Dealing with distribution mismatch in semi-supervised deep learning for COVID-19 detection using chest X-ray images: A novel approach using feature densities.

In the context of the global coronavirus pandemic, different deep learning solutions for infected subject detection using chest X-ray images have been proposed. However, deep learning models usually need large labelled datasets to be effective. Semi-supervised deep learning is an attractive alternative, where unlabelled data is leveraged to improve the overall model's accuracy. However, in real-world usage settings, an unlabelled dataset might present a different distribution than the labelled dataset (i.e. the labelled dataset was sampled from a target clinic and the unlabelled dataset from a source clinic). This results in a distribution mismatch between the unlabelled and labelled datasets. In this work, we assess the impact of the distribution mismatch between the labelled and the unlabelled datasets, for a semi-supervised model trained with chest X-ray images, for COVID-19 detection. Under strong distribution mismatch conditions, we found an accuracy hit of almost 30%, suggesting that the unlabelled dataset distribution has a strong influence in the behaviour of the model. Therefore, we propose a straightforward approach to diminish the impact of such distribution mismatch. Our proposed method uses a density approximation of the feature space. It is built upon the target dataset to filter out the observations in the source unlabelled dataset that might harm the accuracy of the semi-supervised model. It assumes that a small labelled source dataset is available together with a larger source unlabelled dataset. Our proposed method does not require any model training, it is simple and computationally cheap. We compare our proposed method against two popular state of the art out-of-distribution data detectors, which are also cheap and simple to implement. In our tests, our method yielded accuracy gains of up to 32%, when compared to the previous state of the art methods. The good results yielded by our method leads us to argue in favour for a more data-centric approach to improve model's accuracy. Furthermore, the developed method can be used to measure data effectiveness for semi-supervised deep learning model training.

TG/HDL index to identify subjects with metabolic syndrome in the Mexican population.

BACKGROUND: The triglyceride/high-density lipoprotein (TG/HDL) index has been proposed as an indicator of cardiovascular risk. In Mexico, there is a study in young adults that relates it to insulin resistance, but no cutoff point that distinguishes subjects with metabolic syndrome has been defined. OBJECTIVE: To determine the cutoff point for the TG/HDL index that identifies subjects with metabolic syndrome in the Mexican population. METHODS: Metabolic syndrome was diagnosed using the criteria established in the Third Report of the Adult Treatment Panel of the National Cholesterol Education Program adapted to the Mexican population. To identify the TG/HDL index cutoff point, ROC curve analysis and the Youden index were used. RESULTS: 1,318 subjects aged 40.9 ± 13.0 years participated in the study; 65.6% were women and 34.4% men; 41.2% had metabolic syndrome. The TG/HDL index obtained an area under the curve of 0.85 and an optimal cutoff point value ≥ 3.46, with a sensitivity of 79.6% and specificity of 76.4%. CONCLUSIONS: TG/HDL index cutoff point ≥ 3.46 is suitable for identifying subjects with metabolic syndrome in the Mexican population.

ANTECEDENTES: El índice triglicéridos/lipoproteína de alta densidad (TG/HDL) ha sido propuesto como un indicador de riesgo cardiovascular. En México, existe un estudio en adultos jóvenes que lo relaciona con resistencia a la insulina, pero no se ha definido un punto de corte que distinga a sujetos con síndrome metabólico. OBJETIVO: Determinar el punto de corte para el índice TG/HDL que identifique a sujetos con síndrome metabólico en población mexicana. MÉTODOS: El síndrome metabólico se diagnosticó mediante los criterios establecidos en el Tercer Reporte del Panel de Tratamiento para Adultos del Programa Nacional de Educación en Colesterol adaptados a la población mexicana. Para identificar el punto de corte del índice TG/HDL se utilizó el análisis de curvas ROC y el índice de Youden. RESULTADOS: En el estudio participaron 1318 sujetos con edad de 40.9 ± 13.0 años; 65.6 % fuerin mujeres y 34.4 % hombres; 41.2% presentó síndrome metabólico. El índice TG/HDL obtuvo un valor del área bajo la curva de 0.85 y un valor óptimo de punto de corte ≥ 3.46, con sensibilidad de 79.6 % y especificidad de 76.4 %. CONCLUSIONES: El punto de corte ≥ 3.46 para el índice TG/HDL es adecuado para identificar a sujetos con síndrome metabólico en población mexicana.

