Unreactive alpha or alpha-theta coma electroencephalogram patterns with favorable neurologic outcomes: a report of three cases and literature review
Article information
Abstract
Background
Alpha coma and its variants theta coma and alpha-theta coma (ATC) are clinical coma states characterized by electroencephalographic patterns resembling those of wakefulness. The complete (unreactive) form of these states is invariably associated with poor outcomes among post-anoxic comatose patients.
Case Report
Herein, we present three adult comatose patients (aged 21, 45, and 51 years) with unreactive ATC patterns, who exhibited remarkable neurological improvement. Two patients experienced post-cardiopulmonary arrest (due to electrocution and viral myocarditis, respectively), while the third had severe septic encephalopathy. Electroencephalograms (EEGs) taken at least 24 hours after coma onset revealed unreactive ATC patterns. All patients survived, achieving significant neurological improvement. Remarkably, in one patient, the only symptom was minor incoordination during ambulation upon discharge.
Conclusion
These cases challenge the grave prognosis associated with this EEG pattern, particularly in post-anoxic comatose patients. Factors such as younger age, presence of shockable rhythms, and cause of arrest may have contributed to these favorable outcomes.
INTRODUCTION
Alpha coma (AC) refers to a clinical coma state characterized by an electroencephalographic pattern predominantly comprising alpha activity resembling that of wakefulness [1-5]. This is a rare pattern on electroencephalography (EEG), which is typically associated with a poor prognosis for neurologic recovery, particularly after cardiac arrest [2]. AC may also be observed in other clinical circumstances, such as head trauma, brainstem and thalamic lesions, and toxic metabolic abnormalities [2-4,6,7]. Early reports of post-anoxic AC dating back to the late 1960s described a comprehensive fatality rate. Over time, occasional reports of survival with improvements in cognition have emerged [2,8-13]. A related EEG pattern, the theta variant of alpha coma (theta coma [TC]), has also been described, emphasizing its association with poor outcomes. The transition from AC to TC and vice versa (a phenomenon termed alpha-theta coma [ATC]) in post-anoxic coma supports the hypothesis of a common pathophysiologic mechanism [5].
Two forms of AC, TC, and their transition (ATC) have so far been described. In their complete form, these patterns are unreactive to stimuli, monotonous (with few changes in amplitude), continuous, diffuse, or displaying frontally predominant alpha or alpha-theta activity. In contrast, the incomplete form exhibits partial reactivity, lower monotony, and a posterior distribution of alpha activity [5,14]. The prognostic significance of these patterns has predominantly been demonstrated in studies involving post-anoxic comatose patients [3-5,14]. In such cases, the complete form is almost invariably associated with mortality, while the absence of reactivity to verbal and noxious stimuli is the feature commonly cited as a crucial indicator of poor survival [2,3,14]. Conversely, patients with the incomplete form tended to have a higher probability of clinical recovery [2-5,14]. The prognostic implications of these patterns among metabolic or other non-anoxic causes of coma remain relatively poorly established in the literature, but have been generally associated with a more favorable outcome compared with post-anoxic coma cases [3]. Hence, the underlying etiology of coma, rather than the presence of this EEG pattern alone, appears to play a more critical role in the prognostic value [3,6].
Herein, we present the cases of three comatose patients with complete (unreactive, diffuse, and continuous) AC and ATC patterns on EEG who all achieved remarkable neurological improvement. Given the association of this EEG pattern with mortality and poor neurological recovery, a favorable neurological outcome is operationally defined as survival combined with improvement in the sensorium and cognition from the comatose state (Glasgow Outcome Scale of 3–5).
CASE REPORTS
Case 1
A 21-year-old male with no prior medical conditions sustained an electrical injury while working on a circuit breaker. This resulted in an immediate unresponsiveness due to cardiac arrest. He was transported to the hospital within 6 minutes, where chest compressions were promptly initiated. The initial cardiac rhythm revealed ventricular fibrillation. The patient underwent three defibrillation attempts, with amiodarone administered between shocks. Return of spontaneous circulation (ROSC) was achieved after a total downtime of 12 minutes, after which therapeutic hypothermia was initiated.
Following resuscitation, the patient developed post-anoxic myoclonic jerks, which were successfully controlled with midazolam infusion and valproic acid. After 24 hours, rewarming to physiological temperatures was performed; however, the patient’s Glasgow Coma Scale (GCS) score remained at 3, even 24 hours following discontinuation of midazolam, while the brainstem reflexes were preserved. EEG conducted 48 hours after therapeutic hypothermia revealed diffuse and continuous anteriorly predominant medium voltage 8–9 Hz alpha activity, with occasional 6–7 Hz theta activity (Fig. 1). No reactivity to auditory, photic, or pain stimuli was observed.
