The Forthcoming DSM-5, Critical Care Medicine, and Pediatric Neuropsychiatry: Which New Concepts Do We Need?

Consultation–liaison neuropsychiatry revolves around the recognition and treatment of neuropsychiatric disorders due to (critical) illness. It represents neuropsychiatric practice, which brings together psychiatry, as embedded in the most recent version of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR), and neurology.¹ This mix of somatic and mental health perspectives creates an important and clinically challenging area. The cumulative incidence of, for example, child psychiatric disorders in a somatically healthy population is 7%, increasing to 15% in minor pediatric disorders; 30% in severe pediatric conditions, including epilepsy; and up to 45% in children with structural brain pathology.² Consequently, adequate (pediatric) consultation–liaison psychiatric care is required in various clinical settings.³ The DSM facilitates 1) diagnosis; 2) communication; and 3) etiological and clinical research in mental disorders. It is used in medical and psychological disciplines, and across varying settings, for example, primary care, in- and outpatients, and hospital consultation and liaison. Its structural hierarchy is reflected in the presentational order: 1) disorders secondary to organic pathology, for example, dementia, delirium, intoxications; 2) psychotic disorders, including bipolar disorder; 3) mood and anxiety disorders; and 4) personality disorders. Diagnoses rely on five axes, reflecting the interplay between body (Axis III), mind (Axes I and II), environmental exposures (Axis IV), and their functional impact (Axis V).¹ DSM includes some helpful algorithms (“Decision Trees for Differential Diagnosis”) in the case of, for example, Mental Disorders Due to a General Medical Condition.

Subsequent editions of the DSM have progressively proven their value in clinical practice and scientific research, but have also had their problems. In psychiatry, attitudes to diagnosis remain mixed and polarized, and its value is continuously questioned. With revisions to the international diagnostic systems for psychiatry on the horizon (DSM-5 and ICD-11), this deep ambivalence, derived from Cartesian tensions between “mindless” and “brainless” perspectives,⁴ has resurfaced, fueled by disappointing field trials of DSM-5 diagnostics.⁵ The DSM editions since 1952 tried and try to create order by presenting “provisional” uniform classification systems. And, given the lack of hard objective quantitative measures, this had, and continues to be, work in progress for the foreseeable future, until the moment a better understanding of the “stuff” of psychiatric disorder emerges.⁶

The primary focus of this article is whether the shortcomings and flaws of the DSM-IV-TR regarding the neuropsychiatric aspects of critical illness have been met in its next edition, due to come out in 2013: the DSM-5. After careful examination and comparison of both the current and the proposed edition in this respect and after interdisciplinary discussions, we conclude that five important issues at the interface of (pediatric) somatic and fmental health care still remain to be addressed.

Sickness Behavior

The brain frequently becomes a target organ of pathology in the context of critical illness, revealing secondary neuropsychiatric disorders. Disturbances in brain functioning may lead to autonomic nervous system failure, which is strongly associated with increased morbidity and mortality. Five common secondary neuropsychiatric disorders, or “Organic Reaction Types of the Brain,” are 1) sickness behavior, 2) fever, 3) epilepsy, 4) delirium, and 5) catatonia. Sickness behavior is defined as a dramatically changed behavioral repertoire, which occurs almost universally in humans and mammals after infectious diseases, cancer, and chemotherapy.^7,8 Its characteristics are attentional disturbances, anorexia, fatigue, emotional lability, loss of interest, and malaise. Sickness behavior explains the high prevalence of prodromal delirium in critical illness.⁹ Although a clearly-defined syndrome of clinical and scientific importance, its existence as a diagnostic category is neglected in both the DSM-IV-TR and the proposed DSM-5.¹⁰ One solution is to create a new subsection “Organic Reaction Types of the Brain,” with descriptive text and a set of diagnostic criteria in the chapter, titled “Mental Disorders Due to a General Medical Condition.”

Delirium in Children and “Differential Diagnosis of Mental Disorders Due to a General Medical Condition in Children”