Acute vagus nerve stimulation enhances reversal learning in rats.

Cognitive flexibility is a prefrontal cortex-dependent neurocognitive process that enables behavioral adaptation in response to changes in environmental contingencies. Electrical vagus nerve stimulation (VNS) enhances several forms of learning and neuroplasticity, but its effects on cognitive flexibility have not been evaluated. In the current study, a within-subjects design was used to assess the effects of VNS on performance in a novel visual discrimination reversal learning task conducted in touchscreen operant chambers. The task design enabled simultaneous assessment of acute VNS both on reversal learning and on recall of a well-learned discrimination problem. Acute VNS delivered in conjunction with stimuli presentation during reversal learning reliably enhanced learning of new reward contingencies. Enhancement was not observed, however, if VNS was delivered during the session but was not coincident with presentation of to-be-learned stimuli. In addition, whereas VNS delivered at 30 HZ enhanced performance, the same enhancement was not observed using 10 or 50 Hz. Together, these data show that acute VNS facilitates reversal learning and indicate that the timing and frequency of the VNS are critical for these enhancing effects. In separate rats, administration of the norepinephrine reuptake inhibitor atomoxetine also enhanced reversal learning in the same task, consistent with a noradrenergic mechanism through which VNS enhances cognitive flexibility.

Preinjury Use of Marijuana and Outcomes in Trauma Patients.

BACKGROUND: Recent studies have examined the effects of marijuana in various populations; however, there has been limited research on the effect of marijuana use in severely injured trauma patients. We hypothesized that preinjury use of marijuana would be associated with improved outcomes in severely injured trauma patients. METHODS: All adult (18+ y) level I and level II trauma activations who presented to two large regional trauma centers between 2014 and 2018 were reviewed. Delta-9-tetrahydrocannabinol (THC)- indicated absence of drugs confirmed by testing and as THC + confirmed THC without another drug present. RESULTS: Of the 4849 patients included, 1373 (28.3%) were THC+. The THC + cohort was younger, had more males, and was more likely to be injured by penetrating mechanism (P < 0.001 for all) than THC-. THC + patients had shorter median length of stay (LOS) (P < 0.001) and intensive care unit LOS (P < 0.001). Mortality rate was lower in the THC + group (4.3% versus 7.6%, P < 0.001), but not in multivariate analysis. THC + patients with traumatic brain injury had shorter hospital LOS (P = 0.025) and shorter ventilator days (P = 0.033) than THC- patients. In patients with Injury Severity Score ≥16, THC + patients had significantly lower intensive care unit LOS (P = 0.009) and mortality (19.3% versus 25.0% P = 0.038) than drug-negative patients. CONCLUSIONS: Although preinjury use of marijuana does not improve survival in trauma patients, it may provide some improvement in outcomes in patients with traumatic brain injury and those that are more severely injured (Injury Severity Score ≥16). The mechanism behind this finding needs further evaluation.

Machine Learning for Health: Algorithm Auditing & Quality Control.

Developers proposing new machine learning for health (ML4H) tools often pledge to match or even surpass the performance of existing tools, yet the reality is usually more complicated. Reliable deployment of ML4H to the real world is challenging as examples from diabetic retinopathy or Covid-19 screening show. We envision an integrated framework of algorithm auditing and quality control that provides a path towards the effective and reliable application of ML systems in healthcare. In this editorial, we give a summary of ongoing work towards that vision and announce a call for participation to the special issue  Machine Learning for Health: Algorithm Auditing & Quality Control in this journal to advance the practice of ML4H auditing.

Correcting data imbalance for semi-supervised COVID-19 detection using X-ray chest images.

