Should Consistent Command-Following Be Added to the Criteria for Emergence From the Minimally Conscious State?

Abstract

Objective:

To determine whether consistent command-following (CCF) should be added to the diagnostic criteria for emergence from the minimally conscious state (eMCS).

Design:

Retrospective cohort study.

Setting:

Inpatient rehabilitation hospital.

Participants:

Patients (N=214) with acquired brain injury resulting in disorders of consciousness (DoC) admitted to a specialized rehabilitation program.

Main Outcome Measures:

Difference between time to recovery of CCF and time to recovery of functional object use (FOU) or functional communication (FC), the 2 existing criteria for eMCS as measured by the Coma Recovery Scale–Revised (CRS-R).

Results:

Of 214 patients (median age, 53 years [interquartile range {IQR}, 34-66 years], male: 134 [62.6%], traumatic etiology: 115 [53.7%], admission CRS-R total score: 10 [IQR, 7-13]) admitted to rehabilitation without CCF, FOU, or FC, 162 (75.7%) recovered CCF and FOU or FC during the 8-week observation period. On average, recovery of CCF, FOU, and FC was observed within 1 day of one another, approximately 46 days (IQR, 38.25-58 days) post injury. One hundred and sixteen patients (71.6%) recovered FOU or FC prior to or at the same time as CCF.

Conclusions:

In patients recovering from DoC, CCF reemerges around the same time as FOU and FC. This finding may reflect the shared dependency of these behaviors on cognitive processes (eg, language comprehension, attention, motor control) that are essential for effective interpersonal interaction and social participation. Our results support the addition of CCF to the existing diagnostic criteria for eMCS, but further validation in an independent sample should be conducted.

The Aspen Neurobehavioral Conference Workgroup developed diagnostic criteria for the minimally conscious state (MCS) and emergence from MCS (eMCS),¹ 2 key recovery milestones for patients with disorders of consciousness (DoC). MCS is characterized by clearly discernible but fluctuating signs of consciousness (eg, visual tracking and command-following). Recovery of functional object use (FOU) or functional communication (FC) (abbreviated FOU/FC below) signal eMCS.¹ These behaviors were selected as eMCS criteria based on their role in supporting functional independence and effective interpersonal interactions. The diagnostic criteria for eMCS were established by expert consensus and have not undergone evidentiary validation.

The Coma Recovery Scale–Revised (CRS-R)^1,2 is among the most psychometrically robust measures for establishing a DoC diagnosis³ and includes direct evaluation of behaviors consistent with eMCS. However, at least 2 studies suggest that even conscious patients with severe brain injury may have difficulty achieving the operational threshold for eMCS.^4,5 Moreover, the United Kingdom National Guidelines on DoC include visual discrimination, one approach for assessing command-following, among eMCS diagnostic criteria.⁶ Identifying the full range of behaviors associated with recovery of personal agency is consistent with the original intent of establishing criteria for eMCS and may help prevent delayed initiation of comprehensive inpatient rehabilitation services.

We tested the hypothesis that consistent command-following (CCF) recovers at approximately the same time as FOU/FC based on the theoretical premise that expression of these behaviors is dependent on preserved connectivity of the language network and has similar reliance on linguistic decoding, vigilance, response persistence, and motor planning. Like FOU and FC, CCF is essential for active engagement in rehabilitation and is a strong independent predictor of functional recovery.⁷

Methods

This study was approved by the local Institutional Review Board. Data were acquired from the medical record and housed in Research Electronic Data Capture, a secure, web-based data capture tool.⁸ Informed consent was not required because all data were originally obtained clinically.

Measures

The CRS-R is a standardized neurobehavioral assessment instrument designed to evaluate auditory, visual, motor, oromotor, communication, and arousal functions in patients with DoC.² Transition from MCS to eMCS is marked by the presence of either (1) appropriate use of 2 different objects (FOU); or (2) accurate responses to 6 consecutive yes/no situational orientation questions (FC) on 2 consecutive CRS-R examinations. CCF is operationally defined as 4 consecutive accurate responses to 2 different commands for a total of 8 trials.