Electroencephalogram (EEG) of case 1, a 21-year-old male who developed fatal arrhythmia following electrocution. The EEG was performed 24 hours off sedation with a Glasgow Coma Scale of 3. An anteriorly predominant 8–9 Hz alpha and occasional 6–7 Hz theta rhythm can be observed. There was no reactivity to external stimuli. The bipolar montage is shown in this epoch with a sensitivity of 70 µV/cm.
The patient remained hospitalized for another 1 and a half months. During this period, he underwent intensive physical therapy, resulting in gradual improvement in his sensorium. By the time of discharge, he was fully oriented toward time, place, and person, achieving a GCS score of 15. Although he was able to ambulate with minimal assistance, he exhibited a tendency to fall due to incoordination.
Case 2
A 45-year-old female with no prior medical conditions experienced cardiac arrest secondary to viral myocarditis. Chest compressions were initiated immediately, and the patient was transported to the hospital within 10 minutes. Upon arrival, the initial cardiac rhythm revealed pulseless electrical activity, which subsequently progressed to ventricular fibrillation. The patient underwent three defibrillation attempts and received eight doses of epinephrine. ROSC was achieved following a total downtime of 33 minutes. After resuscitation, her GCS score was 3, with preservation of brainstem reflexes.
Following resuscitation, the patient developed post-anoxic myoclonic jerks controlled with midazolam and fentanyl infusions. Therapeutic hypothermia was performed for 24 hours, targeting a temperature of 32–34 °C. Plain cranial computed tomography scan revealed diffuse cerebral edema consistent with a hypoxic-ischemic injury. EEG conducted 24 hours after rewarming and cessation of sedation revealed a diffuse and continuous, well-modulated, medium voltage 8–9 Hz alpha activity, alternating with rhythmic medium voltage 6–7 Hz theta activity (Fig. 2). The background activity showed no change upon the application of noxious or auditory stimuli. The patient remained in hospital for 3 more months, during which time she showed significant improvement in the neurological status. She was ultimately discharged bedridden with a GCS score of 10 (E4 V2 M4).
Electroencephalogram (EEG) of case 2, a 45-year-old female who developed a fatal arrhythmia due to viral myocarditis. A diffuse and continuous background activity of medium voltage 8–9 Hz alpha with 6–7 Hz theta rhythm with absence of antero-posterior gradient can be observed. There was no reactivity to external stimuli. The EEG was performed 24 hours off sedation with a Glasgow Coma Scale of 3. The bipolar montage is shown in this epoch with a sensitivity of 70 µV/cm.
Case 3
A 51-year-old female with hypertension presented with sudden rotatory dizziness, quadriparesis, dysarthria, and dysphagia, and was subsequently diagnosed with bilateral medial medullary infarction. Secondary stroke prevention was initiated with aspirin therapy. However, 2 weeks after hospitalization, she developed septic shock due to hospital-acquired pneumonia, resulting in severe encephalopathy and subsequent coma.
An EEG performed without sedation revealed predominantly diffuse 8–9 Hz alpha activity with occasional 6–7 Hz theta activity (Fig. 3), showing no reactivity to auditory or painful stimuli. The patient remained hospitalized for 1 month, during which time she received antibiotics and bedside physical therapy. Tracheostomy and gastrostomy tubes were inserted to support breathing and nutrition, respectively. Gradual improvement in the patient’s sensorium was noted as the infection was treated, although her motor strength deteriorated to quadriplegia. She was eventually discharged awake, with a GCS score of 11 (E4 V1 M6), able to communicate through head and eye movements.
Electroencephalogram (EEG) of case 3, a 51-year-old female who was comatose due to severe septic shock. The EEG was performed without sedation with a Glasgow Coma Scale of 3. A diffuse and continuous 8–9 Hz alpha admixed with occasional 6–7 Hz theta activity with the absence of antero-posterior gradient can be observed. There was no reactivity to external stimuli. The bipolar montage is presented in this epoch, with a sensitivity of 70 µV/cm.
DISCUSSION
The generally accepted clinical significance of identifying AC patterns on EEG is the association with a poor prognosis. Further, the pathophysiologies of AC, TC, and ATC remain incompletely understood. However, animal data and autopsy findings suggest that the amygdala plays a role in generating alpha frequencies in the setting of thalamocortical dysfunction, particularly due to the relative preservation of the amygdala following hypoxic-ischemic injury [3-5,14]. When thalamocortical networks are disrupted, as in hypoxic-ischemic encephalopathy, the preserved amygdala and other subcortical structures compensate by assuming the role of primary pacemakers in brain electrical activity [3,5]. The extent of neuroanatomical dysfunction required to produce this EEG pattern may account for the generally poor outcomes associated with the complete form of AC [5].
