When the solution to a clinical or scientific puzzle eludes us for more than a century, as with schizophrenia (formerly dementia praecox), we need new ways of thinking about the problem [1
]. Efforts to understand schizophrenia have focused on neurons and, especially, the role of presumed excess dopamine neurotransmission. We believe that genetic, environmental, and stochastic factors combine with epigenetic factors to create episodes of the illness [3
]. Thus, the syndrome of schizophrenia is viewed as an endpoint in a dynamic process
variously conceptualized as degenerative or developmental or alternating at different points in the process [6
Degenerative models imply that after a period of normal development, the organism, or one of its parts, takes a wrongful turn in its trajectory and begins to malfunction. This describes the eventual outcome for all life forms and is a biological restatement of the second law of thermodynamics. Since degeneration is universal, stating that an illness is degenerative is not particularly helpful. What would be helpful is to determine when in the life course the degeneration begins and how the degeneration is initiated and proceeds. Answers to the "when?" and "how?" questions would then describe the degenerative process in developmental terms.
Developmental models of schizophrenia implicate abnormalities of early brain development predisposing to future schizophrenia. The proponents of the model further argue that the perturbations of development are limited to the early times of development and are discontinuous. Without this qualifier, developmental models are indistinguishable from degenerative models where the degeneration commences early in the life span. The early abnormalities are not necessarily the cause of schizophrenia, but, instead, create a state of risk for a future episode of schizophrenia. That is, a diathesis or predisposition is not a disease. Consequently, there must be factors later in life that convert the vulnerability to an illness. These additional factors are presumed to damage development in such a way that a predisposition becomes actualized. To gain a complete understanding of the syndrome, we must again return to the question of " what happens?"
Following this line of reasoning, the distinction between degenerative and developmental models blurs. In fact, a medical-behavioral condition can be both developmental and degenerative as exemplified by Down syndrome [11
]. Individuals born with trisomy 21 exhibit a number of developmental anomalies including cardiac malformations, abnormal dermatoglyphics, skeletal changes, and muscular hypotonia, to name a few. As trisomy 21 infants mature, most exhibit degrees of mental retardation. By about age 50, these individuals invariably develop Alzheimer-like CNS degenerative changes that can be seen at autopsy [13
Schizophrenia involves both developmental and degenerative features. From the time of Bleuler [14
] and Kraepelin[15
], "It is certain that many a schizophrenia can be traced back into the early years of the patient's lives..." [14
] p. 252. The 'follow back' studies of schizophrenia support these views [16
]. Likewise, prospective studies of children at high risk for schizophrenia report developmental anomalies in motor skills, cognition, and attention long before the onset of overt illness [17
]. Overt psychotic symptoms for some individuals usually start in the late teenage years or early twenties, but the illness can start as early as middle childhood [20
] and may, more rarely, start in old age [21
] p 73].
The evidence suggesting early developmental perturbations in schizophrenia is compelling. At the same time, there certainly are examples of deterioration reminiscent of Kraepelin's suggestion for some people with schizophrenia. However, deterioration in clinical course may not indicate CNS deterioration. Instead, the decline could be a secondary consequence of an illness that disrupts education, economic achievement, and social functioning leading to a downward spiral in all aspects of adult life. Consistent with an early degenerative process, there are reports of declining cognitive function preceding onset of psychosis [22
]. Proponents of neurodevelopmental models suggest that the premorbid cognitive abnormalities are developmental risk factors for future schizophrenia (c.f [23
]) and argue that such abnormalities show little evidence of decline after onset [6
]. Whether developmental or degenerative, the premorbid cognitive deficits seen in schizophrenia are also seen in other disorders [25
] and lack specificity and sensitivity thus detracting from the concept that the cognitive abnormalities seen in schizophrenia are useful endophenotypes [26
]. The strongest evidence for a neurodegenerative phenomenon comes from imaging studies showing progressive loss of brain volumes [27
]. Neuropathological studies fail to find widespread classic signs of neurodegeneration such as gliosis though there are exceptions to this generalization [30
]. Observations of abnormal dendritic arborization [31
] are consistent with the neuroimaging evidence suggesting abnormal connectivity between brain regions [29
]. As a cautionary note, most of the neuroimaging and neuropathology results are subject to confounds from the effects of medications and various other treatments, post-mortem intervals, possible effects of diet, smoking habits, as well as a myriad of other potential confounds associated with glucocorticoid mediated stress following chronic illness and associated life's limitations [33
The symptoms of schizophrenia are highly variable. Within families (and thus presuming relative homogeneity of genetic and environmental factors) symptoms can vary widely over time, as illustrated by identical quadruplets concordant for schizophrenia [35
]. Even within affected individuals, symptoms will wax and wane and may even remit [36
] suggesting a life long process.
