Background: Neuroscience can assist clinical understanding and therapy by finding neurobiological markers for mental illness symptoms. Objectives: To quantify biomarkers for schizophrenia and schizoaffective disorder and relate these to discrete symptoms of psychosis. Methods: Within a case-control design with multiple exclusion criteria to exclude organic causes and confounding variables, 67 DSM IV-R diagnosed and 67 control participants from a defined hospital, clinic and community catchment area were investigated for candidate markers. Participants underwent protocol-based diagnostic-checking and symptom rating via Brief Psychiatric Rating Scale and Positive and Negative Syndrome Scale, functional-rating scales, biological sample-collection and sensory-processing assessment. Blood and urine samples were analysed for monoamine neurotransmitters, their metabolites, vitamin cofactors and intermediate-substances related to oxidative stress and metabolism of monoamines. Neurocognitive assessment of visual and auditory processing was conducted at both peripheral and central levels. Biomarkers were defined by Receiver Operating Curve (ROC) analysis. Spearman’s analysis explored correlations between discrete symptoms and the biomarkers. Results: Receiver Operating Curve (ROC) analysis identified twenty-one biomarkers: elevated urinary dopamine, noradrenaline, adrenaline and hydroxy pyrroline-2-one as a marker of oxidative stress. Other biomarkers were deficits in vitamins D, B6 and folate, elevation of serum B12 and free serum copper to zinc ratio, along with deficits in dichotic listening, distance vision, visual and auditory speed of processing, visual and auditory working memory and six middle ear acoustic reflex parameters. Discrete symptoms such as delusions, hostility, suicidality and auditory hallucinations were biomarker-defined and symptom biomarker correlations assumed an understandable pattern in terms of the catecholamines and their relationship to biochemistry, brain function and disconnectivity. Conclusions: In the absence of a full diagnosis, biomarker-symptom-signatures inform psychiatry about the structure of psychosis and provide an understandable pattern that points in the direction of a new neurobiological system of symptom-formation and treatment.
The architecture of schizophrenia is still largely unknown and conventional categories for diagnosing schizophrenia and schizoaffective states are still based on descriptions of symptoms and behaviours [
In the clinical setting, presenting symptoms that fail to join with others to meet sufficient criteria for any firm classified diagnosis cause clinicians considerable stress. Clinicians walk a fine line between knowing what a single symptom such as hallucinations might mean in the longer term and the knowledge that the earlier schizophrenia is confirmed and treatment is able to begin, then the better is the outcome [
Schizophrenia and schizoaffective disorder are allied conditions within the clinical setting, with the DSM- IV-R classification system describing schizoaffective disorder in terms of a major mood disorder episode occurring concurrent with symptoms that meet characteristic symptoms for schizophrenia [
A case-control study design was used to clearly differentiate candidate markers between cases and controls. Multiple exclusion criteria (Section 2.3) were applied to case selection in order to strip psychosis in schizophrenia and schizoaffective disorder down to a functional core uninfluenced by organic causes such as substance abuse, head injury, sensory disability or medication variables that could unduly influence outcome measures.
This study was conducted between May 2010 and December 2014. Participants were enrolled between May 2010 and December 2013 at the Queen Elizabeth Hospital, Woodville, South Australia and at its satellite mental health clinics in the Western Adelaide community catchment area. All participants were informed of the goals, assessment procedures and funding of this study and provided written consent. Ethics permission for the study was obtained from the hospital ethics committee. Participants were from multi-ethnic backgrounds and the age-range was 18 to 60 years. A highly-characterized group of patients diagnosed with schizophrenia or schizoaffective disorder were compared with an age and gender matched group of controls, who had subclinical symptoms only. Similar psychotic symptoms occur in both schizophrenia and schizoaffective conditions [
In order to minimise setting variables participant assessments took place within a 4 day window. After casenote review, a standard interview protocol collected demographic information and information related to development, organic causes or sensory disability. Also recorded were height, weight and absence or presence of developmental difficulty or learning delay, medical co-morbidity, head-injury, family history of mental illness, use of glasses or hearing aids, in order to identify participants with peripheral sensory issues. Other information collected was time of onset of illness and duration of illness, social attachment durability and vision and hearing history.
All participants were rated using standard rating scales [
Non-detained ward patients in partial remission but with residual symptoms of psychosis, were recruited and assessed in the expected last week of their admission, by which time they were sufficiently recovered to give informed consent. Other patients were evaluated in satellite psychiatric outpatient clinics and the nearby research institute setting. A total of 82 symptomatic participants (cases) were recruited and completed assessment. Early statistical analysis of confounders required that 15 participants on SSRI or SNRI antidepressant medication be excluded from the study, due to the masking effect of SSRI and SNRI medication on catecholamine levels, sensory-processing and middle-ear outcome measurements. Therefore the number of patients in the final statistical analysis was 67.
Patients were included in the study if they received a diagnosis of schizophrenia or schizoaffective disorder made by a consultant psychiatrist in the ward or satellite psychiatric outpatient department, according to the DSM-IV-R classification criteria [
The multiple exclusion criteria applied to patient selection are now described: clinical appraisal of a patient’s judgment capacity and orientation in time, place and person was undertaken during recruitment, in order to confirm capacity to consent and exclude delirium. Informal examination to exclude ocular muscle dysfunction, hand, forearm and shoulder dysfunction was also conducted at this stage, in order to ensure that a patient was free to proceed to neuro-physiological tests in the absence of rigidity, dyskinesia, tremor or postural instability as a result of extra-pyramidal side-effects from antipsychotic medication [
A total of 72 control participants were recruited with the assistance of the Population Research and Outcomes Studies (PROS) Unit of the University of Adelaide. These participants were volunteers from the same catchment area as patients affiliated with the Queen Elizabeth Hospital and the North West Adelaide Health Study (NWAHS). Using the same exclusion protocol as for patients, these participants were age-stratified and randomly recruited by phone over the same assessment period as patients. None of the controls reached symptom threshold for schizophrenia or any DSM-diagnosable mental illness, but were rated for reported and observed subclinical symptoms by a psychiatric trained assessor who was not blind to their asymptomatic status, but was blind to biological test results. Two younger controls required for younger age matching, were recruited from a local surf life-saving club within the same NWAHS catchment area. Five control participants were excluded due to their failure to meet exclusion criteria on assessment. The control sample used for the final analysis was drawn from the same catchment area as the patients and final sample consisted 67 controls.
Laboratory tests used to assay biochemical markers in this study were commercially-available blood and urine samples, collected from all participants. Collection methods are documented in
All biochemical testing was conducted by independent laboratories that were blind to participants’ case or control status and all participant symptom raters and neurophysiological assessors were blind to laboratory results. Baseline fasting spot-urine samples were collected for biochemical assays of dopamine, noradrenaline and adrenaline as well as their metabolites homovanillic acid (HVA) and methoxy-hydroxymandelic acid (MHMA) and the serotonin metabolite 5-hydroxyindoleacetic acid (5-HIAA). As part of this analysis, urinary creatinine was determined as a urine concentration standard and this standard was also used for urinary hydroxyhemopyrroline-2-one (HPL). In order to ensure that urine neurotransmitter levels were not affected by hypothalamic pituitary axis (HPA) activation, due to the procedure of blood-taking, urine collection preceded blood-taking, by a minimum of two hours. Fasting blood was also collected for vitamins and mineral cofactors.
Reverse digit span working memory and spatial working memory deficits have a known association with schizophrenia [
Image 1. Sensory processing briefcase equipment (used with permission).
Image 2. Sensory processing speed assessment.
