. 33
( 68 .)


and affective disorders in the category ˜˜psychotic disorders not elsewhere classified™™
and with their own diagnostic criteria, as well as with subtypes, namely bipolar
type and depressive type. In DSM-IV, published in 1994, schizoaffective disorders
belonged to the category ˜˜other psychotic disorders™™ with almost the same diag-
nostic criteria and the same subtypes as in DSM-III-R. This time, the mixed
bipolar symptomatology was recognized as well.
ICD-9 continued the tradition of ICD-8. In ICD-10, after bouncing around like
a ping-pong ball during successive draft publications, schizoaffective disorders
landed in a category of their own within schizophrenia and delusional disorders,
with an extensive description, as well as with five subcategories:
192 A. Marneros et al.

(1) schizoaffective disorders, at the present manic
(2) schizoaffective disorders, at the present depressive
(3) mixed schizoaffective disorder
(4) other schizoaffective disorders
(5) schizoaffective disorders not otherwise specified
The evolution of the concept and definitions of schizoaffective disorders con-
tinues. Many aspects remain to be clarified, and many questions still require
answers. Most diagnostic systems only recognized the concurrent form of schizo-
affective disorders, not the sequential one. But operational research showed no
differences in any investigated dimension between concurrent and sequential
schizoaffective disorders (Marneros et al., 1986; 1988a“c; 1991).
As we previously pointed out (Marneros, 1999), the ongoing evolution of concepts
and definitions of schizoaffective disorder lead to enduring uncertainty. The question:
What are the schizoaffective disorders? remains unanswered, which is also a reason why
onlymeager research results existin the complicated field ofschizoaffective mixed states.

Current research in schizoaffective mixed states

Schizoaffective mixed states in the Cologne study
As mentioned above, for the most part, systematic research on schizoaffective
mixed states has not been published. An exception is research performed as a part
of the Cologne study (1986“1991). The Cologne study, which is a longitudinal
study comparing schizophrenic, affective, and schizoaffective disorders, has been
published in several presentations and in a monograph, which also includes an
extensive English summary (Marneros et al., 1991). A total of 402 patients were
followed-up for an average of 25 years after the onset of their illness. The diagnoses,
obtained longitudinally, were as follows: schizophrenic disorders (n ¼ 148), schizo-
affective disorders (n ¼ 101), affective (mood) disorders (n ¼ 106).
A distinction was made between ˜˜episode™™ (which is only a cross-sectional
diagnosis) and ˜˜illness™™ or ˜˜disorder™™ (which are longitudinal diagnoses). The
˜˜episodes™™ (cross-sectionally defined) were classified according to slightly modi-
fied DSM criteria into schizophrenic, affective (depressive, manic, manic-depressive
mixed), schizoaffective (schizodepressive, schizomanic, schizomanic-depressive
mixed) and non-characteristic episodes.
The diagnoses of an ˜˜illness™™ or ˜˜disorder™™ were made only longitudinally and
took into account all types of episodes occurring during the whole course. The
final diagnoses were longitudinally defined as follows:
* schizophrenic disorders: only schizophrenic episodes during the whole course

* affective (mood) disorders: only depressive, manic, or manic-depressive mixed

episodes during the whole course
193 Schizoaffective mixed states


(27 years)

Bipolar With mixed episodes
104 45 (43%)

Fig. 8.1 Frequency of mixed episodes in bipolar patients in the Cologne study.

