Hospital Acquired Infections in a Medical Intensive Care Unit Naeem Akhtar

ORIGINAL ARTICLE Hospital Acquired Infections in a Medical Intensive Care Unit Naeem Akhtar

ABSTRACT Objective: To determine the frequency of nosocomial infections and causative organisms in medical intensive care unit (ICU) patients and antimicrobial susceptibility pattern of the isolates. Study Design: A cross-sectional study. Place and Duration of Study: The Holy Family Hospital, Rawalpindi, during the period of May 2007 to April 2008. Methodology: Clinical samples from patients having any signs of site-specific infections or fever appearing anytime after 48 hours of admission into ICU were collected. The samples were cultured onto suitable culture media and bacterial isolates were identified using standard biochemical methods. Antimicrobial susceptibility testing to conventional and newer antibiotics was performed on Mueller Hinton agar using disc diffusion method. Frequency percentages were determined. Results: Bacteria or Candida spp. were isolated from 269/440 (60.1%) samples. The most frequent site of infection was respiratory tract (47.95%) followed by urinary tract (25.3%). Pseudomonas (P.) aeruginosa, Klebsiella (K.) pneumoniae, Escherichia (E.) coli and Candida spp. were the commonest organisms. The isolation rate of Gram-positive bacteria was relatively low. Majority (> 50%) of the Gram-negative isolates were resistant to many of the antibiotics tested. Relatively low resistance was only observed against amikacin (21.3%) and imipenem (26.1%). Majority (> 60%) of Gram-negative isolates were resistant to cefotaxime, ceftriaxone and ceftazidime. The isolates showed high resistance to ofloxacin (65.9%) and ciprofloxacin (73.9%). Conclusion: The high frequency of HAIs and antibiotic resistance rate, suggests that more strict infection control practices along with prescription of antibiotics after antibiotic susceptibility testing should be implemented to limit the emergence of antibiotic resistant organisms.

Key words: Hospital acquired infections. Medical intensive care unit. Antibiotic susceptibility. Prevalent microorganisms.

INTRODUCTION spp. have been reported as cause of infection in these 10-12 Patients in intensive care units (ICUs) have a higher risk patients. Recently, Gram-negative bacilli have been of acquiring hospital acquired infections (HAIs) than reported more frequently than Gram-positives in this 10 those in non-critical care areas.1 ICU-acquired infection setting. rate is five to ten times higher than hospital-acquired The ICUs are an area of considerable antibiotic use in infection rates in general ward patients.2 Considerable which antibiotic-resistant organisms are prevalent.13,14 increased mortality and major costs are associated with Their prevalence and rates of resistance can vary ICU-acquired infections.3-5 This results from the acute enormously depending on geographic location as well severity of illness of ICU patients, prolonged hospital as location among ICU types.15 For proper management stay, immunosuppression, increased use of antimicrobials of ICU-infections, it is important to have updated and frequent exposure to therapeutic procedures like knowledge about prevalence of the causative agents endotracheal intubation with mechanical ventilation, and their antimicrobial susceptibility patterns in nasotracheal intubation, intravenous lines, central institution-specific ICUs.16 There are very few published venous lines and urinary catheterization.6-9 studies on the epidemiology of HAIs in ICU patients from The most common HAIs in ICU patients are respiratory Pakistan. Therefore, this study was carried out to tract infections (RTIs), urinary tract infections (UTIs) and determine the prevalent microorganisms in medical ICU- bloodstream infections (BSIs) and are most often patients and antimicrobial susceptibility profile of the associated with the use of invasive devices.7 Both isolates to the commonly used older and newer antibiotics. Gram-positive and Gram-negative bacteria and Candida

