Psychosis and Schizophrenia

Overview

Psychosis, in a broad sense, signifies impaired reality-testing ability. The most important symptoms of psychosis are delusions and hallucinations, although other signs and symptoms (such as disorganized speech or behavior and catatonia) are also considered psychotic phenomena.

Psychosis is not a specific diagnosis, because it can occur in a wide variety of clinical contexts. Of the psychotic disorders, the prototypical condition is schizophrenia. In schizophrenia, psychotic symptoms occur chronically, without gross organic abnormalities of the brain (hence the term functional psychosis ) or a severe medical disturbance (as in a delirium). Moreover, psychosis may occur when the mood is normal, thus differentiating it from bipolar disorder or from psychotic depression. While psychosis is a defining (and often the most apparent) clinical feature of schizophrenia, other symptom clusters (e.g., negative symptoms and cognitive symptoms) are largely responsible for the psychosocial disability that usually accompanies this disorder.

Schizophrenia was originally described at the close of the nineteenth century by Emil Kraepelin (who categorized madness into episodic mood disorders and chronic psychotic illnesses; he termed the latter dementia praecox , which was later renamed schizophrenia by Eugen Bleuler). In the 100 years since the condition was identified, much progress has been made (in large part the result of the discovery of the first antipsychotic, chlorpromazine), turning schizophrenia from an illness treated in asylums (state hospitals) to one treated in community settings ( Figure 28-1 ). The discovery that genetic risk factors are shared between schizophrenia and manic-depressive illness, i.e., bipolar disorder (as well as other psychiatric syndrome), might eventually lead to a new nosology based on genetics and pathophysiology. In the interim, the fundamental Kraepelinian dichotomy between schizophrenia (and related psychotic disorders) and bipolar disorder remains a cornerstone in psychiatric diagnosis.

Epidemiology and Risk Factors

Schizophrenia is a syndrome that occurs in all cultures and in all parts of the world. Epidemiological studies have found incidence rates between 7.7 and 43.0 (median 15.2) new cases per 100,000. The point prevalence is approximately 5 per 1,000 and the life-time morbidity risk is approximately 1%. The mortality rate is 2.5 times that of the general population, and this mortality gap continues to grow. Gender differences also exist; men have a 30%–40% higher life-time risk for schizophrenia than women, and the age of onset is roughly 3–4 years later for females.

The old dogma that cited the identical incidence and prevalence rates of schizophrenia around the globe is not quite correct. Clear geographical differences exist, albeit within a fairly narrow range of twofold to threefold differences. The fact that these modest variations in rates exist between cultures, and within subgroups in cultures, likely reflects different risk factors for schizophrenia in different populations (e.g., more infections in one setting, more drug use in another).

The best-established risk factors are not genetic, but environmental ( Box 28-1 ). Immigrant populations have an increased risk, with the highest risks for the children of first-generation immigrants, followed by the immigrants themselves. Urban living increases risk relative to rural living. Etiological insults during brain development include intra­uterine infections, particularly influenza after exposure during the first half of pregnancy, and nutritional deficiencies, including low maternal folate and elevated homocysteine levels. Maternal starvation has strong support from two epidemiological studies of famines (the Dutch Hunger winter of 1944–1945 and the 1959–1961 Chinese famine ). Both studies found a doubling of risk for schizophrenia in offspring of mothers who starved during the first trimester. Other early environmental risk factors include obstetrical complications. Later environmental risk factors are head injury and substance use (particularly cannabis). Frequent pre-morbid cannabis use (i.e., more than 50 uses) increases the risk for the development of schizophrenia by sixfold.

The majority of patients with schizophrenia lack a family history of the disorder. Nevertheless, the fact that genes matter was shown in now-classic twin and adoption studies of the 1960s. For monozygotic twins, the risk of developing schizophrenia approaches 50% for the unaffected twin if the co-twin has schizophrenia. Having siblings or parents (i.e., first-degree relatives) with schizophrenia increases an individual's risk to approximately 10-fold above that seen in the general population. With greater genetic distance, the risk for schizophrenia decreases to a twofold risk over the population risk for second-degree relatives.

Paternal age increases the risk for schizophrenia in a linear fashion. Compared with the children of fathers who are less than 25 years old the relative risk for children increases steadily with paternal age, to about 2.0 for fathers in the 45 to 49 age group, and to almost 3.0 for fathers older than age 50. This increase in risk with paternal age is consistent with the hypothesis that de novo mutations contribute to the genetic risk in schizophrenia. New mutations could explain why a disorder associated with lower fertility rates has not disappeared; in men as opposed to women, the germ-line cells, spermatogonia, continue to divide throughout life, allowing replication errors to accumulate and to be transmitted to offspring.