Over the years, the prognostic significance of AC, TC, and ATC has been controversial, due to reports of survival among post-anoxic coma patients with these EEG patterns [15]. According to the EEG severity grading system published by the American Academy of Neurology for patients following cardiac arrest, AC or TC is considered grade 3, which institutional experience suggests correlates with a poor neurologic outcome [16]. However, some researchers have inferred that this EEG pattern alone is insufficient for accurate prognostication when considered in isolation [15,16]. Consequently, small-scale prospective and retrospective studies were conducted to re-evaluate the prognostic value of AC, TC, and ATC by considering other EEG factors, such as reactivity and persistence. In one study by Berkhoff et al. [5], which enrolled 14 comatose patients after cardiac arrest, 9 had complete ATC, and all died. The other five patients had incomplete ATC and survived. Three patients regained cognitive function, while the other two remained dependent on all activities of daily living. This led to the conclusion that complete ATC was invariably associated with poor outcomes in post-anoxic comatose patients. In another retrospective study by Fernández-Torre et al. [14], hypoxic-ischemic AC, TC, and ATC were found to be associated with poor prognosis and high mortality rates, with a greater probability of clinical recovery among incomplete forms; however, a significant limitation of the study was that patients did not undergo therapeutic hypothermia. In addition to an unreactive EEG pattern, other factors that appeared to be associated with mortality included old age (>50 years) and asystole as the initial rhythm. This has led to a general consensus that EEG reactivity may be the strongest sign associated with an increased likelihood of a favorable prognosis in patients with AC, TC, and ATC, while those with unreactive and persistent patterns are associated with a grave prognosis in survival and recovery [2-5,14].
Table 1 summarizes the pertinent clinical features of the three cases described herein. The first case is particularly notable for the cause of cardiac arrest, which was due to a high-voltage electrical injury, but ultimately led to a significant improvement in neurological status upon discharge. This is noteworthy, as two remarkably similar cases of cardiopulmonary arrest in 26- and 28-year-old males due to high-voltage electrical shock were reported by Grindal and Suter in 1975 [8] (Table 2). While comatose, the EEG patterns of these two patients displayed maximal alpha activity in the frontocentral regions, with one of them unresponsive to noxious, visual, or auditory stimuli (no stimuli were applied in the EEG recording of the other patient). Both patients exhibited ventricular fibrillation as the initial rhythm, similar to the first case described in our series. Gradual improvement in the neurological status was observed as early as 48 hours after the arrest. At discharge, both patients were able to communicate verbally. One showed residual difficulties in memory and naming objects, whereas the other had some difficulty with spatial and temporal orientations. In contrast, the first patient described in our study was discharged with only impaired coordination during ambulation.
Salient clinical features of the included cases
Clinical features of patients who survived with neurological improvement after complete AC, TC, or ATC, previously reported in literature
Despite the complete AC pattern observed following cardiac arrest in patients reported in the study by Grindal and Suter [8], including the first case presented here, remarkable neurological improvement was noted. Striking similarities in these cases that may have contributed to the favorable outcomes included the cause of injury (electrical injury), age group, and presence of a shockable rhythm during resuscitation. The latter two factors were aligned with the findings of a retrospective study by Fernández-Torre et al. [14], which indicated that younger age and ventricular fibrillation were associated with more favorable outcomes. It is also challenging to determine whether electrical injury as a cause of cardiopulmonary arrest itself contributed to the favorable outcomes, or if this kind of injury leading to cardiac arrest may have been coincidental, given that the patients were young and presented with shockable rhythms favoring more promising outcomes, despite also presenting with the complete forms of AC. Table 2 summarizes the clinical features of the cases reported by Grindal and Suter [8], including other coma patients who survived with neurological improvement, despite having complete forms of AC, TC, or ATC, as described in the literature. Notably, all were in the younger age group, with no patients aged >60 years, and most had shockable rhythms during resuscitation in postcardiac arrest cases.
In the second case presented herein, age and the development of a shockable rhythm during resuscitation were most likely the clinical factors that may have contributed to the remarkable recovery, despite the complete form of ATC. In a retrospective study by Fernández-Torre et al. [14], a 51-year-old male with complete AC experienced ventricular fibrillation during resuscitation with a Glasgow-Pittsburgh Cerebral Performance category outcome of 2 (indicating a good outcome with moderate disability). In the third case, coma was attributed to severe encephalopathy due to septic shock. This recovery is also consistent with the findings of a retrospective study by Kaplan et al. [3] involving 36 patients, in which a better outcome was generally observed in comatose patients with metabolic or septic insults, one of whom also did not exhibit reactivity of the AC pattern in the EEG. Indeed, the same study showed that when the etiology of coma involved cardiorespiratory arrest, the mortality was 8.5 times higher.
