Single Case Reports on Late Recovery from Chronic D Isorders of Consciousness: a Systematic Review and Ethical Appraisal

NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

Original Article

Single case reports on late recovery from chronic d isorders of consciousness: A systematic review and ethical appraisal

Katja Kuehlmeyer,a Corinna Klingler,a Eric Racine,b,c,d Ralf J. Joxa

a Institute of Ethics, History and Theory of Medicine, University of Munich, Germany b Neuroethics Research Unit, Institut de recherches cliniques de Montréal (IRCM), Montréal, Canada c Department of Medicine and Department of Social and Preventive Medicine, Université de Montréal, Canada d Departments of Neurology and Neurosurgery, Medicine and Biomedical Ethics Unit, McGill University, Montreal, Canada

their eyes intermittently but do not show any signs of Abstract responsiveness and arguably no awareness [1, 2]. For Late recovery from the vegetative state/unresponsive wakefulness syn- patients who show signs of responsiveness such as drome (VS/UWS) three months after nontraumatic brain injury and one command-following, visual pursuit of objects or ges- year after traumatic brain injury is considered to be exceedingly rare. tural yes/no communication, the diagnostic category of The VS/UWS is declared permanent after these time frames. Prognosis MCS was introduced in 2002. According to the Aspen of recovery from the VS/UWS is central for decisions about life-sustain- workgroup guideline, a patient emerges from MCS ing treatment and hence of ethical relevance. when he reliably and consistently demonstrates func- Objectives: We aimed to describe single case reports of late recovery tional interactive communication or functional use of from permanent VS/UWS in scientific journals with a focus on the de- two different objects [3]. Evidence on prognosis and scription of improvements and outcomes. potential for recovery in patients with VS/UWS or MCS Methods: We conducted a systematic review of single case descriptions is scant and subject to different forms of biases such as searching PubMed and the Cases Database for synonyms of disorders of the self-fulfilling prophecy [4]. consciousness and the term “recovery”. Single case reports have a specific clinical and ethical Results: We screened 1406 records and identified15 single case reports significance when it comes to late recovery of VS/UWS in 17 scientific journals from 1977 to 2012. Recovery was noticed be- or MCS, especially with regard to the timeframe within tween four and 33 months after brain injury of non-traumatic etiology, which improvements in the patient’s cognition or func- and between 15 months and six years after traumatic brain injury with tional status are observed. In disorders of conscious- outcomes ranging from minimally conscious state to almost full recov- ness, time before recovery has paramount ethical im- ery (re-entering productive work). The reports were heterogeneous, and portance given that decisions about life-sustaining some of them lacked important information. treatment are heavily based on prognosis which, in Discussion: Single case reports on late recovery exist and call into ques- turn, worsens with the duration of the disorder [5]. tion the justificationfor timeframes in established prognostic guidelines. Professional guidelines have offered recommendations We suggest a systematic approach to follow-up on single cases (e.g. a about timeframes after which recovery should be con- prospective case registry) to improve the evidence-base for prognosis. sidered highly improbable. One of the most influential We also propose recommendations for an improved reporting and rec- reports is the consensus statement of the Multi-Society ommend further reflection on the role of case studies of recovery from Task Force (MSTF) on Persistent Vegetative State, is- disorders of consciousness in the discourse on end-of-life decision mak- sued in 1994, which regarded the VS/UWS to be per- ing. manent twelve months after traumatic brain injury and three months after nontraumatic brain injury [6]. After Keywords: unresponsive wakefulness syndrome, vegetative state, prog- this time, recovery of function was considered exceed- nosis, end-of-life decision making, late recovery, ethics ingly rare and the functional outcome was thought to almost invariably be severe disability. The MSTF based their recommendation on the findingthat only fivever - Introduction ifiedcases of late recovery were published in the scientific literature up until 1994 [7–10]. Further case re- Disorders of consciousness refer to conditions occur- ports were identified by the Task Force but in their ring after severe brain injury, such as coma, the vege- opinion provided insufficientinformation [11, 12]. The tative state/unresponsive wakefulness syndrome (VS/ British Royal College of Physicians’ guideline from UWS) and the minimally conscious state (MCS). Coma 2003 considered a VS to be permanent also twelve is a transient state during which the patient does not months after a traumatic brain injury, but only six open his eyes and remains unresponsive to his envi- months after nontraumatic events [13]. The guideline ronment. The VS/UWS refers to patients who open of the German Society for Neurology on hypoxic en-

Bioethica Forum / 2013 / Volume 6 / No. 4 137 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

cephalopathy, including experts from Switzerland and this current paper (see for example [20] for a normative Austria, does not give recommendations on time until ethical analysis on end-of-life decision-making for pa- permanency, but discusses the potential of prognostic tients with disorders of consciousness by RJJ). indicators measured within the first three days after brain injury (e.g. absent pupillary and corneal reflexes; absent somatosensory evoked potentials) [14]. A Euro- Methods pean guideline also refers to time frames (six and twelve months) after which the question of treatment We first devised an informal review protocol that spec- limitation (respectively “further active therapy of the ified the different steps to be undertaken. Following the patient”) may be raised [15]. logic of the PICOS acronym (patients (P), intervention Professional guidelines usually combine an evi- (I), comparison (C), outcome(s) (O), and study design dence-based approach including the systematic review (S)) we decided on the following inclusion criteria. (P) methodology with consensus procedures in order to Patients: We included all single case studies on patients give recommendations on diagnosis, therapeutic pro- that were either in a coma, the VS or the MCS (not pa- cedures or disease management [16]. There are only a tients with Locked-in Syndrome or severe cognitive im- few therapeutic measures that have been evaluated for pairments, but responsiveness exceeding the MCS). (I) this patient group in (randomized) controlled trials. A Intervention/Comparison: We decided to not exclude systematic review of these measures includes studies any article based on interventions administered. (O) on dopaminergic agents (levodopa, amantadine), zolp- Outcome: Only studies reporting late recovery from dis- idem, median nerve stimulation, deep brain stimula- orders of consciousness were included. We definedlate tion, extradural cortical stimulation, spinal cord stimu- recovery as an improvement that takes place three lation and intrathecal baclofen [17]. Some treatment months after a non-traumatic brain injury or twelve approaches improved the responsiveness in certain pa- months after a traumatic brain injury. Recovery was tients, but overall the studies are inconclusive. A Co- defined differently for patients in coma/VS/UWS or chrane review of sensory stimulation programs for in- MCS. For patients being comatose or in the VS/UWS, dividuals in coma or VS/UWS revealed no valid recovery was defined as any sign of responsiveness evidence to prove or rule out their effectiveness [18]. (see criteria for MCS). For patients in MCS recovery To our knowledge, no systematic review of cases of late was not only defined as emergence from MCS through recovery from disorders of consciousness has been regaining the ability to communicate verbally (see cri- conducted yet. If spontaneous late recoveries occur teria for emergence from MCS) but could also include they are often reported by the treating physicians in significant improvements in responsiveness within the case reports, and these may be taken up by the pa- boundaries of MCS from MCS minus to MCS plus [21]. tients’ families and the media, having potentially a pro- (S) Study Design: We only included single case reports found impact on health care practice [19]. There are and excluded case series and intervention studies. In- several potential benefitsto clinicians, family members tervention studies were understood as investigations and ethicists to have an understanding of the single that prospectively planned an intervention, involved a case reports of late recovery observed in clinical prac- study protocol and planned the measurement of out- tice and published in scientific journals. Such a com- come parameters. We only included medical case re- prehensive review a) could provide descriptions of ex- ports or case reports which gave medical details. We emplary courses and outcomes of recovery; b) could excluded short legal case reports as well as commen- indicate how innovative forms of treatment or medical taries on case reports. errors may have an impact on the patients’ recoveries We searched PubMed and the Cases Database (accessi- and c) allow insights into the circumstances under ble: http://www.casesdatabase.com), using a search which patients were followed up and the reports being strategy that consisted of synonyms of the disorders of published. Furthermore, our results could test the va- consciousness (“vegetative state”, “unresponsiveness lidity of the empirical basis of timeframes for prognos- wakefulness syndrome”, “prolonged coma”, “apallic tication of recovery which have a profound and imme- syndrome”, “coma vigil”, “minimally conscious state”, diate impact on end-of-life decision-making and a “consciousness disorders”) connected by the Boolean distal impact on public understanding of patients’ operator “or”. This list of terms was then linked by the prognoses. To shed light on these crucial aspects, we Boolean operator “and” to the term “recovery”. De- conducted a systematic review to answer the following scriptors from the thesauri of the databases were specific questions: (1) What are the single case reports added to the search strategy where applicable. Addi- on recovery from permanent VS/UWS after the time tionally we used filters for species (human) and lan- periods specifiedby the MSTF? (2) How are the late re- guage (English, German) in PubMed. Last searches coveries explained? (3) What are the outcomes of pa- were conducted on May 16 (PubMed) and May 28 tients? We discuss the findings with respect to their (Cases Database), 2013. In addition we screened the ethical implications, setting the stage for further nor- bibliographies of the papers to identify further relevant mative ethical analysis, which is beyond the scope of papers.