An important issue regards the diagnosis of pediatric delirium in critically ill children <16 months, since, at the pediatric intensive care unit, 80% of children are <36 months, and 50% are <12 months. The DSM-IV-TR section on children does not mention pediatric delirium. Whereas diagnosis of delirium in adults relies heavily on neurocognitive symptoms, diagnosis of delirium in children relies more on behavioral symptoms, given the difficulty of assessing neurocognition in infants. The five Pediatric Anaesthesiology Emergence Delirium criteria: eye contact, awareness of surroundings and caregiver, purposeful actions, inconsolability, and restlessness may represent appropriate diagnostic criteria in DSM-5.¹¹ Another issue is that the DSM-IV-TR flow chart “Differential Diagnosis of Mental Disorders Due to a General Medical Condition” has two omissions.⁹ First, there is no consideration of the fact that nearly all pediatric intensive-care unit patients already have disturbances of consciousness and cognition—due to sickness behavior and analgo-sedation, and, thus, already have pre-delirium. Furthermore, for a flow chart to be of use in a (pediatric) intensive-care unit context, its starting-point should be the everyday occurrence of 1) “a sudden emotional and behavioral disturbance in a critically ill patient” and/or 2) “refractory agitation,” rather than the current entry cue of “symptoms that are due to the direct physiological effects of a general medical condition.” The flow chart should discuss in logical order of clinical importance the differential diagnoses of the underlying causes: major somatic complications, discomfort causes, psychosocial deprivation, delirium, and medication-related causes: withdrawal/tolerance/paradoxical excitation. Also, the sequential order of the current flow chart—although logical in itself—seems inadequate, as, in critical illness, several causes frequently operate simultaneously. Our proposed change for DSM-5 is to devote a section to “Emotional and Behavioral Issues in the Critically Ill (Child),” combined with a novel flow chart and pediatric delirium criteria.

Catatonia

Whereas much is known about organic reaction types of the brain in adults and elderly persons, considerably less expertise exists for children, especially regarding catatonia.¹² By conceptualizing catatonic symptoms as part of another disorder, the catatonic state remains underrecognized, particularly in the context of diagnosis and treatment of an underlying critical illness. It has become clear that catatonia deserves recognition as an important and treatable syndrome by itself, in adults and children alike.¹² The DSM-5 Development Proposed Revision section announces a new chapter titled “Catatonia,” which will present a descriptive text and diagnostic criteria. We agree, but are concerned that, up until now, no differential diagnostic criteria have been formulated that aid in making the distinctions among 1) hyperactive delirium; 2) autism; and 3) anti-NMDA–receptor encephalitis (see below). Additional paragraphs in the related chapters presenting these differential diagnostic criteria may be required.

Regressive Disorders in Childhood

DSM-IV-TR recognizes two pediatric pervasive developmental disorders, characterized by a normal development, followed by regression, often leading to severe cognitive decline: Rett’s Disorder and Childhood Disintegrative Disorder. In practice, a large number of children are diagnosed with the category: Pervasive Developmental Disorder, Not Otherwise Specified, including any form of atypical autism. The forthcoming DSM-5 subsumes all these disorders, under “Autism Spectrum Disorders.” Also, DSM-5 encourages clinicians to use a “specifier” in cases where known medical disorders play a role. Since 2007, a new Axis III diagnosis exists: anti-NMDA-receptor encephalitis. In children, its main Axis I differential diagnoses include Childhood Disintegrative Disorder, Early-Onset Schizophrenia, Late-Onset Autism, and Catatonia.¹³ Given the severe neuropsychiatric presentation and integrated psychiatric and somatic approach toward etiology, treatment, and outcome, this new (critical and potentially lethal) illness must not be missed. Therefore, regressive disorders in the context of possible somatic disorders must be given due attention, for example, by creating a section titled “Regressive Disorders in Infancy and Childhood,” and by adding a separate subsection titled “anti-NMDA-receptor encephalitis across all ages,” including diagnostic criteria in the chapter “Mental Disorders Due to a General Medical Condition.”

Age-Related Diagnostic Criteria With Regard to Neuropsychiatric Disorders

DSM-5 proposes that the current DSM-IV-TR chapter devoted specifically to children is omitted. Instead, there will be a chapter titled “Neurodevelopmental Disorders” including most of the childhood diagnostic conditions. In the other DSM-5 chapters, paragraphs presenting age-related symptoms and age-related subtypes will be included where applicable. We applaud these intentions of The Lifespan Developmental Approaches Study Group. However, given 1) the lack of an elementary chapter titled “Children,” and 2) the confusing name of the chapter “Neurodevelopmental Disorders,” we are not reassured. This approach may give rise to a diagnostic maze, especially for non-mental health colleagues working in child health.

Conclusions

By addressing the five issues described above, DSM-5 will incorporate new scientific and clinical insights, paying attention, in some instances for the first time, to common neuropsychiatric aspects and problems. Most of these issues have not been recognized, or not properly appreciated, until recently. By identifying, structuring, and discussing these in the DSM language, “conditio sine qua non” is met for the first time; clinical care for all the critically ill—as well as future scientific research—will be better equipped to blossom.