Where applicable, visual assessment was conducted using the participant’s usual glasses and alternate-cover- test was conducted prior to visual testing, to exclude visual fixation disparity (phoria or tropia) as a potentially confounding variable [
Auditory assessments were conducted in a quiet room (ambient noise level 20 dB) and preceded by examination of the external auditory meatus to exclude obvious pathology or sebum obstruction. Audiometric examination was conducted using the MAICO Audiogram MA 40 [
In an effort to investigate pre-attentive deficits relating to acoustic reflexes and their relationship to central auditory-processing deficits or delays, specific measures were obtained for ear canal volume at threshold auditory response, peak middle ear pressure at threshold auditory response and gradient of the middle ear pressure elevation, using the Tymp GSI 38 by VIASYS Healthcare [
Data analysis on a sample consisting of 67 cases and 67 controls, within which there were 30 female cases with a mean age of 42 years and 37 males with a mean age of 39. In the control sample there were 34 females with a mean age of 45 and 33 males with a mean age of 46. For the total participant sample females totalled 64 and males totalled 70. The mean age for cases was 41 years and the mean age for controls was 46 years, matched on average to ±3 years. Variable distributions were mapped and quantile functions calculated using EasyFit software [
Test | Method, analyser, reagents | Laboratory/reference |
---|---|---|
Neurotransmitters | ||
Biogenic amines: dopamine, noradrenaline and adrenaline, | Spot-baseline (fasting) urinary neurotransmitter testing (second void morning), snap-frozen to minus 30 degrees and analysed by mass spectrometry, using nanomols per millimol of urinary creatinine as a standard. | SA Pathology, Adelaide, South Australia. Whiting MJ. 2009. Simultaneous measurement of urine metanephrines and catecholamines by liquid chromatography with tandem mass spectrometric detection. Annals of Clinical Biochemistry, 46, 129-136. |
Creatinine | Spot urine specimen from the same void as biogenic amines, expressed in millimols per Litre. | SA Pathology, Adelaide SA. |
Oxidative stress: Urinary hydroxyhemopyrroline-2-one | ||
Urinary hydroxyhemopyrroline-2-one (HPL) | Spot urine (second void morning) in ascorbic acid, snap frozen (-30C) and light-protected. Colorimetric method at 540nm, following solvent extraction and reaction with Erich’s reagent. micrograms per decilitre. | Applied Analytical Laboratories, 8/26 Nestor Drive, Meadowbrook, Queensland, Australia. +61 7 3133 1615. |
Nutrition-Biochemistry | ||
Vitamin D (25-OH) | Diasorin Liason assay kit. for use on the Liaison platform nmol/L. | Clinpath Laboratories, 19 Fullarton Rd, Kent Town. South Australia 5067+61 8 8366 2000. |
Serum total Vitamin B12 | Competitive Electrochemiluminescent Immunoassay. Roche Modular E 170. Automated Immunoassay Analyser, using Roche Vitamin B12 Reagent (pmol/L). | Clinpath Laboratories. As above. |
Red Cell Folate | Competitive Electrochemiluminescent Protein Binding Assay, using Roche Modular E 170, using Roche Folate Red Blood Cell (RBC) Reagent and Roche Folate RBC Haemolysing Reagent. on Automated Immunoassay Analyser (nmol/L). | Clinpath Laboratories. As above. |
Vitamin B6 (Pyridoxal-5’-phosphate coenzyme form) | Whole blood High Pressure Liquid Chromatography with fluorescent detection. Chromsystems Vitamin B6 in Whole Blood High pressure Liqid Chromatography Reagent Kit. Waters Alliance 2695 Separations Module. Waters 474 Fluorescence Detector (nmol/L). | Sullivan Nicolaides Pathology 143 Whitmore St, Taringa. Queensland 4068. Australia. +61 7 337 8666 |
Serum Copper | Flame Atomic Absorption Spectrophotometry. Varian AA-240FS (umol/L). | Douglass Hanly Moir Pathology 14 Griffnock Avenue, Macquarie Park. New South Wales 2113. +61 2 9855 5222. |
Red Cell Zinc | Inductively coupled plasma mass spectroscopy (ICP-MS), using 6% n-Butanol reagent and Agileny ICP-MS 7500ce analyser (umol/L). | Sullivan Nicolaides Pathology. As above. |
Serum Ceruloplasmin | Immunoturbidimetric method, using 6K91-30 Multignet Caeruloplamin Kit and Abbott Architect ci16000 analyser (g/L). | Douglass Hanly Moir Pathology. As above. |
Percentage Free Copper/Red Cell Zinc | Percentage of free copper in the serum calculated by an equation based on the molecular and atomic weights of ceruloplasmin and copper (one ceruloplasmin molecule binds to six copper atoms). The ratio of the percentage free copper to red cell zinc was calculated as “percentage free copper”/“Red cell zinc umol/L”. | Calculated by authors |
Intermediate substrates and enzymes | ||
MTHFR Ala222Val (C677T) methyl tetrahydrofolate reductase polymorphism | Real time PCR analysis Roche Diagnostics Light-Cycler 480 kit. Using TecnoBiol reagents, Sigma probes and primers on Roche LC480 analyser. | Douglass Hanly Moir Pathology. As above. |
Plasma homocysteine | Ice transported EDTA sample. Competitive Chemiluminescent Immunoassay, using Seimens Homocysteine reagent on Seimens Advia centaur Automated Immunoassay (umol/L). | SA Pathology. +61 8 8222 3000 As above. |
Serum histamine | Beckman Coulter Radio Immunoassay, using Beckman Coulter R.I.A. Kit on Perkin Elmer Wizard 1470 Automated Gamma Counter (umol/L). | Sullivan and Nicolaides. As above. |
Visual assessments | Method | Reference |
---|---|---|
Near vision acuity test | Sussex Vision test of near vision. Near vision test card SNT-3000-L, 2009-2011. | Sussex Vision test of near vision. Near vision test card SNT-3000-L, 2009-2011. Sussex Vision International Ltd.) http://Zazzle.com.au/near_vision_testcard_visual_acuity_exam-137420023635816730. Sussex Vision International Ltd. |
Visual (symbol) span | Increasing number of symbols are presented in a standardised order from left to right. Test score reported as the absolute number of visual symbols recalled in the correct order. | Visual symbol Span. A subset test of WMS-IV. Weschler Memory Scale, Fourth Edition. 2009. Pearson Inc. Keith RW. Pearson Clinical Assessment P.O. Box 599700 San Antonio, Texas 78259 |
Distance vision (Binocular distance vision acuity) | Right distance vision, then left distance vision, with 20 seconds inter-test interval. | The Snellen-Chart (Snellen 1862) H. Snellen, Probebuchstaben zur Bestimmung der Sehschärfe, Utrecht 1862. |
Threshold visual speed of processing performance as a percentage of age (Expresses visual processing speed in terms of the visual processing system’s relative age) Shortest interval of time a person can notice between the order of presentation of two optical stimuli. Speed of visual order processing increases with age. For adults between the range of 18 and 60 years, the normal range for visual speed of processing is 24 to 72 milliseconds). For adults between the range of 18 and 60 years, the normal range for visual speed of processing is 24 to 72 milliseconds. | Person tested sees two brief flashes of light randomly presented from left-to-right or right-to-left on multiple occasions, and must decide which light flash appeared first. The inter-stimulus time interval (ISI) between the flashes is shortened by computer algorithm, if the answer is correct, otherwise it is lengthened. A performance-age rating, is provided, configured against norms-for-age. Performance-age is subtracted from the individual’s actual age and the result divided by the age of the test subject is multiplied by 100. | Brain Boy Universal Professional instrument (MediTECH 2010). Langer Acker 7. D-30900 Wedemark, Germany. |
Auditory assessment | Method | Reference |
---|---|---|
Reverse digit span (Measures auditory (verbal) working memory) Normal range is 6 to 7 | With gaze aversion by listening participant and tester, digits are read in set sequence. The tested participant is asked to repeat them in reverse order. Reported as the absolute number of digits correctly recalled in reverse order. | Weschler D.: Wechsler Adult Intelligence Scale-III (WAIS-III), 1997. The Psychological Corporation. Pearson Clinical Assessment, P.O. Box 599700, San Antonio, Texas 78259. |
Competing words performance for age as a percentage of age (Intra-cerebral dichotic listening performance for processing of auditory information) Normal ranges vary with age | A voice-over CD and earphones test ability to correctly identify both of two competing-words (CW), delivered separately to the right and left ears. Using this test’s normative-for-age database, the difference between each test subject’s expected and actual performance-for-age was calculated, and this was then divided by the actual age of the test subject, and multiplied by 100. | Keith RW: SCAN-3 test for adolescents and adults. Pearson Clinical and Talent Assessment 2009. Published by Pearson ISBN 0158910176. |
Threshold speed of Auditory processing as a percentage of age (Speed of auditory processing systems relative to age) Speed of auditory processing reduces with age. For adults in the age range of 18 and 60 years, the normal range for auditory speed of processing is 46 to 72 milliseconds. | Person tested hears two clicks, randomly presented from right to left and left to right side, presented through headphones. By pressing a right or left button, a decision must be made from which side the dual-stimulus originates. If the answer is correct, the inter-stimulus interval between flashes (ISI) is shortened, otherwise it is lengthened. The auditory order threshold is the shortest ISI a person can correctly differentiate between two auditory impressions. A read-out of the threshold speed of auditory (order) processing is provided, along with a norm performance-age rating. Auditory speed of (order) processing performance as a percentage of age is calculated by subtracting the norm-for-age from the performance-age, divided by the age of the test subject, multiplied by 100. | The Brain Boy Universal Professional Instrument (MediTECH 2010). By MediTECH Electronic GmbH, Langer Acker 7, 30900 Wedemark, Germany. www.meditech.de |
Middle ear assessments | Method | Reference |
---|---|---|
Ear canal volume at threshold auditory response. Normal range > 0.2 cm∙s3 | The GSI 38 measures combined volume of the ear canal and the middle ear space. An ear canal volume greater than 0.2 cm3 may indicate tympanic membrane perforation due to trauma or abnormal laxity of the tympanic membrane in response to acoustic pressure. | GSI 38 Grason-Stadler 2005-Auto Tymp. Graason-Stadler A division of VIASYS Health Care, 5225. Verona Rd, Building 2, Madison, WI 53711. |
Peak middle ear pressure at threshold auditory response Normal range −150 daPa to + 100 daPa | The GSI 38 measures peak middle ear pressure achieved in response to tympanic membrane change associated with tympanic muscle contraction. This is tested over the range of +200 daPa to −400 daPa. Low pressures may accompany Eustachian tube dysfunction. | GSI 38 (Grason-Stadler 2005) |
Gradient middle ear pressure at threshold auditory response. Normal range 60 - 150 daPa Percentage length of the base of the stapes reflex divided by the total duration of the reflex. (A measure of the strength of the stapes reflex during its maximal period of contraction) The term “threshold” refers to the first reflex response which typically occurs at a frequency of 500Hz in a decibel range of 90 to 110. Projected Stapes amplitude (Alternative measure of stapes contraction strength) Time-to-off-set of the stapes reflex contraction divided by the base length (a measure of any acoustic-reflex offset, advance or delay). | The GSI 38 measures the middle ear pressure gradient, for which the normal range is 50 to 110 daPa. High gradients are indicative of middle ear effusion. The GSI 38 traces the stapes reflex contraction to maximum amplitude of 8 millimetres, after which it traces a basal threshold formed for the maximum portion of the reflex. The length of this basal portion divided by the total duration of the reflex contraction was multiplied by 100. Direct measurement of stapes amplitude at the intersection of projected onset and offset contraction gradients. Threshold base directly measured and compared with the total duration of the stapes reflex, from initiation to time of offset. | GSI 38 (Grason-Stadler 2005). GSI 38 (Grason-Stadler 2005). GSI 38 (Grason-Stadler 2005). GSI 38 (Grason-Stadler 2005) |
biomarker variables is provided in
ROC analysis plots the sensitivity and specificity of the test result against each outcome-measurement, to give an indication of a variable ability to split between cases and controls, which is a test of screening capacity for that variable and thus, its biomarker status [
Though we essentially limited biomarkers identification to candidate markers that met the above criteria 1)-3), the ideal odds ratio of association with diagnosis of schizophrenia or schizoaffective disorder (
Having identified our biomarkers using the above criteria, we then used Spearman’s correlation matrix to explore and rank the relationship of the participants’ symptom-ratings to their ROC-discovered biomarkers.
