The Contribution of Explainable Machine Learning Algorithms Using ROI-based Brain Surface Morphology Parameters in Distinguishing Early-onset Schizophrenia From Bipolar Disorder

RATIONALE AND OBJECTIVES

To differentiate early-onset schizophrenia (EOS) from early-onset bipolar disorder (EBD) using surface-based morphometry measurements and brain volumes using machine learning (ML) algorithms.

METHOD

High-resolution T1-weighted images were obtained to measure cortical thickness (CT), gyrification, gyrification index (GI), sulcal depth (SD), fractal dimension (FD), and brain volumes. After the feature selection step, ML classifiers were applied for each feature set and the combination of them. The SHapley Additive exPlanations (SHAP) technique was implemented to interpret the contribution of each feature.

FINDINGS

144 adolescents (16.2 ± 1.4 years, female=39%) with EOS (n = 81) and EBD (n = 63) were included. The Adaptive Boosting (AdaBoost) algorithm had the highest accuracy (82.75%) in the whole dataset that includes all variables from Destrieux atlas. The best-performing algorithms were K-nearest neighbors (KNN) for FD subset, support vector machine (SVM) for SD subset, and AdaBoost for GI subset. The KNN algorithm had the highest accuracy (accuracy=79.31%) in the whole dataset from the Desikan-Killiany-Tourville atlas.

CONCLUSION

This study demonstrates the use of ML in the differential diagnosis of EOS and EBD using surface-based morphometry measurements. Future studies could focus on multicenter data for the validation of these results.

The association between plasma thyroxine levels and neurocognitive impairment in early-onset schizophrenia and other psychosis spectrum disorders

BACKGROUND/AIM

Limited studies have delved into the association between thyroid hormones and neurocognition in schizophrenia. We aimed to evaluate the relationship between thyroid hormone levels and neurocognitive functions in patients with schizophrenia and other psychosis spectrum disorders (SSD).

METHOD

A total of 135 patients with early-onset SSD were included in the study. The participants underwent a cognitive assessment. Blood samples were collected to measure serum levels of thyroid-stimulating hormone (TSH), free thyroxine (fT4), and free triiodothyronine (fT3). Subgroup analyses were conducted based on the severity of the psychosis.

FINDINGS

The results revealed a significant association between fT4 levels and various cognitive domains, including processing speed, verbal fluency, working memory, verbal learning, verbal memory, and visual memory. However, serum TSH and fT3 levels exhibited no significant association with neurocognitive impairment in adjusted linear regression models. Specifically, the correlation between fT4 levels and global cognition was more pronounced in patients with higher scores.

CONCLUSIONS

Serum fT4 levels were associated with the performance across various cognitive domains in cases of early-onset psychotic disorders. This correlation was accentuated among patients with higher illness severity. Future studies could focus on the effects of specific pathways that can affect the course and progression of psychosis.

Social cognition and neurocognition in first-episode bipolar disorder and psychosis: The effect of negative and attenuated positive symptoms

BACKGROUND

Schizophrenia and bipolar disorder are associated with neurocognitive and social-cognitive impairments. To date very few studies investigated social cognition in first-episode bipolar disorder (FEBD). Our main aim was to investigate the differences in social cognition and neurocognition between FEBD and first-episode psychosis (FEP). Another aim was to investigate neurocognitive correlates of negative symptoms and attenuated psychotic symptoms in FEBD.

METHODS

This study included 55 FEBD, 64 FEP and 43 healthy controls. A comprehensive neuropsychological battery assessing social cognition, processing speed, verbal and visual memory, working memory, sustained attention, and executive functions was administered to all participants.

RESULTS

Both FEBD and FEP were associated with widespread deficits in all neurocognitive domains and social cognition. Both FEP (d = -1.19) and FEBP (d = -0.88) were also impaired in social cognition. In FEP, effect sizes (Cohen's d) of neurocognitive deficits ranged from -0.71 to -1.56. FEBD was also associated with relatively milder but similar neurocognitive deficits (d = -0.61 to-1.17). FEBD group performed significantly better than FEP group in verbal and visual memory, processing speed, and executive function domains (d = -0.40 to-0.52). Negative symptoms and social functioning were associated with neuropsychological impairment in both groups. The severity of attenuated psychotic symptoms was associated with poorer verbal memory in FEBD (r = -0.39, p < 0.01).

LIMITATIONS

The cross-sectional nature of the current study is the main limitation.

CONCLUSIONS

Neurocognitive and social-cognitive deficits are evident in both FEBD and FEP. In FEBD, more severe memory deficits might be markers of clinical overlap and shared neurobiological vulnerability with psychotic disorders.

Neurocognition and social cognition in youth and young adults at ultra-high-risk for psychosis and bipolar disorder

BACKGROUND

Schizophrenia and bipolar disorder are associated with significant deficits in neurocognition and social cognition. Unlike the studies in chronic stages of these disorders, very limited information is available regarding neurocognitive and social-cognitive impairment before the onset of bipolar disorder. Our main aim was to investigate the differences in neurocognition and social cognition between individuals at ultra-high risk for psychosis (UHR-P) and bipolar disorder (UHR-BD).

METHODS

This study included 152 help-seeking individuals identified as UHR-P (n = 78) and UHR-BD (n = 74), who were compared with a healthy control group (n = 43). A comprehensive neuropsychological battery was administered to all participants.