The major behavioral symptoms of schizophrenia include alterations in cognition, memory, perception, thought (inferred from language), motor functions, and affect. People with schizophrenia may show abnormal dermatoglyphics and other minor physical anomalies [37
]. Other oddities to be incorporated in a comprehensive explanation of schizophrenia include highly visible nail fold capillaries [43
] and the rarity of rheumatoid arthritis among schizophrenic persons [45
]. These physical characteristics suggest the need to look beyond the nervous system per se
to have a comprehensive view of the illness.
The fact that the schizophrenia syndrome, as currently defined, is relatively common provides important information about the frequency of causal factors. About 1% of the population will experience schizophrenia during the lifespan. Except for a few rare exceptions, this 1% risk is remarkably constant around the globe regardless of culture, geography, or ethnicity. Men and women are affected equally. These facts mean that the risk factors for schizophrenia must also be common and ubiquitous. Given that the concordance rate for schizophrenia in identical twins [46
] is only about 50%, there must be at least two global risk-increasing categories for schizophrenia, i.e., something(s) genetic and something(s) environmental. Assuming these risk factors are independent of each other, the joint probability of acquiring both risk factors is the product of their population frequencies that, for schizophrenia, equals about .01. To make a simplifying assumption to allow easy calculations, let us say that the two risk factors are present with about equal frequency in the population. With this simplification, straightforward mathematics indicates that the individual frequencies of these factors are close to the square root of the population frequency of 1%. That would mean that about 10% of the population would encounter at least one risk factor. The math indicates that the greater the number of independent risk factors, the more common they are. [See [47
] for further elaboration].
Our challenge is to develop a theory of schizophrenia that can plausibly explain an illness that affects all domains of behavior (thought, affect, motor performance, etc), that has elements of developmental perturbations early in life leaving clues such as minor physical abnormalities, and also has elements of degenerative changes. At the same time, the defect is so subtle that we can't find the cause(s) with our best modern technology. Furthermore, in spite of brain-wide dysfunctions, many individuals with schizophrenia remain sufficiently intact that, with good treatment and a bit of luck, can maintain jobs and function usefully in society. Thus, we need to find frequent and ubiquitous factors that can affect virtually all brain functions as well as creating somatic signs, but they operate in ways that leave these functions only slightly "off kilter" as compared to the complete disruption seen in strokes, or classical degenerative disorders such as Alzheimer, or as seen in Down syndrome where the behavioral pathology is apparent from earliest stages. As we try to explain schizophrenia, we must account for most all of the developmental and degenerative features of schizophrenia.
To account for the panoply of signs and symptoms seen in schizophrenia, any complete theory of schizophrenia must include organism wide systems. In addition to the nervous system, the immune system and the vascular system are defensible candidates. Both are invoked in the following theory: Some schizophrenia psychoses are the result of damage to the micro-vascular system in the brain initiated by genetically influenced abnormal inflammatory processes acting in response to ubiquitous environmental factors that trigger inflammatory responses, including infection, trauma, or hypoxia. It is the relative infrequency of the vulnerable genotypes in the population [48
] that results in only a small proportion developing overt psychosis.
We wish to emphasize that our hypothesis specifically identifies the microvascular system as the critical site of inflammation. We postulate that the inflamed micro-vessels lose their coupling with astrocytes, leading to disrupted regulation of cerebral blood flow and damage to the blood brain barrier. These disruptions in homeostatic mechanisms then lead to abnormal signal processing. Our focus on inflammation of the vessels differentiates our hypothesis from models of widespread parenchymal inflammation such as seen in psychotic syndromes following, for example, encephalitis lethargica, or paraneoplastic syndromes. Many acute inflammatory disorders of the brain involve inflammation of both the parenchyma and the vasculature. By contrast, we are proposing a chronic, smoldering, inflammation of the vessels alone. And, finally, we distinguish our hypothesis from the theories of schizophrenia implicating direct parenchymal infection of the brain (c.f. [49
]) and also differentiates our hypothesis from speculations about schizophrenia that invoke infectious agents altering DNA [50
Many prior debates about inflammation in the brains of people with schizophrenia have focused on the presence of absence of gliosis (see [51
] for review). The consensus opinion is that gliosis, though present in some cases, is not a consistent feature of the neuropathology of schizophrenia. However, as Harrison [51
] points out, evaluating gliosis is fraught with a multitude of problems and is not a definitive indicator of degenerative/inflammatory changes in the brain. More recent efforts have demonstrated activation of microglia in the brains of some individuals with schizophrenia implying an ongoing immunopathological process in addition to what ever happened early in development [52
]. Ongoing neurodegenerative processes are suggested by increased levels of S100B, a small calcium binding astrocytic protein that is involved in inducing apoptosis and modulating proinflammatory cytokines [53
It is likely that the current clinical syndrome of schizophrenia is etiologically heterogeneous. We do not pretend to explain all (DSM or ICD) cases of syndromal schizophrenia. Instead, we put forward our hypothesis as an attempt to define a psychiatric syndrome in terms of a particular pathophysiology. Following this course may then help refine our nosology (see also section on 'specificity' below) and cause us to recalculate basics 'facts' such as prevalence rates.