Participants

We included patients who met the following criteria: (1) had a diagnosis of traumatic or nontraumatic acquired brain injury resulting in DoC; (2) were at least 16 years of age; (3) were admitted to a comprehensive inpatient rehabilitation program; (4) had no evidence of CCF or FOU/FC on initial CRS-R administration; and (5) recovered FOU/FC and CCF during an 8-week observation period that corresponds to the duration of the specialized clinical DoC program. We excluded patients who (1) were not admitted for inpatient rehabilitation immediately after discharge from acute care (n=5); (2) did not have any valid CRS-R examinations during the observation period (n=15 with no examinations; n=8 with only invalid examinations); (3) had evidence of CCF or FOU/FC on admission (n=55); or (4) did not recover CCF or FOU/FC during the observation period (n=108).

Procedures

Trained clinicians administered the CRS-R twice weekly over the 8-week observation period as part of routine care. We divided patients into 5 groups: (1) Group1_CCF+FOU/FC: recovered CCF and FOU/FC on the same assessment day; (2) Group2_CCF→FOU/FC: recovered CCF prior to FOU/FC; (3) Group3_FOU/FC→CCF: recovered FOU/FC prior to CCF; (4) Group4_CCFonly: recovered CCF without recovery of FOU/FC; and (5) Group5_FOU/FConly: recovered FOU/FC without recovery of CCF. For each group, we evaluated the days between injury and the first instance of CCF and FOU/FC.

Statistical analysis

We report descriptive statistics for demographic and clinical characteristics, Kruskal-Wallis tests with Bonferroni correction and chi-square for group comparisons, and Mann-Whitney tests to compare days from injury to recovery of behaviors. We conducted analyses in SPSS v24.^a Statistical significance was concluded if P<.05.

Results

Among 405 patients with DoC admitted for inpatient rehabilitation between 2012 and 2020, a total of 214 (median age, 53 [interquartile range {IQR}, 34-66 years], male: 134 [62.6%], traumatic etiology: 115 (53.7%), admission CRS-R total score: 10 [IQR, 7-13], days between consecutive assessments: 4.8 [IQR, 3-5]) met inclusion criteria. Demographics did not differ between groups (table 1). Admission CRS-R total scores differed between groups (P<.001) and were lowest in Group4_CCFonly.

Table 1.

Study sample characteristics

Characteristic	Sample N=214	Group 1 CCF+FOU/FC n=60 (28%)	Group 2 CCF→FOU/FC n=46 (21.5%)	Group 3 FOU/FC→CCF n=56 (26.2%)	Group 4 CCF only n=21 (9.8%)	Group 5 FOU/FC only n=31 (14.5%)	P Value
Age, y	53 (34-66)	52 (37.75-60.5)	58 (29.5-69.75)	53 (28.5-62.25)	59 (47-66)	50 (31.5-67.5)	.515*
Sex, male, n (%)	134 (62)	37 (61)	34 (73)	34 (60)	13 (61)	16 (51)	.376†
Etiology, TBI, n (%)	115 (53)	33 (55)	21 (45)	38 (67)	7 (33)	16 (51)	.054†
Time from injury to rehab admission, d	26 (21-33)	26 (20-33)	26.5 (22-30.75)	26.5 (20.75-34.5)	24 (20-41)	26 (22.5-33)	.962*
Diagnosis at admission to rehab, VS/MCS minus/MCS plus (%)	25/48/26	18/56/25	43/30/26	14/51/33	52/33/14	12/64/22	.001†
CRS-R total score at admission to rehab	10 (7-13)	11 (8.75-14)	8 (4-11.75)	10 (7.75-13.25)	6 (4-9)	11 (8-13)	<.001*
FOU observed first (%)	40	36	35	43	NA	77	<.001†
FC observed first (%)	30	28	41	41	NA	19
FOU and FC observed for the first time on the same assessment (%)	20	35	24	16	NA	3

NOTE. Data are median (IQR) unless indicated. Group1_CCF+FOU/FC, Group2_CCF→FOU/FC, and Group3_FOU/FC→CCF recovered CCF and FOU/FC within 8 weeks of rehabilitation admission. After Bonferroni correction, admission CRS-R total scores differed between Group1_CCF+FOU/FC and Group2_CCF→FOU/FC, Group1_CCF+FOU/FC and Group4_CCFonly, Group3_FOU/FC→CCF and Group4_CCFonly, and Group4_CCFonly and Group5_FOU/FConly. Group5_FOU/FConly was more likely to demonstrate FOU rather than FC as the first sign of eMCS.