RESULTS

UHR-P was associated with widespread deficits in all neurocognitive and social-cognitive domains. Effect sizes (Cohen's d) of these deficits ranged from -0.57 to -1.34. UHR-BD was associated with significant deficits in processing speed, executive functions, sustained attention and social cognition (d = -0.48 to-0.70, p < 0.05). UHR-P performed significantly worse than UHR-BD in social cognition, processing speed, verbal memory and executive function domains (d = -0.39 to-0.64, p < 0.05). Negative symptoms were associated with impaired social cognition in the UHR-P group and verbal memory deficits in the UHR-BD group. Cognitive impairment was associated with functional impairment in both groups.

CONCLUSIONS

While UHR-P is associated with more widespread cognitive impairment, deficits in processing speed, executive functions, sustained attention and social cognition might be common features of both UHR groups. In early intervention services, cognition should be considered as a target for assessment and intervention not only for individuals at high risk for psychosis but also for bipolar disorder.

Can diffusion tensor imaging have a diagnostic utility to differentiate early-onset forms of bipolar disorder and schizophrenia: A neuroimaging study with explainable machine learning algorithms

BACKGROUND/AIM

Accurate diagnosis of early-onset psychotic disorders is crucial to improve clinical outcomes. This study aimed to differentiate patients with early-onset schizophrenia (EOS) from early-onset bipolar disorder (EBD) with machine learning (ML) algorithms using white matter tracts (WMT).

METHOD

Diffusion tensor imaging was obtained from adolescents with either EOS (n = 43) or EBD (n = 32). Global probabilistic tractography using an automated tract-based TRACULA software was performed to analyze the fractional anisotropy (FA) of forty-two WMT. The nested cross-validation was performed in feature selection and model construction. EXtreme Gradient Boosting (XGBoost) was applied to select the features that can give the best performance in the ML model. The interpretability of the model was explored with the SHApley Additive exPlanations (SHAP).

FINDINGS

The XGBoost algorithm identified nine out of the 42 major WMTs with significant predictive power. Among ML models, Support Vector Machine-Linear showed the best performance. Higher SHAP values of left acoustic radiation, bilateral anterior thalamic radiation, and the corpus callosum were associated with a higher likelihood of EOS.

CONCLUSIONS

Our findings suggested that ML models based on the FA values of major WMT reconstructed by global probabilistic tractography can unveil hidden microstructural aberrations to distinguish EOS from EBD.

Electroconvulsive therapy or clozapine for adolescents with treatment-resistant schizophrenia: an explorative analysis on symptom dimensions

OBJECTIVE

This study sought to compare pre-intervention patient characteristics and post-intervention outcomes in a naturalistic sample of adolescent inpatients with treatment-resistant psychotic symptoms who received either electroconvulsive therapy (ECT) or clozapine.

METHODS

Data of adolescents with schizophrenia/schizoaffective disorder receiving ECT or clozapine were retrospectively collected from two tertiary-care psychiatry-teaching university hospitals. Subscale scores of the Positive and Negative Symptom Scale (PANSS) factors were calculated according to the five-factor solution. Baseline demographics, illness characteristics, and post-intervention outcomes were compared.

RESULTS

There was no significant difference between patients receiving ECT (n = 13) and clozapine (n = 66) in terms of age, sex, and the duration of hospital stay. The ECT group more commonly had higher overall illness and aggression severity. Smoking was less frequent in the clozapine group. Baseline resistance/excitement symptom severity was significantly higher in the ECT group, while positive, negative, affect, disorganisation, and total symptom scores were not. Both interventions provided a significant reduction in PANSS scores with large effect sizes.

CONCLUSION

Both ECT and clozapine yielded high effectiveness rates in adolescents with treatment-resistant schizophrenia/schizoaffective disorder. Youth receiving ECT were generally more activated than those who received clozapine.

Development and validation of Kiddie Schedule for Affective Disorders and Schizophrenia (K-SADS) for Internet Gaming Disorder (IGD) and factor analytic assessment

OBJECTIVE

To develop and validate Kiddie Schedule for Affective Disorders and Schizophrenia, Present and Lifetime version (K-SADS-PL) for Internet Gaming Disorder (IGD) in adolescents.

METHODS

Questions and threshold criteria of the K-SADS-IGD was generated based on the related section of K-SADS-PL. Then, the sample consist of IGD group and matched control group with no significant difference in psychiatric comorbidities from clinical settings were included to assess the psychometric properties of the K-SADS-IGD. Exploratory and Confirmatory Factor analysis were conducted to evaluate and compare DSM model of IGD and two different Models of IGD proposal in adolescents.

RESULTS

Exploratory Factor Analysis of K-SADS-IGD revealed a single factor explaining 61.469% of the total variance. Confirmatory Factor Analysis indicates that although the K-SADS-IGD model fit indices were also acceptable, Model 1, which excluded the 7th criterion of IGD criteria of DSM-5 showed better fit in adolescent population. The Likelihood Ratio Positive and the Likelihood Ratio Negative estimates for the diagnosis of K-SADS-IGD were 31.4 and 0.12, respectively, suggesting that K-SADS-IGD was beneficial for determining the presence and the absence of IGD in adolescents. Also, K-SADS-IGD could detect disordered gamers with significantly low functionality (even after controlling the impact of comorbidities) from non-disordered gamers.