All candidate biochemical variables tested, (with the exception of MTHFR C667T polymorphism, and plasma homocysteine), produced a high area under the curve (AUC > 0.6). High serum B12 ROC AUC was marginal at
Parameter | Number patients | Mean patients | Standard error | Number controls | Mean controls | Standard error |
---|---|---|---|---|---|---|
Age participants | 67 | 40.5 | 1.3 | 67 | 45.7 | 1.4 |
Age males | 37 | 38.6 | 1.4 | 33 | 46.4 | 1.7 |
Age females | 30 | 42.9 | 2.4 | 34 | 45.0 | 2.2 |
Body Mass Index (BMI) | 53 | 30.0 | 1.2 | 66 | 26.7 | 0.6 |
Right hand dominance % | 65 | 92.6 | 2.1 | 67 | 93.1 | 1.7 |
Urine creatinine (mmol/L) | 66 | 9.2 | 0.7 | 67 | 9.5 | 0.7 |
5-Hydroxyl indole acetic acid (5-HIAA) | 66 | 4.3 | 0.8 | 67 | 1.6 | 0.1 |
Plasma homocysteine (umol/L) | 66 | 10.0 | 0.3 | 66 | 9.5 | 0.3 |
Red cell acetylcholine esterase (U/gb Hb) | 61 | 39.8 | 0.7 | 67 | 39.6 | 0.7 |
Hearing threshold (Db) | 60 | 533.3 | 33.3 | 64 | 515.6 | 24.8 |
Visual threshold of near vision | 61 | 6.3 | 0.5 | 67 | 5.2 | 0.1 |
0.565 and high histamine produced a poor quality ROC curve. In the field of sensory processing assessment, all candidate variables tested reached ROC identified biomarker status, with the exception of outcome measures for gap-detection and auditory-figure ground tests within the SCAN test for auditory processing disorder and outcomes for near vision using the Sussex near vision test. Catecholamine metabolites homovanillic acid (HVA) and methoxy-hydroxymandelic acid (MHMA), along with the serotonin metabolite, 5-hydroxyindoleacetic acid (HIAA), also failed to reach significance on ROC analysis.
On ROC analysis, twenty-one outcome measures achieved the required area under the curve (AUC > 0.65), risk of rejection (RR < 10%) and sufficient correlation coefficient (usually with sufficient odds ratio and/or specificity), to achieve status as a biomarker (
In order to draw meaning from the correlation between symptoms and biomarkers, we segregated symptoms in two main ways. Firstly according to their lead (strongest) correlative biomarker and secondly according to the strength of their correlation with the three principal catecholamine components, noradrenaline, adrenaline and dopamine. In the following presentation the P values for Spearman’s correlates are not presented, however significance levels correspond to the following general rules: for rho > 0.80 to 0.40, P < 0.0001, for rho 0.4 to 0.350, P = 0.000, for rho 0.350 to 0.307, P = 0.001, for rho 0.260, P = 0.007, for rho 0.255, P = 0.008, for rho 0.250, P = 0.009, for rho 0. 247, P = 0.010, for rho 0.231, P = 0.016 and for rho 0.195, P = 0.043.
ROC variables | No. of Obs | AUC | ROC P value | SENS | SPEC | % RR-Ho |
---|---|---|---|---|---|---|
High dopamine (DA) | 133 | 0.702 | <0.0001 | 0.379 | 0.940 | 0.01 |
High noradrenaline (NA) | 133 | 0.851 | <0.0001 | 0.742 | 0.881 | 0.01 |
High adrenaline (AD) | 133 | 0.844 | <0.0001 | 0.758 | 0.821 | 0.01 |
High (HPL/creatinine) | 133 | 0.696 | <0.0001 | 0.697 | 0.642 | 0.01 |
High free copper to zinc ratio | 133 | 0.611 | 0.022 | 0.470 | 0.746 | 2.19 |
Low B6 activation | 129 | 0.638 | 0.002 | 0.800 | 0.484 | 0.17 |
High serum vitamin B12 (80% CI) | 134 | 0.563 | 0.0933 | 0.373 | 0.761 | 18.56 |
Low red cell folate | 133 | 0.654 | 0.001 | 0.591 | 0.716 | 0.10 |
Low vitamin D | 132 | 0.651 | 0.001 | 0.462 | 0.791 | 0.12 |
Low visual span | 126 | 0.862 | <0.0001 | 0.831 | 0.821 | 0.01 |
High visual speed of processing (% of age) | 122 | 0.875 | <0.0001 | 0.909 | 0.731 | 0.01 |
High visual speed of processing discrepancy (VSOP) (% of age) | 122 | 0.875 | <0.0001 | 0.909 | 0.731 | 0.01 |
Poor distance vision on right | 128 | 0.597 | <0.0001 | 0.475 | 0.851 | 0.01 |
Low reverse digit span | 127 | 0.810 | 0.000 | 0.900 | 0.552 | 0.02 |
High competing words discrepancy (% of pass score) | 124 | 0.799 | <0.0001 | 0.759 | 0.773 | 0.01 |
High auditory speed of processing discrepancy (ASOP) (% of age) | 121 | 0.874 | <0.0001 | 0.745 | 0.879 | 0.01 |
High threshold ear canal volume | 123 | 0.603 | <0.0001 | 0.367 | 0.825 | 0.01 |
Low threshold peak middle ear pressure | 124 | 0.617 | 0.000 | 0.700 | 0.484 | 0.04 |
High threshold gradient middle ear pressure (90 per cent) | 124 | 0.580 | 0.0674 | 0.370 | 0.891 | 6.74 |
High threshold stapes amplitude projected | 123 | 0.626 | 0.003 | 0.583 | 0.651 | 0.29 |
Low threshold time to offset over base-length | 122 | 0.659 | 0.001 | 0.683 | 0.613 | 0.01 |
High threshold percentage base-length over duration. | 122 | 0.657 | 0.001 | 0.583 | 0.774 | 0.14 |
ROC: Receuver Operating Curve, AUC, No Obs, Number of observations, Area Under Receiver Operating Curve (ideally > 0.6), SENS: Sensitivity, SPEC. Specificity, ROC P value. Receiver Operating Curve variavle P value at 95% Confidence Interval, % RR-Ho. Percentage Risk Ratio for Null Hypothesis (ideally < 0.10), HPL: Urinary hydroxyhemopyrroline-2-one.
When schizophrenia and schizoaffective disorder were segregated and ranked according to their lead (strongest) correlative biomarker, they fell into five main groups, outlined in Tables 7-11.
1) Low visual span Tables 7(a)-(f).
2) Low visual speed of processing Tables 8(a)-(c).
3) Low auditory speed of processing (ASOP) biomarker Tables 9(a)-(d).
