Practice Standards for Multimodality Neuromonitoring

Multimodality neuromonitoring (MNM) refers to the integration, display, and interpretation of more than one physiological measurement source to guide clinical treatment for patients with brain injuries (Box 1). Monitoring cardiac function using electrocardiography (ECG), blood pressure, and volume plethysmography (Spo2) is a standard method in intensive care settings to identify and treat significant changes in a patient’s condition. However, measurements targeted at the brain have not yet been standardized in clinical practice and may vary based on institution, clinician experience, or clinical environment, including patterns that focus on intracranial pressure (ICP) or cerebral blood flow, cerebral metabolism, or cortical function. In adult intensive care, “multimodal monitoring” (commonly referred to as MMM) is almost synonymous with the use of invasive brain monitoring modes, while pediatric intensive care tends to use non-invasive modes more. Any monitor or device used for continuous or frequent bedside sequential measurements can be considered part of what is referred to here as MNM.

Box 1: Definitions Used in the Delphi Process

Multimodality Neuromonitoring (MNM) — “Neuromonitoring” refers to the use of any frequently (preferably continuously) available bedside brain physiological measurement to detect clinically significant events in real time. “Multimodal” refers to a more comprehensive assessment of the brain using more than one measurement source, implying a higher degree of complexity and applicable only to selected high-risk patients, typically in intensive care settings where neurological examinations are limited. MNM is distinct from neurological diagnostic methods such as radiological examinations or those performed infrequently or as needed, such as somatosensory evoked potentials or serum-based biomarkers.

Use Environment — Broadly, the use environment includes users, tasks, devices, and the physical and social environment of the service system used. In this case, the use environment refers to the healthcare environment, including the types of problems patients may encounter that MNM may be useful or helpful for.

Data — A comprehensive dataset of measurements collected during natural monitoring processes, which can be used for review, processing, or analysis.

Device — Technology used to generate brain monitoring measurements through direct connection with the patient, typically hardware such as intracerebral strain gauge catheters, optical imaging pads, or electrodes and amplifiers.

Measurement — Quantifiable parameters that constitute a set of numerical data, which define and set the conditions of a system (here referring to the brain and its neurophysiology).

Patterns — Strategies used to monitor the brain, including routine vital signs monitoring, non-invasive techniques such as scalp EEG, and invasive methods such as intracranial pressure monitoring.

Monitor — Hardware that records the output of local devices, typically within the patient’s room. For example, intracranial pressure monitors, electroencephalography systems, and bedside vital signs monitors.

Technology — Any hardware or software practically applied to multimodal network management.

Literature on the practical application of MMM includes a consensus summary statement, expert opinion recommendations, and surveys of adult and pediatric brain monitoring practices at most U.S. centers. European and Latin American practice surveys focused on traumatic brain injury (TBI) MNM for specific diseases. Additionally, there are narrative reviews on invasive and non-invasive MNM or MNM for patients with traumatic brain injury, aneurysmal subarachnoid hemorrhage (aSAH), and cardiac arrest. However, consensus has not been reached on the following aspects: a) the minimum technical requirements for MNM; b) which patients may benefit from MNM-guided treatment; c) the efforts required to integrate MNM measurements and interpret the resulting data; d) the training required to understand and accurately interpret MNM information. Only by addressing these gaps can we hope to provide consistency in clinical practice, enhance the clinical utility of MNM, and begin systematically improving individualized intensive care management strategies to impact treatment outcomes.

Evidence-based guidelines for emerging applications like MMM are scarce, thus expert opinion is needed to provide implementation standards. We conducted three rounds of the Delphi consensus process, which included clinicians with recognized expertise in treating multiple organ dysfunction, to address these issues and other knowledge gaps.

Practice Standards for Multimodality Neuromonitoring

The Use Environment for Multimodal Monitoring

Experts reached a consensus on important clinical factors when considering the practicality of multiple organ dysfunction, including the patient’s level of consciousness, underlying disease state, and the potential risk of secondary brain injury. Experts unanimously agreed that MNM is best used to guide individualized treatment, including the withdrawal of harmful therapies. Regarding whether pre-existing conditions or age would affect the efficacy of MNM, there was neither consensus nor disagreement among participants.