Bioethica Forum / 2013 / Volume 6 / No. 4 138 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

The titles of the papers were screened by the first au- ies are not dependent on statistical analysis we did not thor to discard those articles that were not concerned have to assess a risk of a statistical bias of results, but with recovery from disorders of consciousness. Ab- focused on reporting quality. As there are no standard stracts of papers that were not discarded were screened reporting guidelines for events of late recovery (like for by two independent reviewers. The teams who were adverse drug events) [22] we assessed whether infor- screening the abstracts always consisted of one medi- mation in the areas identifiedto be relevant (see above) cal student and either the first or the second author to was reported or not. We analyzed whether the report- ensure having both research and clinical competency ing was complete and precise enough to allow the in each team. We then sought access to the articles via reader to appraise the accuracy of the diagnosis and to the local university and the public library. Where nec- get a good impression of the outcome of the recovery. essary, we also contacted the authors. Full texts of the We neither evaluated the credibility of the reports nor articles were screened independently by KK and CK. did we exclude any reports from our analysis due to All cases were discussed with RJJ (neurologist and ex- lack of relevant information. Missing information was pert on disorders of consciousness) before final inclu- noted during data extraction. sion. Relying on the review by Dixon-Woods we could not We developed a data extraction tabloid. Information identify any standards for cross-case analysis in sys- considered relevant to our research question was the tematic reviews that fitted our research question [23]. patient’s age and gender, diagnosis prior to recovery, We extracted information from each single case de- duration until recovery, first signs of recovery, out- scription by means of content analysis using a tabloid come, follow-up, and explanation of recovery. Data was of inductive and deductive categories (within-case extracted as reported by the author (without further in- analysis) as Miles and Huberman have suggested [23]. terpretation) by either KK or CK, while the respective Then we discussed similarities and differences be- other controlled the extracted data. As single case stud- tween the cases, but due to the heterogeneous case reports we did not conduct any quantitative analysis or statistics. Instead we synthesized the data within a nar- rative summary (see table 1) and reported identified Figure 1: categories and missing data.

PRISMA 2009 Flow Diagram Results

n Based on the search strategy, we identified1406 poten- Records identiﬁed through Additional records identiﬁed tio database searching through other sources tially relevant publications. The application of the

ﬁca (n = 1391) (n = 15) screening process yielded a total of 15 cases of late re- nti covery from chronic disorders of consciousness pub- Ide lished in 17 scientificjournals between 1977 and 2012. Records screened (title) Records excluded (n = 1406) (n = 718) We display the search and inclusion process in a PRISMA flow chart (see figure 1). ing een

Scr Records screened (abstract) Records excluded (n = 688) (n = 586) Description of the publications

The cases were reported in twelve different journals, No access to Full-text articles assessed Full-text articles excluded, articles for eligibility with reasons mostly in the fieldof medicine, especially neurology and ed (n = 7) (n = 102) (n = 78) neurologic rehabilitation, except for one report which lud

Inc was published in an ethics journal and written by a phi- losopher [12]. Some articles were published in journals Studies included in Reasons for exclusion: analysis – No case report with high impact factors, such as the New England (n = 17) (e.g. commentary) – Reported more Journal of Medicine or The Lancet. The journal that than one case ty published the most single case reports was Brain Injury – No coma/VS/MCS ili – Artiﬁcially induced (n = 5). The purpose of the articles was not always to re-

gib Cases included in analysis coma (less due to double Eli – No recovery reporting of cases) port late recovery of cognition or improvements in func- – No late recovery (n = 15) tional status. Sometimes the authors intended to report the validity of prognostic markers (mostly by use of electroencephalography) or to report unexpected recov-

From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews ery despite unfavorable results of the prognostic tests. and Meta-Analyses: The PRISMA Statement. PLoS Med 6(6): e1000097. doi:10.1371/journal.pmed1000097 A few authors intended to report results of experimen- For more information, visit www.prisma-statement.org. tal treatment attempts outside of clinical studies or re-

Bioethica Forum / 2013 / Volume 6 / No. 4 139 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

