The intraoperative neurophysiological monitoring team

44.1

Introduction: airliners and operating rooms

After the Institute of Medicine reported up to 98,000 annual deaths in the United States from medical errors, a new focus on error causation and its possible amelioration began . As many as two-thirds of all errors were shown to arise from dysfunctional team dynamics, rather than incompetent care . More recent data in the United States continues to reveal high mortality (100,000 deaths/annum), high annual cost ($19.5 billion), and increased length of hospital stay (4.6 days) attributable to medical errors . Two decades before the 2000 Institute of Medicine report, data had revealed that the increasing number of catastrophic airliner accidents largely resulted from faulty communication and errant decision-making among flight crews . The airline industry, after some initial resistance, opened up crew member communication and flattened in-flight decision-making hierarchies. “Crew resource management” training was adopted to facilitate open communication, teamwork, and safety protocols . For example, United Airlines teaches this formulation: “I’m concerned, I’m uncomfortable, this is unsafe.” Similar skills have been gradually implemented in many health-care settings .

Within medical safety systems, two critical error-reducing preconditions are salient: situational awareness and critical language . Situational awareness has been defined as a, “dialogue, which keeps members of the team up to date with what is happening and how they will respond if the situation changes.” As in the United Airlines example, a safer environment within complex systems (such as an operating room) also means any individual with an ongoing responsibility within the system must use critical language to get everyone to “stop and listen” as soon as the risk for patient harm is identified . In intraoperative neurophysiological monitoring (ION), these preconditions translate to ongoing situationally aware knowledge of pertinent in-room and in-wound events by all members of the monitoring effort . When a threatening context/situation is temporally wedded to an ION alert, causation is more likely inferred, and the use of critical language is not only justified, it is required.

“Talk among physicians” and other advanced degreed providers are integral to any safety regime within acute areas of the hospital like the operating room (OR). Such talk “… is essential in the negotiation of professional relationships, the distribution of responsibility, the inducement of cooperation, and the assessment of competence” . Neither telephonic communication nor e-mail nor e-chat substitute satisfactorily for face-to-face communication. Further, “The current weaknesses in communication in the OR may derive from a lack of standardization and team integration … decisions are often made without all relevant team members present, and much communication is consequently reactive and tension provoking” .

ElBardissi and Sundt have introduced a specific error-reducing intraoperative model: Systems Engineering Initiative to Patient Safety. Like crew resource management, the ElBardissi and Sundt model emphasizes a “teamwork and communication” culture based on an essential premise: superb individual practitioner skill guarantees neither optimal error avoidance nor satisfactory patient outcomes . Specifically, these investigators have shown that 45% of the errors during cardiac procedures may be attributed to poor teamwork. The group cites “miscommunication” and “lack of team familiarity” as prominent factors. Therefore they have proposed: team training, standardized communication, preoperative briefings, and team familiarity .

Sacks et al. have published a systematic review of operating room culture (teamwork, communication, and safety) . Their review included studies measuring team performance metrics, surgical error rates, and outcomes. The authors conclude, “[Operating room] culture improvement appears to be associated with … positive effects, including better patient outcomes ….” Given the broad spectrum of intraoperative settings reviewed, there is no reason whatever to believe ION should be exempted from teamwork optimization.

44.2

ION teamwork and models of care

When individuals work cooperatively, that coordinated team effort may better affect progress toward a common purpose. The operating room team as a whole during ION includes an MD or PhD neurophysiologist, technological support if required, the surgeon, anesthesiologist, and other personnel who contribute to the effort. The physician or PhD neurophysiologist leads the ION team component. However, the idea of an ION team is only an abstraction unless practically translated to the “common purpose”—a reduction in errors and improved patient outcomes. If the medical error avoidance literature is taken seriously, the expert’s neuromonitoring task includes more than the dispassionate interpretation of tests. If required, the neurophysiologist expert must convey the data in a way that leads to the possibility of a clinically effective change in case management .

Each of the diverse models of ION care structurally determines team effectiveness. Throughout much of continental Europe, an ION expert with an advanced degree may personally attend patients in the operating room or may be positioned either in-room or in-house and, in any case, is ready to assist an in-room technologist if summoned. In the United States, these models exist primarily in large academic or referral centers. Much of the in-room ION in the United States is performed by either a technician (limited training, no certification) or a nationally certified technologist. Expert ION supervision is commonly provided remotely by an online physician. That physician may (or may not) hold a neurophysiology practice background or credentials. The online physician may enjoy limited acquaintance with in-room technical or professional staff and may often supervise many cases simultaneously. Even in the instance of a credentialed neurophysiologist physician, this latter model of “remote/online” care may disadvantage the ION expert with respect to absent situational awareness and limited intercollegial communications .