CONCLUSION

K-SADS-IGD was found to be a reliable and valid instrument in adolescents. The model excluding 7th criteria of DSM-5 IGD was found to be more consistent than the current DSM-5 IGD model in the adolescent population. Therefore, the diagnostic criteria might be required to adjust according to the age group since the clinical symptomatology of IGD in adolescents may differ from that in adults. The K-SADS-IGD may meet the need for a certain and standardized tool to assess IGD in this population.

INTERNET GAMING DISORDER AND PROBLEMATIC INTERNET USE: THE LINKS BETWEEN CYBERBULLYING, AGGRESSION, LONELINESS AND PSYCHIATRIC COMORBIDITIES

AIMS

To compare adolescents clinically diagnosed with Internet Gaming Disorder (IGD) and problematic internet use (PIU) in terms of cyberbullying, aggression, and loneliness.

METHODS

Male adolescent patients (N=124, 14.3±1.7 years) with Internet Addiction Scale (IAS) scores ≥50 were clinically interviewed for IGD in utilizing DSM-5 criteria. Patients without full IGD criteria were included as PIU comparisons. Clinical variables were assessed using the second version of the Revised Cyber Bullying Inventory, short-form of the UCLA Loneliness Scale, Buss Perry Aggression Questionnaire, Child Depression Inventory, and Screen for Child Anxiety Related Emotional Disorders.

RESULTS

Compared to individuals with PIU, those with IGD were significantly more likely to have attention-deficit hyperactivity disorder, higher social phobia scores, higher cyberbullying scores, higher loneliness scores, been a cyberbully, and been a cyberbully victim.

CONCLUSION

Male adolescents with IGD have higher rates of psychiatric comorbidity, perceived loneliness, cyberbullying, and being a victim of cyberbullying than those with PIU. Future studies could evaluate these predictors of transition from PIU to IGD in large cohort samples.

Parent rated bedtime resistance and comorbidity may predict levels of attention among Turkish children diagnosed with ADHD in on-line education classes during the COVID-19 outbreak

OBJECTIVE

This study aimed to compare the attention levels, of Turkish children and adolescents with Attention Deficit/ Hyperactivity Disorder (ADHD) in on-line education classes with healthy controls.

METHODS

This study is a cross-sectional, internet-based, case-control study that recruited 6-18 years old patients diagnosed with ADHD and receving treatment and healthy controls from eight centers. The measurements used in the study were prepared in the google survey and delivered to the participants via Whatsapp application.

RESULTS

Within the study period, 510 children with ADHD and 893 controls were enrolled. Parent- rated attention decreased significantly in both groups during on-line education classes due to COVID-19 outbreak (p<0.001; for each). Children and adolescents with ADHD had significantly elevated bedtime resistance, problems in family functioning difficulties than control children according to parental reports (p=0.003; p<0.001; p<0.001, respectively). Furthermore, bedtime resistance and comorbidity significantly predicted attention levels in on-line education.

CONCLUSION

Our findings may underline the need to augment student engagement in on-line education both for children without attention problems and those with ADHD. Interventions shown to be effective in the management of sleep difficulties in children as well as parent management interventions should continue during on-line education.

Investigating associations between appetite-regulating hormones, aggression and craving in males with cannabis use disorder

BACKGROUND

Aggression and craving are common and important withdrawal symptoms in cannabis use disorder. The present study investigated the association between appetite-regulating hormones, aggression, and craving during cannabis withdrawal syndrome (CWS).

METHODS

Fifty-six male subjects diagnosed with cannabis withdrawal and 45 healthy males were included in the study. The Substance Craving Scale, the Buss-Perry Aggression Questionnaire, and the State-Trait Anxiety Inventory were implemented at baseline. Blood samples were drawn to measure ghrelin, leptin, adiponectin, and resistin levels in the serum. Then, the Point Subtraction Aggression Paradigm (PSAP) was applied. Bloodwork and psychometric assessment procedures were re-implemented after the PSAP. At the 7-day follow-up, psychometric assessments and hormone measurements were repeated in the CWS group.

RESULTS

Baseline serum ghrelin and adiponectin levels were lower in the CWS group than controls at baseline. After PSAP, there was a significant increase in ghrelin levels of patients with CWS compared to controls. Patients yielded higher aggression scores, while there was no significant correlation between hormonal changes and PSAP findings. At 7-day follow, ghrelin and resistin levels significantly increased, while serum leptin decreased in patients with CWS. Finally, there was a positive association between craving and resistin levels.

CONCLUSIONS

Our results present the changes in appetite-regulating hormones. Long-term follow-up studies are needed to shed light on neuroendocrinological aspects of cannabis withdrawal.

Investigating the effects of age, IQ, dosing, and anthropometric measures on the treatment persistence in long-term methylphenidate use

OBJECTIVES

This study aimed to determine anthropometric and clinical correlates of persistence to methylphenidate (MPH) treatment in Turkish youth with attention-deficit hyperactivity disorder (ADHD).

METHODS

Data from medical records of 518 children and adolescents with ADHD were recorded between March 2012 and January 2022. Clinical variables of patients persistent to MPH ≥ 2 years were compared with those of the non-persistent group. Children and adolescent age groups were compared using Kaplan-Meier estimates for treatment drop-outs. Cox regression analysis until the treatment drop-out was implemented to calculate hazard ratios (HRs) for gender, age, full-scale IQ, and anthropometric measures. Weight, height, and body mass index (BMI) z-scores were calculated per national guidelines.