4) Elevated noradrenaline
5) Elevated adrenaline
Biomarker ROCs | Cognitive disorganisation rho for alpha 0.05 |
---|---|
Low visual span (n 126) | 0.632 |
High noradrenaline (n 133) | 0.583 |
Low auditory speed of processing (% of age) (n 121) | 0.561 |
Low competing words score (% of pass score) (n 124) | 0.520 |
Low visual speed of processing (% of age) (n 122) | 0.510 |
High adrenaline (n 133) | 0.459 |
High dopamine (n 133) | 0.405 |
High distance vision score on right (n 128) | 0.341 |
Low reverse digit span (n 128) | 0.335 |
High HPL/creatinine (n 133) | 0.332 |
High serum B12 (n 134) | 0.283 |
Low red cell folate (n 133) | 0.258 |
High threshold gradient middle ear pressure (n 124) | 0.248 |
Long threshold per cent base length/ duration (n 122) | 0.242 |
Low vitamin D (n 132) | 0.233 |
Low threshold time to offset/base length (n 122) | 0.195 |
Low activated vitamin B6 (n 126) | 0.180 |
Biomarker ROCs | Suspiciousness rho for alpha 0.05 |
---|---|
Low visual span (n 126) | 0.604 |
Low visual speed of processing (% of age) (n 122) | 0.573 |
Low auditory speed of processing (% of age) (n 121) | 0.568 |
High noradrenaline (n 133) | 0.533 |
Low competing words score (% of pass score) (n 124) | 0.477 |
High adrenaline (n 133) | 0.458 |
Low reverse digit span (n 128) | 0.412 |
High distance vision score on right (n 128) | 0.390 |
Long threshold per cent base length/duration (n 122) | 0.303 |
High dopamine (n 133) | 0.299 |
Low threshold peak middle ear pressure (n 124) | 0.284 |
Low red cell folate (n 133) | 0.283 |
High HPL/creatinine (n 133) | 0.268 |
Low vitamin D (n 132) | 0.244 |
High threshold stapes amplitude projected (n 123) | 0.236 |
Low activated vitamin B6 (n 126) | 0.228 |
Low threshold time to offset/base length (n 122) | 0.205 |
High Serum B12 (n 134) | 0.199 |
High threshold gradient middle ear pressure (n 124) | 0.196 |
High threshold ear canal volume (ECV) (n 123) | 0.189 |
High percent free copper to zinc ratio (n 133) | 0.181 |
Biomarker ROCs | Poor attention rho for alpha 0.05 |
---|---|
Low visual span (n 126) | 0.575 |
Low auditory speed of processing (% of age) (n 121) | 0.551 |
Low visual speed of processing (% of age) (n 122) | 0.546 |
High noradrenaline (n 133) | 0.488 |
High adrenaline (n 133) | 0.473 |
Low competing words score (% of pass score) (n 124) | 0.455 |
Long threshold per cent base length/duration (n 122) | 0.384 |
High HPL/creatinine (n 133) | 0.350 |
Low reverse digit span (n 128) | 0.336 |
Low red cell folate (n 133) | 0.284 |
High threshold stapes amplitude projected (n 123) | 0.266 |
High threshold gradient middle ear pressure (n 124) | 0.261 |
Low threshold time to offset/base length (n 122) | 0.257 |
High distance vision on Right (n 128) | 0.247 |
High dopamine (n 133) | 0.232 |
Low vitamin D (n 132) | 0.227 |
High threshold ear canal volume (ECV) (n 123) | 0.191 |
High Serum B12 (n 134) | 0.190 |
Biomarker ROCs | Hostility rho for alpha 0.05 |
---|---|
Low visual span (n 126) | 0.572 |
High noradrenaline (n 133) | 0.517 |
Low auditory speed of processing (% of age) (n 121) | 0.504 |
Low visual speed of processing (% of age) (n 121) | 0.419 |
Low competing words score (% of pass score) (n 124) | 0.415 |
High adrenaline (n 133) | 0.393 |
Low reverse digit span (n 128) | 0.390 |
High dopamine (n 133) | 0.321 |
Low vitamin D (n 132) | 0.291 |
Low red cell folate (n 133) | 0.284 |
High HPL/creatinine (n 133) | 0.282 |
Low activated vitamin B6 (n 126) | 0.270 |
Low threshold peak middle ear pressure (n 124) | 0.262 |
High threshold gradient middle ear pressure (n 124) | 0.254 |
High distance vision score on right (n 128) | 0.241 |
Low threshold time to offset/base length (n 122) | 0.233 |
Long threshold per cent base length/duration (n 122) | 0.277 |
High threshold stapes amplitude projected (n 123) | 0.195 |
High threshold ear canal volume (ECV) (n 123) | 0.179 |
Biomarker ROCs | Blunted affect rho for alpha 0.05 |
---|---|
Low visual span (n 126) | 0.567 |
Low auditory speed of processing (% of age) (n 121) | 0.548 |
High noradrenaline (n 133) | 0.513 |
Low competing words score (% of pass score) (n 124) | 0.489 |
Low visual speed of processing (% of age) (n 122) | 0.482 |
High adrenaline (n 133) | 0.441 |
Low reverse digit span (n 128) | 0.411 |
Long threshold per cent base length/duration (n 122) | 0.294 |
High distance vision score on right (n 128) | 0.283 |
High HPL/creatinine (n 133) | 0.273 |
High threshold gradient middle ear pressure (n 124) | 0.264 |
Low vitamin D (n 132) | 0.233 |
Low threshold time to offset/base length (n 122) | 0.211 |
High threshold stapes amplitude projected (n 123) | 0.205 |
Low red cell folate (n 133) | 0.193 |
Low activated vitamin B6 (n 126) | 0.187 |
High dopamine (n 133) | 0.182 |
Biomarker ROCs | Self neglect rho for alpha 0.05 |
---|---|
Low visual span (n 126) | 0.509 |
High noradrenaline (n 133) | 0.441 |
High adrenaline (n 133) | 0.419 |
Low visual speed of processing (% of age) (n 121) | 0.411 |
Low competing words score (% of pass score) (n 124) | 0.373 |
Low auditory speed of processing (% of age) (n 121) | 0.350 |
High HPL/creatinine (n 133) | 0.321 |
Low reverse digit span (n 128) | 0.311 |
Long threshold per cent base length/duration (n 122) | 0.269 |
High serum B12 (n 134) | 0.265 |
High distance vision score on right (n 128) | 0.260 |
High dopamine (n 133) | 0.250 |
Low vitamin D (n 132) | 0.241 |
Low threshold time to offset/base length (n 122) | 0.223 |
High threshold gradient middle ear pressure (n 124) | 0.187 |
High threshold stapes amplitude projected (n 123) | 0.187 |
Low red cell folate (n 133) | 0.175 |
Biomarker ROCs | Judgment and insight impairment rho for alpha 0.05 |
---|---|
Low visual speed of processing (% of age) (n 121) | 0.598 |
Low visual span (n 126) | 0.595 |
Low auditory speed of processing (% of age) (n 121) | 0.587 |
High noradrenaline (n 133) | 0.580 |
High adrenaline (n 133) | 0.493 |
Low competing words score (% of pass score) (n 124) | 0.465 |
Low reverse digit span (n 128) | 0.417 |
High distance vision score on right (n 128) | 0.397 |
Long threshold per cent base length/duration (n 122) | 0.327 |
High dopamine (n 133) | 0.324 |
High HPL/creatinine (n 133) | 0.316 |
Low red cell folate (n 133) | 0.276 |
Low threshold time to offset/base length (n 122) | 0.226 |
High threshold gradient middle ear pressure (n 124) | 0.223 |
Low vitamin D (n 132) | 0.222 |
Low threshold peak middle ear pressure (n 124) | 0.213 |
Low activated vitamin B6 (n 126) | 0.210 |
High threshold ear canal volume (ECV) (n 123) | 0.201 |
High percent free copper to zinc ratio (n 133) | 0.193 |
High threshold stapes amplitude projected (n 123) | 0.182 |
Biomarker ROCs | Delusions rho for alpha 0.05 |
---|---|
Low visual speed of processing (% of age) (n 121) | 0.594 |
Low auditory speed of processing (% of age) (n 121) | 0.586 |
Low visual span (n 126) | 0.578 |
Low competing words score (% of pass score) (n 124) | 0.503 |
High noradrenaline (n 133) | 0.491 |
High adrenaline (n 133) | 0.418 |
High distance vision score on right (n 128) | 0.410 |
Low reverse digit span (n 128) | 0.396 |
High dopamine (n 133) | 0.344 |
Long threshold per cent base length/duration (n 122) | 0.325 |
Low red cell folate (n 133) | 0.318 |
High HPL/creatinine (n 133) | 0.258 |
Low vitamin D (n 132) | 0.241 |
Low activated vitamin B6 (n 126) | 0.225 |
High threshold stapes amplitude projected (n 123) | 0.223 |
High serum B12 (n 134) | 0.221 |
Low threshold peak middle ear pressure (n 124) | 0.209 |
Low threshold time to offset/base length (n 122) | 0.197 |
Biomarker ROCs | Anxiety rho for alpha 0.05 |
---|---|
Low visual speed of processing (% of age) (n 121) | 0.497 |
Low visual span (n 126) | 0.491 |
Low auditory speed of processing (% of age) (n 121) | 0.521 |
High noradrenaline (n 133) | 0.454 |
High adrenaline (n 133) | 0.418 |