schizoaffective disorders:

a. at least one schizoaffective episode within the course (schizodepressive,
schizomanic, schizomanic-depressive mixed episode)
b. all sequential manifestations of schizophrenic and affective episodes, inde-
pendent of the type and number of other episodes
This definition has been proved empirically. No differences concerning investigated
levels were found (Marneros et al., 1991).
Mood, schizoaffective, and schizophrenic disorders were compared longitudin-
ally (average follow-up 25.1 years from onset). Both mood and schizoaffective
disorders were divided into bipolar and unipolar types, and they were investigated
both as a voluminous group (bipolar affective and schizoaffective disorders as a
single group), and individually (bipolar affective versus bipolar schizoaffective:
Marneros et al., 1989a“c; 1990a, b). The main results regarding mixed episodes in
bipolar patients in the Cologne study can be summarized as follows.
At the end of the observation time (for patients with mood disorders, 27 years),
145 patients were classified as unipolar and 104 as bipolar. Of the bipolar patients,
43% showed at least one mixed episode (either affective or schizoaffective mixed
episode; Fig. 8.1).
The 104 bipolar patients showed 685 episodes during the investigation period.
One hundred and seventeen episodes (17%) were mixed bipolar episodes, affective
and schizoaffective (Fig. 8.2).
Patients with mixed episodes were found to have the following characteristics:
* more frequently females

* more frequently a family history of affective disorders

* more and longer episodes

* more often a poorer long-term outcome

The syndromatological instability during course is interesting. Only 1% of the
patients showed a monomorphous affective-mixed course, which means that
during the whole course, only affective mixed episodes occurred. Four percent
194 A. Marneros et al.

Unipolar course Episodes
141 593

Affective and
(27 years)

104 685 117 (17%)
Bipolar course Episodes Mixed bipolar

Fig. 8.2 Frequency of mixed episodes in bipolar patients in the Cologne study.

of the patients showed a course dominated by affective mixed episodes, which
means that the majority of the episodes fulfill the criteria of affective mixed states.
Similarly, only 1% of the bipolar patients had a monomorphous schizoaffective
mixed course, which means that only schizoaffective mixed episodes occurred
during the whole course. Four percent of the bipolar patients had a course
dominated by schizoaffective mixed states, which means that the majority of
episodes fulfill the criteria of schizoaffective mixed episodes.

The Halle Bipolarity Longitudinal Study
The Halle Bipolarity Longitudinal Study (HABILOS) (Marneros et al., 2004) exam-
ines, amongst others, schizoaffective mixed states. The study population involved
inpatients of the Department of Psychiatry and Psychotherapy of the Martin-Luther
University Halle-Wittenberg, Germany. It is part of a voluminous Affectivity
Project and involves all inpatients with manic, schizomanic, affective, and schizo-
affective mixed episodes treated between 1993 and 2000. The diagnostic criteria of
the above-mentioned episodes were: ICD-10, DSM-IV criteria and, additionally for
mixed states, Cincinnati (McElroy et al., 1992) and Pisa criteria (Perugi et al., 1997).
The characteristics of the study population are given in Table 8.3.
All patients were first assessed as inpatients and later re-examined as outpatients
(mean follow-up time: 5.1 years). Several instruments, which assessed psycho-
pathology, illness history, personal history, disability, functioning, quality of life,
and other constructs, were applied (Table 8.4).


Characteristics of episodes
Episodes were defined according to ICD-10, DSM-IV, and Cincinnati criteria.
Nevertheless, findings presented in this chapter regarding episodes are limited to
195 Schizoaffective mixed states

Table 8.3 Characteristics of the study population

Total number of patients 277
Female 135 (48.7%)
Age at first illness episode (mean, range) 32.3 years (13.0“66.2)
Age at the end of the follow-up period (mean, range) 47.9 years (20.6“90.1)
Duration of the illness (mean, range) 15.7 years (1.2“58.3)
Prospective period (mean, range) 5.1 years (1.0“9.7)
Total number of episodes 2119

Table 8.4 Instruments of the Halle Bipolarity Longitudinal Study (HABILOS)