A key factor in the fight against viral diseases such as the coronavirus (COVID-19) is the identification of virus carriers as early and quickly as possible, in a cheap and efficient manner. The application of deep learning for image classification of chest X-ray images of COVID-19 patients could become a useful pre-diagnostic detection methodology. However, deep learning architectures require large labelled datasets. This is often a limitation when the subject of research is relatively new as in the case of the virus outbreak, where dealing with small labelled datasets is a challenge. Moreover, in such context, the datasets are also highly imbalanced, with few observations from positive cases of the new disease. In this work we evaluate the performance of the semi-supervised deep learning architecture known as MixMatch with a very limited number of labelled observations and highly imbalanced labelled datasets. We demonstrate the critical impact of data imbalance to the model's accuracy. Therefore, we propose a simple approach for correcting data imbalance, by re-weighting each observation in the loss function, giving a higher weight to the observations corresponding to the under-represented class. For unlabelled observations, we use the pseudo and augmented labels calculated by MixMatch to choose the appropriate weight. The proposed method improved classification accuracy by up to 18%, with respect to the non balanced MixMatch algorithm. We tested our proposed approach with several available datasets using 10, 15 and 20 labelled observations, for binary classification (COVID-19 positive and normal cases). For multi-class classification (COVID-19 positive, pneumonia and normal cases), we tested 30, 50, 70 and 90 labelled observations. Additionally, a new dataset is included among the tested datasets, composed of chest X-ray images of Costa Rican adult patients.

Sample-Size Determination Methodologies for Machine Learning in Medical Imaging Research: A Systematic Review.

PURPOSE: The required training sample size for a particular machine learning (ML) model applied to medical imaging data is often unknown. The purpose of this study was to provide a descriptive review of current sample-size determination methodologies in ML applied to medical imaging and to propose recommendations for future work in the field. METHODS: We conducted a systematic literature search of articles using Medline and Embase with keywords including "machine learning," "image," and "sample size." The search included articles published between 1946 and 2018. Data regarding the ML task, sample size, and train-test pipeline were collected. RESULTS: A total of 167 articles were identified, of which 22 were included for qualitative analysis. There were only 4 studies that discussed sample-size determination methodologies, and 18 that tested the effect of sample size on model performance as part of an exploratory analysis. The observed methods could be categorized as pre hoc model-based approaches, which relied on features of the algorithm, or post hoc curve-fitting approaches requiring empirical testing to model and extrapolate algorithm performance as a function of sample size. Between studies, we observed great variability in performance testing procedures used for curve-fitting, model assessment methods, and reporting of confidence in sample sizes. CONCLUSIONS: Our study highlights the scarcity of research in training set size determination methodologies applied to ML in medical imaging, emphasizes the need to standardize current reporting practices, and guides future work in development and streamlining of pre hoc and post hoc sample size approaches.

Culture-independent identification and quantification of Gallibacterium anatis (G. anatis) by real-time quantitative PCR.

Gallibacterium is a genus within the family Pasteurellaceae characterized by a high level of phenotypic and genetic diversity. No diagnostic method has yet been described, which allows species-specific identification of Gallibacterium anatis. The aim of this study was to develop a real-time quantitative PCR (qPCR) method allowing species-specific identification and quantification of G. anatis. A G. anatis specific DNA sequence was identified in the gyrase subunit B gene (gyrB) and used to design a TaqMan probe and corresponding primers. The specificity of the assay was tested on 52 bacterial strains. Twenty-two of the strains represented all of the presently available 13 phenotypic variants of G. anatis originating from different geographical locations. Nine strains represented each of the additional six Gallibacterium species and 21 strains represented other poultry associated bacterial species of the families Pasteurellaceae, Enterobacteriaceae and Flavobacteriaceae. Regarding specificity none of non-G. anatis strains tested positive with the proposed assay. To test and compare the qPCR method's ability to detect G. anatis from field samples, the sensitivity was compared to a previously published conventional PCR method and culture-based identification, respectively. The detection rates were 97%, 78% and 34% for the current qPCR, the conventional PCR and the culture-based identification method, respectively. The qPCR assay was able to detect the gene gyrB in serial dilutions of 10(8) colony forming units (CFU)/ml to as low as 10(0) CFU/ml copies. The proposed qPCR method is thus highly specific, sensitive and reproducible. In conclusion, we have developed a qPCR method that allows species-specific identification of G. anatis.

Depression increases the risk for uncontrolled hypertension.