RESULTS

Persistent and non-persistent study groups had similar full-scale IQ, weight, height, and BMI z-scores at treatment onset. The mean MPH dose was significantly higher in the persistent group compared to the non-persistent counterparts (31.43 ± 10.70 vs. 24.28 ± 9.60 mg/d, p < 0.001, d = 0.70). Compared to children, the adolescents showed earlier treatment drop-outs in males (p < 0.001) but not in females (p = 0.110). Younger age showed a positive effect on treatment persistence. Conversely, baseline BMI and IQ scores were not associated with long-term persistence.

DISCUSSION

Our study demonstrated lower daily doses and older age-onset were associated with early drop-outs in MPH treatment. These findings supported the notion that effective dosing strategies at younger ages could increase the sustainability of the treatment with MPH in the Turkish population.

The effects of catechol-O-methyltransferase single nucleotide polymorphisms on positive and negative symptoms of schizophrenia: A systematic review and meta-analysis

The catechol-O-methyltransferase (COMT) gene is thought to have an important role in the etiopathogenesis of schizophrenia, but there are conflicting results regarding its role in clinical presentation. We aimed to elucidate the relationship between the single nucleotide polymorphisms (SNPs) in the COMT gene and the severity of positive and negative symptoms. In order to investigate the relationship, the PubMed, PubMed Central, Scopus, and Cochrane CENTRAL databases were screened for eligible articles. Thirty-eight studies, including 4443 adult patients with schizophrenia, were included in the quantitative analyses, and four studies were qualitatively assessed. Quantitative analyses were performed for acutely ill and clinically stable patient subgroups regarding the different genotypes of rs4680 SNP. Our results showed that the severity of negative symptoms was higher in patients who were rs4680 Met homozygous compared to Val/Met heterozygotes only in acutely ill samples. There was no other significant difference between genotypes. Meta-regression did not reveal any significant moderator effect on the difference in negative symptoms. General psychopathology, positive, negative, and total psychotic symptom levels also were similar between Val homozygotes and Met carriers. Nonetheless, there are some limitations in the study. First, SNPs except for rs4680 were under-researched because of the limited number of studies. Second, high heterogeneity across studies was the main concern. Our results suggested that the COMT rs4680 Met allele was associated with higher levels of negative symptoms within acutely ill patients. Future studies should focus on specific patient subgroups to reveal the moderating effects of SNPs.

Index depressive episode and antidepressant exposure were associated with illness characteristics of pediatric bipolar disorder

OBJECTIVE

Pediatric bipolar disorder (PBD) is a serious, recurrent disorder leading to severe functional impairment. As a first mood episode, index episode could affect the long-term course of the illness. This study aimed to investigate the clinical characteristics of youth with PBD from our multicenter, nationwide, naturalistic follow-up samples and to identify (i) the effects of index mood episode and (ii) the effect of previous antidepressant treatments on the age at mania onset of PBD.

METHOD

The study sample consisted of 271 youth with BD-I followed by the child and adolescent psychiatry clinics of seven different university hospitals and three research state hospitals, representing six geographic regions across Turkey. All diagnoses were made according to structured interviews, and all data were retrospectively obtained from clinical records by the clinicians.

RESULTS

When patients with index depressive/mixed episodes (IDE, n=129) and patients with index (hypo)manic episodes (IME, n=142) were compared, the total number of mood episodes and rapid cycling feature were significantly higher in the IDE group than in the IME group. The Cox regression analysis adjusted for sociodemographic and illness characteristics revealed female adolescents in the IDE group treated with antidepressants were more likely to have an earlier onset of mania (hazard ratio=2.03, 95% confidence interval=1.31-3.12, p=0.001).

CONCLUSION

This is the first large-scale nationwide follow-up study in Turkey that indicated prior antidepressant treatments were associated with an earlier onset of mania in youth, particularly in adolescent females. Larger prospective studies are needed to identify neurodevelopmental processes underlying PBD and initiate prevention approaches.

Comparison of inflammation markers and severity of illness among patients with COVID-19, acute psychiatric disorders and comorbidity

Background and objectives

Neutrophil, lymphocyte counts, lactate dehydrogenase (LDH), D-dimer, fibrinogen, and comorbid illness are associated with the course and prognosis of COVID-19. However, the course of acute severe psychiatric disorders overlapping with COVID-19 infection was not investigated and remained as an unclarified research area. This study aimed to demonstrate inflammatory markers and the course of patients suffering from both conditions.

Methods

Thirty-eight inpatients with COVID-19 and comorbid acute psychiatric disorders (COVID-19+PD), 31 inpatients with COVID-19, and 38 inpatients with an acute psychiatric disorder (PD) were included in the study. Neutrophil, lymphocyte counts, serum ferritin, lactate dehydrogenase (LDH), D-dimer, fibrinogen, Systemic immune-inflammation index (SII), neutrophil/lymphocyte ratio (NLR), and C-reactive protein (CRP) were compared to evaluate inflammation levels.

Results

Patients with SARS-CoV-2 infection had older age compared to the PD group. CALL (Comorbidity, age, lymphocyte, lactate dehydrogenase) scores which predict the progression risk in patients with COVID-19 pneumonia, of both COVID-19 groups were found similar. The COVID-19+PD had higher SII in the study sample. Additionally, the COVID-19+PD group had higher NLR, ferritin, and CRP levels than those of the PD group.

