schuhmann et al 2008, Kim et al 2009a, Weerakkody et al 2011, Lalou et al 2020).Generally, shunting improved compliance; however, in the case of PTH, no significant difference between shunted and non-shunted was reported (Lalou et al 2020).Additionally, malfunctioning shunts demonstrated poorer compliance (Schuhmann et al 2008, Weerakkody et al 2011).
The association of continuous CC/CCR with different neurological conditions consistently demonstrated a prevalent trend characterized by diminished compliance (Haubrich et al 2013, Lalou et al 2020, Levrini et al 2021).An exception was atrophy, where the patients demonstrated ample compliance during the infusion test (Weerakkody et al 2011).However, there was no substantial difference in the quantitative value of compliance between TBI, iNPH and LIAS (Haubrich et al 2013, Lalou et al 2020, Levrini et al 2021).Recording after infusion proved effective in differentiating PTH and atrophy (Levrini et al 2021).Various researchers extensively explored the relationship between compliance with CPP (Czosnyka et al 1994, Portella et al 2005, Kim et al 2012, Capel et al 2014, Zeiler et al 2018a).The majority of the analysis aligns with the findings that compliance showed a negative correlation with CPP becoming more intact with the increase of CPP (Czosnyka et al 1994, Portella et al 2005, Capel et al 2014, Zeiler et al 2018a).However, compliance was reduced substantially when CPP exceeded 100 mmHg (Portella et al 2005).It is worth noting that one study deviated from this pattern, as it failed to identify a significant correlation between their defined compliance metric and CPP (Kim et al 2012).
While compliance demonstrated association with several parameters, some did not show any or significant correlation.These are sedative and vasopressor agents (Froese et al 2020(Froese et al , 2021)), elastance and PVI measured through lumbar infusion test (Speil et al 2012), CrCp, WT or CM (Petrella et al 2008) and body position (Raabe et al 1999).

Limitations/future direction
This systematic review was conducted on 76 studies, which varied in patient population, patient type, and different continuous measurement techniques of CC/CCR.In the case of the RAP metric, using two different ICP sensors can result in two different RAP values, which are affected by local differences (Eide andSorteberg 2013, Calviello et al 2018).Additionally, wrong RAP values can be measured using intraparenchymal frontal cortex sensors in patients with infratentorial lesions, which needs to be evaluated.As a consequence of using an anonymized database, that evaluation may be hampered (Calviello et al 2018).The data set size is a crucial factor in a compliance study, but due to the effect of surgery like decompressive craniectomy, confounding factors are introduced, which can affect the patient outcome.As a result, the data set becomes small, and the analysis range becomes narrower (Howells et al 2012).Furthermore, some of the studies showed a good correlation with the outcome or essential parameters to predict the outcome, but few studies did not conclude if it could be used clinically (Yau et al 2002, Ng et al 2005, Steiner et al 2005, Kim et al 2009a, Pineda et al 2018).Another limitation of RAP is that researchers may find it unclear how RAP will be affected by an open intraventricular drain, as it depends on a closed intracranial space (Steiner et al 2005).During an infusion test, a paradoxical decrease of RAP was noticed for some cases, which could not be interpreted (Czosnyka et al 1988).Moreover, in a dataset, a difference in age can affect the RAP result since ICP is a function of age.In addition, studies found an association between continuous CC/CCR with age (Kiening et al 2002(Kiening et al , 2005)).In terms of Ca and Ci, the calibration of these two compliance metrics cannot be completed due to the unknown cross-sectional area of arterial vessels.This makes the comparison between patients impossible, and thereby, only relative changes can be observed (Kim et al 2009a, Kazimierska et al 2021).While calculating Ca, assuming venous outflow being constant is not correct either (Capel et al 2014).Besides, changes in MCA diameter can impact the measure of CBFV and, thereby, Ca (Carrera et al 2011b).
This scoping review only includes the methods that involve human subjects.The animal studies were excluded in this review.Consequently, potential methods, crucial associations, findings, or predictive aspects related to continuous CC/CCR might not be encompassed in this review.
The methods for calculating continuous CC/CCR require advancements in precision and accuracy for practical application in clinical settings.RAP is the most researched topic in this field, indicating a potential area for improvement by focusing on its patterns and identifying characteristics and artifacts to improve its precision.Furthermore, there is an evident gap in exploring the neuroimaging sector, with limited studies discussing the correlation between neuroimaging and CC/CCR metrics.A promising avenue for enhancement lies in integrating ML with neuroimaging, potentially simplifying the determination of CC/CCR and their predictability.But in that case, the model needs to provide high accuracy and precision.

Conclusion
To conclude, this systematic review provides a thorough overview of continuous compliance measurement methods, emphasizing research related to RAP.At the same time, described the Spiegelberg Compliance Monitor, TCD feature, centroid metrics, and distinct peaks of ICPW prominently as well.Notably, a majority of studies focus on TBI and hydrocephalus patients, whereas most of the hydrocephalus cases encompass infusion studies and analysis on the effect of shunt.Another crucial aspect highlighted in this review is the numerous associations between these metrics and various physiological parameters, as well as patient outcomes, underscoring the utility of continuous CC/CCR in diverse medical contexts and its potential clinical applications.However, it is noteworthy that many studies recommended further analysis before widespread clinical application.