Bacterial profile and antimicrobial susceptibility pattern in catheter related nosocomial infections.MS Tullu, CT Deshmukh, SM Baveja
Department of Paediatrics, Seth G. S. Medical College, Mumbai. , USA
This prospective study was carried out over a period of 6 months in the Paediatric Intensive Care Unit (PICU) of a tertiary care teaching hospital. The aim of the study was to determine the organisms causing catheter related nosocomial infections in the PICU and to study their antimicrobial susceptibility pattern. Patients with endotracheal intubation, indwelling urinary catheters and central venous catheters (CVC)/venous cutdown catheters were included in the study. Colonization of the endotracheal tube, urinary catheter related infections (UCRI) and colonization of the CVC/venous cutdown catheters was studied. E. coli was the commonest organism colonizing the endotracheal tube tip with maximum susceptibility to cefotaxime and amikacin. E. coli was also was the commonest organism causing UCRI with maximum susceptibility to nitrofurantoin and amikacin. Acinetobacter was the commonest organism colonizing the CVC/venous cutdown catheters with maximum susceptibility to ciprofloxacin. All these sites of catheter related infections considered together, E. coli and Klebsiella were the commonest nosocomial organisms. Both had maximum susceptibility to amikacin. Methicillin resistant Staphylococcus aureus (MRSA) was isolated only from one culture. All the organisms had a poor susceptibility to cefazolin and amoxycillin. A knowledge of the resident microbial flora and their antimicrobial susceptibility pattern is necessary for formulating a rational antibiotic policy in an ICU.
Keywords: Antibiotics, therapeutic use,Catheterization, Central Venous, adverse effects,Cross Infection, drug therapy,etiology,microbiology,Drug Resistance, Microbial, Human, Intubation, Intratracheal, adverse effects,Prospective Studies, Urinary Catheterization, adverse effects,
The incidence of nosocomial infections in intensive care units (ICU) is showing a rising trend mainly because of increasing invasive procedures performed in the ICU. The therapeutic interventions that have been recognized as associated with infectious complications include-indwelling urinary catheters, sophisticated life support, intravenous fluid therapy, cardiovascular prosthetic devices, implantable orthopaedic prosthesis, immunosuppressive therapy etc. Changes in the population at risk, changes in spectrum of available pathogens and an increased use of sophisticated diagnostic and therapeutic modalities including broad spectrum antibiotics have contributed to the evolution of problems with nosocomial infections.
The incidence of nosocomial infection rates in paediatric hospitals ranges from 2.8% and 6.5%. This incidence is even higher in the intensive care units. Staphylococci and gram-negative bacilli are responsible for most nosocomial infections.
Bacterial resistance to commonly used antimicrobial agents is a commonly encountered problem in the ICU. A knowledge of the common bacterial flora of an ICU and their antibiotic susceptibility pattern helps in deciding a rational antibiotic policy. The purpose of the present study was to determine the common bacteria responsible for various catheter related nosocomial infections and to study their antibiotic susceptibility pattern.
This prospective study was conducted in the PICU of a tertiary care teaching hospital over a period of 6 months.
All patients with endotracheal tubes (with or without mechanical ventilation), indwelling urinary catheters and central venous catheters/venous cut-down catheters were included in the study.
The common indications for insertion of the endotracheal tube were for maintenance of airway patency and for mechanical ventilation. The endotracheal tube was inserted and removed using strict aseptic technique. After extubation, the tip of the endotracheal tube was cut with a sterile blade and sent in a sterile tube for bacterial culture. The qualitative method used for culture included incubation in glucose broth for four hours followed by smear examination and culture on sheep blood agar and MacConkey’s agar. The culture plates were incubated overnight and examined for growth. Organisms were identified on the basis of colony characteristics and biochemical reactions. `Colonization’ of endotracheal tube was defined as positive growth from the endotracheal tube tip. ‘Nosocomial pneumonia’ was diagnosed when all of the following four criteria were met: New and persistent infiltrates more than 48 hours (not otherwise explained) appearing on chest radiograph; positive bacterial culture growth from the endotracheal tube tip; fever - temperature more than 38?C; and elevated leucocyte count (more than 10,000 cells per cubic mm).