Department of Pathology, Rawalpindi Medical College, METHODOLOGY Rawalpindi. The study was carried out at the eight-bed medical ICU Correspondence: Dr. Naeem Akhtar, House No. A-6, (MICU) of The Holy Family Hospital in Rawalpindi, from Rawalpindi Medical College Staff Colony, Rawalpindi Road, May 2007 to April 2008. The patients were observed for Rawalpindi. any signs of site-specific infections or fever appearing E-mail: [email protected] anytime after 48 hours of admission into ICU. Patients Received October 22, 2009; accepted April 14, 2010. admitted into ICU with any signs of infections were

386 Journal of the College of Physicians and Surgeons Pakistan 2010, Vol. 20 (6): 386-390 Hospital acquired infections in a medical intensive care unit excluded from the study. Clinical specimens including isolation rate of other Gram-negative bacteria, tracheal aspirates, sputum, urine, urine catheter tips, Staphylococcus (S.) aureus and methicillin-resistant S. aureus central venous line tips, blood, body fluids, pus and (MRSA), was relatively low. others were collected from the patients and were Details of the antibiotic resistance pattern in Gram- processed at microbiology laboratory of the hospital. positive and Gram-negative isolates are shown in The specimens were cultured onto suitable culture Table I and Table II respectively. None of the isolates, media like MacConkey agar, sheep blood agar, except vancomycin against Gram-positive isolates, were chocolate agar and Sabouraud’s agar. Plates were 100% sensitive to the antibiotics tested. Relatively low incubated aerobically for 24-48 hours. The isolates were resistance was only observed against amikacin (21.3%) identified by colonial morphology, gram-staining, and imipenem (26.1%). Resistance in Gram-negative biochemical tests like catalase, coagulase, oxidase, bile isolates was more marked than in Gram-positive solubility and API-10S (Biomeroux, France). isolates. Resistance to commonly used antibiotics was Antimicrobial susceptibility testing was performed on very high. Majority of the Gram-negative and to a lesser Mueller Hinton agar using disc diffusion method in extent, Gram-positive isolates showed > 50% resistance accordance with Clinical and Laboratory Standard to many of the antibiotics tested. Out of two Institute Guidelines.17 Zone sizes of each antimicrobial fluoroquinolones tested, higher resistance was agent were recorded and interpreted as resistant, observed in ciprofloxacin (73.9% and 69.2%) than intermediate or susceptible. Intermediately susceptible ofloxacin (65.9% and 61.5%) in Gram-negative and isolates were considered resistant. Susceptibility was Gram-positive isolates respectively. Among third tested against following antimicrobials according to their generation cephalosporins tested, >60% of the majority spectrum of activity against Gram-positive or Gram- of Gram-negative isolates were resistant. Among the negative bacteria. Antimicrobials tested against Gram- aminoglycosides tested, resistance