Conclusions

The prognosis of COVID-19 is not worse when accompanied by a psychiatric disorder. Laboratory assessment can guide clinicians to distinguish those infected with SARS-CoV-2 within psychiatric inpatient units. The biochemical assessment did not robustly support higher inflammatory levels in the comorbid COVID-19 and psychiatric disorder group compared to the COVID-19 group.

Association Between Anterior Cingulate Cortex Neurochemical Profile and Clinical Remission After Electroconvulsive Treatment in Major Depressive Disorder: A Longitudinal 1H Magnetic Resonance Spectroscopy Study

BACKGROUND

The aim of the study was to assess anterior cingulate cortex (ACC) neurochemical profile of patients with unipolar major depressive disorder (MDD) before and after electroconvulsive therapy (ECT) by using 1H magnetic resonance spectroscopy (1H-MRS).

METHOD

Using 1H-MRS, the metabolite levels of choline, glutamate + glutamine (Glx), myo-inositol, N-acetylaspartate, and total creatine were measured in ACC before and after 4-week ECT. The Montgomery-Åsberg Depression Rating Scale (MADRS) was implemented by blind raters to evaluate the efficacy of the treatment. Electroconvulsive therapy-remitter (ER) and nonremitter groups were compared using the 1-way repeated measures analysis of variance.

RESULTS

Thirty patients with unipolar MDD (aged 41.3 ± 10.0 years, 66.7% female) were included in the study. The ER group (n = 16, 53.3%) and NR group did not differ regarding baseline Global Assessment of Functioning and MADRS scores. At the end of 4-week ECT treatment, results did not suggest any significant difference for metabolite levels in ACC. When compared with the NR group, the ER group had higher baseline levels of Glx (8.8 ± 1.8 vs 6.3 ± 2.0, P = 0.005) and total creatine (5.3 ± 0.6 vs 4.7 ± 0.5, P = 0.010). In addition, elevated baseline Glx (r = -0.68, P = 0.002) was associated with lower MADRS scores at the end treatment. Finally, the change in Glx levels was correlated with change in MADRS scores after ECT (r = 0.47, P = 0.049).

LIMITATIONS

Modest sample size and 1H-MRS at 1.5 Tesla are limitations of the study.

CONCLUSIONS

Results suggested that Glx levels could be a predictor of remission. Studies with larger samples should explore neurochemical correlates of ECT in unipolar MDD.

Investigating Predictors of Clozapine Response in Adolescents with Schizophrenia and Schizoaffective Disorder

Objectives: We aimed to determine the clinical predictors of clozapine response in patients with early-onset schizophrenia (EOS)/schizoaffective disorder and the effect of substance use disorder (SUD) on treatment outcomes. Methods: Medical records of patients with treatment-resistant EOS receiving clozapine were identified for data analysis dated between January 2015 and April 2020. Patients on clozapine were followed in an inpatient unit of a tertiary care mental health hospital. Using the Positive and Negative Symptom Scale (PANSS), ≥30% reduction was defined as the response criteria after clozapine treatment. Results: Of 50 subjects (age: 16.3 ± 1.3 years, 36.0% female), 22 subjects (44.0%) met the defined response criteria. Clozapine responder (CLZ-R) and clozapine nonresponder (CLZ-NR) groups were similar regarding age at illness onset, sex, and duration of untreated psychosis. The CLZ-R group had higher baseline positive PANSS scores (24.8 ± 8.1 vs. 17.1 ± 6.6, p = 0.001, d = 1.0) and total PANSS scores (94.8 ± 17.2 vs. 80.1 ± 19.8, p = 0.008, d = 0.8) compared with the CLZ-NR counterparts. The duration of hospital stay was longer in the CLZ-NR group (122.3 ± 48.2 vs. 87.3 ± 36.2 days, p = 0.007). Among 32 male patients, the presence of SUD (n = 9, 28.1%) was associated with a less reduction in total PANSS scores (F = 7.5, p = 0.010) and higher levels of positive symptoms at the end of the treatment (12.8 ± 4.1 vs. 18.8 ± 7.4, p = 0.006, d = 1.0). Synthetic cannabinoids were the most common substance type used among males with treatment-refractory EOS (25.0%). Conclusions: Our results did not support the role of sociodemographic variables in clozapine response. Positive symptoms and SUD yielded a prognostic value in patients receiving clozapine.

Longitudinal Course of Depressive, Anxiety, and Posttraumatic Stress Disorder Symptoms After Heart Surgery: A Meta-Analysis of 94 Studies

OBJECTIVE

This study aimed to analyze the longitudinal course of depression, anxiety, and posttraumatic stress disorder (PTSD) symptoms in patients with cardiac disease after heart surgery (HS).

METHODS

We conducted a systematic review and random-effects meta-analysis of cohort studies in patients undergoing HS, measuring anxiety, depressive, and PTSD symptoms before and at least 30 days thereafter. Subgroup and meta-regression analyses, investigation of publication bias, and quality assessment were undertaken.

RESULTS

We included 94 studies relating to 15,561 patients. HS included coronary artery bypass graft surgery, valve replacement, implantable cardioverter-defibrillator placement, left ventricular assist device placement, heart transplantation, and other types of HS. Across studies, symptoms of depression (g = 0.32; 95% confidence interval [CI] = 0.25 to 0.39; p < .001) and anxiety improved after HS (g = 0.52; 95% CI = 0.43 to 0.62; p < .001), whereas PTSD symptoms worsened (g = -0.42; 95% CI = -0.80 to -0.04; p = .032). The reduction of depression and anxiety levels was more pronounced for patients with underlying coronary artery disease and heart failure and persisted for 1 year after HS, whereas the increase in PTSD symptoms returned to baseline after 6 months. Depression improvement was inversely associated with older age, diabetes, hypertension, and dyslipidemia and positively with baseline heart failure. No additional clinical or demographic variables were associated with the course of anxiety symptoms. Quality of included studies was low overall. Publication bias was nonsignificant.