The major indications for insertion of urinary catheters were to monitor urine output in hemodynamically unstable patients and in life threatening diseases like shock, multiorgan failure etc. The urinary catheters were inserted and removed using strict aseptic technique. After removal, the tip of each catheter was cut using a sterile blade and sent in a sterile tube for bacterial culture. Qualitative method was used for isolation of organisms from the urinary catheter tip. The urinary catheter tip was flushed with one ml of glucose broth. A loopful of the broth was taken and inoculated on blood agar and MacConkey’s agar. The plates were incubated overnight. The organisms were identified on the basis of colony characteristics and biochemical reactions. Simultaneously, a gram stained smear was also prepared. Also, urine was collected through the draining portal of the urinary catheter using aseptic precautions (after 48 hours of insertion of the catheter) and cultured. The urine culture was done by semiquantitative method using standard loop method and a colony count of more than 105 colony forming units (CFU) per ml was considered significant. ‘Colonization’ of the urinary catheter was defined as positive growth from the urinary catheter tip culture. ‘Bacteriuria’ was defined as positive culture from the urine collected through the urinary catheter (collected after 48 hours of catheter insertion). Urinary catheter related infection (UCRI) was defined as positive growth from the urinary catheter tip (colonization) and/or positive growth from urine collected through the urinary catheter (bacteriuria).
The indications for central venous catheterization/venous cut-down catheterization were- for intravenous access and/or for central venous pressure monitoring. The catheter insertion was performed under strict aseptic precautions. A short section (approximately 5 cm) of the catheter (including the area directly beneath the skin) was aseptically cut off and sent to the laboratory in a sterile tube for culture. This section of the catheter was rolled across the surface of an agar plate with sterile forceps. After overnight incubation, the colonies were counted. A positive culture was defined as more than/equal to 15 CFU3. Also, blood was collected through the venous catheter 48 hours after it’s insertion and cultured. ‘Colonization’ of the central venous catheter (CVC)/venous cut-down catheter was defined as positive growth from the tip of the catheter and/or the blood culture collected through the catheter.
The antibiotic susceptibility of the organisms (isolated from different sites) to various antibiotics was determined using the Kirby-Bauer method (disk diffusion technique) and the results were interpreted as per National Committee for Clinical Laboratory Standards (NCCLS) guidelines.
Sixty-nine patients had undergone endotracheal tube insertion, 44 patients had undergone urinary catheterization and 33 patients had undergone central venous/venous cut-down catheterization. Six patients had endotracheal tube and urinary catheter both. Eleven patients had endotracheal tube and CVC/venous cut-down catheter. Eight patients had urinary catheter and CVC/venous cut-down catheter. Eight patients had undergone all the three procedures. The mean age of the study group was 4.05 years (35 days to 12 years). The average stay in the PICU was 7.65 days per patient (two days to 39 days).
Sixty-nine patients underwent endotracheal intubation. The average number of days of intubation was 3.46 days per patient (12 hours to 15 days). The average duration of change of each endotracheal tube was 2.71 days. Colonization with bacteria was seen in 70 out of 88 endotracheal tips cultured. The total number of organisms isolated from 70 endotracheal tube tips was 96. Two organisms were isolated from each of 26 endotracheal tube tips. Nosocomial pneumonia developed in 19 out of the 69 patients. Nosocomial pneumonia developed only in patients with endotracheal tube colonization who had undergone mechanical ventilation. The organisms commonly isolated from endotracheal tube tips were- E. coli (33.33%), Klebsiella (29.16%), Pseudomonas (11.46%), Proteus (10.41%) and Acinetobacter (5.21%). The various organisms colonizing the endotracheal tube and their antibiotic susceptibility pattern are presented in [Table - 1]. E. coli was the commonest organism with maximum susceptibility to cefotaxime and amikacin.
