Psychopharmacology and the Classification of Functional Psychoses
By Thomas A. Ban and Bertalan Pethö
Four-Dimensional Classification
Endogenous Psychoses
The place of "delusional psychoses" is not clearly defined within the two major schemes of psychiatric classifications, i.e., ICD-9 and DSM-III, both of which have their focus on disorders traditionally subsumed under the category of endogenous psychoses. The same applies to the DCR (Pethö, Ban and Kelemen et al., 1984).
Within a two-dimensional model of psychiatric classification "endogenous psychoses" are like mirror images of psychogenic-reactive psychoses. Although in some instances there might be a precipitating factor, the severity of the event is usually insufficient to explain the emergence of the psychosis and the subject matter of the psychosis is unrelated to the traumatizing experience. In this context, the content of the psychosis remains incomprehensible. Endogenous psychoses do not have an identifiable purpose. They appear as intruding events which undermine the livelihood of affected patients.
Attempts to classify endogenous psychoses range between two extremes. At one extreme, there is the "individual psychosis," unique, occurring once only in the particular form, the end-product exclusively of its own internal and external components, such as constitution, age, sex, character, milieu, situation and life experience; while at the other extreme there is the "unitary psychosis," completely amorphous and undefined, the general outcome of an interaction between an endogenous predisposition to mental illness and some form of exogenous provoking factor, with idiosyncratic structural elements giving the illness its own peculiar stamp (Birnbaum, 1923, 1974). This concept of "unitary psychosis" is different from Neumann's (1859) who contends that different forms of illness are only different stages of one and the same disease process.
The greatest impetus for the conceptual development of endogenous psychoses was Kahlbaum's (1874) formulation of the notion of "nosological entity" which postulates a close correspondence between etiology, brain pathology, symptom pattern and outcome picture (Jablensky, 1981) (Figure 4). The origin of the notion dates back to Bayle's now classic work on the Diseases of the Brain and Its Membranes (Traite des Maladies du Cerveau et de ses Membranes) published in 1826. He put forward the thesis that general paralysis has one cause, i.e., "chronic arachnitis," which is clearly defined in terms of pathological anatomy; has a specific symptomatology which combines motor and mental signs; and most important, has a specific pattern of development comprising three phases, each marked by different symptoms. These phases are delire monomaniaque characterized by prevailing exaltation; delire maniaque characterized by prevailing delusions and overvalued ideas; and etat de demence, the terminal phase, which is characterized by dementia.
Although virtually forgotten outside of France, Bayle's treatise provided fertile ground for the development of psychiatry as a medical discipline. By linking psychopathology to cerebral pathology (chronic arachnitis) he opened the path for German "somaticism." One of the most prominent exponents of this approach was Griesinger (1845, 1876) who in his famous "treatise" published in 1845 declared that "mental diseases are somatic diseases, that is, diseases of the brain." He also asserted that there are no differences between organic and functional disorders, that psychiatry and neuropathology are not merely two closely related fields, but they are one field in which "one language should be spoken" and in which "the same laws prevail" even if "there are many physicians even whole schools of psychiatry, who demand proof." Griesinger's contentions were elaborated in the work of Westphal (1871a,b), who made a systematic attempt to establish the nature of cerebral lesions in psychoses and whose work, according to Pichot (1983) "formed a bridge between Griesinger, for whom organo-clinical correlations were a particular article of faith, and psychiatrists like Meynert and Wernicke, who made use of their anatomical discoveries to work out and formulate general conceptions in psychiatry."
It has not been recognized sufficiently that Bayle's treatise yielded our present conception of "dementia," i.e., a specific end-state which is the result of chronic organic-neurological changes in the brain regardless of their cause or etiological specificity. It is even less well known that it was Bayle's treatise that focused attention on the importance of the course of illness determined in his view by the brain pathology intrinsically linked to etiology. It is this three-dimensional model, i.e., etiology, cross-sectional psychopathology and course of illness that has served as the basis for Kahlbaum's (1874) concept of "nosological entity." It is this three-dimensional model that allowed for the assumption that seemingly disperate clinical states and syndromes may belong together in a "disease." It also opened the way for Kraepelin (1896) to group together several clinical syndromes into the nosological entities of manic-depressive insanity and dementia praecox. By employing a three-dimensional approach and separating phasic-remitting affective psychoses from processual-progressing schizophrenias, Kraepelin succeeded in laying down the foundation for the dichotomy of endogenous psychoses, which, according to Jablensky (1981) served as the basis for our classification of these disorders to date (Table VII).