CONCLUSIONS

Depressive and anxiety symptoms improve for 1 year after HS, whereas PTSD symptoms might worsen. Older patients and those with metabolic comorbidities, valve disease, or ventricular arrhythmias are at higher risk for continued depressive and anxiety symptoms and should be monitored closely.

Relationship between carotid artery intima-media thickness and brachial artery flow-mediated dilation in peritoneal dialysis patients

BACKGROUND AND AIM

Carotid artery intima-media thickness (CIMT) and brachial artery flow-mediated dilation percentage (FMD%) are two commonly used parameters for detecting subclinical atherosclerosis. However, studies investigating the relationship between CIMT and brachial artery FMD% in different populations have produced conflicting results. The aim of this study was to determine the relationship between CIMT and brachial artery FMD% in patients on peritoneal dialysis (PD) METHODS: Fifty-two PD patients without known cardiovascular disease and 30 age-gender matched controls were included in the study. Endothelial function was determined using ultrasonography (US) to measure the FMD of the brachial artery, and this parameter was expressed as the percentage change from the baseline diameter of the brachial artery (FMD%). We also measured CIMT by US and analysed the relationship between CIMT and brachial FMD%.

RESULTS

The CIMT was significantly higher in patients than in the control group (0.84 +/- 0.08 vs. 0.75 +/- 0.06 mm, P < 0.01), whereas brachial artery FMD% was lower in patients than in the controls (8.2 +/- 5.0 vs. 11.7 +/- 5.5%, P < 0.01). There was no significant correlation between CIMT and FMD% (r = -0.004, P = 0.94).

CONCLUSION

Although PD patients are known to be characterized by an impaired flow-mediated vasodilatation of brachial artery and increased in CIMT, we did not find a significant correlation between FMD% and CIMT in our PD patient cohort. One possible explanation for our results is that each method measures a different aspect and stage of atherosclerosis.

Ruptured aneurysm of noncoronary sinus of Valsalva: demonstration with magnetic resonance imaging

There are few published reports regarding imaging findings of sinus of Valsalva aneurysms (SVA) with magnetic resonance imaging (MRI). We present an unusual form of ruptured SVA, emphasizing the diagnostic importance of MRI among a range of imaging techniques. This case report describes a case of idiopathic (thought to be congenital in origin), acutely symptomatic ruptured noncoronary SVA diagnosed by MRI and confirmed with conventional angiography and surgery. MRI accurately showed aneurysm size, location, and rupture into the right atrium and provided valuable information about this rare lesion.

Oxidative stress and asymmetric dimethylarginine is independently associated with carotid intima media thickness in peritoneal dialysis patients

BACKGROUND

Oxidative stress (OS) and asymmetric dimethylarginine (ADMA) are accepted as nonclassical cardiovascular risk factors in end-stage renal disease patients. To clarify the role of these factors in the atherosclerotic process, we investigated if OS and ADMA are associated with common carotid artery intima media thickness (CIMT) in peritoneal dialysis (PD) patients.

METHODS

Thirty PD patients without known atherosclerotic disease and classical cardiovascular risk factors as well as age- and gender-matched 30 healthy individuals were included. We measured serum thiobarbituric acid-reactive substances (TBARS), malondialdehyde (MDA), advanced glycation end product (AGE), pentosidine, advanced oxidation protein products (AOPP), ADMA and CIMT in each subjects.

RESULTS

TBARS, MDA, AOPP, AGE, pentosidine and ADMA levels were significantly higher in PD patients than in controls (p < 0.001). CIMT in patients was higher than in the control group (0.83 +/- 0.09 vs. 0.77 +/- 0.06 mm; p < 0.01). CIMT was independently correlated with TBARS (beta = 0.33, p < 0.01), MDA (beta = 0.27, p < 0.01), AOPP (beta = 0.22, p < 0.02), AGE (beta = 0.45, p < 0.01), pentosidine (beta = 0.56, p < 0.01) and ADMA (beta = 0.54, p < 0.01).

CONCLUSIONS

OS markers and serum ADMA levels independently predict the CIMT level in PD patients.

Catecholamine response during haemodynamically stable upright posture in individuals with and without tilt-table induced vasovagal syncope

AIM

Changes in circulating catecholamine concentrations during vasovagal faints have been the subject of considerable study. However, whether catecholamines are part of the triggering mechanism, or principally reflect attempted compensation for an evolving circulatory crisis is unknown. To address this issue, we determined whether the circulating catecholamine response to upright posture differs among patients with and without inducible vasovagal faints at a time when there is no detectable haemodynamic compromise.

METHODS AND RESULTS

Blood samples for measurement of adrenaline and noradrenaline (Norepi) concentrations were obtained in the baseline state, and at both 2-3 min and 4-6 min of upright posture in 22 patients undergoing head-up tilt-table testing for evaluation of syncope of unknown cause. In 11 individuals tilt-testing induced syncope at >5 min head-up posture (Group 1). In 11 other individuals tilt testing did not result in syncope (Group 2). Supine arterial catecholamine levels were comparable in the two groups. However, adrenaline concentrations during upright posture tended to be greater at 2-3 min and were significantly greater at 4-6 min in Group 1 than in Group 2 (P< 0.01). These differences occurred in the absence of significant intergroup differences in mean arterial pressure or cardiac cycle lengths. Norepi concentrations also increased in both groups, but without significant differences.