Forty-four patients underwent urinary catheterization with insertion of 51 catheters. Average days of catheter in-situ was 5.16 days per patient (two days to 16 days). Colonization of urinary catheter tip was seen in 21 out of 51 samples (41.18%) with isolation of 28 organisms. Bacteriuria was seen in ten out of 51 samples (19.61%) with isolation of 14 organisms. UCRI was seen in 24 out of 51 samples (47.06%). The commonest organisms colonizing the tip of the urinary catheter were E. coli (46.43%), Pseudomonas (17.86%), Klebsiella (17.86%), Citrobacter (10.71%) and Acinetobacter (7.14%). E. coli colonizing the urinary catheter tip had maximum susceptibility to amikacin and cefuroxime while Pseudomonas was susceptible to quinolones and amikacin. Common organisms isolated from the urine collected through the catheter (bacteriuria) were E. coli (64.29%), Pseudomonas (14.29%), Proteus (7.14%), Citrobacter (7.14%) and Klebsiella (7.14%). E. coli thus isolated had maximum susceptibility to nitrofurantoin, amikacin and cefuroxime while Pseudomonas was maximally susceptible to amikacin. The commonest organisms causing UCRI (colonization and/or bacteriuria) and their antibiotic susceptibility pattern are presented in [Table - 2]. E. coli (52.38%) and Pseudomonas (16.67%) were the commonest organisms causing UCRI.
Thirty-three patients had CVC/venous cut-down catheter insertion. The total number of catheters inserted was 38. Each catheter was in-situ for an average of 4.95 days (two day to 14 days). Eleven catheter tip cultures were positive out of 38 samples (28.95%) with isolation of 15 organisms. The number of positive blood cultures through the CVC/venous cut-down catheters was six out of 38 (15.79%) with isolation of six organisms. ‘Colonization’ was seen in 12 out of 38 catheters (31.58%). Common organisms isolated from the catheter tip were Acinetobacter (26.67%), E. coli (20%), Klebsiella (20%), Proteus (13.33%), Pseudomonas (6.67%) and methicillin resistant Staphylococcus aureus (6.67%). Common organisms isolated from the blood culture collected through the catheter were Acinetobacter (50%), Pseudomonas (16.67%), Enterobacter (16.67%) and coagulase negative Staphylococcus (16.67%). The organisms colonizing the CVC/venous cut-down catheters and their antibiotic susceptibility pattern are presented in [Table - 3]. The commonest organisms colonizing the catheter were Acinetobacter (33.33%) and E. coli (14.29). Acinetobacter thus isolated, had maximum susceptibility to ciprofloxacin.
The commonest organisms isolated from various sites i.e. endotracheal tube tip, urinary catheter tip, urine culture collected through the urinary catheter, CVC/venous cut-down catheter tip & blood culture collected through the CVC/venous cut-down catheter together with their antibiotic susceptibility pattern are presented in [Table - 4]. E. coli and Klebsiella were the commonest organisms isolated. E. coli had maximum susceptibility to amikacin (55.36%), cefuroxime (52.62%) and cefotaxime (49.12%). Klebsiella had maximum susceptibility to amikacin (67.57%), cefotaxime (67.57%) and ciprofloxacin (62.5%). Pseudomonas was susceptible to amikacin (45%), cefotaxime (40%) and ceftriaxone (36.84%). Proteus was maximally susceptible to amikacin (76.92%), ceftriaxone (53.85%) and cefuroxime (46.15%). Acinetobacter was susceptible to ciprofloxacin (63.64%), amikacin (57.14%), cefuroxime (57.14%) and ceftriaxone (57.14%).
Nosocomial infections are becoming an increasing problem for hospitalized patients, especially in the ICU. Nosocomial infections mainly affect the lower respiratory tract, urinary tract, surgical wounds, skin and blood (bacteremia). The common interventions done in an ICU are endotracheal intubation, urinary catheterisation and insertion of central venous lines and not surprisingly, these are responsible for the nosocomial infections. Such invasive procedures are routinely done in ICU and these ICU are becoming reservoirs of multiple drug resistant bacteria. Excessive and injudicious use of broad spectrum and higher antibiotics adds to the growing problem. Also, the mortality is higher with resistant organisms like MRSA and gram negative organisms in critically ill patients.
We isolated E. coli, Klebsiella and Pseudomonas as commonest organisms colonizing the endotracheal tube. Most of the previous studies had Pseudomonas aeruginosa isolated as the commonest organism,,,. Other studies have also isolated Serratia marcescens, Staphylococcus aureus and gram negative bacilli,,.