Scope Instrument

Axis I diagnosis DSM-IV SCID-I (Wittchen et al., 1997)
Axis II diagnosis DSM-IV SCID-II (Fydrich et al., 1997)
History of illness Rating of episodes (Marneros et al., 1991)
Medication Documentation of medication
(Marneros et al., 1991)
Depressive symptoms CDRS (Mason et al., 1993)
Manic symptoms BDI (Beck et al., 1961)
YMRS (Young et al., 1978)
Psychotic symptoms MSS (Kruger et al., 1997)
Personality PANSS (Kay et al., 1987)
Temperament NEO-FFI (Costa and McCrae, 1989)
Social biography TEMPS-A Questionnaire (Akiskal et al., 2002)
Premorbid functioning SOBI (Marneros et al., 1991)
Social disability PAS (Cannon-Spoor et al., 1982)
Social functioning DAS-M (Jung et al., 1989)
Global functioning SOFAS (American Psychiatric Association, 1994)
Quality of life GAS (Spitzer et al., 1978)
WHOQOL-Bref (WHOQOL Group, 1998)

DSM-IV, Diagnostic and Statistical Manual of Mental Disorders; SCID-I, Structured Clinical
Interview for DSM-IV (Axis I); CDRS, Cornell Dysthymia Rating Scale; BDI, Beck Depression
Inventory; YMRS, Young Mania Rating Scale; MSS, Manie-Selbstbeurteilungsskala (English,
SRMI “ Self-Report-Mania Inventory); PANSS, Positive and Negative Syndrome Scale; NEO-
FFI, Neo-Five Factor Inventory; TEMPS-A, Temperament Scale “ Autoversion; SOBI,
Sociobiographical Interview; PAS, Premorbid Adjustment Scale; DAS-M, Disability Assessment
Schedule; SOFAS, Social and Occupational Functioning Assessment Scale; GAS, Global
Assessment Scale; WHOQOL, WHO Quality of Life.
196 A. Marneros et al.

Manic 486 (22.9%)

62 (2.9%)
Mixed affective

Schizomanic 312 (14.7%)

Schizodepressive 133 (6.3%)

Mixed schizoaffective 56 (2.6%)

Schizophrenic 251 (11.8%)

Others 206 (9.7%)

0 100 200 300 400 500 600 700
Fig. 8.3 Types of episode (n ¼ 2119).

the ICD-10 definition. Out of the 2119 episodes evaluated, the most frequent type
of episode was the depressive one (n ¼ 613; 29% of all episodes). Mixed episodes
were found to be the rarest type of episodes: there were only 118 affective and
schizoaffective mixed episodes (ICD-10) (6%) and the frequency of mixed affec-
tive episodes (n ¼ 62; 3%) was almost identical to that of mixed schizoaffective
episodes (n ¼ 56; 3%; Fig. 8.3).
These findings of HABILOS confirm the results from other studies that the most
frequent type of episodes in bipolar disorders is the depressive type (Marneros and
Brieger, 2002; Marneros et al., 1990a, b; 1991; see chapter 2).
Interestingly, although the inclusion criterion for the study population was the
presence of ˜˜bipolar episodes™™ (i.e., manic, schizomanic and mixed), 251 episodes
(12%) fulfilled the ICD-10 criteria of a schizophrenic episode, in addition to the
depressive and schizodepressive episodes. This is compatible with findings show-
ing a syndromal instability in the course of bipolar disorder (Marneros et al.,
1990b; 1995; Marneros and Pillmann, 2004). This finding repeatedly demonstrates
that ICD and DSM criteria of disorders not involving syndrome changes during
longitudinal course are not sufficient to define ˜˜diseases.™™ According to the long-
itudinal criteria proposed by Marneros et al. (1986, 1991), mood disorders with
˜˜syndrome shift™™ into schizophrenic syndromes and vice versa fulfill the criteria of
the ˜˜sequential type™™ of schizoaffective disorders.

Duration of episodes
Based on methodological considerations, we decided to evaluate, as a first step, the
length of episodes according to length of inpatient treatment. Obviously, this is
not the real duration of an episode, but it certainly serves as an indicator for the
real duration of an episode. Mixed episodes are, according to classical but also new
197 Schizoaffective mixed states




. 33
( 68 .)