The two diagnostic end-points provided by the dichotomy (i.e., schizophrenic and affective psychoses), received support at first from the differential responsiveness within the endogenous psychoses, to antipsychotic-neuroleptics and mood-stabilizer lithium salts. Later on it received support from the demonstration that antipsychotics are therapeutically more effective than other agents in the treatment of schizophrenias, whereas mood-stabilizer lithium salts and tricyclic antidepressants are therapeutically more effective than other agents in the treatment of manic-depressive psychoses and endogenous depressions respectively.
By following the pharmacological profile of chlorpromazine-type of antipsychotics (neuroleptics) and imipramine-type of antidepressants, a large number of psychoactive drugs, with similar action mechanisms have been synthesized and introduced into clinical practice for the treatment of schizophrenic and affective (primarily depressive) disorders. Although there is no agreement as to what extent these drugs fulfill therapeutic expectations, there has been an increasing consensus that they cannot be considered specific for a particular illness and do not have equal therapeutic effectiveness during the different developmental stages of the same illness. Because they do not fulfill these two prerequisites, neither the action mechanism of antipsychotics nor of antidepressants can provide sufficiently reliable and valid clues for the generation of hypotheses relevant to the understanding of schizophrenic and/or affective psychoses. Regardless of their immediate contribution to psychiatric theory, systematic studies designed to reveal the action mechanism of antipsychotics and antidepressants contributed greatly to the development of the technology necessary for in vivo exploration of brain mechanisms and brain functions. It remains to be seen, however, whether the newly developed technology will provide the necessary basis of meaningful research in the schizophrenias and affective psychoses within a four-dimensional model of psychiatric diagnosis.
Similarly, regardless of its relevance to psychopharmacology, Kraepelinian nosology proved to be a long step forward by introducing order in a chaotic field. In spite of its limitations it has provided the necessary diagnostic orientation for the initial development of drugs with therapeutic effects in the functional psychoses. It has not succeeded, however, in providing the necessary end-points for the identification of individual patients who are responsive to one or another drug. By definition a three-dimensional model of diagnosis does not take into consideration the last developmental stage of psychiatric illness. Wernicke (1894, 1900), recognizing the limitations of Kraepelinian nosology, pointed out the low prognostic validity of the Kraepelinian diagnoses. He also developed a finely differentiated symptomatology of the endogenous psychoses. This symptomatology provided a description of the psychiatric syndromes that was considerably more adequate than Kraepelin's (Leonhard, 1961). Kleist (1921), a pupil of Wernicke was able to separate many of Wernicke's syndromes as distinct forms of psychiatric illness. He also revealed, that some of these highly differentiated forms of illness -- subsumed under the heading degenerative psychoses -- could only partially be allotted to one or the other of Kraepelin's two major diagnostic groups (Table VIII). Included among them are a number of "affective," "conative" and "intellectual" illnesses with a "shift-like" or a "phasic" course following a simple-monopolar or a multiform-bipolar pattern (Silveira, 1961).
Recognition of the difficulties in separating schizophrenic (dementia praecox) and affective (manic depressive) psychoses on the basis of cross-sectional assessment of psychopathological symptoms created a diagnostic crack with a considerable percentage of patients not fitting either of the two diagnoses. Similarly, recognition of the low prognostic validity of the two diagnoses created a diagnostic gap in which patients suffering from a number of different illnesses are given the same diagnosis.
There are some indications that at least some of these problems might be overcome by adopting the principles of the Wernicke school -- a four- dimensional classification with emphasis on the end-state (subtypes) -- as developed through the work of Kleist by Leonhard and his collaborators. Leonhard's (1979) classification of endogenous psychoses recognizes within affective or phasic (unipolar and bipolar) and schizophrenic (systematic and unsystematic) psychoses a number of different illnesses and identifies a third group of psychoses, referred to as cycloid psychoses, resembling the group of phasic psychoses in their course and the group of unsystematic schizophrenias in their content (Figure 5).
At first sight Leonhard's classification appears to be bewilderingly complex. If its basic tenets are understood, however, the system becomes rational and simple. The classification is based on a four-dimensional model of psychiatric diagnosis with emphasis on the final stage of psychiatric illness which in the ultimate analysis ranges from full recovery (phasic psychoses) to moderate or marked defect states (systematic schizophrenias) with personality transformation (cycloid psychoses) and mild to moderate defect (nonsystematic schizophrenias) in between.