For invasive and non-invasive monitoring strategies, the use of MNM in comatose patients with TBI, aSAH, or subdural hemorrhage (ICH) was consistent. For case-based scenarios involving patients with Glasgow Coma Scale scores of 9-12, experts neither agreed nor disagreed on the utility of invasive versus non-invasive MNM. No case-based scenario reached consensus on the sole use of non-invasive MNM. However, pediatric experts differed from adult experts regarding the utility of MNM: a) early myoclonic status epilepticus after cardiac arrest (median 7.5, IQR 7-8 vs adult experts median 7, IQR 6-8; p=0.03), b) patients requiring veno-arterial extracorporeal membrane oxygenation (ECMO) (pediatric experts median 7.5, IQR 7.5–8.0 vs adult experts median 6, IQR 5–7; p<0.01), and c) patients requiring veno-venous ECMO (pediatric experts median 7, IQR 7–7 vs adult experts median 6, IQR 4–7; p=0.02).

Minimum Required Monitors, Devices, and Technologies

Ten invasive and non-invasive monitors and devices were deemed critical for MNM, many of which were also considered very important in several specific clinical usage environments (Table 2; and Supplementary Fig 1, http://links.lww.com/CCM/H385). Participants unanimously agreed that indices of cerebrovascular autoregulation (such as pressure reactivity index or tissue oxygenation index), optimal cerebral perfusion pressure (CPPopt), and quantitative electroencephalography were also important, but consensus was not reached. Only pediatric experts agreed on the use of near-infrared spectroscopy (NIRS) and prolonged transcranial Doppler (TCD).

The consensus-based technologies required for MNM include bedside display of single and multiple values, as well as time-locked trends displayed on a single visual monitor; access to high-resolution data; and the ability to manipulate the displayed data at the bedside (i.e., scrolling or zooming in and out of different time scales). The functions of bedside display summarizing data, setting thresholds or alarms, and real-time annotating MNM data to represent clinical events have reached consensus. The need for real-time remote review of MNM data while annotating clinical events and the ability to access high-resolution data independent of bedside devices have also reached consensus.

Minimum Necessary Clinical Efforts

It was unanimously agreed that integrating and interpreting multiple organ dysfunction data requires specific skills and expertise, and that intensive care unit staff (in general) do not have enough time in their daily clinical practice to integrate and interpret multiple organ dysfunction data. The ability to integrate brain and cardiopulmonary measurement data and process, analyze, and interpret this data to provide clinical relevance is crucial. Participants unanimously agreed that most ICU staff believe that regularly writing summaries and interpretations of MNM data helps inform decisions in daily clinical treatment.

Implementing Neuromonitoring

Consensus on the implementation of multiple organ dysfunction includes: the need for bedside interfaces to facilitate clinical understanding of multiple organ dysfunction information, and investing in educating bedside users (such as clinical nursing teams) to better understand and respond to multiple organ dysfunction information. Experts unanimously agreed that dedicated personnel with technical expertise (such as neuromonitoring technology specialists) and MNM “readers” help implement MNM, but this viewpoint has not reached consensus. The consensus reached is that consistency is needed in determining which patients require invasive or non-invasive MNM and which specific technologies should be used. The consensus suggests that local practice standards should be developed by engaging multidisciplinary stakeholders involved in clinical treatment to ensure the quality of care for patients receiving medical treatment.

Training Background and Educational Opportunities

Participants unanimously agreed that specialized training is needed to develop the expertise to interpret multimodal neuromonitoring data, but there was no consensus on the types of foundational training programs currently offered. Educational methods include practical workshops, guided operations, and clinical practice. Participants reached a consensus on nine core concepts that are essential for building a sufficient knowledge base for interpreting MMM data (Figure 2). Finally, participants unanimously agreed that professional skills should be certified through task forces or established societies, but consensus was not reached.