Low reverse digit span (n 128) | 0.324 |
Low competing words score (% of pass score) (n 124) | 0.330 |
High HPL/creatinine (n 133) | 0.288 |
Low vitamin D (n 132) | 0.248 |
High percent free copper to zinc ratio (n 133) | 0.245 |
High distance vision score on Right (n 128) | 0.249 |
High dopamine (n 133) | 0.234 |
Low red cell folate (n 133) | 0.219 |
Low threshold peak middle ear pressure (n 124) | 0.219 |
High threshold gradient middle ear pressure (n 124) | 0.220 |
Long threshold per cent base length/duration (n 122) | 0.192 |
High threshold stapes amplitude projected (n 123) | 0.183 |
Low threshold time to offset/base length (n 122) | 0.109 |
Biomarker ROCs | Symptom intensity rating (SIR) (severity measure) rho for alpha 0.05 |
---|---|
Low auditory speed of processing (% of age) (n 121) | 0.634 |
Low visual span (n 126) | 0.632 |
Low visual speed of processing (% of age) (n 121) | 0.616 |
High noradrenaline (n 133) | 0.575 |
High adrenaline (n 133) | 0.521 |
Low competing words score (% of pass score) (n 124) | 0.498 |
Low reverse digit span (n 128) | 0.446 |
High distance vision score on right (n 128) | 0.368 |
Long threshold per cent base length/duration (n 122) | 0.365 |
High HPL/creatinine (n 133) | 0.342 |
High dopamine (n 133) | 0.304 |
High threshold gradient middle ear pressure (n 124) | 0.282 |
Low red cell folate (n 133) | 0.278 |
Low vitamin D (n 132) | 0.254 |
High threshold stapes amplitude projected (n 123) | 0.242 |
Low threshold time to offset/base length (n 122) | 0.239 |
High threshold Ear Canal Volume (ECV) (n 123) | 0.226 |
High serum B12 (n 134) | 0.194 |
Low activated vitamin B6 (n 126) | 0.187 |
Biomarker ROCs | Unusual thought content rho for alpha 0.05 |
---|---|
Low auditory speed of processing (% of age) (n 121) | 0.581 |
Low visual speed of processing (% of age) (n 121) | 0.580 |
Low visual span (n 126) | 0.567 |
Low competing words score (% of pass score) (n 124) | 0.456 |
High noradrenaline (n 133) | 0.447 |
High adrenaline (n 133) | 0.386 |
High distance vision score on right (n 128) | 0.363 |
Low reverse digit span (n 128) | 0.351 |
High HPL/creatinine (n 133) | 0.325 |
Long threshold per cent base length/duration (n 122) | 0.297 |
Low red cell folate (n 133) | 0.287 |
High serum B12 (n 134) | 0.249 |
Low threshold peak middle ear pressure (n 124) | 0.240 |
High dopamine (n 133) | 0.237 |
High threshold stapes amplitude projected (n 123) | 0.223 |
High threshold ear canal volume (ECV) (n 123) | 0.194 |
Low vitamin D (n 132) | 0.194 |
High percent free copper to zinc ratio (n 133) | 0.192 |
Low activated vitamin B6 (n 126) | 0.178 |
Biomarker ROCs | Abstract thinking impairment rho for alpha 0.05 |
---|---|
Low auditory speed of processing (% of age) (n 121) | 0.546 |
Low visual speed of processing (% of age) (n 121) | 0.545 |
Low visual span (n 126) | 0.536 |
Low competing words score (% of pass score) (n 124) | 0.505 |
High noradrenaline (n 133) | 0.460 |
High adrenaline (n 133) | 0.405 |
High HPL/creatinine (n 133) | 0.395 |
Long threshold per cent base length/duration (n 122) | 0.367 |
Low reverse digit span (n 128) | 0.319 |
High dopamine (n 133) | 0.272 |
High threshold gradient middle ear pressure (n 124) | 0.267 |
High threshold stapes amplitude projected (n 123) | 0.261 |
High distance vision score on right (n 128) | 0.240 |
Low threshold time to offset/base length (n 122) | 0.236 |
Low red cell folate (n 133) | 0.230 |
High serum B12 (n 134) | 0.219 |
High percent free copper to zinc ratio (n 133) | 0.191 |
Low vitamin D (n 132) | 0.186 |
Biomarker ROCs | Hallucinations rho for alpha 0.05 |
---|---|
Low auditory speed of processing (% of age) (n 121) | 0.545 |
Low visual speed of processing (% of age) (n 121) | 0.502 |
Low visual span (n 126) | 0.473 |
High noradrenaline (n 133) | 0.454 |
High adrenaline (n 133) | 0.397 |
Low competing words score (% of pass score) (n 124) | 0.388 |
Low reverse digit span (n 128) | 0.387 |
High distance vision score on right (n 128) | 0.355 |
Long threshold per cent base length/duration (n 122) | 0.287 |
Low red cell folate (n 133) | 0.272 |
High HPL/creatinine (n 133) | 0.258 |
High threshold stapes amplitude projected (n 123) | 0.251 |
High threshold ear canal volume (ECV) (n 123) | 0.250 |
Low threshold time to offset/base length (n 122) | 0.230 |
High dopamine (n 133) | 0.219 |
Low vitamin D (n 132) | 0.200 |
Low activated vitamin B6 (n 126) | 0.190 |
Biomarker ROCs | Poor impulse control rho for alpha 0.05 |
---|---|
High noradrenaline (n 133) | 0.575 |
Low visual span (n 126) | 0.515 |
High adrenaline (n 133) | 0.485 |
Low competing words score (% of pass score) (n 124) | 0.467 |
Low auditory speed of processing (% of age) (n 121) | 0.451 |
High dopamine (n 133) | 0.432 |
Low visual speed of processing (% of age) (n 121) | 0.402 |
Low reverse digit span (n 128) | 0.362 |
Low vitamin D (n 132) | 0.305 |
High distance vision score on right (n 128) | 0.304 |
Long threshold per cent base length/duration (n 122) | 0.300 |
Low activated vitamin B6 (n 126) | 0.278 |
Low red cell folate (n 133) | 0.273 |
Low threshold time to offset/base length (n 122) | 0.264 |
High HPL/creatinine (n 133) | 0.242 |
High threshold stapes amplitude projected (n 123) | 0.232 |
Biomarker ROCs | Excitement rho for alpha 0.05 |
---|---|
High noradrenaline (n 133) | 0.522 |
Low auditory speed of processing (% of age) (n 121) | 0.493 |
High adrenaline (n 133) | 0.437 |
Low visual speed of processing (% of age) (n 121) | 0.437 |
Low visual span (n 126) | 0.422 |
Low competing words score (% of pass score) (n 124) | 0.382 |
High dopamine (n 133) | 0.366 |
High distance vision score on right (n 128) | 0.357 |
Low reverse digit span (n 128) | 0.326 |
Low red cell folate (n 133) | 0.291 |
Long threshold per cent base length/duration (n 122) | 0.277 |
Low activated vitamin B6 (n 126) | 0.268 |
Low threshold time to offset/base length (n 122) | 0.223 |
High threshold stapes amplitude projected (n 123) | 0.218 |
High threshold ear canal volume (ECV) (n 123) | 0.206 |
Biomarker ROCs | Emotional withdrawal rho for alpha 0.05 |
---|---|
High adrenaline (n 133) | 0.503 |
Low visual speed of processing (% of age) (n 121) | 0.470 |
Low visual span (n 126) | 0.467 |
High noradrenaline (n 133) | 0.432 |
Low competing words score (% of pass score) (n 124) | 0.413 |
Low auditory speed of processing (% of age) (n 121) | 0.407 |
Low reverse digit span (n 128) | 0.374 |
Long threshold per cent base length/duration (n 122) | 0.350 |
High threshold stapes amplitude projected (n 123) | 0.317 |
High HPL/creatinine (n 133) | 0.294 |
Low threshold time to offset/base length (n 122) | 0.277 |
High threshold gradient middle ear pressure (n 124) | 0.271 |
Low vitamin D (n 132) | 0.257 |
High threshold ear canal volume (ECV) (n 123) | 0.227 |
High distance vision score on right (n 128) | 0.222 |
Low red cell folate (n 133) | 0.214 |
Low activated vitamin B6 (n 126) | 0.197 |
No dopamine |
Biomarker ROCs | Suicidality rho for alpha 0.05 |
---|---|
High adrenaline (n 133) | 0.341 |
High noradrenaline (n 133) | 0.308 |
Low visual span (n 126) | 0.264 |
Low auditory speed of processing (% of age) (n 121) | 0.252 |
Low visual speed of processing (% of age) (n 121) | 0.240 |
Low reverse digit span (n 128) | 0.229 |
Low competing words score (% of pass score) (n 124) | 0.225 |
Low vitamin D (n 132) | 0.225 |
High threshold ear canal volume (ECV) (n 123) | 0.222 |
Low red cell folate (n 133) | 0.217 |
High dopamine (n 133) | 0.198 |
Low activated vitamin B6 (n 126) | 0.186 |
When symptoms were segregated according to significant positive and negative correlative strengths for the three principal ROC catecholamines noradrenaline (NA), adrenaline (AD) and dopamine (DA), symptom formation for schizophrenia and schizoaffective disorder fell into seven finely tuned graduated levels of catecholamines.