BACKGROUND: Because hypertension and depression share common pathways, it is possible that each disease has an impact on the natural history of the other. OBJECTIVE: To determinate whether depression influences blood pressure control in hypertensive patients. METHODS: Forty hypertensive patients undergoing antihypertensive treatment, excluding beta-blockers and central-acting agents, self-measured their blood pressure several times a day for three days using a validated, commercially available device. All patients also completed the Zung Self-rating Depression Scale survey for depression. Associations between the results of the blood pressure and depression tests were determined using the Spearman correlation coefficient; RR was also measured. RESULTS: Of the 40 patients, 23 were depressed, and 21 of these 23 had poor control of their blood pressure. The RR for uncontrolled hypertension in depressed patients was 15.5. A significant correlation between systolic (r=0.713) and diastolic (r=0.52) blood pressure values and depression was found. CONCLUSION: Depression is common in patients with uncontrolled hypertension and may interfere with blood pressure control. Screening for depression in hypertensive patients is a simple and cost-effective tool that may improve outcomes.

Mental health during the COVID-19 pandemic in a longitudinal study of Hurricane Katrina survivors.

While the COVID-19 pandemic is known to have caused widespread mental health challenges, it remains unknown how the prevalence, presentation, and predictors of mental health adversity during the pandemic compare to other mass crises. We shed light on this question using longitudinal survey data (2003-2021) from 424 low-income mothers who were affected by both the pandemic and Hurricane Katrina, which struck the U.S. Gulf Coast in 2005. The prevalence of elevated posttraumatic stress symptoms was similar 1-year into the pandemic (41.6%) as 1-year post-Katrina (41.9%), while elevated psychological distress was more prevalent 1-year into the pandemic (48.3%) than 1-year post-Katrina (37.2%). Adjusted logistic regression models showed that pandemic-related bereavement, fear or worry, lapsed medical care, and economic stressors predicted mental health adversity during the pandemic. Similar exposures were associated with mental health adversity post-Katrina. Findings underscore the continued need for pandemic-related mental health services and suggest that preventing traumatic or stressful exposures may reduce the mental health impacts of future mass crises.

Índice de inmunidad-inflamación sistémica (IIS) como marcador pronóstico de mortalidad en pacientes con COVID-19.

OBJECTIVE: To determine if the systemic immune-inflammation index (SII) is a prognostic marker of mortality in COVID-19 patients. METHOD: Retrospective study that included patients admitted to a general hospital in Mexico City with diagnostic of COVID-19, confirmed by quantitative polymerase chain reaction from nasopharyngeal swab specimens in addition to characteristic symptomatology and computerized thoracic tomography imaging. Upon admission an hematic biometry was taken to calculate the SII (neutrophils × platelets/lymphocytes). The optimal cut-off point was determined from a ROC curve; the chi-square test was used to evaluate the association of SII with mortality, the strength of the association was estimated through the odds ratio (OR) and, finally, a multivariate binary logistic regression analysis was performed. RESULTS: 140 individuals were included, 86 (61.4%) men and 54 women (38.6%), the mean age of patients was 52 (± 13.81) years old. The best prognostic cut-off point found was 2332.30 × 109 (area under the curve: 0.68; 95% confidence interval [95% CI]: 0.59-0.77; p < 0.05). The OR was 3.78 (95% CI: 1.83-7.82; p < 0.05). CONCLUSIONS: We demonstrated that the SII is an easily available tool, effective and a prognostic marker of mortality in hospitalized COVID-19 patients.

OBJETIVO: Determinar si el índice de inmunidad-inflamación sistémica (IIS) es un marcador pronóstico de mortalidad en pacientes con COVID-19. MÉTODO: Estudio retrospectivo que incluyó pacientes que ingresaron con diagnóstico de COVID-19 a un hospital general de la Ciudad de México, confirmado mediante prueba de reacción cuantitativa en cadena de la polimerasa con transcriptasa inversa de muestras de hisopado nasofaríngeo, además de la sintomatología característica y los hallazgos de la tomografía computarizada de tórax. A su ingreso se les realizó biometría hemática para el cálculo del IIS (neutrófilos × plaquetas/linfocitos). Mediante una curva ROC se determinó el punto de corte óptimo del IIS. Para evaluar la asociación del IIS con la mortalidad se utilizó la prueba de ji al cuadrado, la fuerza de la asociación con la razón de momios (OR, odds ratio) y se realizó un análisis multivariado de regresión logística binaria. RESULTADOS: Se incluyeron 140 individuos, de los cuales 86 (61.4%) eran hombres y 54 (38.6%) mujeres, con una media de edad de 52 (± 13.81) años. El mejor punto de corte pronóstico fue 2332.30 × 109 (área bajo la curva: 0.68; intervalo de confianza del 95% [IC95%]: 0.59-0.77; p < 0.05). La OR fue de 3.78 (IC95%: 1.83-7.82; p < 0.05). CONCLUSIONES: El IIS mostró ser una herramienta de fácil disponibilidad y un marcador pronóstico de mortalidad al ingreso en pacientes hospitalizados con COVID-19.