In Canada a revealing “model of care” debate is underway. The debate appears to be free from fee-for-service considerations that are present in the United States. Other jurisdictions around the world might benefit by an examination of the evolving Canadian experience.

On the one side, David Houlden and the Canadian Association of Neurophysiological Monitoring (CANM) support the “expert in the room” model. Because that “expert” is a qualified technologist at present, a major push is underway to credential these monitorists in nontechnical skills: (1) fluency with an advanced body of ION knowledge and (2) an intercollegial communication competence that may permit optimized discussions of differential diagnosis and possible interventions with the surgeon . This earnestly proposed “Certified Intraoperative Neurophysiology Practitioner” (CINP) model must be viewed as aspirational rather than current practice. In fact, the enhanced “technologist-only” CINP model has already been criticized as an online educational program rather than formal and thoroughgoing professional training with independent regulation .

On the other side, Jonathan Norton has surveyed Canadian orthopedic spine surgeons and neurosurgeons who order neuromonitoring. Currently, many self-monitor or rely on a technologist to perform and interpret the data. However, the overwhelming majority of these surgeons, if given the choice, would prefer a PhD neurophysiologist (92%–95%) or a neurologist (63%–75%) to interpret ION data . Although Norton has noted a “generally negative perception in Canada” of the US remote ION model, it is unclear how surgeons will be otherwise supplied with the desired numbers of advanced degreed experts. In a recent CANM newsletter article, Norton discusses “remote telepresence,” during which dedicated audiovisual connections may possibly provide the remotely placed expert an acceptable minimum degree of virtual situational awareness and face-to-face communication .

Later in this chapter, the challenges to effective ION teamwork, which are structurally embedded in some models of care, are detailed. Suggestions are made to redress these challenges where they are evident.

44.3

Injury prediction versus injury prevention

A major systematic review has shown that spine ION testing regimes can, indeed, predict neurological injuries . Using evidence-based methodology of the American Academy of Neurology, four Class I studies of transcranial electrical motor evoked potentials (MEPs) and/or somatosensory evoked potentials (SEPs) in various settings were reviewed. The three MEP-era Class I studies specifically reported the MEP alert criterion as 50%–60% MEP amplitude loss. This very sensitive alert indeed generated, as expected, no false negative reports among the studies (sensitivity=1.0). The combined positive predictive value (PPV) for serious neurologic injury was 0.30. This “low” PPV is compatible with good test accuracy in low injury prevalence settings . Unfortunately, a prediction analysis that looks at differences in outcomes (no injuries with negative testing, 30% risk of serious injury with positive testing) indirectly informs us about the role ION might play in injury prevention. An MEP alert does not simply serve as a proxy or end point for neurological injury. An alert often indicates the need to reverse or modify a surgical maneuver to avoid a true positive whenever possible. However, there is no assurance whatsoever that surgeons are bound to intervene after an ION alert. Wiedemayer examined surgeons’ responses to ION alerts. Only about 50% of the alerts were followed by an intervention .

The case for using test accuracy/prediction as a proxy for clinical effectiveness (injury prevention) is further weakened by studies that have found that accurate test performance is poorly associated with clinical efficacy . In a review relevant to the Wiedemayer findings, the authors conclude, “… improvements in test accuracy will not benefit patients unless they lead to changes in diagnoses and patient management, requiring evaluations of the effect of improved accuracy on decision making” .

Even in spine deformity correction, the reliable progression from test interpretation to an effective outcome may require further examination. In a retrospective review of 108,419 surgeon-reported Scoliosis Research Society (SRS) cases, which included a mix of deformity and other extradural spine surgery, the sensitivity for postoperative spinal cord deficit when both SEP and MEP were recorded was only 0.43 (false negative rate=57%) . This level of ION inaccuracy has not been previously reported in cases series or their systematic review, which identify sensitivity rates of 81%–100% . The SRS review’s explanations for poor ION sensitivity include underreporting by surgeons, unstandardized ION alert levels (especially MEP), and variations in “… how [signal changes] are reported to the operating surgeon ….” Surgeon underreporting and how changed ION data is reported to the surgeon are related phenomena that may indicate the need to consider and improve the culture of communication and collaboration between the neurophysiologist and surgeon.