Abbreviation: d days; MCS minus, minimally conscious state without evidence of language function; MCS plus, minimally conscious state with evidence of language function; NA, not applicable; TBI, traumatic brain injury; VS, vegetative state.

^*Kruskal-Wallis Test.

^†Pearson χ² test.

For patients in Group1_CCF+FOU/FC, Group2_CCF→FOU/FC, and Group3_FOU/FC→CCF (n=162), days from injury to recovery of CCF and FOU/FC did not differ (injury to CCF=46.5 days [IQR, 39.25-59 days]; injury to FOU/FC=46 days [IQR, 38.25-58 days], P=.563) (fig 1), and 71.6% recovered CCF either concurrently with or after FOU/FC. Group4_CCFonly (n=21) recovered CCF 71 days (IQR, 54-86 days) after injury. Group5_FOU/FConly (n=31) recovered FOU/FC 46 days (IQR, 39.5-62 days) after injury. Group5_FOU/FConly was more likely to demonstrate FOU rather than FC as the first sign of eMCS (P<.001).

Fig 1.

(A) Temporal association between recovery of CCF and FOU/FC. On the x-axis, day 0 represents day of recovery of FOU/FC. Sixty patients recovered CCF on the same day as FOU/FC. Patients left of day 0 recovered CCF before FOU/FC; patients to the right of day 0 recovered CCF after eMCS. (B) Timeline of recovery of CCF and eMCS (ie, FOU/FC). Days are reported using medians. During the observation period, Group1_CCF+FOU/FC recovered CCF and FOU/FC on same assessment day, Group2_CCF→FOU/FC recovered CCF before FOU/FC, and Group3_FOU/FC→CCF recovered FOU/FC before CCF.

Discussion

To explore whether criteria for eMCS should be extended to include CCF, we assessed the temporal relationship between recovery of CCF and FOU/FC. Consistent with a prior study reporting that CCF and FOU/FC frequently co-occur,⁹ we found that CCF recovered at approximately the same time as FOU/FC. Currently, command-following, evidenced by either CCF or reproducible movement to command (ie, correct responses to at least three of four trials of 1 command), meets diagnostic criteria for MCS plus, an MCS subgroup characterized by behavioral signs of consciousness that include evidence of preserved language function.¹⁰ However, the cognitive demands of CCF are substantially higher, requiring correct responses on 4 consecutive trials of 2 different commands. Therefore, like FOU and FC, CCF may support return of effective communication and active participation in rehabilitation. Moreover, CCF, FOU, and FC all rely on language comprehension and attentional control, behaviors that are mediated by common pathways located within frontoparietal association cortices.¹¹

Approximately 10% of patients recovered CCF but not FOU/FC (Group4_CCFonly) during the observation period. These patients had the lowest admission CRS-R scores, suggesting a slower recovery trajectory. Approximately 15% of patients recovered FOU/FC but not CCF (Group5_FOU/FConly). This was the only group in which most patients (77%) demonstrated FOU as the first sign of eMCS, possibly reflecting impairment in language function.

Study limitations

Our sample is composed of patients admitted to a specialized inpatient rehabilitation program for DoC and may not generalize to other settings. In addition, more frequent CRS-R assessment, a longer observation period and alternate analytical approaches (eg, item response theory) may have identified additional patterns of recovery of CCF and FOU/FC. Finally, the immediate and long-term functional correlates of recovering CCF and FOU/FC across different time scales are unknown. Although these behaviors reemerge together, their effect on subsequent recovery of independence requires further study.

Conclusions

Recovery of CCF appears to follow the same trajectory as FOU/FC. These behaviors have similar processing demands and rely on proximally located regions of association cortex. This finding, which requires validation, supports extending the current diagnostic criteria for eMCS to include CCF. Including CCF in the eMCS diagnostic criteria may facilitate early detection of eMCS, improve individualized treatment planning, facilitate prognostication, and help avoid delayed initiation of comprehensive inpatient rehabilitation services.¹²

List of abbreviations:

CCF

consistent command-following

CRS-R

Coma Recovery Scale–Revised

DoC

disorders of consciousness

eMCS

emergence from minimally conscious state

FC

functional communication

FOU

functional object use

IQR

Interquartile Range

MCS

minimally conscious state