vealed that an improvement followed a change in stan- Explanation dard care. The firstauthors of the reports were working in the following countries at the time of the publication An explanation for the unexpected recovery was miss- submission: USA (n = 8), Italy (n = 3), Germany (n = 2), ing in six case descriptions. Most authors hypothesized UK, Netherlands and Oman (n = 1 respectively). that different therapeutic measures might have ad- vanced the patient’s recovery (cranioplasty, drugs such as intrathecal baclofen, an experimental immunologi- Course of recovery cal treatment, the use of heated air for ventilation, ag- gressive efforts by the nursing staff which was not de- We identified ten cases of late recovery after brain in- scribed in detail). Some authors hypothesized that jury of nontraumatic causes (see table 1) [8, 12, 24–33] spontaneous changes in the brain might have taken and five cases of late recovery following traumatic place (axonal regrowth or neurochemical change kick- brain injury (see table 2) [7, 11, 34–36]. Among the started by preceding seizures) leading to recovery cases were two children [28, 33]. Recovery was noticed while others discovered misdiagnosis, although behav- between four and 33 months after brain injury of non- ioral responses to stimuli had been absent. traumatic etiology, and between 15 months and six years after traumatic brain injury. In one case the exact point of recovery was not identified,but the patient had Quality appraisal improved when the doctors assessed him 12 months after brain injury [28]. Fourteen patients were reported Given our aims, the most important aspects of the case to have been in a VS and one in a prolonged coma. In descriptions were (1) the description of the patient’s di- two articles both diagnoses VS/UWS and MCS had been agnosis prior to the recovery, (2) the description of the reported prior to recovery. Eight reports were pub- firstsigns of recovery and (3) the description of the out- lished before 2002. In one publication, the diagnosis come. (1) A good report allowed the appraisal of the di- (VS/UWS) and origin of the patient’s impairments was agnosis. Despite mentioning the diagnosis, the guide- unclear because the stroke described seemed not to be line followed for diagnosis was not regularly mentioned. responsible for the patients’ clinical deterioration [24]. An accepted behavioral test instrument (e.g. the The authors also mentioned a dementia state, but its CRS-R), which has been available during some of the role remains vague. The reported recovery was either period covered by the reports, was rarely used to con- from VS to MCS (mostly in the newer publications), firm the clinical evaluation. Often the doctor who re- within the spectrum MCS (from MCS minus to MCS ported the recovery seemed not to have been involved plus) or from VS/UWS or MCS to functional communi- in the diagnosis. Even though most authors explicitly cation. stated the diagnosis in the title and described them in The outcomes were heterogeneously reported with re- more detail in the text, it did not always become clear gard to the perspective of the authors and the level of whether the patient was really in the state mentioned the patient’s improvement. Many authors only de- or whether the recovery was rather a discovery of a scribed improvements in responsiveness (to auditory misdiagnosed patient. In two cases the authors even or visual stimuli); others reported motor improvements acknowledged uncertainty concerning diagnosis (VS or in oral feeding, as well as regained cognitive competen- MCS) [26, 29–31]. (2) Reporting of first signs of recov- cies and remaining deficits. Different test instruments ery was generally good. The firstsigns of recovery were were used to measure the outcome: the Glasgow Out- often observed by the author or a team member, iden- come Scale (GOS), the Glasgow Coma Scale (GCS), the tifiedand described in detail. If they reported a gradual Functional Independence Measure (FIM), the disability recovery, however, it was sometimes unclear at which rating scale, a participation index or intelligence tests point the recovery process started. The time between (e.g. the Wechsler Adult Intelligence Scale – R). brain injury and recovery was mentioned precisely (in The outcomes were found to be related to the follow-up months) in all but one paper, however, in four cases the period. The most detailed and accurate description was way the case was portrayed made it hard to under- available for “Kate” who made a remarkable recovery stand exactly when those events had taken place or after severe encephalomyelopathy and was followed-up what event had been identifiedas firstsign of recovery. fiveyears after recovery [29–31]. Only one patient was Even in the articles published after 2002, when guide- observed to make a recovery that allowed him to re-en- lines for the diagnosis of MCS were broadly adopted ter productive work [36]. His cognitive capabilities [3], it was not always reflected by the author whether even allowed him to resume his university studies. He it was an improvement to MCS, within MCS or an was followed for seven years after a car accident. Other emergence from MCS that was identifiedas a meaning- outcomes were rather unfavorable and the patients ful recovery. (3) The outcome was generally reported in probably remained dependent on medical and nursing a cursory fashion and rarely described in detail and en- care though the potential for further improvement was riched by validated outcome scales that were appropri- rarely discussed. ate to the patient’s condition. The application of un-

Bioethica Forum / 2013 / Volume 6 / No. 4 140 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

specific outcome scales often lead to floor or ceiling on late recovery from disorders of consciousness. effects, which means they were not able to accurately While brain damage due to hypoxic brain injury occurs indicate the patient’s improvements (either because a twice as often as traumatic brain injury [15], recovery large proportion of heterogeneous patients score high from hypoxic brain injury is known to be rarer than re- or low on the test). The data were hardly comparable covery from traumatic brain injury [5, 6]. However, we across different cases and attempts to standardize the found more case reports on late recovery from brain outcome by means of a secondary analysis using the injury of nontraumatic causes. This might be explained GOS failed due to a lack of description of the functional by the shorter time after which permanency can be as- status in the articles. In addition to that, most authors – sumed in non-traumatic cases. If we had used the six especially those who only described short follow-up pe- month cut-off for permanency of nontraumatic brain riods – did not discuss further potentials for improve- injuries according to the British and European guide- ment which would also have been of interest. lines. this would have reduced our cases of unexpect- edly late recovery from fifteen to six [8, 24, 27, 28, 32, 33], because the other patients would have recovered Discussion within the expected timeframe. If recovery occurs within the first six months after the injury, it may be We combined the method of systematic review with the easier detected because the patients may still be in hos- content analysis of qualitative single case studies to pital care and not yet discharged home or to long-term provide data on late recovery from brain injury. To our care facilities. In addition, because recovery from non- knowledge this is the firstsystematic single case review traumatic brain injury is known to be less probable,

Ta ble 1: Cases of late recovery from nontraumatic brain injury

Reference Age at Gen- Cause of the Diagnosis Interval First signs of Outcome* Last brain der brain injury prior to between recovery follow- injury M/F recovery brain up since (years) (Coma; injury recovery VS/UWS; and (months) MCS; LiS; recovery other) (months) Fellerhoff 73 at FStroke, maybe VS/UWS 7 Opened eyes on Responsiveness: looked at the children 8 2012 stroke; viral or but uttering demand; moved but no proof that she recognized them, 82 at bacterial “Maria”; them in the oral: swallow independently; motor: change infection or “deep requested direction; e.g. grab for the bar of the bed or the dementia sleeplike eyes followed hands of the grandchildren; later: death state”; person; turn head as final outcome “dementia state” Pistoia 52 M Cardiac arrest, VS/UWS 4 Sustained visual Responsiveness: minimally conscious; no 2008 CPR pursuit, reproduci- test: CRS-R score of 13/23, motor: ble visual fixation, tetraplegia might have hindered voluntary acts: e.g. performance of object manipulation followed visual or (key indicator of emergence from MCS) auditory stimuli, commands Al-Adawi 52 MSubarachnoid VS/UWS 6Increased Overall: significant improvements; 2 2006 hemorrhage and MCS awareness of widespread improvement in his (aneurysm); surroundings, cognitive and functional abilities; tests: cardiac arrest: verbalizes more and PPI improved from 0% to 50–60%, FIM anoxic brain better oriented in improved from 10% to 60–70% injury possible time and place Ford 53 FRespiratory VS/UWS 33 Respond to Communication: reply to questions no 2006 failure and after being questions with a with “I don’t wanna”; “I no power”; cardiac arrest responsive simple yes/no verbalization sporadic, and interrupted by months of silence Goodwin 6MStrangulation; VS/UWS Unclear; Clearly related to Vigilance: Patient alert and interactive, Unclear; 1993 cardio re-as- environment, test: GCS=15; oral: feed himself, (2 years pulmonary sessment responded motor: walk with walker; continued after the resuscitation after 12 appropriately to improvement in muscle function, accident) voice and receptive communication: receptive language language normal, expressive language limited by muscle spasticity; most of language could be understood

Bioethica Forum / 2013 / Volume 6 / No. 4 141 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

Macniven 26 FSevere VS/UWS, 6Respond to Motor: severe physical disabilities that 60 2003; encephalo- later MCS environment; initial require a wheelchair cognitive: Wilson myelopathy signs that level of intellectual, executive and memory skills 2001; consciousness was largely within the normal range; difficulty Menon improving, for in face and emotion expression 1998 example articulation recognition; almost complete cognitive of single words and recovery. communication: severe phrases dysarthria, use of a communication board Rosen- 43 MCerebral VS/UWS 17 Alert patient Oral: oral praxis normal; visual: a right 24 berg anoxia due to followed two-step homonymous hemianopia, motor: 1977 cardiac arrest commands by severely physically disabled; moved blinking, attempting fingers and some control of legs; name to talk, moving his parts of the body that were touched, right hand; full and proprioception intact, cognitive: voluntary eye intellectual functions partly recovered; movements and oriented (person, place, year and time), good movements memory performance impaired, amnesia of his tongue, but for time before and after admission; able bilateral facial to name simple objects but could not weakness recognize complex collections; recognize some letters, test: score 100 on verbal section of the WAIS with particular deficits noted on tasks requiring concentration, recent memory skills and learning new information. communication: tell stories and jokes Sara 44 MSpontaneous VS/UWS 19 a) or a) Appearance of Oral: eating ice-cream, motor: gradually Unclear; 2007 subarachnoid 21 b) spontaneous regained motor abilities with both 2 or 4 hemorrhage unintentional hands; starting to manipulate objects; months due to a movements in the unwrapping parcels, peeling tangerines, ruptured second and third turning on and off a portable radio aneurysm finger of the right receiver, trying to move his wheelchair, hand; never communication: able to articulate the reached the level of name of his son, to give yes/no answers an intentional to very simple questions movement

b) patient started to show aroused emotional responses to external stimuli, mild voluntary activity and partial recovery in cognition, spontaneous vocalization Steinbock 85 FStroke VS/UWS 4.5 Took small amounts Oral: dependent on feeding tube, no 1989 of food by mouth communication: able to communicate, and engaged in but often inconsistent in her responses conversation Tsao 5MNear- VS/UWS 7. 5 Gradually became Cognition: test: SB: at lower limit of the 9 1991 drowning aware of parents mildly mentally retarded range; social: good socialization skills; interacted well with friends in games and plays; communication: more recently producing 3–4 word phrases sponta- neously; overall: mildly retarded