Beneficial effect of combination therapy using an angiotensin-converting enzyme inhibitor plus verapamil on circulating resistin levels in hypertensive patients with type 2 diabetes.

BACKGROUND: Resistin levels are strongly correlated with insulin resistance and vascular inflammation. Type 2 diabetic and hypertensive patients have higher circulating levels of resistin, which is associated with endothelial dysfunction. OBJECTIVE: To compare the effect of trandolapril (T) and its fixed-dose combination with verapamil (FDTV) on resistin levels in hypertensive, type-2 diabetic patients. METHODS: Forty type-2 diabetic patients with never-treated hypertension were randomly assigned to two groups. One group received FDTV 2 mg/180 mg once per day; the other group received T 2 mg once per day. Study drugs were administered for three months in both groups. Resistin levels were measured using ELISA at the beginning of the study and at study end. Patients were evaluated monthly for blood pressure, fasting serum glucose levels and adverse events. Statistical analysis was performed using ANOVA. RESULTS: All patients experienced a significant reduction in blood pressure. Both therapeutic regimens reduced resistin levels; however, FDTV treatment resulted in a greater decrease in resistin levels (mean [± SD] 25.5±13 ng/mL to 17.2±10 ng/mL) when compared with T treatment (22.4±12 ng/mL to 18.5±8 ng/mL) (P<0.05). None of the patients experienced an adverse event. CONCLUSION: Results showed that FDTV resulted in a greater reduction in resistin levels than T treatment alone.

[Validation of diagnosis scales for differencing hemorrhagic from ischemic cerebral vascular disease]. / Escalas clínicas para diferenciar enfermedad vascular cerebral hemorrágica de isquémica.

BACKGROUND: differentiating hemorrhagic from ischemic cerebral vascular disease (CVD) is the starting point for the treatment. The aim was to compare the diagnostic accuracy of the scales that differentiate hemorrhagic from ischemic stroke. METHODS: we applied the scale of Siriraj Stroke Score (SSS) and Greek Stroke Score (GSS) to patients with stroke. The results were described as means and frequencies. For significant variables odds ratio was calculated. We calculated the validity of both scales compared to the head computed tomography. RESULTS: ninety one patients had ischemic stroke and 28 were hemorrhagic. The mean systolic blood pressure in ischemic stroke was 138.94 mmHg (SD ± 26.90) and hemorrhagic was 165.55 mmHg (SD ± 36.40) p = 0.0007. The atherogenic index (AT) in ischemic stroke was 4.52 (SD ± 1.52) and in hemorrhagic was 4.84 (SD ± 2.01) p = 0.87. The specificity of the SSS for hemorrhagic stroke is 85.5 % and 96.7 % for the GSS. CONCLUSIONS: the GSS has a high specificity for hemorrhagic stroke.

A real use case of semi-supervised learning for mammogram classification in a local clinic of Costa Rica.

