Emergency presentations of melioidosis in Hospital USM : a ten-year review

Introduction Melioidosis emerged as a global problem, however, few studies have specifically described the clinical characteristics and outcomes when patients with melioidosis are treated at an Emergency Department (ED). Melioidosis is a tropical infectious disease caused by Gram-negative bacteria Burkholderia pseudomallei. The prompt diagnosis of melioidosis is a great challenge to ED physicians because of the disease’s diverse clinical presentations. The most common organ affected by this organism is lungs. However, it is very difficult to differentiate clinically the pneumonia causes by melioidosis with other organisms. It is postulated that heavy rainfall with strong winds brings buried B. pseudomallei isolates to the soil surface; thereafter, they are more likely to be inhaled at high concentrations of the bacteria resulted in severe septic shock and early mortality. This study aimed to determine the clinical characteristics and laboratory profiles of patients with melioidosis that attended ED, Hospital Universiti Sains Malaysia(HUSM), Kelantan. Methodology This is a retrospective cohort study (record review). All patients presented to ED, HUSM from January 2001 till December 2011 with positive culture for melioidosis were included in the study. Data collected via hospital’s computerized microbiology database (WHO-net program) for candidate individuals with melioidosis. Data entered and analyzed by using SPSS version 19.0 to generate descriptive and analytical statistics. Results A total of 86 patients were reviewed. Median age of the cases was 51 years old, males (79.1%) and Malays (96.5%). About 91.9% of the patients presented with fever, followed with cough (62.8%), and shortness of breath (25.6%). Majority of the organ involved were lung (65.1%), soft tissue (19.8%), liver (18.6%), and joints (11.6%). Base on symptoms, there were significant association for cough (p=0.02), shortness of breath (p=0.001) and abscess (p=0.030) with the outcomes. Even though the most common risk factors were diabetes (79.1%), followed with rainy season (55.8%) and soil exposure (36%), there was no significant associations between risk factors and outcomes. All physical findings is significantly associated with the outcomes except for temperature (p>0.05). Overall, all laboratory profiles of melioidosis patients were deranged or prolonged. The only significant association were urea (p=0.001), creatinine (p=0.001), AST (p=0.002) and PT/INR (p=0.001). In term of empirical antibiotic usage and time lag before initiation of therapy,there was no significance associations (p>0.05) with the outcomes. The absence of shortness of breath (p=0.017), systolic blood pressure (p=0.049), diastolic blood pressure (p=0.018) and PT/INR (p=0.012) were the main predictor for the patients to recover from melioidosis. Therefore, predictor model for recovery among melioidosis patient is: Recovery (z) = 5.161 (No SOB) -0.074 (SBP) + 0.186 (DBP) – 7.010 (PT/INR) The Nagelkerke R Square shows that about 76% of the variation in the outcome variable (recovery) is explained by this logistic model and the overall accuracy of this model to predict subjects to recover (with a predicted probability of 0.5 or greater) is 87.7%. Conclusion The clinical presentations and laboratory profiles can predict the outcomes of patient with melioidosis. There was no significant association between empirical usage of antibiotics and time lag before initiation of antimelioid therapy with the outcomes of melioidosis patients