Most recently published articles are described first; *categories derived from inductive content analysis; Abbreviations: M = Male; F= Female; CPR =Cardio- pulmonary resuscitation, Test scales: FIM=Functional Independence Measure, CRS-R=Coma Recovery Scale – Revised; PPI=Patient Participation Index; GCS=Glasgow Coma Score; WAIS= Wechsler Adult Intelligence Scale; SB= Stanford-Binet-Intelligence Scale:

Bioethica Forum / 2013 / Volume 6 / No. 4 142 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

Ta ble 2: Cases of late recovery from traumatic brain injury

Reference Age at Gen- Cause of Diagnosis Interval First signs of Outcome* Last brain der: the brain prior to between recovery follow-up injury M/F Injury recovery brain since (years) (Coma; VS/ injury recovery UWS; MCS; and (months) LiS; other) recovery (months) Sancisi 22 MCar accident VS/UWS 19 Consistent behavioral ADL: lives in the community, speaks, Approx. 2009 responses; visual walks, undertakes activities of daily 84 pursuit, appropriate living and outdoor leisure activity smiling, obeying independently; resumes university simple commands studies and has a competitive part-time with left arm job test: GOS=Moderate disability Faran 28 MCar accident VS/UWS 6 a) a) ERPs in EEG or Responsiveness: completely aware of 8 2005 or 20 b) b) articulating words his situation, recognized his family and friends, cognitive: anterograde amnesia ADL test: DRS=17 Childs 18 FMotor VS/UWS 15 a) or a) possible leg flexion Oral: oral feeding, cognitive: attention 43 1996 vehicle 17 b) and eye closure on span limited, consistent orientation to accident two occasions in person only, communication: follow response to conversations; communicating by commands words and short phrases; emotive: enjoyed pampering, and her mood was b) responses were usually euphoric; ADL: remained progressively more wheelchair-bound, disabled and consistent dependent for care Arts 18 FCar accident VS/UWS 30 Able to nod and Cognitive: regained mental capacity 42 1985 lift her leg with the exception of a severely defective short-term memory; read and watch television, communication: able to comprehend, communicate; write by means of a typewriter, social: establish relations with well-known persons. ADL: completely ADL-dependent; Further improvements: expected Tanheco 25 FMotorcycle Coma 72 Opened eyes, Vigilance: alert, oral: feed herself 14 1982 accident responded to objects cognitive: decreased retention, and conversation. memory, integration, communication: She was able to nod speech; ADL: groom herself, moderate and shake her head assistance for dressing and transfers accurately in needed, greater level of independence response to in activities and vocational productivity questions was precluded by the severities of her injuries; lives with family

Most recently published articles are described first; *categories derived from inductive content analysis; Abbreviations: ADL: Activities of daily living; DRS=Disability rating scale (score); GOS=Glasgow-Outcome Scale

physicians could be more surprised when it occurs, of single cases to be able to learn about the reporting of and thus be more inclined to publish these cases. This spontaneous recovery and to get detailed information emphasizes the ethical relevant consequences of guide- on the courses and outcomes of the recoveries, so we lines’ definitions of timeframes for declaring a condi- ruled out case series and cohort studies that also con- tion permanent: if it is set too narrow, some patients tain valid cases of late recovery (e.g. [9, 45–48]). In one may die from withdrawal of treatment who could have of the most recent prospective studies, late recovery of otherwise recovered significant functions. Due to our responsiveness was identified in ten patients with inclusion criteria to only report single case studies we long-lasting VS/UWS and six of them further pro- had to exclude reports on a famous case of late recov- gressed to consciousness [45]. First results of a new ery [37–40], because this case was published jointly and ongoing prospective study using data of a joint pa- with the description of another patient suffering from tient registry of five German rehabilitation centers as- TBI (like in other reports e.g. [41–44]). This famous pires to contribute further data on the prognosis of dis- case describes the recovery of a patient from MCS 19 orders of consciousness [49]. Prospective cohort studies years after brain injury. Upfront, this points out a cru- would have been favorable if we were interested in fre- cial limitation of our study. We focused on rich reports quencies of late improvements, but they have been crit-

Bioethica Forum / 2013 / Volume 6 / No. 4 143 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

icized for being biased by missing cases that were not prone to overestimate the probability of late recovery registered because early decisions on treatment limita- when consulting case reports like these. When overes- tions have been made. Yet, not including articles that timating the probability for late recovery in a case reported more than one case can only be justified by a where further improvements will not occur, one could practical reason: we had to decide how many cases do harm to patients and their families by deciding to were acceptable per report (two or ten?). This line had prolong the patient’s life without the patient having the to be drawn arbitrarily so we decided to only include potential to ever reaching a level of meaningful recov- single case studies. ery. There is also a risk that an overestimated potential A further limitation is that we conducted our search in for recovery could lead physicians to paternalistically a limited number of databases with a limited variety of overrule the patient’s autonomous wishes written in an keywords and language restriction to German and En- advance directive or expressed orally. As an interview glish and there were papers that we could not access study with next of kin of patients with disorders of (n = 7). Although we screened the relevant papers in consciousness in the long-term care setting indicated, pairs of two we consulted the neurologist only in the a perceived potential for further recovery might be last step of the inclusion process and cases might have weighed stronger by the patients’ family caregivers than been missed. the patients’ will [54]. A futile prolongation of life would Another important case was excluded from our analy- lead to costs without benefits for families and society. sis because the first signs of recovery were observed within the first year after traumatic brain injury, yet these signs of progression to MCS at first were not in- International guidelines, publication bias terpreted as signs of recovery by the treating physi- and bias of perspective cians [50–53]. This case is important for the discourse on end-of-life decision making because there was a le- The timeframes in the guidelines developed by the gal proceeding asking the court for permission to termi- MSTF have been based on only five identified and ver- nate the feeding by gastrostomy tube. It was reported ifiedcases of late recovery which served as evidence to that the patient had previously expressed verbally that propose that prognosis is poor after three or respec- she would not wish to continue living in the case of tively twelve months. This evidence should not be con- s evere brain injury [50]. As a result of a thorough neu- sidered sufficientto determine the probability of recov- ropsychological assessment by two independent evalu- ery and was in fact acknowledged to be suboptimal by ators the patient was found to be responsive, “sentient members of the Task Force themselves. Due to a publi- and wanted to live”. Remarkably, the patient was fol- cation bias, the amount of published case studies can- lowed-up several times up to ten years after brain not answer squarely the question when a VS/UWS injury, where she was living in the community with should be considered permanent. There are barriers to 24-hour care, was capable of communication, oral publishing case reports in peer-reviewed scientific feeding, and using a wheelchair. Although this case is journals like the researchers’ motivation and the not a proof of the potential for late recovery in other peer-review process. A researcher needs to observe the cases, it demonstrates how important a thorough anal- recovery, which probably requires a physician in a ysis of the individual’s chances for recovery is. double role as researcher and doctor. The clinician has to decide to report, which requires him to be interested in the topic, aware of the value this information could Appraisal of the clinical evidence have for colleagues, having the resources and the access to all the relevant information on the patient. In The evidence compiled in our study could be used ei- addition, the journal has to accept his report, which ther to justify continuing life-sustaining treatment be- could be hindered by objections against the evidence yond a certain timeframe by stating that these cases oc- level of single case reports. Those problems are mir- cur or to justify limiting life-sustaining treatment by rored in the case reports we identified: the quality of pointing out how rarely they are reported in scientific the publications was heterogeneous and some reports journals and how unfavorable their outcomes are. We were lacking information, due to either lack of access propose a third option of encouraging physicians to to important data or selective reporting. Furthermore, make detailed single case descriptions available which we speculate that researchers in countries where legal allow: a) the identificationof hypotheses for prognostic cases of treatment withdrawal were recently discussed markers and indicators for late recovery; b) the identi- (e.g. USA, Italy, and Germany) could be more motivated fication of treatment errors that could have prevented to publish cases of late recovery than researchers in the patient from recovering earlier; and c) to get insight other countries. into the long-term courses of VS/UWS and more impor- Prognostic guidelines recommending timeframes for tantly of MCS, where there is a lack of long-term data permanency can lead to self-fulfilling prophecies, because the diagnostic category has only been in use meaning that late recoveries are not observed because for a bit more than the last decade. Physicians might be physicians have decided to stop treatment without