It is unlikely that a single gene with a significantly large effect will be found that can cause schizophrenia. Most likely, schizophrenia is similar to other non-Mendelian complex disorders where many different common variants of genes each make a small, yet important, contribution to disease vulnerability, or occasionally rarer variants of genes make larger contributions. Such vulnerability gene variants convey susceptibility without directly causing the disease: the disease is only expressed when combined with other genes or certain environmental factors (e.g., infection or drug use). The number of susceptibility genes is unknown; however, many genes can confer risk in a population. Since 2007 when the first genome-wide association studies (GWAS) in schizophrenia was published, GWAS of increasingly larger sample sizes have added susceptibility genes that confer risk for schizophrenia (for a partial list, see Table 28-1 ). The most promising candidate genes are involved in brain development, frontal lobe function, myelination, synaptic function, and glutamate transmission. Moreover, it appears likely that most gene variants are not specific for schizophrenia; instead, they confer risk for neuropsychiatric disorders that cut across current clinical diagnoses. Several genetic disorders, such as 22q11 deletion syndrome (velo-cardio-facial syndrome [VCFS] or DiGeorge syndrome) or Klinefelter syndrome (XXY syndrome), increase the risk for psychosis in affected individuals. In addition, rare structural variants, including microdeletions and microduplications (copy number variants, or CNVs), may also contribute to the illness pathophysiology by disrupting signaling pathways important in neurodevelopment. An interaction between susceptibility genes and environment forms the basis for the neurodevelopmental hypothesis of schizophrenia in which a clinically silent, latent propensity toward schizophrenia (e.g., a genetic vulnerability or insult during brain development, such as intrauterine infections or maternal starvation) gets uncovered when the brain matures (e.g., during the naturally occurring pruning of excessive synapses) or additional insults (e.g., cannabis use) push a vulnerable brain toward psychosis. Since the clinical picture (phenotype) is not solely determined by the gene sequence (genotype), but rather by which genes are expressed (the epigenotype), some gene–environment interactions may occur via epigenetic modifications of the genome. Functionally, a disruption of coordinated activity among regions within (and between) several, distinct networks in the brain has been suggested to account for the emergence of clinical symptoms in this developmental model.

Pathophysiology

Evidence from neurochemistry, cellular neuropathology, and neuroimaging studies supports the idea that schizophrenia is a brain disease: that is, a disease manifest by abnormalities in brain structure, brain function, or both. The observed abnormalities are, thus far, of limited specificity and sensitivity, and have therefore not yet had clinical relevance for diagnosis and treatment, and are of only limited value in prognostication. Furthermore, there is no single universally accepted theory regarding the brain dysfunction seen in schizophrenia, but rather there are a host of competing and overlapping models.

Changes in Brain Structure

From the time the first human brain images were obtained, the size of the brain and its ventricular system in schizophrenia has been measured, first using the indirect method of pneumoencephalography, and later with computed tomography (CT) and magnetic resonance imaging (MRI). Soon after the CT became widely available, the landmark study of Johnstone and colleagues revealed that patients with schizophrenia have larger ventricular volumes than demographically matched healthy control subjects. This finding, which has subsequently been replicated many times using MRI, has had a sustained impact on the field and the public perception of the disorder, because it provided the first incontrovertible, visually apparent evidence that schizophrenia is a disease of the brain ( Figure 28-2 ). However, in the decades that followed that initial finding, it became clear that ventricular enlargement is neither specific nor sensitive enough to use as a diagnostic tool; there is a great deal of overlap between the distribution of ventricular sizes of patients with schizophrenia and healthy people (a person with schizophrenia can have ventricular volume well within the normal range). Also, the increase in ventricular volume found in schizophrenia is relatively diffuse (i.e., the entire ventricular system is somewhat affected). Thus, there is no clearly localizable change in brain structure associated with this abnormality.