The implementation of deep learning-based computer-aided diagnosis systems for the classification of mammogram images can help in improving the accuracy, reliability, and cost of diagnosing patients. However, training a deep learning model requires a considerable amount of labelled images, which can be expensive to obtain as time and effort from clinical practitioners are required. To address this, a number of publicly available datasets have been built with data from different hospitals and clinics, which can be used to pre-train the model. However, using models trained on these datasets for later transfer learning and model fine-tuning with images sampled from a different hospital or clinic might result in lower performance. This is due to the distribution mismatch of the datasets, which include different patient populations and image acquisition protocols. In this work, a real-world scenario is evaluated where a novel target dataset sampled from a private Costa Rican clinic is used, with few labels and heavily imbalanced data. The use of two popular and publicly available datasets (INbreast and CBIS-DDSM) as source data, to train and test the models on the novel target dataset, is evaluated. A common approach to further improve the model's performance under such small labelled target dataset setting is data augmentation. However, often cheaper unlabelled data is available from the target clinic. Therefore, semi-supervised deep learning, which leverages both labelled and unlabelled data, can be used in such conditions. In this work, we evaluate the semi-supervised deep learning approach known as MixMatch, to take advantage of unlabelled data from the target dataset, for whole mammogram image classification. We compare the usage of semi-supervised learning on its own, and combined with transfer learning (from a source mammogram dataset) with data augmentation, as also against regular supervised learning with transfer learning and data augmentation from source datasets. It is shown that the use of a semi-supervised deep learning combined with transfer learning and data augmentation can provide a meaningful advantage when using scarce labelled observations. Also, we found a strong influence of the source dataset, which suggests a more data-centric approach needed to tackle the challenge of scarcely labelled data. We used several different metrics to assess the performance gain of using semi-supervised learning, when dealing with very imbalanced test datasets (such as the G-mean and the F2-score), as mammogram datasets are often very imbalanced. Graphical Abstract Description of the test-bed implemented in this work. Two different source data distributions were used to fine-tune the different models tested in this work. The target dataset is the in-house CR-Chavarria-2020 dataset.

Improving Uncertainty Estimation With Semi-Supervised Deep Learning for COVID-19 Detection Using Chest X-Ray Images.

In this work we implement a COVID-19 infection detection system based on chest X-ray images with uncertainty estimation. Uncertainty estimation is vital for safe usage of computer aided diagnosis tools in medical applications. Model estimations with high uncertainty should be carefully analyzed by a trained radiologist. We aim to improve uncertainty estimations using unlabelled data through the MixMatch semi-supervised framework. We test popular uncertainty estimation approaches, comprising Softmax scores, Monte-Carlo dropout and deterministic uncertainty quantification. To compare the reliability of the uncertainty estimates, we propose the usage of the Jensen-Shannon distance between the uncertainty distributions of correct and incorrect estimations. This metric is statistically relevant, unlike most previously used metrics, which often ignore the distribution of the uncertainty estimations. Our test results show a significant improvement in uncertainty estimates when using unlabelled data. The best results are obtained with the use of the Monte Carlo dropout method.

Glycemic disorders and their impact on lung function. Cross-sectional study. / Trastornos glucémicos y su asociación con la función pulmonar. Estudio transversal analítico.

BACKGROUND AND OBJECTIVE: Diabetes mellitus can affect the lungs, in its various structures and functions. Current research is being conducted to establish the clinical impact of hyperglycaemia on lung function. The objective of this study is to determine if the glycaemic state (euglycaemic, prediabetes or diabetes) is associated with a decrease in lung volume, determined by spirometry. PATIENTS AND METHODS: An analytical cross-sectional study was carried out at the Ticomán General Hospital in Mexico City. Glucose and glycosylated haemoglobin concentration were used as the parameters to determine if the subjects had a glycaemic disorder. They were further categorised into euglycaemic, prediabetic and diabetic subjects according to ADA criteria guidelines. The subjects underwent forced spirometry testing, obtaining expiratory volume at the first second (FEV1), forced vital capacity (FVC), FEV1/FVC ratio, and peak expiratory flow (FEP). The lung volumes between the groups were compared. RESULTS: A total of 55 subjects were studied; 43 women, and 12 men; 14 euglycaemic, 9 prediabetic, and 32 with diabetes. Diabetic individuals presented a %FEP decrease compared to the prediabetic and euglycaemic subjects. The fasting serum glucose values correlated with decrease of %FEV1, FEV1/FVC and %FEP, while the HbA1c concentration only correlated with the decrease of %FEP. CONCLUSIONS: Subjects with diabetes have a lower %PEF than euglycaemic and prediabetic subjects, while the %FEV1, %FVC and the FEV1/FVC ratio do not vary between the different glycaemic states. Acute glycaemic non-control correlated with a decrease in more spirometric parameters than chronic glycaemic non-control.

Association of KIR2DL2 gene with anti-cyclic citrullinated protein antibodies for serodiagnosis in rheumatoid arthritis.