Bioethica Forum / 2013 / Volume 6 / No. 4 144 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

knowing how the patient’s condition would have pro- nent VS/UWS does not benefit the patients because ceeded [5, 19]. The low expectation of recovery might they are permanently unaware and hence a prolonga- lead to a so called “neglect syndrome” and “therapeu- tion of life in a chronic VS/UWS is not in the best in- tic nihilism” meaning that patients with a poor progno- terest of the patient [58]. The German chamber of sis are treated with less effort and thus their conditions p hysicians does not accept a chronic disorder of con- worsen [55]. This is ethically relevant because it might sciousness alone as a prerequisite for the limitation of bias decisions about life-sustaining treatment towards treatment, and refers to the “medical indication” – the termination and impede research on long-term out- treatment proposal of the treating physician when con- comes of these patients. A clinical culture dominated by sidering the patients best interest [59]. Accordingly “therapeutic nihilism” – where guidelines about time- they assume that even patients in chronic VS/UWS can frames for permanency were being accepted by the still benefit from further treatment because there might treating physician given the lack of evidence-based be – at least in some cases – potential for improvement. therapeutic measures – could have further hindered In Great Britain the legal case of “M” stimulated a dis- the publication of cases of late recovery. cussion on the value of life in a MCS, when a judge de- Although our research cannot determine the statistical clined the patient’s family’s request for withdrawal of probabilities for late recovery, it demonstrates that the treatment – which in their perspective reflectedthe pa- established guidelines should be reexamined as late re- tient’s prospective will – with the reason that M recoveries are (in spite of the described obstacles) still sponded positively to some stimuli, in which she could observed and reported. It could be understood from be further supported, and “the importance of preserv- our study that a timeframe of three months is too short ing life” as being the decisive factor [60, 61]. It could to declare a VS/UWS of non-traumatic cause perma- furthermore be argued that the principle of justice nent because we discovered ten single-case descrip- would demand that patients and their families should tions of late recovery after this timeframe. Our results have fair chances to get treatment which might not be at least provide evidence that questions the assumption the case when the resources for long-term care facili- that these cases occur extremely rarely. That is not to ties are rare [20]. In the case of conflicting principles say that our study is able to give information on the they need to be weighed against each other. In the legal chances for late recovery three months after nontrau- case of M the principle of beneficence was weighed matic brain injury, which only prospective studies are higher by the judge than the principle of respect for au- able to. Furthermore, we assume that a lot of patients tonomy but it was argued that the court’s reasoning remain in the VS/UWS and die from complications but was flawed and that continued artificial nutrition and are never published extensively in journals. hydration was not in this patient’s best interests and thus should have been withdrawn [60]. How do our results relate to the discourse on end-of-life Appraisal of ethical implications decision making? In disorders of consciousness, pa- of the clinical evidence tient autonomy could only be represented through prospective autonomy or substitutive autonomy because To provide ethical guidance on decisions about life-sus- whenever the patient will be able to communicate, he taining treatment for patients with disorders of con- will have improved beyond a MCS. Prospective and sciousness we apply the principles of biomedical ethics substitutive autonomy rely on the availability of valid [56]. These principles – respect for autonomy, benefi- information to the patient himself or his surrogate de- cence, non-maleficence,and justice – need to be speci- cision maker who is most often a non-clinician without fied to be applicable for a concrete situation and have expert knowledge on disorders of consciousness. to be weighed and balanced against each other. The physicians’ judgment of the patient’s best interest In the context of life-sustaining treatment for disorders as a potential informant for patients or their relatives of consciousness, the principle of respect for autonomy and important actor in the decision-making process may be decisive when the patient prospectively ex- can only be assessed based on the background of valid pressed a self-determined and well-reflected will (e.g. information about the chances of recovery for analo- by advance directive or orally expressed wishes) gous patients which requires the availability of valid against life support in a permanent VS/UWS. The prin- clinical evidence. The provision of prognostic informa- ciples of beneficenceand non-maleficencedemand that tion has high ethical relevance in these cases. For ex- the potential benefit that the patient will reap from fur- ample, if a citizen – a potential patient – wants to ex- ther treatment does outweigh the harms of the treat- press his treatment wishes for a future state VS/UWS, ment. Potential harms could be the side effects of the he needs to balance the potential benefits of a future treatment or suffering caused by the sequelae of the treatment against the potential harms without knowl- brain injury (e.g. pain caused by spasticity). The prin- edge of his own future condition. It could be assumed ciple of nonmaleficence could especially guide a deci- that most patients would not want to reject treatment sion when further treatment is judged to be futile [57]. measures in acute coma (because chance for recovery It has been argued in legal cases that life in a perma- and good outcome are reasonably high), but would

Bioethica Forum / 2013 / Volume 6 / No. 4 145 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