Subsequent work has attempted to identify specific structural changes within the cerebral cortex and subcortical nuclei in patients with schizophrenia. Structural MRI studies have found abnormal reductions in regional brain volume or cortical thickness in a large number of areas in schizophrenia, consistent with post-mortem findings of diminished volumes of cortical and subcortical regions, and increased pyramidal cell packing density. For example, meta-analyses of volumetric analyses of imaging data indicate that, on average, hippocampal volumes are 4% smaller in patients with schizophrenia than in matched controls. Other imaging techniques that measure the integrity of the fiber bundles of the brain, such as diffusion tensor imaging (DTI), have found changes in white matter tracts in schizophrenia, suggesting that there may be molecular abnormalities affecting the fiber connections between brain regions in the disorder. Longitudinal imaging studies have found that, although progressive reductions in brain volume normally occur during human brain maturation, the rate of volume change (brain tissue loss) in patients with schizophrenia is more than twice that of healthy subjects. Some of these volume changes might occur during the development of psychosis or during psychotic relapses, and some (including smaller volumes of the anterior cingulate, insular, temporal, parahippocampal cortices) may be present prior to the onset of the illness (during the prodromal state) and predict the later development of full-blown psychosis. However, some of these progressive reductions in brain volume seen in patients with schizophrenia may represent effects of being in poorer physical health, smoking tobacco, or being treated with antipsychotic medication. Distinguishing the changes in the brain that result from the life-style changes and treatments associated with having the illness from those related to the fundamental pathophysiology of schizophrenia remains a challenge for ongoing research.

Changes in Brain Function

Event-related electroencephalography (event-related potentials [ERPs]) and functional magnetic resonance imaging (fMRI) are two technologies used to examine brain function that have distinct advantages: ERPs have excellent temporal resolution, whereas fMRI has superior spatial (anatomical) resolution. Using ERPs, several abnormal physiological effects, sometimes called endophenotypes (if they are also seen in first-degree relatives and appear to be heritable), have been well replicated in patients with schizophrenia. The P ₅₀ , for example, is an ERP waveform that measures the ability to suppress irrelevant information (sensory gating). The P ₅₀ is often attenuated in patients with schizophrenia, as well as in their clinically healthy relatives. This endophenotype is linked to genetic polymorphisms of the alpha ₇ nicotinic receptor, a potential target of treatments for the cognitive deficits of schizo­phrenia. Another measure of sensory gating, prepulse inhibition (PPI) has been linked to haploinsufficiency of the Tbx1 gene. This gene is one of the affected genes in the 22q11 deletion syndrome that is associated with increased risk for schizophrenia. Abnormalities in other ERP measurements, such as the P300 and “mismatch negativity” have also been found consistently in patients with schizophrenia.

Functional neuroimaging studies have found that frontal cortical regions, including the dorsolateral prefrontal and anterior cingulate cortices, function abnormally in schizophrenia, consistent with the findings of reduced frontal volumes, and impairment in the cognitive domains (executive function, planning, task switching) subserved by these areas, in schizophrenia. However, the direction of findings has been variable; some studies have found reductions in prefrontal responses, whereas other studies have found abnormal increases. These discrepancies appear to depend on the level of difficulty of the cognitive task performed, suggesting that the prefrontal cortex in schizophrenia is in fact “inefficient” in its use of neural resources to execute cognitive tasks. Abnormalities in the function of other regions, such as the hippocampus or lateral temporal or parietal cortex, which receive projections from the prefrontal cortex, has led to the proposal that distributed networks of regions (rather than one or two specific brain areas), as well as the connections among these regions, are disrupted in schizophrenia. Thus, whereas cognitive impairment in schizophrenia has been linked to dysfunction of regions known to mediate executive and memory processes, impaired emotional function in schizophrenia (e.g., abnormal assessments of emotional salience or meaning in psychotic states and the impaired motivation seen in patients with negative symptoms) has been linked to changes in networks involved in generating emotional responses and emotional learning and memory. In addition, there is evidence for abnormalities in basic sensory processing in schizophrenia. Thus, multiple domains of brain functioning are compromised in schizophrenia to varying extents, suggesting that one or more fundamental physiological process is disrupted across several networks.

The theoretical conceptualization of schizophrenia as a “disconnection syndrome” has received recent support from studies of functional connectivity—an fMRI measure of the degree to which regions within a network exhibit coordinated (i.e., correlated) activity. This type of correlated activity or functional coupling within a network is thought to reflect in part the structural integrity of the associated anatomical connections. Patients with schizophrenia exhibit widespread abnormalities (both increases and decreases compared to healthy subjects) in functional connectivity. Functional connectivity changes in a frontotemporal circuit have also been found in healthy people and patients with schizophrenia who have a variant of a gene associated with schizophrenia risk. Also, several studies have shown that a network that is more active during “resting” states (called the default mode network [see Chapter 72 ]), which includes the medial prefrontal, posterior cingulate, and lateral temporal-parietal cortices, as well as medial temporal lobe structures, may be disrupted early on in schizophrenia and in people with genetic risk factors for the disorder. For example, Figure 28-3 shows that functional connectivity between two central nodes of the default mode network, the medial prefrontal and posterior cingulate cortices, which are each involved in self-referential thinking and other types of social cognition, is markedly diminished in patients with schizophrenia, in comparison to demographically matched healthy control subjects. Future studies will determine the nature of the relationship between changes in functional and structural connectivity measures in schizophrenia and the underlying molecular mechanisms that give rise to these abnormalities.

