Repurposing Propofol as a Prognostic Probe for Return of Consciousness.

There is a growing realization that behavioral and neurophysiologic responses to general anesthesia can provide useful information regarding underlying brain health. For example, in response to concentrations of propofol or ether-derived inhaled anesthetic agents (e.g., sevoflurane and isoflurane) associated with loss of consciousness, people with healthy brains often have prominent anterior electroencephalographic α and θ spindles, which exhibit phase–amplitude coupling with slow δ waves (1). In contrast, surgical patients with preexisting cognitive impairment often fail to develop these typical, prominent anterior electroencephalographic spindles in response to propofol or inhaled general anesthetics (2). Furthermore, in patients with underlying neurologic vulnerability who experience postoperative delirium, electroencephalographic suppression tends to occur at relatively low concentrations of anesthetic agents (3, 4). General anesthetic agents might therefore be viewed as a neural stress test revealing brain health or vulnerability that might not otherwise be overtly manifest.

In this issue of the Journal, Duclos and colleagues (pp. 171–182) present results from a small clinical study in which they administer propofol to 12 patients who are in a minimally conscious state or who have unresponsive wakefulness syndrome, previously referred to as a persistent vegetative state (5). The idea behind this study is that the function and integrity of certain brain networks can be inferred by analyzing data from a combination of electroencephalographic channels. Propofol typically causes disruption of certain networks, which have been proposed to undergo adaptive reconfiguration, or in other words, a reorganization of functional connections that might help preserve overall network homeostasis. The authors leverage this concept, which has been demonstrated in both awake individuals performing a task and in the spontaneous activity of the anesthetized brain. They propose the adaptive reconfiguration index as a surrogate for network resilience; when there is little change in network configuration on exposure to propofol, the index is low, and when brain networks respond like that of a healthy individual, the index is high (5). Interestingly, three patients who had high adaptive reconfiguration indices in response to propofol all fully regained consciousness within three months. None of the patients with low adaptive reconfiguration indices regained consciousness. The intriguing hypothesis emerges that the combination of propofol with a readily derived, standard 18-channel–based electroencephalographic index can be used as a pragmatic, bedside prognosticator in the ICU for patients in a minimally conscious state (5).

Adaptive reconfiguration of brain networks during general anesthesia was first proposed on the basis of a study of propofol in healthy volunteers (6). Despite clear changes in functional connectivity in the anesthetized state, large-scale EEG-based networks still maintained a hub-like organization, much like a network of airports (6). It was later found that this preservation of global organization was likely driven by a reconfiguration of network hubs in the brain, with a reversal of posterior and anterior areas with rich functional connectivity (7). Returning to the airport analogy, it would be as if the major hub airport in Michigan shifted rapidly from Detroit to Lansing—the general type of network organization would be the same (there would still be a hub airport and a more local one), but the location of the high-traffic hub areas would be different. Such a shift in airport hubs would no doubt lead to different patterns or directions of air travel. Similarly, by using the more principled approach of computational modeling, it was found that the location of the functional hubs in the brain define the directionality of functional connectivity (8). In a past case report by some of the authors on the study and us, it was found that propofol induced the same characteristic shifts in hub location and directional connectivity in a patient with unresponsive wakefulness syndrome as it typically did in healthy volunteers (9). This was unexpected, but even more surprising was that this patient recovered, prompting the question of whether a reconfiguration of hub topology while he was still unconscious indicated that his brain networks were still healthy enough to respond flexibly.

Duclos and colleagues have now addressed this question in a larger patient cohort, demonstrating the further promise of using propofol to probe the response and resiliency of brain networks in patients with pathological states of unconsciousness (5). Although the adaptive reconfiguration index was 100% accurate in this study in predicting the return of full consciousness within three months, this finding should still be regarded as preliminary because only 3 of 12 patients regained full consciousness in the time frame of interest (5).

Importantly, the adaptive reconfiguration index is only informative in the context of a brief and reversible brain perturbation, leading to network disruption and allowing network reconfiguration. Such perturbation is reliably provided by a propofol challenge. The adaptive reconfiguration index, unlike alternative analytical approaches such as the perturbational complexity index (10, 11), performs poorly at diagnosing specific disorders of consciousness. Indeed, the very basis of a high index is that networks respond in a pathologic state as healthy individuals do. Thus, the adaptive reconfiguration index is proposed to have only prognostic rather than diagnostic value. One important question that is not sufficiently addressed by Duclos and colleagues is how the adaptive reconfiguration index performs for people who have covert consciousness, also known as cognitive–motor dissociation (11). Such individuals are conscious of the world and are aware of sensations, including pain (11). Hence, they are uniquely vulnerable. In our quest to alleviate pain and suffering, we would not wish to miss this diagnosis.

In conclusion, this provocative study provides a striking example of innovative and safe drug repurposing. Propofol was introduced in the 1970s as an intravenous sedative and anesthetic agent (12). Today, it is used extensively and safely for these purposes in operating rooms, procedural settings, and ICUs. Duclos and colleagues have performed a small, but rigorous, assessment suggesting that, after 60 years of rendering millions of people reversibly unconscious, propofol might now be able to reinvent itself as an oracle that can reliably reveal whose persistent unconsciousness is reversible and whose is not.