want to reject treatment when recovery becomes highly for these patients. Psychologically, it could lead to more unlikely. Accordingly, advance directive forms often conservative decision making (waiting longer) because provide the option rejecting treatment in a “perma- doctors could be afraid of being held responsible legally nent”or “irreversible” VS/UWS. But this begs the ques- when the case registry monitors decision-making for tion when and how chronicity can be established. the registered cases. Our results show that the existing guidelines which use Yet, a registry would allow us to collect better data on time frames (e.g. three months after nontraumatic the “natural” course of recovery. A computer-based brain injury) to determine permanency are biased be- case registry is not a new idea. “The National Trau- cause they initially justified their timeframes with bi- matic coma data bank” was initially founded by univer- ased evidence (scientifically published reports of cases sity hospital centers in the US in 1979 for the registra- of late recovery available at that time). tion of patient data followed-up after traumatic brain Clearly, from an ethical and clinical standpoint, we injury [64, 65] and is meanwhile the largest aggrega- need better evidence to more accurately declare the tion of trauma registry data ever assembled (see [66] ). time point and clinical condition where the prognosis This data bank only allows the inclusion of cases after of a patient with a disorder of consciousness is not un- traumatic brain injury and is to our knowledge limited certain anymore but definitely hopeless. Citizens can to US-American facilities. We recommend an interna- then use this information to better express their pro- tional registry that also includes patients who experi- spective wishes thereby allowing physicians to respect enced nontraumatic brain injuries and we recommend their appraisal of different outcomes of brain injury like a focus on disorders of consciousness. VS/UWS, MCS and others such as recovery above MCS. Naturally there are organizational barriers, like a lack Our long-term goal should be to improve the evidence of funding, and research ethics issues to consider (e.g., for a valid prognostication of recovery from brain in- balancing the storage of private data and the lack of injury. Our short term suggestion, until the evidence base formed consent from the single patient against the high will have improved, is to thoroughly inform the general benefit of generating knowledge on affected patients). public and families about the different diagnostic Further discussion is needed to determine who should groups in disorders of consciousness, the potentials for have access to the data, e.g. professionals only or pro- improvements for the respective patient groups, the viding a platform to inform the general public on over- probable outcomes and the uncertainty under which all outcomes and recovery. Such a registry could im- they might have to make decisions about the refusal of prove the standards of patient follow-up and therefore treatment [19]. They might have to think about time- improve the evidence base for end-of-life decision mak- frames themselves and assess the impact of the dura- ing and public discourse. tion of a disorder of consciousness as one important variable. This would give patients the chance to bal- (2) Quality of case reporting of brain injuries ance for themselves for how long they would prefer the Given this ethical significance of prognostic informa- potential of improvement against the potential harm of tion, developing a standard for reporting cases with living in a condition where not being able to communi- disorders of consciousness after severe brain injuries cate and potentially suffering from pain which is a clin- would be advisable. Otherwise, the quality of the evi- ical aspect of high ethical relevance [62]. dence base could be too limited to usefully inform clinical and ethical discussions. From the appraisal of the (1) An international prospective registry for patients quality of the reports we derived initial recommenda- with disorders of consciousness after severe brain tions for the reporting of single cases of brain injury injuries (see figure 2) that would need to be discussed further In order to ground end-of-life decisions on a reliable with experts from different fieldssuch as neurorehabil- prognosis for the patient, more information is needed itation. on the possibility of late recovery and on the outcomes of those late recoveries. The best way to retrieve this information would be to perform prospective observa- Conclusion tional studies. This, however, is methodologically diffi- cult in such a heterogeneous and relatively rare spec- Single case reports of late recovery from disorders of trum of disorders [63]. An alternative option is an consciousness are published rarely, but they are not a international prospective registry for patients with dis- sufficientsource of evidence to conclude that after cer- orders of consciousness after severe brain injury: it tain timeframes recovery of consciousness can be ruled could a) lower the barriers for reporting a case of un- out. More information needs to be gathered about the expected recovery, creating an opportunity to also re- course of recovery of patients with disorders of con- port cases with unfavorable outcomes and b) pool data sciousness as a ground for end-of-life decision making. to improve the information base for decisions about Having reliable information on the probability of recov- treatment limitation. It could be argued that such a reg- ery is key to the formulation of an autonomous will for istry may already alter the practice of decision making or against future treatment and to the judgment of the

Bioethica Forum / 2013 / Volume 6 / No. 4 146 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

Figure 2: Reporting recommendations VS/SRW wird nach diesen Zeiträumen für chronisch As single case reports are an important source of information, initial recom- bzw. permanent erklärt. Die Rehabilitationsprognose mendations for reporting those cases were derived from our results and their ist beim VS/SRW zentral für Entscheidungen über le- analysis. benserhaltende Maßnahmen und daher von ethischer • Information on patient • Treatment administered Relevanz. o Date of injury • Description of recovery o Age at injury (if it applies) Ziel: Unser Ziel war es, Einzelfallberichte von später o Gender o Date of recovery o Comorbidity Erholung aus dem permanenten VS/SRW in wissen- o Timeframe until recovery schaftlichen Zeitschriften zu sammeln und dabei den • Type of injury o Firsts signs of recovery o TBI/NTBI o Course of recovery Schwerpunkt auf die Beschreibung der Rehabilitations- o Pathological anatomy o Final health state/ ergebnisse zu legen. residual deficits • Condition of patient o Follow-up period o Outcome of Glasgow Coma Scale Methoden: Wir führten ein systematisches Review von o Explanation: Hypothesis o Eye opening regarding causal mecha- Einzelfallbeschreibungen durch und suchten in Pub- o Outcome of behavioral tests nism for improvement Med und einer Fall-Datenbank nach Synonymen von o Outcome of neuroimaging o Potential for further techniques Bewusstseinsstörungen («chronic disorders of con- recovery o Outcome of neurophysiologic sciousness») und dem Begriff Erholung («recovery»). examinations o Outcome of laboratory tests Ergebnisse: Wir prüften 1406 Einträge und identifizier- o Frequency of examinations ten 15 Einzelfallbeschreibungen in 17 wissenschaft- lichen Zeitschriften zwischen 1977 und 2012. Erholun- gen wurden vier bis 33 Monate nach nicht-traumatischer patients’ best interest. Given the low prevalence of dis- Hirnschädigung und 15 Monate bis sechs Jahre nach orders of consciousness like the VS/UWS and MCS, sin- traumatischer Hirnschädigung beschrieben. Die Ergeb- gle case studies can play a role in informing those de- nisse («outcomes») reichten vom minimal bewussten cisions not only as preliminary information that could Zustand (MCS) bis zur beinahe vollständigen Rehabili- be pooled and statistically analyzed, but also to inform tation (Wiedereintritt in die produktive Arbeit). Die stakeholders on the courses and outcomes of disorders B erichte waren heterogen und einigen fehlten wichtige of consciousness. I nformationen. Diskussion: Einzelfallberichte über späte Erholung existieren und stellen etablierte prognostische Leitli- Acknowledgements nien in Frage. Wir empfehlen ein systematisches Vor- gehen, um Einzelfälle zu verfolgen (z.B. ein prospekti- We are grateful to Christine Marhold, Tobias Budick ves Fallregister), um die Evidenz für prognostische and Hendrik Terwort for their help with the literature Aussagen zu erhöhen. Wir formulieren Empfehlungen research and screening. We thank Gian Domenico für eine verbesserte Berichterstattung von Einzelfällen. Borasio for early feedback on a preliminary presenta- Die Rolle von Fallberichten später Erholung für den tion of a first impression of the data of the systematic Diskurs über Therapieentscheidungen am Lebensende review and the colleagues at the Institute of Ethics, His- sollte Gegenstand weiterer Reflexionen sein. tory and Theory of Medicine at the LMU, foremost Georg Marckmann and Sebastian Schleidgen for their suggestions to edit the manuscript. We would like to Resumé sincerely thank Galia Assadi for her help with the revi- sion of the paper. Parts of this study was funded by the Histoires de cas individuels souffrant de pathologies German Ministry of Education and Research (Neuro- chroniques de la conscience et qui se sont rétablis: ethics initiative: Neuroethics of Disorders of Conscious- une revue systématique et une évaluation éthique. ness; GP0801). Les rétablissements tardifs d’états végétatifs chroniques ou de syndrome d’éveil non répondant (VS/UWS) plus Zusammenfassung de trois mois après un accident cérébral non traumatique ou plus d’une année après une lésion cérébrale traumatique sont considérés comme excessivement Einzelfallberichte über späte Erholung von Bewusst- rares. L’état d’un patient VS/UWS est déclaré perma- seinsstörungen. Systematisches Review und ethische nent après ce laps de temps. Le pronostic de rétablisse- Bewertung ment d’un état VS/UWS est central pour les décisions Eine späte Erholung von einem Vegetativen Status bzw. concernant les traitements de maintien en vie. einem Syndrom reaktionsloser Wachheit (VS/SRW) wird drei Monate nach einer nicht-traumatischen Hirn- Objectifs: Nous avons visé à décrire les cas individuels schädigung und ein Jahr nach einer traumatischen de rétablissement tardif d’états VS/UWS permanents Hirnschädigung als extrem selten eingeschätzt. Der rapportés dans les revues scientifiques,avec une atten-