was higher to positive bacteria were: ampicillin, co-amoxiclav, gentamicin than amikacin. cephradine, cephalexin, cefaclor, cefotaxime, amikacin, Table I: Antibiotic resistance pattern of Gram-positive bacterial gentamicin, tetracycline, chloramphenicol, co-trimoxazole, clinical isolates from patients admitted in medical intensive ofloxacin, ciprofloxacin, vancomycin and imipenem. care unit (MICU). Methicillin resistance in MRSA was tested using Antibiotics Organisms isolated oxacillin discs. Antimicrobials tested against Gram- S. aureus MRSA S. pneumoniae Total negative bacteria were co-amoxiclav, cefotaxime, cefta- (n=13) (n=19) (n=07) (n=39) zidime, ceftriaxone, amikacin, gentamicin, tetracycline, No resistant No resistant No resistant No resistant chloramphenicol, co-trimoxazole, ofloxacin, ciprofloxacin, (%) (%) (%) (%) imipenem and aztreonam. Ampicillin 12 (92.3) ---- 04 (57.1) 16 (80)* Co – amoxyclav 09 (69.2) 11 (57.9) 03 (42.9) 23 (59) Microsoft Excel spread sheet was used to analyze the Cephradine 03 (23.1) ---- 02 (28.6) 5 (25)* data in the form of percentages. Approval for the study Cephalexin 06 (46.2) ---- 04 (57.1) 10 (50)* was obtained from Research and Ethics Committee, Cefaclor 08 (61.5) ---- 04 (57.1) 12 (60)* Rawalpindi Medical College and Allied Hospitals, Cefotaxime 06 (46.2) ---- 03 (42.9) 9 (45)* Rawalpindi. Ofloxacin 05 (38.5) 17 (89.5) 02 (28.6) 24 (61.5) Ciprofloxacin 08 (61.5) 17 (89.5) 02 (28.6) 27 (69.2) Amikacin 01 (7.7) 01 (5.3) 01 (14.3) 03 (7.7) RESULTS Gentamicin 10 (76.9) 16 (84.2) 04 (57.1) 30 (76.9) Out of total 440 specimens received, bacteria or Candida Tetracycline 08 (61.5) 15 (78.9) 05 (71.4) 28 (71.8) spp. were isolated from 269/440 (60.1%) samples. The Chloramphenicol 02 (15.4) 10 (52.6) 05 (71.4) 17 (43.6) most frequent site of infection was the respiratory tract Co – trimoxazole 09 (69.2) 18 (94.7) 02 (28.6) 29 (74.4) Imipenem 07 (53.8) 06 (31.6) 02 (28.6) 15 (38.5) (47.95%) followed by urinary tract (25.3%), wounds Vancomycin 0 (0) 0 (0) 0 (0) 0 (0) (8.2%) and blood stream (7.06% BSI). K. pneumoniae *No (%) of isolates resistant to the tested antibiotic out of total of 20 (S. aureus=13 and (30.2%), P. aeruginosa (28.7%), and E. coli (19.4%) were S. pneumoniae=07). the commonest organisms isolated from respiratory DISCUSSION tract samples. Similarly, K. pneumoniae (25%), E. coli (23.5%), P. aeruginosa (16.2%) and Candida spp. (14.7%), HAIs are seen worldwide but are less studied and are were isolated from urinary tract samples. Organisms given less emphasis in developing countries like causing BSIs were P. aeruginosa (31.6%), E. coli (31.6%), Pakistan. This study reports a high rate of HAIs, as well Streptococcus pneumoniae (21.1%), and K. pneumoniae as high antibiotic resistance, in ICU settings of a tertiary (10.5%). These BSI organisms were isolated from care hospital in Rawalpindi. RTIs (47.95%) and UTIs cultures of blood and the central venous line tips. The (25.3%) were the commonest infections from these