CONCLUSION

Circulating adrenaline concentrations in posturally induced vasovagal faints rise more rapidly in vasovagal fainters than in comparably posturally stressed non-fainters, and were significantly greater in fainters prior to either detectable haemodynamic compromise or diminution of circulating Norepi levels. These findings suggest that a premonitory rise in adrenaline concentrations occurs in vasovagal fainters unassociated with an evolving circulatory crisis.

Influence of Treatment of Type II Leaks on the Aneurysm Surface Area

OBJECTIVE

to determine whether interventional treatment of type II endoleaks leads to a decrease in aneurysm surface area.

MATERIAL AND METHOD

type II endoleaks were detected in a group of 14 male patients (median age: 70.2 years) following endovascular repair of a total number of 160 infrarenal aneurysms of the abdominal aorta. The surface area of the aneurysm was determined by computed tomography (CT) pre- and postoperatively and at subsequent follow-up examinations. If type II endoleaks were documented at CT, patients underwent treatment by means of coil embolisation.

RESULTS

interventional treatment resulted in successful occlusion of type II endoleaks in eight patients. One of the cases exhibited spontaneous occlusion. Occlusion was associated with an average decrease in aneurysm surface area of 3.3 cm(2)( p =0.01). In one of these patients, treatment resulted in a temporary occlusion of the endoleak, also with associated decrease in aneurysm size. After recurrence of the type II endoleak, however, the patient experienced an increase in aneurysm surface area. In the remaining four patients the type II endoleaks persisted, resulting in a non significant increase in aneurysm surface area.

CONCLUSION

only complete occlusion of endoleaks results in decrease in the size of the aneurysm sac. Because of endotension and the risk of rupture we favour an early interventional treatment of type II endoleaks.

Endoleaks after endovascular repair of aortic aneurysm: are they predictable?-initial results

PURPOSE

To evaluate the predictability of endoleak.

MATERIALS AND METHODS

Thirteen women and 60 men (mean age, 69.8 years) underwent transfemoral insertion of endoluminal stent-grafts for treatment of aortic aneurysms. Follow-up included helical computed tomography (CT) at 3-month intervals. In the cases of endoleak, angiography also was performed to document the number of leak sites, their size and position, the feeding artery, the size of the aneurysm, the amount of thrombus, and the visualization of the lumbar arteries and inferior mesenteric artery. These data were correlated (Student t test) with the probability of endoleak.

RESULTS

A total of seven (10%) endoleaks were identified at CT in 68 patients. The feeding vessels were lumbar arteries in three cases, the inferior mesenteric artery in three cases, and the median sacral artery in one case. Of all factors, only the number of lumbar arteries visualized preoperatively (P <.005) had a significant correlation with probability of endoleak. In 71% (five of seven patients) of the cases of lumbar endoleak, four lumbar arteries were patent, whereas among the 61 patients with successfully repaired aneurysm, only eight (13%) had four patent lumbar arteries. Endoleaks were never found in the primarily thrombosed sections of an aneurysm.

CONCLUSION

Prediction of endoleaks with absolute certainty remains elusive. The single correlating risk factor identified from the data was patency of four or more lumbar arteries visualized preoperatively at CT.

Does successful embolization of endoleaks lead to aneurysm sac shrinkage?

PURPOSE

To determine whether embolization of endoleaks after endovascular repair of aortic aneurysms is associated with a decrease in the diameter of the aneurysm.

METHODS

Fifteen patients (13 men; median age 76 years) demonstrated primary endoleaks persisting > or =3 months after endovascular abdominal aortic aneurysm repair with a variety of commercial endografts. In 8 patients, coil embolization failed to completely obliterate the leak, whereas embolotherapy proved successful in the remaining 7 patients. Surveillance of endoleaks and the effect of their embolization on aneurysm size were followed with serial computed tomographic (CT) scans. Follow-up after embolization extended for at least 12 months.

RESULTS

Patients with persistent leak exhibited a slight increase in mean aneurysm area from 21.41+/-4.25 to 22.47+/-6.70 cm2. Medians differed from 20.03 (range 16.59-28.73) to 23.03 (range 14.14-33.69) cm2 (p = 0.0078, 95% confidence interval [CI] = 0.0075 to 0.0081). Successfully treated patients, however, showed a mean decrease from 20.58+/-3.63 (median 19.87) to 16.36+/-6.46 cm2 (median 18.10) at 1 year (p = 0.0156, 95% CI = 0.0151 to 0.0160). The medians differed significantly between groups (p < 0.05).

CONCLUSIONS

Persistent endoleaks after endovascular aortic aneurysm treatment led to an increase in the diameter of the aneurysm, whereas complete occlusion was associated with a significant decrease in aneurysm diameter.

Treatment of leaks after endovascular repair of aortic aneurysms

PURPOSE

To evaluate leaks after the endovascular repair of aortic aneurysms and treat them with occlusive therapy.