Bioethica Forum / 2013 / Volume 6 / No. 4 147 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

tion particulière portée à la description des améliora- 8. Rosenberg GA, Johnson SF, Brenner RP. Recovery of Cognition af- tions et des résultats. ter Prolonged Vegetative State. Annals of Neurology. 1977;2: 167–8. Méthodes: Nous avons conduit une revue systématique 9. Higashi K, Hatano M, Abiko S, Ihara K, Katayama S, Wakuta Y, et al. Five-year follow-up study of patients with persistent vegetative des descriptions de cas individuels. Nous avons mené state. J Neurol Neurosurg Psychiatry. 1981;44:552–4. notre recherche sur les sites de PubMed et de Cases Da- 10. Snyder BD, Cranford RE, Rubens AB, Bundlie S, Rockswold GE. tabase pour les synonymes de pathologies de la Delayed Recovery from Postanoxic Persistent Vegetative State. An- nals of Neurology. 1983;14:152. conscience et pour l’expression « rétablissement ». 11. Tanheco J, Kaplan PE. Physical and Surgical Rehabilitation of Pa- tient After 6-Year Coma. Archives of Physical Medicine and Reha- Résultats: Nous avons examiné 1406 rapports et iden- bilitation. 1982;63:36–8. tifié 15 cas dans 17 revues scientifiques de 1977 à 12. Steinbock B. Recovery from persistent vegetative state? The case of Carrie Coons. Hastings Cent Rep. 1989;19:14–5. 2012. Le rétablissement a été observé dans la période 13. Royal College of Physicians. The Vegetative State: Guidance on di- allant de 4 à 33 mois après un accident cérébral non agnosis and management. [Report of a working party of the Royal traumatique, et dans la période allant de 15 mois à 6 College of Physicians]. London: Royal College of Physicians; 2003. 14. Deutsche Gesellschaft für Neurologie. Hypoxische Enzephalopathie ans après une lésion cérébrale traumatique, avec des (HE) résultats allant d’un état minimalement conscient 15. von Wild K, Gerstenbrand F, Dolce G, Binder H, Vos P, Saltuari L, (MCS) à un rétablissement presque complet (retour sur et al. Guidelines for Quality Management of Apallic Syndrome / Vegetative State. European Journal of Trauma and Emergency Sur- le marché du travail). Les compte rendus étaient hété- gery. 2007;33:268–92. rogènes et il manquait des informations importantes 16. Burgers JS, Grol R, Klazinga NS, Makela M, Zaat J. Towards evi- dans certains d’entre eux. dence-based clinical practice: an international survey of 18 clinical guideline programs. Int J Qual Health Care. 2003;15:31–45. Conclusions: Des histoires de rétablissement tardif 17. Georgiopoulos M, Katsakiori P, Kefalopoulou Z, Ellul J, Chroni E, Constantoyannis C. Vegetative state and minimally conscious state: existent et mettent en question les directives acceptées a review of the therapeutic interventions. Stereotactic Functional concernant les pronostics. Nous suggérons un suivi Neurosurgery.88:199–207. plus systématique des cas individuels (p. ex. un registre 18. Lombardi F, Taricco M, De Tanti A, Telaro E, Liberati A. Sensory stimulation of brain-injured individuals in coma or vegetative state: international) pour améliorer le pronostic. Nous propo- results of a Cochrane systematic review. Clinical Rehabilitation. sons aussi des recommandations pour une améliora- 2002;16:464–72. 19. Jox RJ, Bernat JL, Laureys S, Racine E. Disorders of consciousness: tion dans l’annonce de ces cas et pensons qu’il faut responding to requests for novel diagnostic and therapeutic inter- pousser la réflexion sur le rôle des cas de rétablisse- ventions. Lancet Neurol. 2012;11:732–8. ment tardif dans l’optique des prises de décision 20. Jox RJ. End-of-life decision making concerning patients with disorders of consciousness. Res Cogitans [Internet]. 2011 8 Nov 2011; concernant la fin de vie. 1(8):[43-61 pp.]. Available from: http://www.klinikum.uni-muen- chen.de/Projekt-Neuroethik/download/de/forschung/Res_Cogi- tans_Ralf_Jox.pdf. 21. Bruno MA, Vanhaudenhuyse A, Thibaut A, Moonen G, Laureys S. From unresponsive wakefulness to minimally conscious PLUS and Correspondence functional locked-in syndromes: recent advances in our under- Katja Kuehlmeyer standing of disorders of consciousness. Journal of Neurology 2011; Ludwig-Maximilians-Universität München 258:1373–84. 22. Kelly WN, Arellano FM, Barnes J, Bergman U, Edwards IR, Fernan- Institut für Ethik, Geschichte und Theorie der Medizin dez AM, et al. Guidelines for submitting adverse event reports for Lessingstr. 2 publication. Pharmacoepidemiol Drug Saf. 2007;16:581–7. D-80336 München 23. Dixon-Woods M, Agarwal S, Jones D, Young B, Sutton A. Synthesis- ing qualitative and quantitative evidence: a review of possible E-Mail: katja.kuehlmeyer[at]med.lmu.de methods. J Health Serv Res Policy. 2005;10:45–53. 24. Fellerhoff B, Laumbacher B, Wank R. Responsiveness of a patient Manuscript submitted: 3.7.2013 in a persistent vegetative state after a coma to weekly injections of Revisions submitted: 29.10.2013 autologous activated immune cells: a case report. Journal of Medi- Accepted: 29.10.2013 cal Case Reports [Internet]. 2012 26.05.2013 6(6). Available from: http://www.jmedicalcasereports.com/content/6/1/6. 25. Pistoia F, Sacco S, Palmirotta R, Onorati P, Carolei A, Sara M. Mis- References match of neurophysiological findings in partial recovery of con- 1. Laureys S, Celesia GG, Cohadon F, Lavrijsen J, Leon-Carrrion J, sciousness: a case report. Brain Inj. 2008;22:633–7. Sannita WG, et al. Unresponsive wakefulness syndrome: a new 26. Al-Adawi S, Burke DT, Mastronardi SE. Seizure heralding func- name for the vegetative state or apallic syndrome. BMC Medicine. tional recovery in a patient with apallic syndrome: A case report 2010;8:68. with retrospective-prospective observation. Epilepsy Behav. 2. Jennett B, Plum F. Persistent vegetative state after brain damage. 2006;8:776–80. A syndrome in search of a name. Lancet. 1972;1:734–7. 27. Ford GP, Reardon DC. Prolonged unintended brain cooling may in- 3. Giacino JT, Ashwal S, Childs N, Cranford R, Jennett B, Katz DI, et hibit recovery from brain injuries: case study and literature review. al. The minimally conscious state: definition and diagnostic crite- Med Sci Monit. 2006;12:CS74–9. ria. Neurology. 2002;58:349–53. 28. Goodwin SR, Toney KA, Mahla ME. Sensory evoked potentials ac- 4. Becker KJ, Baxter AB, Cohen WA, Bybee HM, Tirschwell DL, New- curately predict recovery from prolonged coma caused by strangu- ell DW, et al. Withdrawal of support in intracerebral hemorrhage lation. Crit Care Med. 1993;21:631–3. may lead to self-fulfilling prophecies. Neurology. 2001;56:766–72. 29. Macniven JA, Poz R, Bainbridge K, Gracey F, Wilson BA. Emotional 5. Bernat JL. Chronic disorders of consciousness. Lancet. 2006;367: adjustment following cognitive recovery from ‘persistent vegetative 1181–92. state’: psychological and personal perspectives. Brain Inj. 6. Medical aspects of the persistent vegetative state (2). The Multi-So- 2003;17:525–33. ciety Task Force on PVS. N Engl J Med. 1994;330:1572–9. 30. Wilson BA, Gracey F, Bainbridge K. Cognitive recovery from “per- 7. Arts W, van Dongen HR, van Hof-van Duin J, Lammens E. Unex- sistent vegetative state”: psychological and personal perspectives. pected improvement after prolonged posttraumatic vegetative Brain Inj. 2001;15:1083–92. state. J Neurol Neurosurg Psychiatry. 1985;48:1300–3. 31. Menon DK, Owen AM, Williams EJ, Minhas PS, Allen CM, Boniface