Journal of the College of Physicians and Surgeons Pakistan 2010, Vol. 20 (6): 386-390 387 Naeem Akhtar

Table II: Antibiotic resistance pattern of Gram-negative bacterial clinical isolates from patients admitted in medical intensive care unit (MICU). Antibiotics Organisms Isolated

E. coli (n=53) K. pneumoniae (n=72) Citrobacter spp. (n=03) Proteus spp. (n=09) P. aeruginosa (n=74) Total (n =211) No resistant (%) No resistant (%) No resistant (%) No resistant (%) No resistant (%) No resistant (%) Co-amoxiclav 45 (84.9) 60 (83.3) 03 (100) 08 (88.9) 53 (71.6) 169 (80.1) Cefotaxime 33 (62.3) 43 (59.7) 02 (66.7) 05 (55.6) 49 (66.2) 132 (62.6) Ceftazidime 38 (71.7) 34 (47.2) 02 (66.7) 06 (66.7) 63 (85.1) 143 (67.8) Ceftriaxone 38 (71.7) 45 (62.5) 03 (100) 09 (100) 62 (83.8) 157 (74.4) Ofloxacin 40 (75.5) 38 (52.8) 02 (66.7) 04 (44.4) 55 (74.3) 139 (65.9) Ciprofloxacin 40 (75.5) 53 (73.6) 02 (66.7) 05 (55.6) 56 (75.7) 156 (73.9) Amikacin 13 (24.5) 21 (29.2) 02 (66.7) 02 (22.2) 07 (9.5) 45 (21.3) Gentamicin 26 (49.1) 33 (45.8) 01 (33.3) 05 (55.6) 40 (54.1) 105 (49.8) Tetracycline 38 (71.7) 44 (61.1) 02 (66.7) 09 (100) 53 (71.6) 146 (69.2) Chloramphenicol 34 (64.2) 42 (58.3) 02 (66.7) 05 (55.6) 44 (59.5) 125 (59.2) Co -trimoxazole 34 (64.2) 64 (88.9) 02 (66.7) 08 (88.9) 51 (68.9) 159 (75.4) Imipenem 08 (15.1) 15 (20.8) 01 (33.3) 01 (11.1) 30 (40.5) 55 (26.1) Aztreonam 17 (32.1) 05 (6.9) 02 (66.7) 05 (55.6) 45 (60.8) 74 (35.1) patients and this is in agreement with other studies from with other report from US NNIS,22 but in present study Pakistan11,12,18 and other countries.3,9 In a study this was not associated with urinary catheterization. conducted at Karachi, the frequency of RTIs and UTI In this study the most commonly used antibiotics was 21% and 44.6% respectively.12 In another study belonging to penicillins, cephlosporins, fluoro-quinolones, carried out at Hyderabad, the frequency of RTIs and aminoglycosides, co-trimoxazole and carbapenems UTIs was 30.1% and 39.1% respectively. This difference were tested against the bacterial isolates to have an could possibly be due to the difference in antibiotic update current status of the resistance pattern. The prescribing practices and variations in sample collection, higher resistance in these patients is probably due to culture and susceptibility testing practices. Most RTIs different antibiotic prescribing practices as it varies were associated with mechanical ventilation (82.2%) as enormously.23 Very high resistance rate to the most the later is a major factor for RTIs in ICUs.1,19 Higher rate commonly used antibiotics including; fluoroquinolones is of UTIs (25.3%) in these patients is obvious due to most likely due to misuse, overuse and over-the-counter chronic illnesses like stroke and diabetes. This was availability of these antibiotics. HAIs particularly in ICUs associated with urinary catheterization (67.6%) and admitted patients who are at higher risk get primary prolonged urinary catheterization is a major risk factor infection with multi-resistant organisms. This makes the for UTIs in ICU patients.15 These findings are in close treatment of these infections progressively more agreement with those reported by the U.S. National complicated16 and is also associated with increased Nosocomial Infections Surveillance (NNIS) system and patient charges or hospital costs and mortality.5 local studies.7,12,18 Relatively lower rate of BSIs (7.06%) Furthermore, a major contributing factor to this in this study as compared to other local studies 23.7% increased resistance is use of the antibiotics without and 27%,12,18 may partly be due to the reason, that culture and susceptibility testing as even basic facilities blood cultures were manually processed or primary for culture and antibiotic susceptibility testing are not BSIs are less prevalent in our setting. The relatively high available at majority of healthcare settings in our set up. incidence of HAIs observed in this study may be a Relatively lower resistance rate to imipenem and reflection of the higher severity of illness, poor nutritional amikacin is probably due to lesser use of these status, more interventions, fewer staff, poor adherence antibiotics and is in agreement with another local to aseptic measures and high proportion of mechanical study.12 Vancomycin remains the only option for MRSA ventilation.4,9 as most of the other antibiotics tested appear ineffective; Gram-negative bacteria i.e. however, chances of development of vancomycin- P. aeruginosa, K. pneumoniae resistant are there. and E. coli were the commonest organisms isolated from S. aureus RTIs, UTIs and BSIs. This is in agreement with previous In view of this present situation one approach is to opt studies from Pakistan12,20 and other countries.9,10 for newer and more expensive antibiotics like linezolid, Prevalence of other Gram-negative bacteria, S. aureus dalfopristin-quinopristin and tigecycline for Staphylococcal and MRSA is relatively low as in general Gram-negative infections and cefoperazone-sulbactam for Gram- bacterial infections are increasing in ICUs than Gram- negative infections. Better and a more cost effective positive bacterial infections.10,21 A markedly higher approach will be, use of the antibiotics after culture and isolation rate of Candida spp. from UTIs is in agreement antibiotic susceptibility testing particularly in ICU