E. coli was the commonest organism causing UCRI in our study. This is similar to the observation by Schaeffer et al and Igra et al,. Schaeffer et al isolated Staphylococcus aureus, Staphylococcus epidermidis and Streptococcus fecalis in addition.
Our study showed Acinetobacter, E. coli and Klebsiella to be commonest organisms colonizing the CVC/venous cut-down catheters. Bentley et al, Darbyshire et al, Dillon et al and Moran et al have isolated Staphylococcus as the commonest organism colonizing the CVC,,,. Moran et al have also isolated Enterococcus, Proteus and Pseudomonas. Acinetobacter was the commonest organism isolated by us. Though Acinetobacter is commonly isolated from skin and throat of healthy people, it is also known to colonize respirators, respiratory tubings and intravenous catheters in ICU. Serious and often fatal opportunistic infections caused by Acinetobacter are also known. We could isolate Staphylococcus only in two cultures - one was from the venous catheter tip with MRSA being isolated and the other was coagulase negative Staphylococcus from blood culture collected through the catheter.
A knowledge of the antibiotic susceptibility of the organisms isolated in the ICU helps to formulate an antibiotic policy for the ICU. This also avoids un-necessary use of broad-spectrum empirical antibiotics and prevents emergence of drug resistant bacterial strains.
E. coli and Klebsiella isolated from the endotracheal tube tip were maximally susceptible to cefotaxime and amikacin in our study. These would be the drugs of choice in treating nosocomial pneumonias. E. coli was the commonest organism causing UCRI in our study & had maximum susceptibility to nitrofurantoin and amikacin. Acinetobacter was the commonest organism colonizing the CVC/venous cut-down catheters as seen by us and was maximally susceptible to ciprofloxacin. We had only a single isolate of MRSA.
The susceptibility of various bacteria to cefazolin was uniformly poor in our study [Table - 1], [Table - 2], [Table - 3]. This was probably due to the overuse of cefozolin in the indoor and ICU patients. Cefazolin being easily available free of cost to the patients from poor socioeconomic class, it was being used as first line drug from the cephalosporin class and resistance emerged due to overuse of the drug.
A similar study was done by Ojha et al in postoperative patients. The commonest nosocomial isolates seen by them were - Staphylococcus aureus (methicillin resistant - 32%, methicillin sensitive - 20%), Pseudomonas (22.5%), E. coli (5.6%), Enterobacter (5.6%), other gram negative bacilli (4.0%) and miscellaneous (10%) including fungi. MRSA was susceptible to vancomycin (100%), clindamycin (80%-89%), rifampicin (63%-89%), amikacin (59%-80%) and ciprofloxacin (35%-59%). Pseudomonas was susceptible to amikacin (74%-80%), ceftazidime (66%-82%) and ciprofloxacin (53%-60%). E. coli showed susceptibility to amikacin (100%), ceftizoxime (74%-100%) and cefotaxime (73%-75%). The differences in organisms and their antibiotic susceptibility pattern as seen by Ojha et al and in our study reflects the difference in the resident flora (and their antibiotic susceptibility pattern) isolated from their post-operative wards and our PICU.
Prevention of nosocomial infections is an important aspect in the management of patients in an ICU. Rather than the use of more sophisticated catheters like antibiotic coated catheters, simpler measures of asepsis while insertion of the catheter and general improvement of hygiene can decrease the incidence of nosocomial infections. Also, reviewing antibiotic susceptibility pattern of common organisms isolated from the ICU and formulating appropriate antibiotic protocols for ICU and hospital help in managing nosocomial infections. This would prevent emergence of drug resistant strains.
Gram negative bacteria are common causes of catheter related nosocomial infections. The organisms that we isolated were uniformly sensitive to commonly used antibiotics like third generation cephalosporins and aminoglycosides. Each ICU should have an annual review of its microbial flora and their antibiotic susceptibility pattern, which would help in formulation of a rational antibiotic policy.
[Table - 1], [Table - 2], [Table - 3], [Table - 4]