Bioethica Forum / 2013 / Volume 6 / No. 4 148 NEUROETHIK / NEUROÉTHIQUE / NEUROETHICS

SJ, et al. Cortical processing in persistent vegetative state. Wolfson 49. Grill E, Klein AM, Howell K, Arndt M, Bodrozic L, Herzog J, et al. Brain Imaging Centre Team. Lancet. 1998;352:200. Rationale and design of the prospective german registry of outcome 32. Sara M, Sacco S, Cipolla F, Onorati P, Scoppetta C, Albertini G, et al. in patients with severe disorders of consciousness after acute brain An unexpected recovery from permanent vegetative state. Brain injury. Arch Phys Med Rehabil. 2013;94:1870–6. Inj. 2007;21:101–3. 50. McMillan TM. Neuropsychological assessment after extremely se- 33. Tsao CY, Ellingson RJ, Wright FS. Recovery of cognition from per- vere head injury in a case of life or death. Brain Inj. 1997; sistent vegetative state in a child with normal somatosensory 11:483–90. evoked potentials. Clin Electroencephalogr. 1991;22:141–3. 51. McMillan TM, Herbert CM. Neuropsychological assessment of a po- 34. Childs NL, Mercer WN. Brief report: late improvement in con- tential “euthanasia” case: a 5 year follow up. Brain Inj. 2000;14: sciousness after post-traumatic vegetative state. N Engl J Med. 197–203. 1996;334:24–5. 52. McMillan TM, Herbert CM. Further recovery in a potential treat- 35. Faran S, Vatine JJ, Lazary A, Ohry A, Birbaumer N, Kotchoubey B. ment withdrawal case 10 years after brain injury. Brain Inj. Late recovery from permanent traumatic vegetative state heralded 2004;18:935–40. by event-related potentials. Journal of Neurology, Neurosurgery 53. Shiel A, Wilson BA. Assessment after extremely severe head injury and Psychiatry. 2006;77:998–1000. in a case of life or death: further support for McMillan. Brain Inj. 36. Sancisi E, Battistini A, Di Stefano C, Simoncini L, Montagna P, Pip- 1998;12:809–16. erno R. Late recovery from post-traumatic vegetative state. Brain 54. Kuehlmeyer K, Borasio GD, Jox RJ. How family caregivers’ medical Inj. 2009;23:163–6. and moral assumptions influence decision making for patients in 37. Voss HU, Uluc AM, Dyke JP, Watts R, Kobylarz EJ, McCandliss BD, the vegetative state: a qualitative interview study. J Med Ethics. et al. Possible axonal regrowth in late recovery from the minimally 2012. conscious state. J Clin Invest. 2006;116:2005–11. 55. Fins JJ. Constructing an ethical stereotaxy for severe brain injury: 38. Wijdicks EF. Minimally conscious state vs. persistent vegetative balancing risks, benefits and access. Nat Rev Neurosci. 2003;4: state: the case of Terry (Wallis) vs. the case of Terri (Schiavo). Mayo 323–7. Clinical Proceedings. 2006;81:1155–8. 56. Beauchamp T, Childress J. Principles of Biomedical Ethics: Oxford 39. Laureys S, Boly M, Maquet P. Tracking the recovery of conscious- University Press, 2013. ness from coma. J Clin Invest. 2006;116:1823–5. 57. Jox RJ, Schaider A, Marckmann G, Borasio GD. Medical futility at 40. Fins JJ, Schiff ND, Foley KM. Late recovery from the minimally the end of life: the perspectives of intensive care and palliative care conscious state: ethical and policy implications. Neurology. clinicians. J Med Ethics. 2012;38:540–5. 2007;68:304–7. 58. Wade DT. Ethical issues in diagnosis and management of patients 41. Avesani R, Gambini MG, Albertini G. The vegetative state: a report in the permanent vegetative state. BMJ. 2001;322:352–4. of two cases with a long-term follow-up. Brain Inj. 2006;20: 59. Grundsätze der Bundesärztekammer zur ärztlichen Sterbebeglei- 333–8. tung. Deutsches Ärzteblatt. 2011 108:A 346–8. 42. Strens LH, Mazibrada G, Duncan JS, Greenwood R. Misdiagnosing 60. Glannon W. Burdens of ANH outweigh benefits in the minimally the vegetative state after severe brain injury: the influenceof med- conscious state. J Med Ethics. 2013. ication. Brain Inj. 2004;18:213–8. 61. Huxtable R. ‘In a twilight world’? Judging the value of life for the 43. Matsuda W, Matsumura A, Komatsu Y, Yanaka K, Nose T. Awaken- minimally conscious patient. J Med Ethics. 2012. ings from persistent vegetative state: report of three cases with par- 62. Demertzi A, Racine E, Bruno MA, Ledoux D, Gosseries O, Van- kinsonism and brain stem lesions on MRI. J Neurol Neurosurg Psy- haudenhuyse A, et al. Pain Perception in Disorders of Conscious- chiatry. 2003;74:1571–3. ness: Neuroscience, Clinical Care, and Ethics in Dialogue. Neuro- 44. Kriel RL, Krach LE, Jones-Saete C. Outcome of children with pro- ethics. 2013;6:37–50. longed unconsciousness and vegetative states. Pediatr Neurol. 63. Stepan C, Haidinger G, Binder H. Prevalence of persistent vegeta- 1993;9:362–8. tive state/apallic syndrome in Vienna. Eur J Neurol. 2004;11: 45. Estraneo A, Moretta P, Loreto V, Lanzillo B, Santoro L, Trojano L. 461–6. Late recovery after traumatic, anoxic, or hemorrhagic long-lasting 64. Marshall LF, Becker DP, Bowers SA, Cayard C, Eisenberg H, Gross vegetative state. Neurology. 2010;75:239–45. CR, et al. The National Traumatic Coma Data Bank. Part 1: Design, 46. Levin HS, Saydjari C, Eisenberg HM, Foulkes M, Marshall LF, Ruff purpose, goals, and results. J Neurosurg. 1983;59:276–84. RM, et al. Vegetative state after closed-head injury. A Traumatic 65. Marshall LF, Toole BM, Bowers SA. The National Traumatic Coma Coma Data Bank Report. Arch Neurol. 1991;48:580–5. Data Bank. Part 2: Patients who talk and deteriorate: implications 47. Dubroja I, Valent S, Miklic P, Kesak D. Outcome of post-traumatic for treatment. J Neurosurg. 1983;59:285–8. unawareness persisting for more than a month. J Neurol Neuro- 66. American College of Surgeons. National Trauma Data Bank® surg Psychiatry. 1995;58:465–6. (NTDB). Available from: https://www.ntdbdatacenter.com/Default. 48. Andrews K. Recovery of patients after four months or more in the aspx, [October 28 2013]. persistent vegetative state. BMJ. 1993;306:1597–600.

Bioethica Forum / 2013 / Volume 6 / No. 4 149