388 Journal of the College of Physicians and Surgeons Pakistan 2010, Vol. 20 (6): 386-390

Hospital acquired infections in a medical intensive care unit

patients. The best approach would be to improve and 8. Kollef MH, Fraser VJ. Antibiotic resistance in the intensive care implement optimal infection prevention practices in ICUs unit. Ann Intern Med 2001; 134:298-314. Comment in: Ann Intern Med 2002; 136:173; author reply 173. of low resource countries and expertise9 as it is believed that one third of these hospital-acquired infections can 9. Habibi S, Wig N, Agarwal S, Sharma SK, Lodha R, Pandey RM, Epidemiology of nosocomial infections in medicine be prevented by optimal infection prevention practices.25 et al. intensive care unit at a tertiary care hospital in northern India. Trop Doct 2008; 38:233-5. CONCLUSION 10. Gaynes R, Edwards JR. National Nosocomial Infections Bacteria or Candida spp. were isolated from 60.1% Surveillance System. Overview of nosocomial infections caused samples taken from supposedly infected patients by gram-negative bacilli. Clin Infect Dis 2005; 41:848-54. Epub admitted in medical ICU. The most frequent site of 2005 Aug 16. Comment in: Clin Infect Dis 2006; 42:577-8; author infection was the respiratory tract followed by urinary reply 578. tract. P. aeruginosa, K. pneumoniae and E. coli were the 11. Wahid F, Masood N, Jafri A. Nosocomial pneumonia in commonest organisms followed by MRSA and S. aureus. mechanically ventilated patients. Pak Armed Forces Med J 2005; 3: The high frequency of HAIs and high bacterial 25-8. resistance rate in ICU patients indicate the importance 12. Rizvi MF, Hasan Y, Memon AR, Abdullah M, Rizvi MF, Saleem to study HAIs and suggest, that more strict measures S, et al. Pattern of nosocomial infection in two intensive care units regarding infection control practices along with of a tertiary care hospital in Karachi. J Coll Physicians Surg Pak prescription of antibiotics according to antibiotic 2007; 17:136-9. Comment in: J Coll Physicians Surg Pak 2008; 18:134-5. susceptibility testing should be implemented to limit the emergence of antibiotic resistant organisms. 13. Albrich WC, Angstwurm M, Bader L, Gartner R. Drug resistance in intensive care units. Infection 1999; 27:S19-23. Acknowledgement: This study is a part of research 14. Fridkin SK, Gaynes RP. Antimicrobial resistance in intensive project funded by Higher Education Commission of care units. 1999; 20:303-16. Pakistan, Islamabad. Thanks to my colleagues Mr. Yusuf Clin Chest Med Rajput, Dr. Umair Mahmood, Dr. Amber Habib, Dr. 15. Bagshaw SM, Laupland KB. Epidemiology of intensive care unit- Abubakr, Dr. Waqas, Mr. Tariq Baig, Mr. Arthur Manzoor, acquired urinary tract infections. Curr Opin Infect Dis 2006; 19: 67-71. Mr. Muhammad Mushtaq and Mr. Sajjad for their facilitation in this study in their capacity. 16. Doshi RK, Patel G, MacKay R, Wallach F. Healthcare-associated infections: epidemiology, prevention and therapy. Mt Sinai J Med REFERENCES 2009; 76:84-94. 17. Clinical and Laboratory Standards Institute. Performance 1. Richards M, Thursky K, Buising K. Epidemiology, prevalence, standards for antimicrobial disk susceptibility tests: approved and sites of infections in intensive care units. Semin Respir Crit Care standard. 