Rheumatoid arthritis is a clinical autoimmune syndrome that causes joint damage. The positive or negative anti-cyclic citrullinated protein (CCP) antibodies serodiagnosis differentiates two subsets of the disease, each with different genetic background. Previous studies have identified associations between KIR genes and rheumatoid arthritis but not with anti-CCP serodiagnosis. Therefore, we investigated the proportion of patients seropositive and seronegative to anti-CCP and its possible association with KIR (killer cell immunoglobulin-like receptor) genes. We included 100 patients with rheumatoid arthritis from western Mexico, who were determined for anti-CCP serodiagnosis by ELISA, and 16 KIR genes were genotyped by PCR-SSP. The proportion of seropositive anti-CCP patients was 83%, and they presented a higher frequency of KIR2DL2 genes than the seronegative group (73.6% vs. 46.2%, p = 0.044) which, in turn, presented a higher KIR2DL2-/KIR2DL3+ genotype frequency than the first ones (46.2% vs. 17.2%, p = 0.043). These results suggest different KIR genetic backgrounds for each subset of the disease according to anti-CCP serodiagnosis.

La artritis reumatoide es un síndrome clínico autoinmune que causa daño en las articulaciones. El serodiagnóstico positivo o negativo para anticuerpos proteicos anticíclicos citrulinados (CCP) diferencia dos subconjuntos de la enfermedad, cada uno con diferente fondo genético. Estudios previos han identificado asociaciones entre los genes killer cell immunoglobulin- like receptor (KIR) y la artritis reumatoide, pero no con el serodiagnóstico de anti-CCP. Por lo tanto, investigamos la proporción de seropositividad y seronegatividad anti-CCP y su posible asociación con genes KIR. Se incluyeron 100 pacientes con artritis reumatoide del occidente de México, a quienes se les determinó su serodiagnóstico anti-CCP por ELISA y también se les realizó genotipificación de 16 genes KIR por PCR-SSP. La proporción de pacientes seropositivos anti-CCP fue del 83% y presentaron una mayor frecuencia génica KIR2DL2 que el grupo seronegativo (73.6% vs. 46.2%, p = 0.044), estos últimos presentaron mayor frecuencia genotípica KIR2DL2-/KIR2DL3+ que los primeros (46.2% vs. 17.2%, p = 0.043). Los resultados sugieren diferente fondo genético KIR para cada subconjunto de la enfermedad, de acuerdo con el serodiagnóstico anti-CCP.

Hemagglutinating activity of serovar reference strains of Ornithobacterium rhinotracheale.

In the present study, the hemagglutinating activity of 9 reference strains (serovars A-I) of Ornithobacterium rhinotracheale was investigated by using fresh erythrocytes from 15 different species: chicken (broiler, rooster, hen), turkey, pigeon, quail, duck, Harris hawk (Parabuteo unicinctus), house finch (Carpodacus mexicanus), cow, sheep, horse, dog, rabbit, pig, human (groups A, B, AB, and O), and rainbow trout (Oncorhynchus mykiss). All 9 strains agglutinated rabbit erythrocytes. None of the strains was able to agglutinate hen, cow, horse, or rainbow trout erythrocytes. The number of positive reactions among the remaining species varied. Results indicate that the use of rabbit erythrocytes is better suited for testing the hemagglutinating activity of O. rhinotracheale.

[Effect of the components of the metabolic syndrome on pulmonary function. The unexpected role of high-density lipoprotein cholesterol]. / Efecto de los componentes del síndrome metabólico en la función pulmonar. El inesperado papel del colesterol ligado a lipoproteínas de alta densidad.