Practice Standards for Multimodality Neuromonitoring

Figure 2. Core Concepts Required for Clinical Neuromonitoring. Participants were asked to select each core concept, which were deemed to have a sufficient knowledge base for clinicians to understand and interpret multimodal neuromonitoring information. Dark gray concepts reached consensus, meaning that more than two of the three participants selected these concepts; light gray concepts did not reach a consensus threshold. The y-axis reflects the number of participants selecting each concept among a total of 35 participants. EEG = electroencephalography, sd = spread depolarization, Sz = seizure

Discussion

The Delphi process identified several broad areas of consensus regarding MMM practices. We reached consensus on the practicality of MNM for clinical decision-making and the necessity of maintaining consistency in patient selection and clinical treatment approaches. In the consensus process, we identified the fundamental technologies required for providing MMM. Experts unanimously agreed that MNM should be accompanied by the integration and interpretation of MNM information, which requires time and expertise independent of daily clinical treatment. Based on expert opinions, we summarized our findings into a framework for non-invasive MMM practice standards (Table 3).

We identified important clinical considerations regarding the utility of MNM. While some guidelines and an increasing number of clinical studies emphasize the practicality of MNM in guiding clinical management, experts in the Delphi process also recognized MNM as particularly helpful in avoiding or reducing potentially harmful therapies. This consensus is supported by benchmark evidence from South American trials: results from intracranial pressure treatment studies found that the use of invasive ICP monitoring can reduce or limit potentially harmful treatments such as muscle relaxants or increased sedation. The Brain Oxygen Optimization II study in severe traumatic brain injury reported that the MNM ICP/Pbto2 group (n=867 interventions) had fewer interventions than the ICP-only group (n=933 interventions).

In the consensus process, we also identified the indications for MNM — using invasive or non-invasive modes — reaffirming existing literature that primarily focuses on comatose patients with severe TBI (sTBI), aSAH, or ICH. However, experts did not reach consensus on the preferred non-invasive mode in any case-based scenario. An international group predominantly composed of adult intensive care experts had previously reached consensus that invasive ICP monitoring is a prerequisite for interpreting MNM, while a pediatric intensive care survey found that non-invasive modes are more commonly used. In pediatric intensive care, the indications for MNM are broader, including patients with hypoxic-ischemic injury or those on ECMO. In our study, experts unanimously agreed that MNM may be useful in these situations, but only pediatric experts reached consensus. If non-invasive modes are adopted more widely, the indications for MNM in other intensive care unit environments, including adult patients on ECMO or those at risk of secondary brain injury from other critical care conditions, may expand.

Participants unanimously agreed that invasive (ICP, cerebral perfusion pressure, Pbto2) or non-invasive (cEEG, quantitative pupil measurement) modes combined with cardiopulmonary measurements (arterial blood pressure [ABP], ECG, temperature, Spo2, and end-tidal carbon dioxide [ETco2]) are necessary for MNM. There was consensus on the importance of measuring cerebral autoregulation (including indices and calculations of CPPopt), but consensus was not reached. This may reflect recent findings from the Delphi process, which emphasized that while these measures are crucial for adults with sTBI, experts lack consensus on whether and how to incorporate these measures into clinical practice. In contrast, these measures and other specific measures such as cortical electrography and cerebral microdialysis (CMD) received unanimous recognition in specific circumstances. There is significant variability in the use and availability of brain-specific modes. In a European multicenter practice survey focused on TBI, less than a quarter of institutions considered invasive monitoring for comatose patients with negative CT scans or those with CT abnormalities that could not be evaluated, and in these cases, we unanimously agreed that MNM would be useful.

In a survey primarily targeting North American academic centers (72%), 58% reported using cEEG, while less than 5% of European traumatic brain injury centers used cEEG. Between 38-49% of centers reported using TCD, 15-26% used Pbto2, 0-19% used NIRS, 4-5% used rCBF, 3-9% used jugular venous oximetry, and 2-13% used CMD, with centers using autoregulation indices being less than 5%. There were also physician-dependent variations: neurocritical care physicians were more likely than other intensivists to use TCD, Pbto2, or rCBF. However, the use of cardiopulmonary parameters was consistent, as most centers in all surveys used ABP (77-93%), ETco2 (up to 90%), or cardiac output monitoring to guide management (26-53%).