9th ed. Wayne (PA): Clinical and Laboratory Standards Med 2003; 24:3-22. Institute; 2006. 2. Weinstein RA. Nosocomial infection update. Emerg Infect Dis 1998; 18. Shaikh JM, Devrajani BR, Shah SZ, Akhund T, Ishrat B. 4:416-20. Frequency, pattern and aetiology of nosocomial infections in 3. Vincent JL, Bihari DJ, Suter PM, Bruining HA, White J, Nicolas- intensive care unit: an experience at a tertiary care hospital. Chanoin, The prevalence of nosocomial infections in et al. J Ayub Med Coll Abbottabad 2008; 20:37-40. intensive care units in Europe. Results of the European Prevalence of Infection in Intensive Care (EPIC) Study. EPIC 19. Centers for Disease Control and Prevention. Guideline for prevention of nosocomial pneumonia. 1994; International Advisory Committee. JAMA 1995; 274:639-44. MMWR Recomm Rep 39:1191-236. Comment in: JAMA 1996; 275:362. 4. Chen YC, Lin SF, Liu CJ, Jiang DD, Yang PC, Chang SC. Risk 20. Izhar M, Khan S, Naqvi A. Anti-microbial resistance among Gram-negative bacteria prevalent in intensive care units. factors for ICU mortality in critically ill patients. J Formos Med Assoc Pak J 2001; 100:656-61. Surg 2001; 17:23-6. 5. Shorr AF. Review of studies of the impact on Gram-negative 21. Kohlenberg A, Schwab F, Geffers C, Behnke M, Rüden H, bacterial resistance on outcomes in the intensive care unit. Gastmeier P. Time-trends for Gram-negative and multidrug- resistant Gram-positive bacteria associated with nosocomial Critical Care Med 2009; 37:1463-9. infections in German intensive care units between 2000 and 6. Girou E, Stephan F, Novara A, Safar M, Fagon JY. Risk factors 2005. 2008; 14:93-6. Epub 2007 Nov 22. and outcome of nosocomial infections: results of a matched Clin Microbiol Infect case-control study of ICU patients. Am J Respir Crit Care Med 1998; 22. Fridkin SK, Welbel SF, Weinstein RA. Magnitude and prevention 157:1151-8. Comment in: Am J Respir Crit Care Med 1999; 159: of nosocomial infections in the intensive care unit. Infect Dis Clin 341-2. North Am 1997; 11:479-96. 7. Richards MJ, Edwards JR, Culver DH, Gaynes RP. Nosocomial 23. Monnet DL, Archibald LK, Phillips L, Tenover FC, McGowan JE infections in combined medical-surgical intensive care units in Jr, Gaynes RP. Antimicrobial use and resistance in eight US the United States. Infect Control Hosp Epidemiol 2000; 21:510-5. hospitals: complexities of analysis and modeling. Intensive care

Journal of the College of Physicians and Surgeons Pakistan 2010, Vol. 20 (6): 386-390 389 Naeem Akhtar

antimicrobial resistance epidemiology project and national patients in the United States between 1993 and 2004. J Clin nosocomial infections surveillance system hospitals. Infect Control Microbiol 2007, 45:3352-9. Hosp Epidemiol 1998; 19:388-94. 25. Haley RW, Culver DH, White JW, Morgan WM, Emori TG, Munn 24. Lockhart SR, Abramson MA, Beekmann SE, Gallagher G, VP, et al. The efficacy of infection surveillance and control Riedel S, Diekema DJ, et al. Antimicrobial resistance among programs in preventing nosocomial infections in US hospitals. Gram-negative bacilli causing infections in intensive care unit Am J Epidemiol 1985; 121:182-205.

GGGGG*GGGGG

390 Journal of the College of Physicians and Surgeons Pakistan 2010, Vol. 20 (6): 386-390