1) Symptoms with significant correlates for both elevated NA, AD and a lower correlate for DA.
(Severity (SIR) (NA 0.575, AD 0.521) > DA 0.304).
2) Symptoms with significant positive correlates for NA and AD, but low DA correlates.
3) Symptoms with elevated NA (n 133, rho 0.5, P < 0.0001), compared with weaker AD and DA.
4) Symptoms with strongest correlation for adrenaline (AD) in a setting of a weak dopamine correlate.
5) Decreasing correlative strengths for dopamine (DA) n 133, (together with strong NA and AD correlates) and symptoms demonstrating no significant correlates for the dopamine biomarker.
6) Symptoms demonstrating an absence of significant correlates for the dopamine biomarker.
7) Symptoms with significant positive correlates for elevated dopamine biomarker.
8) Symptoms with generalised low or absent correlates for all catecholamines.
These graduated levels of catecholamines are presented in relationship to their biomarker correlates in
Elevated HPL/creatinine levels correlated with high DA ROC (n 131, rho 0.270, P 0.002) and high HIAA ROC (AUC 0.677 at 70% level of significance, n 131, rho 0.200, P 0.022) and high NA ROC (n 131, rho 0.232, P 0.008) and high AD ROC (n 131, rho 0.241, P 0.006). Significant positive correlates were obtained for HPL with clinical severity (SIR index) and a number of negative symptoms: Abstract thinking impairment 0.395, poor attention 0.350, clinical severity (SIR) rho 0.342, cognitive disorganisation 0.332, unusual thought content 0.325, disturbed volition 0.327, lack of spontaneous conversation 0.325, self-neglect 0.321, judgment and insight impairment 0.316, social avoidance 0.314, poor rapport 0.305, emotional withdrawal 0.294. Most notably, HPL held a significant correlation with the symptom of anxiety (n 133, rho 0.245, P 0.004) and marginally-significant correlations with reverse digit span (working memory) deficit (n 127, rho 0.204, P 0.022) and competing words (dichotic listening) deficit (n 123, rho 0.183, P 0.043). Additional marginal correlates were found for high free copper:zinc ratio (n 129, rho 0.171, P 0.053).
Impaired long distance vision (on right) was significantly related to all symptoms with the exception of negative
Catecholamine ROC variables | Symptom correlates (rho) at 95% confidence level. |
---|---|
1) Symptoms with significant correlates for both elevated NA, AD and DA, but DA correlates were lower. | Severity (SIR) (NA 0.575, AD 0.521) > DA 0.304 Delusions (n 133, NA 0.491, AD 0.418) > DA 0.344 |
2) Symptoms with significant positive correlates for NA and AD, but low DA correlates. | Poor attention (NA 0.488, AD 0.473) > DA 0.232 Abstract thinking impairment (NA 0.460, AD 0.405) > DA 0.272 Unusual thought (NA 0.447, AD 0.386) > DA 0.237 Self-neglect (NA 0.441, AD 0.419) > DA 0.250 Anxiety (NA 0.454, AD 0.418) > DA 0.234 Suicidality (AD 0.341, NA 0.308) > DA 0.198 |
3) Elevated NA n 133, rho 0.5, P, 0.0001 compared with weaker AD and DA | Cognitive disorganisation NA 0.583 > (AD 0.495 > DA 0.405) Judgment and Insight impairment NA 0.580 > (NA 0.493 > DA 0.324) SIR 0.575 Poor impulse control NA 0.575 > (AD 0.485 > DA 0.432) Excitement NA 0.552 > (AD 0.437 > DA 0.366) Suspiciousness NA 0.533 > (AD 0.458 > DA 0.299*) Hostility NA 0.517 > (AD 0.373 > DA 0.321) Blunted affect NA 0.513 > (AD 0.441> DA 0.182*) Hallucinations NA 0.454 > (AD 0.397 > DA 0.219*) |
4) Strongest correlative strength is for adrenaline (AD) in a setting of weak dopamine correlate. | SIR 0.521 Emotional withdrawal AD 0.503 > NA 0.432 > DA 0.00* Suicidality AD 0.341, NA 0.308) > DA 0.198* |
5) Decreasing correlative strengths for dopamine (DA) n 133, (in a setting of strong NA and AD correlates demonstrating gradation from excitement to suicidality and emotional withdrawal. | Motor hyperactivity 0.451 Poor impulse control 0.432 Cognitive disorganisation 0.405 Excitement 0.366 Bizarre behaviour 0.363 Grandiosity 0.361 Elated mood 0.385 Hostility 0.321 Self-neglect 0.250 Unusual thought content 0.237 Anxiety 0.234 Judgment and Insight impairment 0.234 Poor attention 0.232 Suspiciousness 0.229 Hallucinations 0.219 Suicidality 0.198 Emotional withdrawal 0.00 |
6) Symptoms demonstrating an absence of significant correlates for the dopamine biomarker. | Emotional withdrawal, blunted affect, depressed mood, disturbed volition, motor retardation, passivity/apathy, symptom severity (SIR), lack of spontaneous conversation, stereotypic thinking, mannerisms and posturing, guilt, dissociative symptoms (experiencing blank periods, unreal feelings, being outside of oneself and other identities within self), ideas of reference and control, distractibility, thought pre-occupation, lack of spontaneous conversation, poor rapport, social avoidance and a history of abuse. |
7) Symptoms with significant positive correlates for elevated dopamine biomarker. | Motor hyperactivity 0.345. Elevated mood 0.258. Grandiosity 0.243. Motor hyperactivity 0.345. |
8) Symptoms with generalised low or absent correlates for all catecholamines. | Experiencing blank periods, (usually considered a dissociative symptom). |
symptoms of depressed mood, disturbed volition, suicide and dissociative symptoms. In contrast, low, significant correlates were largely found for positive symptoms with delusions 0.410, thought preoccupation 0.400, judgment and insight impairment 0.397, suspiciousness 0.390, motor-hyperactivity 0.371, symptom intensity rating (SIR) 0.368, grandiosity 0.368, unusual thought 0.363, excitement 0.357, hallucinations 0.355, cognitive disorganisation 0.341, judgment and insight impairment 0.313, bizzare behaviour 0.311 and poor impulse control 0.304.
Middle ear dysfunction biomarkers demonstrated correlates with schizophrenia and schizoaffective disorderthat included symptoms of disturbed communication:
1) Low attainment of threshold peak middle ear pressure (sensitivity 70%, SIR n 124, rho 0.184, P 0.04), correlated with symptoms of suspiciousness 0.284, hostility 0.262, unusual thought content 0.240, delusions 0.221, thought pre-occupation 0.219, judgment and insight impairment 0.213. This variable also correlated significantly with pre-aged, delayed auditory speed of processing (n 117, rho 0.243, P 0.008) and with elevated noradrenaline levels (n 123, rho 0.220, P 0.014).
2) High external canal volume (ECV) held a biomarker specificity of 82% and for 123 observations, correlated with hallucinations 0.250, emotional withdrawal 0.227, SIR 0.226, depressed mood 0.225, social avoidance 0.244, suicidality 0.222. This variable also correlated specifically with delayed auditory speed of process- ing (n 116, rho 0.284, P 0.02).
3) High threshold gradient middle ear pressure, may be compensative to 1. This biomarker held a ROC specificity of 89% and was significantly correlated with SIR (n 124, 0.282, P 0.005) and with symptoms of poor rapport 0.411, lack of spontaneous conversation 0.282, suspiciousness 0.284, hostility 0.262 and unusual thought 0.240.
4) There was a tendency for increased strength and duration of stapes sound-dampening, acoustic reflex (possibly compensative to 1 and 2), to be also related to negative symptoms:
a) Long % base-length compared with total duration of the maximal contraction of the stapes muscle at threshold is indicative of a strong, long-duration stapes contraction. This biomarker held a sensitivity of 77% for 122 observations and correlated positively with poor attention 0.384, delusions 0.325, suspiciousness 0.303, poor impulse control 0.342, anxiety 0.341, abstract thinking impairment 0.367, SIR 0.365, distractibility 0.331, lack of spontaneous conversation 0.339, stereotypic thinking 0.336, judgment and insight impairment 0.327, poor impulse control 0.303, disturbed volition 0.301, social avoidance 0.301. This variable representing increased duration of stapes contraction also significantly correlated with elevated levels of both noradrenaline (n 121, rho 0.212, P 0.020) and adrenaline (n 121, rho 0.264, P 0.003), respectively.
b) High stapes projected amplitude at threshold contraction response relates to increased strength of stapes contraction as it dampens sound by landing on the oval window of the cochlea. This biomarker was also significantly correlated with SIR (n 123, rho 0.242, P 0.009, and with symptoms of emotional withdrawal (n 123, rho 0.317, P 0.001) and stereotypic thinking (n 123, rho 0.290, P 0.009).
c) Low threshold time to offset/base length (n 122) indicates a protracted stapes contraction, that also correlated significantly emotional withdrawal 0.277 and with similar symptoms to a), such as distractibility 0.268, poor impulse control 0.264, poor rapport 0.259, poor attention 0.257, disturbed volition 0.251, anxiety 0.243, social avoidance 0.241, abstract thinking impairment 0.236 and SIR 0.239. This variable also came close to significant correlation with delayed-for-age auditory processing speed (n 115, rho 0.161, P 0.071).
