INTRODUCTIONSevere acute respiratory syndrome (SARS) affected 8096 individuals in 29 countries, with 774 deaths. In Singapore, there were 238 cases of SARS with 33 deaths. A retrospective analysis was performed to identify predictors of poor outcome in patients with SARS locally.

MATERIALS AND METHODSClinical, laboratory and outcome data of 234 patients admitted to Tan Tock Seng Hospital and Singapore General Hospital were collected and analysed. Only data collected at the time of admission were used in the analysis for predictors of poor outcome. Adverse events were defined as admission to the intensive care unit or death.

RESULTSClinical (temperature, FiO2) and laboratory [leukocyte, lymphocyte, neutrophil, platelet, lactate dehydrogenase (LDH), albumin] trends in groups with and without an adversarial event were presented. Fifty patients experienced an adverse event. On univariate analysis, male gender, advanced age, presence of comorbidities, neutrophilia, lymphopaenia, hyponatraemia, hypoalbuminaemia, transaminitis and elevated LDH or C-reactive protein were found to be significant predictors. On multivariate analysis, predictors of poor outcome were increased age [odds ratio (OR) 1.73 for every 10-year increase; 95% CI, 1.35 to 2.21], neutrophilia (OR 1.06 for every 1 x 10(9)/L increase; 95% CI, 1.02 to 1.11) and high LDH (OR 1.17 for every 100 U/L increase; 95% CI, 1.02 to 1.34). None of the 12 paediatric patients had an adverse event.

CONCLUSIONAdvanced age, neutrophilia and high LDH predict poor outcomes in patients with SARS.

Adolescent ; Adult ; Aged ; Aged, 80 and over ; Antibodies, Viral ; analysis ; Child ; Child, Preschool ; DNA, Viral ; analysis ; Female ; Fluorescent Antibody Technique ; Humans ; Incidence ; Infant ; Male ; Middle Aged ; Retrospective Studies ; Reverse Transcriptase Polymerase Chain Reaction ; SARS Virus ; genetics ; immunology ; Severe Acute Respiratory Syndrome ; epidemiology ; virology ; Severity of Illness Index ; Singapore ; epidemiology ; Survival Rate

Objectives. Polymorphisms in metabolic genes which alter rates of bioactivation and detoxification have been shown to modulate susceptibility to colorectal cancer. This study sought to evaluate the colorectal cancer risk from environmental factors and to do polymorphism studies on genes that code for Phase I and II xenobiotic metabolic enzymes among Filipino colorectal cancer patients and matched controls. Methods. A total of 224 colorectal cancer cases and 276 controls from the Filipino population were genotyped for selected polymorphisms in GSTM1, GSTP1, GSTT1, NAT1 and NAT2. Medical and diet histories, occupational exposure and demographic data were also collected for all subject participants.Results. Univariate logistic regression of non-genetic factors identified exposure to UV (sunlight) (OR 1.99, 95% CI: 1.16-3.39) and wood dust (OR 2.66, 95% CI: 1.21-5.83) and moldy food exposure (OR 1.61, 95% CI:1.11-2.35) as risk factors; while the NAT2*6B allele (recessive model OR 1.51, 95% CI :1.06-2.16; dominant model OR 1.87, 95% CI: 1.05-3.33) and homozygous genotype (OR 2.19, 95% CI: 1.19-4.03) were found to be significant among the genetic factors. After multivariate logistic regression of both environmental and genetic factors, only UV radiation exposure (OR 2.08, 95% CI: 1.21-3.58) and wood dust exposure (OR 2.08, 95% CI: 0.95-5.30) remained to be significantly associated with increasing colorectal cancer risk in the study population.Conclusion. This study demonstrated that UV sunlight and wood dust exposure play a greater role in influencing colorectal cancer susceptibility than genotype status from genetic polymorphisms of the GST and the NAT` genes.
Colorectal Neoplasms ; Polymorphism, Genetic