The surgeon’s decision to change surgical management after an ION alarm may crucially depend on the collegiality with and trust of the communicating neurophysiologist. In fact, as we have seen (and shall see) from multiple literature threads, trusted communication likely determines the efficiency of the link between ION test accuracy (prediction) and reduced intraoperative injuries (prevention).

44.4

Systematic review of intraoperative teamwork

Appropriately changed case management after an alarm depends on well-functioning intercollegial communication. The earlier summarized structured review by Sacks et al. found a broad range of positive results when teamwork was optimized . Given those results, we have performed a systematic review that more narrowly compares surgical patient outcomes in optimized versus suboptimal teamwork environments.

One of us (RNH) conducted a literature search of the EMBASE and MEDLINE databases. SEARCH QUERY: 1992–2017 “operating room” and (“teamwork” or “medical education” or “surgical technique” or “surgical training”) and (“surgical error” or “surgical risk” or “patient safety” or “surgical mortality” or “outcome assessment”). SAS reviewed the 1259 papers found by our search. Our group focused on papers that primarily compared clinical outcomes among patients whose procedures were performed in environments typified by an optimized versus less than optimized culture of teamwork. Studies could also be included if discovered among the citations within papers revealed by the literature search. Studies were excluded if the outcome measure did not bear directly on clinical outcome. Other reasons for exclusion included self-assessed or tested perceptions of intraoperative communication or safety, various measures of surgical errors not including clinical outcomes, descriptions of education courses or simulation training, measures of operative duration or flow disruption, or preoperative discussion/checklist-alone related outcomes. Metaanalysis was used to obtain a summary intervention effect and 95% confidence interval (CI) across studies. An inverse variance method for a random-effects model was employed. Chi-squared and I ² statistics were obtained as measures of the consistency of study results (heterogeneity). A Z test was used for the null hypothesis of no effects for teamwork and communication on outcomes. Although study size did not influence inclusion, studies were size weighted within our final metaanalysis.

No randomized controlled trials were found. Four studies met our criteria. Included studies (1) retrospectively compared a teamwork-trained cohort to a contemporaneous untrained cohort, (2) compared outcomes before and after OR team training, (3) compared outcomes in well optimized versus poorly optimized teamwork conditions, or (4) prospectively measured a validated outcome index overtime as patient safety and teamwork programs were introduced .

Pettker et al. studied the effects of a gradually introduced safety and teamwork culture in both delivery and operating rooms. With respect to teamwork, these environments were adjudged by us as entirely similar to other operating rooms . Although the Forse study demonstrated significantly improved postoperative morbidity and mortality after teamwork training, the authors did not report sufficient variance data to include in our forest plot .

Chi-squared statistics for inconsistency between studies were not significant ( P =.22). I ² (34%) indicated that variability in effect estimates from heterogeneity may be unimportant.

Our metaanalysis further supports and establishes the crucial link between intraoperative teamwork and clinical outcome. The odds ratio of 0.60 (95% CI: 0.37–0.97, P =.04) means that there is a 37% decreased risk of postoperative adverse outcomes when procedures are performed in an environment that actively optimizes communication and teamwork, assuming a 10% baseline risk (40% decreased risk at 1% baseline risk) ( Fig. 44.1 ; Table 44.1 ).

44.5

Medical error avoidance

The error-reducing architecture of complex systems (such as ION) employs forgiving design (reversibility of signal loss after a risky surgical maneuver) and system redundancy (the reduced error rate when multiple tests are applied to the same neural system) . The latter instance may be exampled by fourth ventricular floor multimodality ION: the facial nerve nucleus may be monitored by direct stimulation, corticobulbar MEP, and free-running electromyography (EMG).

As we have seen from our systematic review, the application of forgiving design and redundancy to complex systems (“captained” by an experienced airline pilot or highly skilled surgeon, for example) must be associated with optimized intercollegial teamwork to obtain the best outcomes. Failed communication may result in errors ranging from harmless to catastrophic . The preexisting mind-set (bias) of copractitioners can be assessed and, if needed, mitigated to prevent an avoidable harm.