Nutritional-Biochemical variables significantly correlated with neurotransmitter ROC variables for high dopamine (n 124, rho 0.301, P 0.001), high noradrenaline (n 124, rho 0.381, P < 0.0001 and high adrenaline (n 124 rho 0.302, P < 0.0001). In this setting, low, significant positive correlations for symptoms in relationship to biochemical and nutritional biomarkers were found at 95% confidence level. These included:
1) Elevated vitamin B12 ROC n 134, which correlated with symptoms of cognitive disorganisation 0.283, self-neglect 0.265, abstract thinking impairment 0.219, bizarre behaviour 0.251, unusual thought content 0.249, but had a low correlate with severity (SIR 0.194).
2) Low vitamin D ROC (n 132), which correlated with symptoms of poor rapport 0.349, uncooperativeness 0.315, poor impulse control 0.305, social avoidance 0.300, hostility 0.291 and SIR 0.254.
3) Low red cell folate (n 133), which correlated with symptoms of delusions 0.318, excitement 0.291, uncooperativeness 0.290, SIR 0.278.
4) Low activated B6 (n 126), which correlated with poor impulse control 0.278, hostility 0.270, excitement 0. 268, suspiciousness 0.228, uncooperativeness 0. 242, grandiosity 0.206 and SIR 0.187.
5) High percent free copper: zinc ratio biomarker (n 133), which marginally correlated with symptoms of suspiciousness (rho 0.181, P 0.038), abstract thinking impairment (rho 0.191, P 0.028), anxiety (rho 0.228, P 0.008), unusual thought (rho 0.192, P 0.027), judgment and insight impairment (rho 0.193, P 0.026), thought preoccupation (rho 0.188, P 0.030), distractibility (rho 0.192, P 0.027) and SIR (rho 0.179, P 0.040).
This research brings together ROC-defined biomarkers for schizophrenia and schizoaffective psychosis and investigates their correlation profiles in relation to discrete symptoms of these conditions.
The study indicates that finely-tuned relative-strengths between noradrenaline, adrenaline and dopamine (NA, AD and DA) levels are critical in differentiating symptom-formation in schizophrenia and schizoaffective dis- order. The graduated difference between the elevated NA biomarker and the level of DA appears to turn symp- tom correlates from manic (with high DA) to disorganized and distracted (high NA and somewhat lower AD and DA), to depressive and negative (low DA ± high NA, or low DA alone), to dissociative symptoms such as ex- periencing “blank periods” that are characterised by generally low or absent catecholamine correlates. DA gives high frequency conditioning to glutaminergic synapses in order to potentiate goal-directed sensory information flow around the cortex, whilst high NA raises arousal and vigilance and AD promotes flight [
Elevated noradrenaline’s ability to disrupt DA’s promotion of the sensory signal around the cortex from frontal to temporal-hippocampal areas, fits well with the findings of this study where high NA correlates with symp- toms of poor attention, disorganisation, self-neglect and delusions, provide residual evidence of front-hippo- campal disconnectivity. At the same time noradrenergic fronto-parietal sensory signal disruption [
Urinary HPL is thought to arise under conditions of extreme oxidative stress where it is excreted as a by- product of haeme-related porphyrin synthesis [
In keeping with their various influences on catecholamine biochemistry (
Evidence of significant correlative relationships between elevated catecholamines and auditory and visual processing speed and working memory deficits are postulated to occur via a vicious cycle of catecholamine excess, whereby NA is progressively elevated until fronto-temporal disconnectivity occurs [
qualify as biomarkers, which inter-correlate with each other, with elevated catecholamines, with delay in both auditory and visual speed of processing and also with clinical symptom severity within the research patient population.
In keeping with this noradrenergic fronto-temporal disconnection hypothesis, low visual span may represent ramped-up noradrenaline-induced disconnectivity, reducing the action of the higher visual cortical magnocellular pathway and forcing reliance on the lower parvocellular visual pathway within the default brain network [
Middle ear ROC reflex findings of elevated tympanic contraction gradient and increased volume of the external auditory canal―yet collateral attainment of low middle ear peak pressure―indicate that middle ear patency problems associated with tympanic membrane pathology or Eustachian tube dysfunction are an integral part of schizophrenia and schizoaffective disorder [
In order to exclude confounding factors and isolate the functional core of schizophrenia, many exclusion criteria were applied to sample selection. This reduced available participants resulting in a small discovery data-set. Small sample size together with non-normal data distribution for many variables precluded use of principal component, multivariate and cluster analysis.
Though case-control studies are suitable for discovery projects for low prevalence disorders such as schizophrenia, their cross-sectional design has inherent susceptibility to prevalence and selection bias [
Although most ROC variables identified in our study achieved a robust odds ratio of much over 2, some statisticians claim that odds of association results may be inflated by up to three times with a case-control design. [
The application of this design to only one bracket of psychotic disorders (schizophrenia and schizoaffective disorder) in this project, limits understanding of disease specificity across other mental illness states. A large cohort study with fully-blinded investigators or a prospective, multi-site-clinic trial on a single series consecutive patients with an emphasis on collecting symptomatic data from ultra-high risk medication naïve subjects, will therefore be required to validate these findings.
Though elevated urinary monoamine biomarkers identified in this study related to psychiatric symptoms in understandable ways, fully-synthesized monoamines do not cross the blood-brain barrier even though their precursor substance, L-Dopa, is capable of this transition [
Although rigorous efforts were made to exclude substance-related diagnoses, it is known that alcohol misuse is often under-reported by patients [
This research occurred with antipsychotic medicated patients who met strict criteria for such medication.
Nevertheless, the effect of medication on catecholamine results cannot be fully excluded as a confounding factor. In support of the fact that medications have not influenced these catecholamine results is the fact that there is widespread research evidence that antipsychotic medications increase catecholamine metabolic turnover and reduce dopamine and noradrenaline levels [
Though it is expected that hundreds of biomarkers for schizophrenia will eventually be discovered, this project provides an advanced understanding of psychosis by correlating the mood, perceptual and behavioural symptoms of schizophrenia and schizoaffective disorder with twenty-one quantified biomarkers.
Examining symptoms in relationship to biomarkers reveals that schizophrenic and schizoaffective disorder symptoms consist of a conglomerate of biochemical and neurophysiological dysfunctions related to dysregulation of visual and auditory sensory processing and neurochemistry relating to the biosynthesis and metabolism of catecholamines. Through the remote and direct action of cofactors associated with catecholamine synthesis and metabolism, noradrenaline levels are generally elevated which delays and disrupts cortical visual and auditory sensory processing pathways. This in turn activates the hypothalamic pituitary adrenal axis to produce further catecholamines. In this process, elevated free copper facilitates noradrenaline synthesis at the expense of dopamine and also promotes oxidative stress, whilst vitamin B6 over-utilisation reflects HPA pressure for dopamine synthesis and such lack also accentuates oxidative stress. Low folate remotely influences the ability of catecholamines to be metabolized by COMT via its cofactor, SAMe. Under the variable influence of these cofactors, altering ratios between noradrenaline, adrenaline and dopamine correlate meaningfully with mood, attention and behavioural effects of these catecholamines. In this setting, a predominance of very high-noradrena- line causes fronto-temporal disconnectivity resulting in symptoms of delayed, disrupted and disorganised visual and auditory attention, mixed with frontal disconnection and disinhibition symptoms of poor judgement, poor executive function, poor working memory and poor impulse control. Isolated temporal regions are then left to function alone with impaired insight and delusions, whilst the isolated amygdala within the unmasked default network manifests as symptoms of agitation, aggression and hostility.
This research provides a new understanding of the substructure of many symptoms within the context of psychosis. In particular, biomarkers correlate and combine to underlie symptoms and behaviours of key relevance within the clinical setting, such as suicidality, hostility and auditory hallucinations.
In this context, suicidality is seen to relate to high adrenaline levels representing the flight impulse, together with low-dopamine-related symptom of emotional withdrawal. In addition high noradrenaline levels convey the capacity to act-out on the basis of these impulses and symptoms, in the setting of a psychotic break in sensory processing with dichotic listening deficit, visual and auditory processing deficits, middle ear pathology and deficiency of all three nutrients―D, B6 and folate.