Exposure to sedative and anesthetic agents, such as propofol, midazolam, or fentanyl, is a crucial factor when using EEG patterns to evaluate prognosis and recovery. In addition to their residual effects, these agents, particularly propofol, have been reported to increase the alpha power frontally and the low-frequency EEG power [7,17]. However, when consciousness is lost, these effects result in frontal-predominant slowing with an overriding alpha, which is distinguishable from the AC pattern. The effects of fentanyl are consistent with the induction of high-voltage delta waves, which enhance the slow component of the EEG pattern, unlike the AC pattern [7]. In the first two patients described herein, midazolam was administered as a sedative and antiepileptic. This drug is known to induce fast activity in the beta frequency range and decrease alpha activity in healthy individuals. Further, at high doses, it may rarely induce an AC pattern, as shown in a study by Herkes et al. [18] involving eight acutely ill patients, in which an unreactive AC pattern was noted in a patient who had been given midazolam infusion for 9 days. Induction of this pattern appeared to be short-lived, halting 1 day after midazolam was discontinued, while the EEG rhythm showed excessive slow-wave activity with reactivity. In the first two cases described here, midazolam was administered at therapeutic doses for 24 hours to control post-anoxic myoclonus, while the infusion was discontinued at least 24 hours prior to EEG recording. Nevertheless, knowledge of the effects of these agents that are commonly administered to neurocritically ill patients is paramount, and must be considered when interpreting EEG patterns, which may require repeat determinations, especially when evaluating recovery and neurologic outcomes.
Advances in medical technology have enabled critically ill patients to survive longer. However, there is a growing recognition that continued aggressive treatment in critical care units may not always be beneficial [19]. Determining a proper balance between these approaches thus remains a significant challenge. Brain injury is the primary determinant of functional outcomes in critically ill patients, and most deaths associated with brain injury following cardiac arrest result from active withdrawal of life-sustaining treatment (WLST), as a poor neurological outcome is predicted [20]. Verkade et al. [19] previously highlighted that physicians’ predictions of a low likelihood of survival were major determinants in the decision to withdraw mechanical ventilation. As such, efforts to improve prognostic accuracy are crucial to avoid inappropriate WLST, or pursue futile treatments for patients with inevitably poor projected outcomes. Regarding the utilization of EEG, an independent association with patient outcomes even after adjusting for baseline neurological examinations has been reported [21]. While complete AC, TC, and ATC have been indicated to have grave prognostic significance, the cases presented here challenge the currently accepted implications of these electroencephalographic patterns, particularly in post-anoxic coma. This suggests that prognostication should be performed following a more individualized approach for each patient, considering all demographic, clinical, and neurophysiological parameters are augmented by knowledge from the existing literature. Because AC and its variants may also be transitory patterns leading to rhythms with a more favorable prognosis [8,13,14], it is reasonable to conduct serial EEGs to determine the persistence of patterns with poor prognostic implications, which may also be coupled with other neurophysiological parameters, such as somatosensory evoked potentials (SSEP). In the cases presented herein, repeat EEGs and SSEP were no longer performed, as a gradual improvement in the sensorium had already been observed. However, for patients in whom clinical recovery remains unapparent, serial EEGs to confirm the transitory state of the initial ominous pattern appear rational.
The three cases presented here challenge the traditionally-accepted prognostic implications of complete AC and its variants, particularly among post-anoxic comatose patients. Despite the complete, unreactive EEG patterns observed in all three patients, each demonstrated significant neurological improvement, indicating that favorable outcomes are possible under certain conditions. Factors such as younger age, the presence of a shockable rhythm during cardiac arrest, the specific cause of the arrest (such as electrical injuries), and the etiology of the coma appear to influence recovery. These findings further emphasize the importance of a more individualized approach to the prognostication of comatose patients with unreactive AC patterns, by placing equal importance on the clinical context and patient-specific factors. Moreover, performing serial EEGs and other neurophysiological assessments appears reasonable for providing further insight into the potential for recovery, indicating that AC and its variants could represent a transitory state, rather than a definitive marker of poor prognosis. Further research focusing on larger prospective studies to validate these observations will be essential to refine prognostic models and better understand the mechanisms underlying these EEG patterns, potentially improving prognostic accuracy and ultimately providing improved guidance for informed and individualized clinical decision-making in critically ill patients.
Notes
Ethics statement
In accordance with the institutional review board policies, the need for approval was waived for this study owing to its nature as a case series involving fewer than five patients. Similarly, the requirement for informed consent was waived, and the manuscript did not contain any identifiable information.
Conflict of interest
No potential conflict of interest relevant to this article.
Author contributions
Conceptualization: all authors. Data curation: all authors. Writing - original draft: MJSA. Writing - review & editing: all authors. All authors read and agreed to the published version of the manuscript.