Clinical Features and Diagnosis

The diagnosis of schizophrenia is made clinically, based on a typical combination of symptoms (present cross-sectionally and longitudinally), in the absence of other psychiatric or medical conditions that would explain the symptoms (see below). The exact number and combination of symptoms, as well as the required duration of symptoms to make a diagnosis of schizophrenia, differ depending on the classification system used ( Table 28-2 , see Box 28-2 for full DSM-5 diagnostic criteria). Making a diagnosis based on clinical symptomatology and course alone, without the help of genetic markers or biomarkers, can lead to different diagnoses over time (even in the same patient if the clinical picture changes).

Schizophrenia is a disorder with an onset in late adolescence or early adulthood. Most patients (80%) present with schizophrenia between 15 and 45 years of age. Onset during childhood or in late life is possible, but not common. Onset at the extremes of the age range shows continuity with typical-onset schizophrenia, although the onset of psychosis after age 50 should raise suspicion for a secondary psychosis (i.e., secondary to a non-psychiatric medical disorder).

The onset of schizophrenia can be acute (i.e., symptoms develop over a few days) or subacute (i.e., symptoms develop over a month), although a more insidious onset with signs of the illness beginning many months or even years before frank psychosis is usually seen ( Figure 28-4 ). A non-specific prodromal phase can often be ascertained retrospectively. Prodromal symptoms include attenuated psychotic symptoms (e.g., suspiciousness, perceptual distortions, or perplexity), depression and suicidality, obsessive thinking, and sleep problems. Role failure and loss of social competence are typical features. However, the progression to the full syndrome of schizophrenia for patients in a putative prodromal state (also referred to as “ultra-high risk state”) is not inevitable, with average conversion rates of 18% after 6 months, 22% after 1 year, 29% after 2 years, and 36% after 3 years. Notably, many help-seeking patients who do not convert to psychosis continue to experience mild symptoms and psychosocial difficulties.

Unfortunately, even after the development of clear-cut psychosis, the patient and his or her family and friends often do not recognize it. Even when recognized, the affected individual often resists treatment. This results in untreated psychosis that lasts on average almost 2 years. The duration of untreated illness, which includes the prodromal period and the duration of untreated psychosis (DUP), can last several years, and typically results in disrupted psychological and social development and impaired role function. A shorter DUP is associated with improved symptom-response to antipsychotics and less impairment in overall functioning and better quality of life. The prodromal period is an area of active research, since the functional impairment and cognitive decline seen in schizophrenia occurs before the onset of psychosis and remains stable even after recovery from psychosis. Exactly when the cognitive decline begins remains uncertain, but early signs of the disturbance (as judged by scholastic records) point toward onset in middle school or even earlier, possibly with further worsening around the time of psychosis and with cumulative damage due to relapse.

The hallmarks of acute schizophrenia are hallucinations and delusions, sometimes grouped together as positive symptoms. Hallucinations are perceptions without an external stimulus, and they can occur in any sensory modality ( Box 28-3 ). However, by far the most common type of hallucination is auditory, occurring in at least two-thirds of patients over the course of their illness. Certain types of third-person (Schneiderian) hallucinations are frequently encountered: several voices talking about the patient, often in a derogatory way; a voice giving a running commentary on what the patient is doing; or a voice repeating what the patient is thinking. While hallucinations in other modalities are possible, visual or olfactory hallucinations in particular should raise the suspicion for an organic etiology of the hallucinations.

Delusions are false, non-culturally sanctioned beliefs that are held with great conviction, even in the face of overwhelming evidence to the contrary. Box 28-4 outlines common delusional themes. The delusional idea puts the patient at odds with his or her culture or subculture. The content of delusions can be non-bizarre (feasible) or bizarre (impossible based on the laws of physics). Bizarre delusions suggest schizophrenia and are less characteristic for delusional disorder.