Regarding the course of illness, Leonhard contends that endogenous psychoses may follow an episodic (phasic psychoses, cycloid psychoses and unsystematic schizophrenias) or continuous (systematic schizophrenias) course; and phasic psychoses may appear as unipolar (mania or depression) or bipolar (manic-depressive psychosis) illnesses.
Insofar as cross-sectional psychopathology is concerned, Leonhard follows the principles of Wernicke's (1881, 1894, 1900, 1906) structure-analysis. Wernicke considered the reflex arc as his functional working unit and saw the cerebral cortex as the organ of "associations." In his "reflex pathology" he dealt with the three functional aspects of the reflex arc, i.e., sensory input, interneuronal associations, and motor output. He maintained that any of the three could be disturbed separately and also in various combinations. Within this model psychopathological syndromes are perceived as a decrease, increase, or a dysfunction in the activity of these structures which become manifest through the effect of the disturbance of "transcortical associations." A further elaboration of this model was put forward by Nyiro (1958, 1962) for whom the three aspects of the reflex arc represented three "pyramidal structures," i.e., perceptual-cognitive (sensory input), relational-affective (intrapsychic) and motor-adaptive (motor output) (Figure 6). Within this frame of reference Leonhard distinguished among psychopathological syndromes which affect primarily one of three structures, e.g., systematic schizophrenias, and psychopathological syndromes which affect primarily more than one structure, e.g., nonsystematic schizophrenias.
He also distinguished between disease groups characterized by dissociation among the three "structures" (nonsystematic and systematic schizophrenias) and disease groups in which the functioning of the three "structures" is congruent or harmonious (phasic and cycloid psychoses). In addition to these general features he brought to attention that both groups of schizophrenias and some of the cycloid psychoses are characterized by prevailing changes in perceptual-cognitive structures (systematic paraphrenias, cataphasia and confusion psychosis). Other disorders and groups of disorders are characterized by prevailing changes in relational affective structures (systematic hebephrenias, affect-laden paraphrenia and anxiety-elation psychosis) and others again by prevailing changes in motor-adaptive structures (systematic catatonias, periodic catatonia and motility psychosis). In phasic psychoses as a rule the primary disturbance is in the relational-affective structure (mood). In some of the subtypes (manic-melancholic psychosis, pure melancholia and pure mania), however, there are corresponding disturbances in the other two structures; and in some of the other subtypes (pure euphorias and pure depressions), there are disturbances in one additional structure, i.e., either perceptual-cognitive or motor-adaptive (Table IX).
Although Leonhard's system of classification is based on Wernicke's model of structure-analysis in psychopathology he also paid considerable attention to Kleist's (1929, 1939) contributions. His diagnoses provide for detailed analyses of psychopathological syndromes with emphasis on the identification of the prevailing structure in the holistic picture. They open a natural path for studies with advanced brain imaging techniques, e.g., PET, SPECT, and MIR.
The same applies to the DCR (Pethö, Ban, Kelemen et al., 1984) which differs from Leonhard's classification in that it separates a group of systematic and non- systematic diseases within both affective and schizophrenic psychoses. It also distinguishes between the characteristics of experience and behavior and emphasizes the importance of the characteristic changes in form (Gestalt) and overall picture. Further, the DCR shifts the point of departure from the end-state to the first or index psychosis, and takes into consideration the course of illness and the characteristics of the outcome not only in terms of psychopathological symptoms, but also in terms of personality characteristics and social adjustment (Figure 7). Within the frame of reference of this classification an important prerequisite for the diagnosis of schizophrenic psychoses is the dissociation among the perceptual-cognitive, affective-relational and motor-adaptive structures that appears to be a "split" to an outside observer. It is difficult to understand the subject matter because it evolves in a catathymic manner or through other mechanisms. Contrary to common belief, hallucinations and/or delusions are not obligatory symptoms of schizophrenic psychoses. Considerably more specific are manifestations of formal thought disorder that disturbs comprehensibility (primary incoherence, tangential thinking, blocking, derailment, desultory thinking, onomatopoesis), affective changes (blunted, inadequate, inappropriate) and changes in personality (abandonment of habits, change in lifestyle, incomprehensibility of behavior, autistic behavior).