BACKGROUND: Metabolic syndrome is a condition that predisposes to cardiovascular disease and diabetes mellitus. In addition, it can have effects over neoplastic pathologies, liver and pulmonary function. Our objective is to analyze the effect of the metabolic syndrome and its components on pulmonary function. METHOD: 110 subjects from Mexico City were evaluated and anthropometric measurements, glucose determination, triglycerides and high-density lipoprotein (HDL) cholesterol were made. They underwent a simple spirometry. Diagnosis of metabolic syndrome was made following the NCEP-ATPIII criteria. RESULTS: Of 110 individuals, 90 (82%) were women and 20 men (18%); 71 subjects (65%) presented metabolic syndrome. Subjects with central obesity had a forced vital capacity (FVC) lower than subjects without central obesity (2.72 vs. 3.11 liters; p < 0.05). Those with low HDL had better spirometric results than subjects with normal HDL (FEV1 2.36 vs. 1.85 liters; p < 0.05), FVC (2.95 vs. 2.45 liters; p < 0.05) and FEV1/FVC ratio (0.78 vs.74; p < 0.05). Hypertensive subjects presented lower volumes in FEV1 (1.91 vs. 2.38; p < 0.05) and FVC (2.49 vs. 2.99; p < 0.05). CONCLUSION: There is no difference between the spirometry volumes of patients with metabolic syndrome versus the metabolically healthy subjects. The only factors associated with a decrease in FEV1 and FVC are central obesity and arterial hypertension. An unexpected finding was the negative correlation between HDL levels and lung function.

ANTECEDENTES: El síndrome metabólico es un estado que predispone a enfermedad cardiovascular y diabetes mellitus. Además, puede repercutir en la función hepática, en patologías neoplásicas y en la función pulmonar. Nuestro objetivo es analizar el efecto del síndrome metabólico y sus componentes sobre la función pulmonar. MÉTODO: Se evaluaron 110 sujetos de la Ciudad de México a quienes se realizaron mediciones antropométricas, determinación de glucosa, triglicéridos y colesterol ligado a lipoproteínas de alta densidad (HDL). Se les practicó una espirometría simple. Se realizó el diagnóstico de síndrome metabólico siguiendo los criterios NCEP-ATPIII. RESULTADOS: De 110 individuos, 90 (82%) fueron mujeres y 20 hombres (18%), y 71 (65%) presentaron síndrome metabólico. Los sujetos con obesidad central tuvieron una capacidad vital forzada (CVF) menor que aquellos sin obesidad central (2.72 vs. 3.11 l; p < 0.05). Los que presentaron colesterol HDL bajo tuvieron mejores resultados espirométricos que los sujetos con colesterol HDL normal (volumen espiratorio forzado en el primer segundo [VEF1] 2.36 vs. 1.85 l; p < 0.05), mejor CVF (2.95 vs. 2.45 l; p < 0.05) y mejor relación VEF1/CVF (78 vs. 74; p < 0.05). Los sujetos hipertensos presentaron menores volúmenes en VEF1 (1.91 vs. 2.38; p < 0.05) y CVF (2.49 vs. 2.99; p < 0.05). CONCLUSIÓN: No existe diferencia en los volúmenes espirométricos de pacientes con síndrome metabólico al compararlos con sujetos metabólicamente sanos. Solo la obesidad central y la hipertensión arterial se asocian con disminución del VEF1 y la CVF. Un hallazgo inesperado es la correlación negativa entre los valores de colesterol HDL y la función pulmonar.

Correlation between epicardial fat thickness and biochemical markers of metabolic risk. / Correlación entre el grosor de la grasa epicárdica y los marcadores bioquímicos de riesgo metabólico.

INTRODUCTION AND OBJECTIVE: Epicardial fat has been associated with increased cardiovascular risk and the development of atherosclerosis. Transthoracic echocardiography provides a reliable measurement of epicardial fat thickness (EFT). The aim of this study is to evaluate the relationship between EFT and biochemical parameters of metabolic risk. MATERIAL AND METHOD: We assessed 211 patients who underwent echocardiography; EFT was measured by two cardiologists. In addition, patients' glycaemia, lipid profile and serum uric acid were measured. Statistical analysis was performed with the Pearson coefficient test and Odds ratio. RESULTS: A positive correlation between EFT with glycaemia (r=.064), total serum cholesterol (r=.0056), high density lipoproteins (r=-.038), or with triglycerides (r=.118) was not observed. However, we did find a significant positive correlation between EFT and serum uric acid (r=.415, P<.00001). The odds ratio for EFT>3mm in patients with hyperuricemia was 6.26 (IC 95 2.79-14, P<.0001). CONCLUSION: Hyperuricemia is strongly associated with EFT in Mexican patients; EFT is a useful tool for global cardiovascular risk calculation.