In our study, the consensus reached by experts was that MNM data should be integrated and provided in a single visual display at the bedside and remotely for ease of review across time scales. In a survey of pediatric intensive care centers, only 8 out of 52 centers (15%) used platforms that integrated multimodal information, indicating a lack of recognition of this need in the use of MNM. We also reached consensus that integrating multimodal information and its clinical interpretation requires specific skills and expertise. Previous Delphi statements focused on single devices, supporting standardized implementation and interpretation by trained experts. Previous consensus also supported interpretation of MNM by professionals with rigorous training and expertise. Currently used multiparameter monitors rely on limited evidence, and conclusions about the relationships between multiple measurements are contradictory. The risk of using inaccurate information in automated analysis algorithms and smart alerts suggested in previous expert consensus recommendations exists. The consensus reached by our experts is that courses including the identified core concepts will allow for the development of management processes, determination of physiological thresholds, and creation of a standardized lexicon for patterns that emerge in integrated MNM data. This idea of utilizing standardization and training paradigms to achieve clinical relevance for new observations has been successful in critical care EEG, where standardized terminology has facilitated several practice-changing studies.

Finally, experts unanimously agreed that the expertise required to use and maintain the necessary technologies, interpret MNM information, and report clinical observations requires efforts independent of daily clinical care. A multicenter survey targeting international traumatic brain injury centers reported that the time and resource costs of MNM are often underestimated. Experts unanimously agreed that the daily work in intensive care cannot meet the time required for adequate interpretation of MNM information, and existing procedural codes do not capture the content of interpreting and reporting MNM information to clinical teams. Experts agreed on the need to develop a standardized framework for the interpretation and reporting of MNM to support multidisciplinary treatment teams. This model has recently been implemented in pediatric intensive care populations, where MNM has had a consistent impact on nursing decisions, shortening the duration of invasive monitoring and mechanical ventilation.

Our study has several limitations. First, there are no formal requirements for the Delphi process in the era of remote conferencing. The response rate of invited experts was below the 90% reported in systematic reviews of Delphi consensus processes in healthcare (IQR, 80-100%). However, the 35 participants in our study were twice the median of 17 participants in other Delphi processes (IQR, 11-31). We chose to exclude continuity non-invasive ICP measurements and other emerging technologies that lack clinical validation. Familiarity with existing devices may influence expert preferences, and the limited evidence base for MNM may introduce bias in consensus about the use environment. Our experts were broadly representative, but the ability to generalize consensus areas to pediatric intensive care is limited. We did not address emerging economies with unique needs and resources. Finally, we acknowledge that not all brain monitoring modes are equivalent: there are different risk/benefit thresholds between invasive and non-invasive modes, and some modes may be more helpful in assessing specific pathologies.

Conclusion

This Delphi process provides expert consensus supporting the practice standards for the use of multimodality monitoring to guide personalized intensive care for high-risk patients. We provide a framework based on the consensus areas to guide the selection of technologies required for MNM and the efforts needed to integrate and interpret MNM information independently from daily clinical treatment. This consensus statement also provides pathways to enhance MNM implementation and supports its broader use through training and education.

Key Points

Question: What is the consensus on the use of multimodality neuromonitoring (MNM) in clinical treatment?

Findings: MNM requires skills and expertise to utilize a minimal set of technologies to integrate and interpret consistent information across various clinical situations to guide and achieve personalized management.

Implications: Practice standards need to be established to optimally use MNM, and we will provide expert consensus to support these standards.

Source:Practice Standards for the Use of Multimodality Neuromonitoring: A Delphi Consensus Process

Critical Care Medicine

DOI: 10.1097/CCM.0000000000006016

Practice Standards for Multimodality Neuromonitoring

Leave a Comment Cancel reply