In the context of predicting and preventing aggression in psychosis, the biomarker profile of hostile, aggressive behaviour is of particular interest. Hostility is seen to relate to high noradrenaline levels and has a strong correlate with the symptom of blunted affect. Hostility also relates to impaired visual span and auditory processing difficulty and has underlying pathology relating to vitamin D deficiency and middle ear pathology. Such findings may be helpful for prediction of hostility in a clinical or forensic context.
Auditory hallucinations are seen to relate to high levels of both noradrenaline and adrenaline with delays in both auditory and visual processing speed, together with reduced visual span. In addition, dichotic listening deficit, middle ear pathology, low folate, low vitamin D and B6 contribute to the constellation of disorders and dysfunctions that subtly accumulate to produce this distressing symptom.
Anxiety is also seen to relate to elevated noradrenaline and adrenaline levels, with impairment of visual span and visual processing speeds to a greater extent than delayed auditory processing. In addition, auditory working memory and dichotic listening deficits impair sensory processing. Underlying this dysfunction resides peripheral abnormalities of long distance vision and the middle ear function, in addition to oxidative stress and low vitamin D and folate.
Biomarkers of acoustic reflexes relating to middle ear pathology, distance vision deficits, dichotic listening disorder and biochemical-nutritional deficits generally represent unmet needs within the clinical population assessed. This implies that there is scope for broader clinical assessment and remediation of persons with schizophrenia and schizoaffective disorder.
Biomarkers that correlate with discrete symptoms or sets of symptoms have therapeutic potential for targeted correction towards cure. They inform about the structure of psychosis, symptom-formation and the patterns in which symptoms present, allowing clinicians greater confidence in diagnosis and management of discrete, resistant, difficult or dangerous symptoms. Such multi-domain understandings provide a template for new biological system of symptom-interpretation in serious mental illness states. In the wider social context, the demonstration of clear biological underpinnings for schizophrenia and schizoaffective psychosis will reduce social stigma associated with these conditions and improve outcome expectations for patients and their families.
The authors acknowledge the additional support of Ms. Natasha J. Radcliffe (MHumNutr, GradDipSocSci, GradCertDisStud, BA (Sociology)) who undertook control stratification and enrolment with analysis of variable distributions, proxy standard errors, ROC analysis, odds ratio analysis and Spearman’s correlation analysis. Statistician Graeme Tucker (B.Sc.) also conducted final review statistical analysis. Queen Elizabeth Hospital academic and secretarial staff, nursing staff, Registrars and Consultants. Dr Geoffrey Schrader MBBS, M Clin Sc, PhD, FRANZCP and Dr Helen Tingay FRANZCP. Western Mental Health Service, Janet Grant (Population Research and Outcome Studies (PROS) and North West Adelaide Health Study assistance with participant selection. Laboratory support from Dr. Michael Metz BS, MD, FAAP, MAACB, FRCPA at Clinpath Laboratories and Dianne Zercher (TQEH Biochem Laboratory) and Dr Malcolm Whiting at SA Pathology. General support from Professor Robert Adams Head of Department of Medicine TQEH, Professor Richard Ruffin AM, Judith Snowden, Helen Goldsack, Dr Marilyn Dyson MBBS, MHP, DRCOG, Dr Sinclair Bode MB BS FACNEM, Veronica Steer B AppSc (OT), Jan Pollard (technical audiology support), Dr N. Williams (Ophthalmologist FRANZCO). Graphic support Adlab SA.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
A provisional patent application was filed by the author at the conclusion of data-collection and initial analysis, in December 2013.
As Chief Researcher, Dr Stephanie Fryar-Williams MB BS. BSc (Biochem/Pharmacol) FRANZCP is an Honorary Research Fellow of the University of Adelaide. Dr Fryar-Williams conceived the project, selected the candidate markers and designed the research protocols, oriented staff and raters in ward and clinic settings, managed and supervised the recruitment of case and control participants, prepared the data-set for analysis by data-transformation, supervised ROC and other statistical analysis, directed and interpreted data analysis within biochemical, neurological and psychiatric theory and wrote the paper.
As a previous Clinical Director of Psychiatry at the Queen Elizabeth Hospital, Woodville SA., Dr Jorg Strobel MD (Bonn). Spec Psych, Psychotherapy FRANZCP facilitated the laboratory, ward and community organisation and coordination components of the project and revised past and final papers for publication.
ROC variables and domains | No. Obs | AUC | Odds ratio | Odds ratio P | ROC variables and domains | No. Obs | AUC | Odds ratio | Odds ratio P |
---|---|---|---|---|---|---|---|---|---|
Low visual span | 126 | 0.862 | 22.46 | <0.0001 | High dopamine | 133 | 0.702 | 9.60 | <0.0001 |
High visual speed of processing discrepancy (% of age) | 122 | 0.875 | 27.22 | <0.0001 | High noradrenaline | 133 | 0.851 | 21.25 | <0.0001 |
Poor distance vision on right | 128 | 0.597 | 5.17 | 0.0001 | High adrenaline | 133 | 0.844 | 14.32 | <0.0001 |
Visual domain | 120 | 0.915 | 41.48 | <0.0001 | Catecholamine domain | 133 | 0.859 | 4.12 | <0.0001 |
Reverse digit span | 126 | 0.862 | 11.00 | <0.0001 | |||||
High competing words discrepancy (% of pass score) | 124 | 0.875 | 10.69 | <0.0001 | Low vitamin B6 activated | 129 | 0.638 | 3.75 | 0.0009 |
High auditory speed of processing discrepancy (% for age) | 121 | 0.874 | 21.23 | <0.0001 | High free copper to zinc ratio | 133 | 0.611 | 2.60 | 0.0104 |
Auditory domain | 119 | 0.891 | 29.57 | <0.0001 | Low red cell folate | 133 | 0.654 | 3.64 | 0.0005 |
Low vitamin D | 132 | 0.651 | 3.24 | 0.0026 | |||||
High (HPL/creatinine) Model | 133 | 0.696 | 4.12 | <0.0001 | High serum B12 (80% CI) | 134 | 0.565 | 1.89 | 0.0933 |
Oxidative stress domain | 133 | 0.696 | 4.12 | <0.0001 | Nutrition-biochemistry | 126 | 0.797 | 8.5 | <0.0001 |
ROC variable | No Obs | Obs. without missing data | Minimum | Maximum | Mean | SD | Median | MAD mean absolute deviation | Fitted distribution |
---|---|---|---|---|---|---|---|---|---|
Visual domain | |||||||||
Visual span | 134 | 126 | 0.00 | 8.00 | 5.476 | 1.35 | 6.00 | 1.00 | Pert |
Visual speed of processing discrepancy (% of age) | 134 | 122 | −90.00 | 207.69 | 6.01 | 54.44 | −5.55 | 30.24 | Log normal |
Distance vision on right | 134 | 128 | 0.00 | 36.00 | 7.98 | 6.03 | 6.00 | 1.50 | Chi-square |
Auditory domain | |||||||||
Reverse digit span | 134 | 127 | 2.00 | 8.00 | 4.22 | 1.33 | 4.00 | 1.00 | Gamma (3P) |
Competing words discrepancy (% of pass score) | 134 | 124 | −69.23 | 50.00 | 0.55 | 22.61 | 3.84 | 15.38 | Johnson SB |
Auditory speed of processing discrepancy (% of age) | 134 | 121 | −100.00 | 220.00 | −3.83 | 56.68 | −18.00 | 31.12 | Gen. extreme value |
Catecholamine domain | |||||||||
Dopamine | 134 | 133 | 45.00 | 358.00 | 142.47 | 53.64 | 129.0 | 32.00 | Log-logistic |
Noradrenaline | 134 | 133 | 3.00 | 106.00 | 25.278 | 18.53 | 19.00 | 9.00 | Johnson SB |
Adrenaline | 134 | 133 | 0.00 | 27.00 | 4.413 | 5.10 | 2.00 | 1.00 | Log-logistic |
HPL/creatinine model | 134 | 133 | 0.35 | 40.04 | 4.586 | 5.94 | 2.47 | 1.18 | Burr |
---|---|---|---|---|---|---|---|---|---|
Nutrition-biochemistry domain | |||||||||
Free copper to Zinc ratio | 134 | 133 | −1.85 | 1.60 | 0.267 | 0.52 | 0.31 | 0.31 | Log-logistic |
B6 activation | 134 | 129 | 12.80 | 1570.0 | 140.44 | 164.6 | 90.00 | 25.00 | Gen. extreme value |
Red cell folate | 134 | 133 | 506.00 | 3291.0 | 1788.9 | 448.7 | 1733. | 236.00 | Log-logistic |
Serum B12 | 134 | 134 | 42.00 | 1388.0 | 406.15 | 178.5 | 367.0 | 104.00 | Gen. extreme value |
Vitamin D | 134 | 132 | 13.00 | 149.00 | 52.462 | 22.15 | 52.00 | 14.00 | Dagum |