Antibiogram of Medical Intensive Care Unit at Tertiary Care Hospital Setting of Pakistan

Objective: To determine the frequency of micro-organisms causing sepsis as well as to determine the antibiotic susceptibility and resistance of microorganisms isolated in a medical intensive care unit. Materials and methods: This is a cross-sectional analysis of 802 patients from a medical intensive care unit (ICU) of Shifa International Hospital, Islamabad, Pakistan over a one-year period from August 2015 to August 2016. Specimens collected were from blood, urine, endotracheal secretions, catheter tips, tissue, pus swabs, cerebrospinal fluid, ascites, bronchoalveolar lavage (BAL), and pleural fluid. All bacteria were identified by standard microbiological methods, and antibiotic sensitivity/resistance was performed using the disk diffusion technique, according to Clinical and Laboratory Standards Institute (CLSI) guidelines. Data was collected using a critical care unit electronic database and data analysis was done by using  the Statistical Package for Social Sciences (SPSS), version 20 (IBM SPSS Statistics, Armonk, NY). Results: Gram-negative bacteria were more frequent as compared to gram-positive bacteria. Most common bacterial isolates were Acinetobacter (15.3%), Escherichia coli (15.3%), Pseudomonas aeruginosa (13%), and Klebsiella pneumoniae (10.2%), whereas Enterococcus (7%) and methicillin-resistant staphylococcus aureus (MRSA) (6.2%) were the two most common gram-positive bacteria. For Acinetobacter, colistin was the most effective antibiotic (3% resistance). For E.coli, colistin (0%), tigecycline (0%), amikacin (7%), and carbapenems (10%) showed low resistance. Pseudomonas aeruginosa showed low resistance to colistin (7%). For Klebsiella pneumoniae, low resistance was seen for tigecycline (0%) and minocycline (16%). Overall, ICU mortality was 31.3%, including miscellaneous cases. Conclusion: Gram-negative infections, especially by multidrug-resistant organisms, are on the rise in ICUs. Empirical antibiotics should be used according to the local unit specific data. Constant evaluation of current practice on basis of trends in multidrug resistance and antibiotic consumption patterns are essential.