Unlike the schizophrenic psychoses, which are characterized by an irreversible process, cycloid psychoses are characterized by complete recovery from each phase, although personality changes may occur. Patients with cycloid psychoses are diagnosed on the basis of a number of distinctive characteristics in the DCR. Among them probably the most important is that the whole field of patient's experience is transformed (protopathic change of Gestalt), with a change in the state of mind and depth of emotions. However, experience, behavior and performance remain in harmony (congruent). Onset is acute or subacute while the psychosis as a rule is multifold (polymorphous) and fluctuating with contradictory influences of healthy and pathological tendencies creating a feeling of uncertainty and/or confusion. Nevertheless, there is a strong emotional involvement in symptoms. Other manifestations include ideas of reference, delusional perceptions, holothymic delusions, misidentifications, thematic incoherence and changes (increase or decrease) of reactive and expressive movements. Similar to the cycloid psychoses which usually display complete recovery from each phase, affective-phasic psychoses are characterized by a remitting course with periodicity and/or rhythmicity. An essential prerequisite, although not necessarily exclusive criterion for this diagnosis is that experience, behavior and performance are in keeping with mood. The same applies to delusions. Nevertheless, the disease picture may be simple (monomorphous) or multiform (polymorphous).
Application of sophisticated technology is warranted only in biologically homogenous populations which meet optimal contemporary standards. There is sufficient evidence to believe that the diagnoses based on Leonhard's system fulfill these requirements. The "here do familial" distinctness of unipolar and bipolar illnesses within the phasic psychoses has been supported by Angst and Perris (1968, 1972); and the "genotypical" distinctness of cycloid psychoses and systematic schizophrenias by Ungvari (1984). This genetic distinctness was also substantiated by multiple threshold analysis which rejected the possibility of identical liability and confirmed the separateness of these two diagnostic categories. In the same study the nonsystematic schizophrenic category displayed a considerable genotypical overlap with the cycloid and the systematic schizophrenic categories ("as if it were a connecting link"). In favor of Leonhard's schizophrenic sub- types are the findings of a high correlation between the distribution of subtypes in the original sample of Leonhard from the late 1930s and early 1940s and in Astrup's sample 20 years later (Wilson and Ban, 1983). In favor also are the statistically significant correlations in rank order of frequency of occurrence of the six paraphrenic, four hebephrenic and six catatonic subtypes in eight countries, in a study carried out in the late 1970s and early 1980s (Ban, Guy and Wilson, 1984a). The finding of differential therapeutic responsiveness between the systematic and the non- systematic schizophrenic populations and within both populations among the different types of patients by Fish (1964c) is also in support of Leonhard's system.
Leonhard's classification is important from a psychopharmacological point of view because it might provide the possibility of identifying patients therapeutically responsive to antipsychotics, antidepressants and mood stabilizer-lithium salts. It might also provide new end-points for research development. While a three-dimensional psychiatric classification is suitable to establish the therapeutic efficacy of drugs with an accepted level of statistical probability within a particular population, a four-dimensional model of psychiatric classification might be suitable to identify treatment responsive patients, i.e., diagnostic subtypes with a reasonable clinical accuracy. Similarly, while within a three-dimensional model of classification psychotropic drug development is restricted to the modification of speed on onset, efficacy and toxicity of drugs which share common pharmacological properties with conventional psychotropics, within a four-dimensional model of classification there are new end-points that open new paths for the development of new drugs with different pharmacodynamic properties. Thus, a four-dimensional model of classification could facilitate the development of a new class of drugs, e.g., "transition compounds" (les produit de transition). Drugs which belong to this category include carbamazepine, an anticonvulsant which is structurally related to tricyclic antidepressants and which may have a place in the treatment of bipolar affective disorders and/or cycloid psychoses; and carpipramine and chlorcarpipramine, i.e., dibenzazepines (tricyclic antidepressant structure) with a butyrophenone (antipsychotic) side chain, which may have a place in the treatment of cycloid psychoses and/or affect-laden paraphrenia, one of the three diagnostic types of the nonsystematic schizophrenias. The same applies to the dibenzoxazepines, such as loxapine, an antipsychotic which demethylates in part to amoxapine, an antidepressant, which in turn hydroxylates in part into a dopamine receptor blocking antipsychotic drug. The dibenzoxazepines today are profiled as a traditional antipsychotic (loxapine) and a traditional antidepressant (amoxapine) because of their common pharmacological properties with prototype antipsychotics and antidepressants. By this, special emphasis is placed on some of their possible adverse effects without full appreciation of their unique therapeutic potential in certain diagnoses that require the combined administration of an antipsychotic with an antidepressant for optimal treatment.