MATERIALS AND METHODS

Seventy patients (11 women, 59 men), aged 26-82 years (mean, 69.2 years), underwent transfemoral insertion of endoluminal stent-grafts for treatment of aortic aneurysms. Indications were traumatic pseudoaneurysms (n = 5) or arteriosclerotic aneurysms (n = 65). Aneurysms were thoracic (n = 5) or infrarenal (n = 65). To exclude the possibility of leaks, spiral computed tomography (CT) was performed at 3-month intervals. Patients with leaks that persisted unchanged longer than 3 months were referred for angiography and occlusive therapy.

RESULTS

At CT, 21 leaks were identified in 17 of 70 patients (24%). Only 11 of those 17 patients (65%) had leaks identified with conventional aortography. Selective angiography, however, depicted all of these. Eighteen of 21 leaks proved amenable to occlusive treatment: surgery (n = 1), further stent implantation (n = 4), or embolization (n = 13). In one leak, spontaneous occlusion occurred after 3 months. Two leaks in either the iliolumbar or the median sacral artery were inaccessible; one remained untreated, and the other was unsuccessfully treated. Mean follow-up of occlusive therapy was 6.8 months (range, 2-14 months).

CONCLUSION

Successful occlusion of perigraft leaks is feasible in most cases and can be performed without major complications.

Angiography of leaks after endovascular repair of infrarenal aortic aneurysms

OBJECTIVE

We examined whether leaks that persist after stent grafting are associated with outflow arteries.

SUBJECTS AND METHODS

Selective angiography was performed in 21 patients with persistent leaks after undergoing endovascular repair of infrarenal aneurysms of the abdominal aorta. Late leaks occurred in five patients whose prostheses were originally sealed. Before angiography, the size and position of leaks were determined with CT and color Doppler sonography.

RESULTS

Superselective angiography was successful in 19 of 21 patients. In two patients, angiography was performed over the afferent artery supplying the leak. We found one outflow artery at the site of the leak in 10 patients (47%); two outflow arteries in five (23.8%); and as many as five outflow arteries in three (14%). Angiography overlooked outflow arteries in three patients (14%). The lumbar and inferior mesenteric, urethral, and testicular arteries were identified as outflow arteries.

CONCLUSION

Other than feeder arteries, persistent leaks are associated with outflow vessels that contribute to the patency of leaks.

Leakages after endovascular repair of aortic aneurysms: classification based on findings at CT, angiography, and radiography

PURPOSE

To ascertain whether the configuration and location of leakages identified at computed tomography (CT) could provide evidence of their angiographically and fluoroscopically confirmed causes.

MATERIALS AND METHODS

Fifty patients aged 26-79 years underwent endovascular repair of traumatic (n = 4) or arteriosclerotic (n = 46) aortic aneurysms (four thoracic, 46 infrarenal). Radiographic examinations in three planes and helical CT were performed 1 week after implantation and every 3 months thereafter. Angiography was performed when there was evidence of a leakage at CT.

RESULTS

CT demonstrated evidence of leakages in 13 patients. Broad-based leakages immediately adjacent to the prosthesis were termed "perigraft leakages." If the area most affected by the leakage lay along the border of the aneurysm, then retrograde leakages were apparent at angiography. If the leakage was ventral to the prosthesis, then its source was the inferior mesenteric artery; if it was dorsolateral, then it was supplied by either the lumbar arteries or the median sacral artery through the hypogastric artery. One circumferential leakage could not be evaluated adequately at CT or angiography. Radiography depicted a rupture of the stent mesh in the middle of the prosthesis. Selective angiography demonstrated all types of leakages and permitted CT classification.

CONCLUSION

The cause of a leakage can be determined with CT on the basis of its configuration and location in the majority of cases.

Effect of intraaortic balloon counter-pulsation on QT dispersion in acute anterior myocardial infarction

The present prospective non-randomized study aimed to examine whether intraaortic balloon counter-pulsation (IABP) has a favorable effect on QT dispersion in patients with acute anterior MI. Patients with acute anterior MI who presented within 6 h after the symptom onset were assigned to the IABP + streptokinase or streptokinase (STK) group. The IABP + STK group was consisted of 26 men and two women (mean age 52.9+/-10.2). The STK group was consisted of 19 men and two women (mean age 54.4+/-10.8). In the IABP + STK group, mean QT interval dispersion significantly shortened 6 h after treatment (50.9+/-15.6 ms before STK, and 36+/-13.9 ms 6 h after STK; P = 0.001) and did not significantly change 24 h after STK (35.6+/-11.2 ms). In the STK group, mean QT interval dispersion did not vary significantly before and 6 h after STK (57.14+/-13.2 ms before STK, and 56.07+/-13.3 ms 6 h after STK; P > 0.05) but 24 h after STK it significantly shortened to 40.42+/-10.8 ms (P < 0.001). Before STK, mean QT interval dispersions in the IABP + STK and STK groups were 50.9+/-15.6 ms and 57.14+/-13.2 ms, respectively (P > 0.05), 6 h after STK, mean QT interval dispersions were 36+/-13.9 ms and 56.07+/-13.3 ms, respectively (P = 0.0001) and 24 h after STK, mean QT interval dispersions were 35.6+/-11.2 ms and 40.42+/-10.8 ms, respectively (P > 0.05). In conclusion this study demonstrates that the adjunct of IABP to thrombolytic therapy, in the setting of acute anterior MI, significantly decreases QT interval dispersion at 6 h and this effect might be secondary to accelerated reperfusion and/or other beneficial effects of IABP.