Causal relationship between the AHSG gene and BMD through fetuin-A and BMI: multiple mediation analysis.

UNLABELLED: Using mediation analysis, a causal relationship between the AHSG gene and bone mineral density (BMD) through fetuin-A and body mass index (BMI) mediators was suggested. INTRODUCTION: Fetuin-A, a multifunctional protein of hepatic origin, is associated with bone mineral density. It is unclear if this association is causal. This study aimed at clarification of this issue. METHODS: A cross-sectional study was conducted among 1,741 healthy workers from the Electricity Generating Authority of Thailand (EGAT) cohort. The alpha-2-Heremans-Schmid glycoprotein (AHSG) rs2248690 gene was genotyped. Three mediation models were constructed using seemingly unrelated regression analysis. First, the ln[fetuin-A] group was regressed on the AHSG gene. Second, the BMI group was regressed on the AHSG gene and the ln[fetuin-A] group. Finally, the BMD model was constructed by fitting BMD on two mediators (ln[fetuin-A] and BMI) and the independent AHSG variable. All three analyses were adjusted for confounders. RESULTS: The prevalence of the minor T allele for the AHSG locus was 15.2%. The AHSG locus was highly related to serum fetuin-A levels (P < 0.001). Multiple mediation analyses showed that AHSG was significantly associated with BMD through the ln[fetuin-A] and BMI pathway, with beta coefficients of 0.0060 (95% CI 0.0038, 0.0083) and 0.0030 (95% CI 0.0020, 0.0045) at the total hip and lumbar spine, respectively. About 27.3 and 26.0% of total genetic effects on hip and spine BMD, respectively, were explained by the mediation effects of fetuin-A and BMI. CONCLUSIONS: Our study suggested evidence of a causal relationship between the AHSG gene and BMD through fetuin-A and BMI mediators.

Multidrug-resistant tuberculosis: response to treatment.

OBJECTIVE: We investigated clinical response and mortality rates in 18 HIV- negative patients with multidrug-resistant tuberculosis. METHOD: We obtained detailed data from our hospital on patients with multidrug - resistant tuberculosis defined as tuberculosis resistant at least to isoniazid and rifampicin. Therapeutic responses were evaluated by both microbiologic and clinical data. RESULTS: From January 1985 to December 1990, 18 HIV-negative patients were identified and treated. One patient committed suicide 1 week after receiving antituberculous drugs. Of the 17 patients, 10 (58.8%) had clinical response as well as microbiologic response, 6 (35.3%) were persisters during 24 months of follow-up, and 1 (5.8%) died after 30 months of treatment. The median follow-up for the patients was 87 weeks (range, 24 to 212). For those who responded well, there was no relapse.

The course of clinical COPD in Thailand.

Spirometry and outcome of 96 cases who were clinically diagnosed as having COPD were studied. Ninety-two cases had significant airway obstruction. Of these 92 cases, initial bronchodilator responses were evaluated in 73 cases. Twenty-six cases were responders (CRAO), while 47 cases were nonresponders (COPD). The average median survival of the whole group was 4 years and 5-year survival was 30 per cent. The annual change in FEV1 in both COPD and CRAO group were fluctuating, with a tendency to decrease in the former and increase in the latter. Spirometric parameters were found to be different between the survivors and nonsurvivors, these included initial PFEV1, PFVC, PFEF 25-75 per cent and postbronchodilator FEV1 and FEF 25-75 per cent. Inspite of the differences, an initial response to bronchodilator could not predict a better outcome for CRAO as compared to COPD. It was concluded that most clinical COPD who had progressive symptoms had significant airway obstruction and shorter survival. Due to fluctuating FEV1 during the course, the initial spirometry and degree of bronchodilator response were not accurate enough to predict subsequent outcome. Long-term follow-up on spirometry and response to bronchodilator should be individually evaluated.

The compromised host and pulmonary infiltration: a prospective study.

If immediate sputum examination does not result in a diagnosis, broad spectrum antibiotics should be administered while awaiting cultures. Three days later, if the patient deteriorates and has a negative hemoculture, bronchoalveolar lavage should be done before open lung biopsy or antifungal drugs are considered. We have demonstrated that rational empiric therapy in immunosuppressed patients with pneumonia is beneficial. Diagnostic tests must be performed so that treatment can be modified later when the etiological agent has been identified.

Pulmonary infections in immunocompromised hosts.

Between February 1985 and January 1987, sixty cases of pulmonary infection in immunocompromised host were seen at the Department of Medicine, Ramathibodi hospital. The underlying causes were: thirty-four cases with hematologic malignancies, nine with autoimmune diseases, seven with aplastic anemia, three with non-hematologic malignancies, and seven following corticosteroid therapy. Infectious agents were identified in thirty-two cases; nineteen bacterial, three fungal, four nocardial, three strongyloidiasis and three tuberculosis. In nineteen cases, the infectious agents were classified as "probable", and in nine cases, "unknown" etiology. Eleven of the twenty hospital infected patients and ten of the forty community infected patients died. Mortality appeared to be related to the underlying diseases, infection acquired in hospital and the infectious agent.

Use of bronchoalveolar lavage to diagnose pulmonary infiltrates in immunocompromised hosts.

Bronchoalveolar lavage is a simple and safe technique that aids in making the diagnosis of pulmonary infiltrates in immunosuppressed patients. The procedure can be done even in thrombocytopenic patients and those requiring ventilatory support.

Autopsy proven cases of pulmonary thromboembolism: 18-year study at Ramathibodi Hospital.

Twelve cases of pulmonary embolism were found among 4,896 autopsies during 18 years in Ramathibodi hospital. This gives an incidence of 0.24 per 100 necropsies. Deep vein thrombosis occurred in two of the twelve cases and was not the main cause of pulmonary embolism in this series.