Bloodstream infections and antibiotic sensitivity pattern in intensive care unit

Introduction

Bloodstream infection (BSI) remains one of the most important cause of morbidity and mortality globally.1 Bloodstream infection (BSI) is defined by a positive blood culture in a patient with systemic signs of infection and can be either secondary to a documented source or primary without identified origin.2 A wide spectrum of organisms have been described to cause BSI, and this spectrum is subject to geographical differences.3 BSI are among the most difficult problems confronting clinicians managing intensive care unit (ICU) patients. Reported infection incidence is 28% in ICUs.3 Septicaemia accounts for 19% of total ICU infections, being third after urinary and respiratory infections. Excessive use of broad-spectrum antibiotics, patients being immunocompromised, the use of indwelling catheters and a multiplicity of invasive procedures make ICU patients most susceptible to colonisation by these highly resistant pathogens.4

Inappropriate use of antibiotics in treatment of BSI increases the mortality of patients and increases the risk of emergence of drug resistance strains. These organisms cause prolonged patient admission, increased healthcare costs and higher morbidity and mortality rates.1 The prevalence of antibiotic resistance in blood borne isolates is gradually increasing and is a worldwide concern.3 Thus, regular surveillance of BSI is needed. In India, where the burden of infectious disease is highest in the world, the inappropriate and irrational use of antimicrobial lead to an increase in the development of antimicrobial resistance (AMR). Furthermore, poor financial conditions, inadequate infrastructure, a high burden of disease and unregulated over-the-counter sales of cheap antibiotics have amplified the crisis of AMR in India.5,6 Awareness of hospital-specific baseline microbial resistance protects against irrational use of antibiotics. This may help progress towards preventing the spread of antibiotic resistance and could be termed proper antibiotic stewardship.2,4,6 Our study was undertaken to identify the various organisms causing BSI in our hospital ICU and to evaluate the antimicrobial susceptibility pattern of isolated strains.

Materials and methods

Our prospective observational study was conducted over a period of six months from August 2018 to January 2019 at Mahatma Gandhi Medical College and Hospital, Jaipur which is a tertiary level teaching institute. All samples were collected following strict aseptic precautions (thorough cleaning of the venous site with 70% alcohol, followed by povidone iodine and then subsequently with alcohol (Triple Swab technique) preferably from two different sites (20 min apart) wherever possible). The rubber cap of each culture bottle-containing brain heart infusion broth was immediately cleaned with 70% alcohol, and the collected blood samples (average 8 ml per site) were immediately inoculated into commercially supplied BD BACTEC (Becton Dickinson, automated blood culture system) aerobic blood culture vials containing 0.025% of sodium polyethanol sulfonate as anticoagulant. In paediatric cases, 1–2 ml of blood was collected and inoculated into BD BACTEC paediatric blood culture bottles. After collection, these bottles were immediately incubated into BD BACTEC FX-40 fully automated blood culture system for detection of growth in blood culture, which gives colour-coded beep alarms on positive detection. At this point, a Gram-stained smear was made, and the microorganism was identified on microscopic screening. Further, the bottle blood was sub-cultured on blood agar and MacConkey agar subsequently and incubated overnight at 37℃ following standard procedures. The colony characteristics and the Gram stain of the growth were assessed and further subjected to species identification and detection of antimicrobial susceptibility pattern on the VITEK2 (fully automated Advanced Expert Phenotypic System) by Biomeriux. For this, a 0.5 McFarland suspension of the colonies was prepared and then loaded following standard operating procedures. The antimicrobial susceptibility pattern and the extended spectrum beta-lactamase (ESBL) status (ESBLs defined as β-lactamases capable of conferring bacterial resistance to the penicillin, first-, second- and third-generation cephalosporins and aztreonam (but not to cephamycins or carbapenems) by hydrolysis of these antibiotics, and which are inhibited by β-lactamase inhibitors such as clavulanic acid)7 was determined by the latest Clinical and Laboratory Standard Institution guidelines and was not subjected to any further testing. MRSA was determined on the basis of cefoxitin sensitivity tested through the VITEK2 system. Contaminants were defined as those isolates or microbes that are presumed as introduced into the culture during either specimen collection or processing which are not routinely known to be pathogenic for the patient, such as diptheroids (any bacterium of Corynebacterium spp. including the diphtheria bacillus, especially one that does not cause disease) or Gram-positive bacilli (e.g. Bacillus spp., viridans group streptococci, Propionibacterium spp. and Micrococcus spp).8

Results

During the study period, a total of 1508 (1242 adult and 266 paediatric) samples were received from 887 suspected cases of septicemia admitted in the ICU. Of these, 350 (23.2%) were positive for bacterial growth and yielded microbial growth. Of these, 327 (93.4%) showed mono-microbial growth and 4 (1.2%) showed growth of two different pathogens, while 19 (5.4%) were deemed contaminants. Patients’ mean age was 35.8 years (range 1–90) of whom 68% were males.

Among the true positives, Gram-negative bacterial isolates were more common in 148 (45.2%) than Gram-positive: 143 (43.9%). The two most common microorganisms isolated were Staphylococcus haemolyticus (coagulase negative staphylococci (CONS)) 60 (18.34%) followed by Escherichia coli 37 (11.36%) (Table 1).

OPEN IN VIEWER

Table 1.

Organisms isolated during the period of study.

Organism isolated	N (%)
Staphylococcus haemolyticus	60 (18.3)
Escherichia coli	37 (11.4)
Candida spp.	34 (10.4)
Klebsiella pneumoniae	32 (9.8)
Staphylococcus epidermidis	30 (9.2)
Acinetobacter baumanii	30 (9.2)
Acinetobacter lwoffii	17 (5.2)
Pseudomonas aeruginosa	17 (5.2)
Enterococcus faecalis	15 (4.6)
Enterococcus faecium	15 (4.6)
Staphylococcus aureus	13 (4.0)
Others (Elizabethkingia meningoseptica, Providencia stuartii, Burkholderia cepacia, Enterococcus avium, Pseudomonas stutzeri, Staphylococcus sciuri, Staphylococcus hominis, Klebsiella oxytoca and Streptococcus agalactiae	27 (8.2)

Among the Gram-positive isolates, CONS were 100% susceptible to daptomycin, vancomycin, teicoplanin, linezolid and tigecycline. All S. aureus isolates were sensitive to vancomycin, teicoplanin, linezolid, tetracycline, doxycycline, daptomycin and tigecycline. Enterococccus sp. was 100% sensitive to cefoperazone-sulbactam, tigecycline and linezolid (Tables 2 and 3).

OPEN IN VIEWER

Table 2.

Antimicrobial sensitivity pattern of Gram-positive bacteria to various antibiotics with sensitivity percentage.

Bacteria	Penicillin	Ceftazidime	Cefepime	Cefoxitin	Cefazolin	Cefoperazone- sulbactam	Piperacillin- tazobactam	Co- trimoxazole	Ciprofloxacin	Levofloxacin
CONS	5.2	Resistant	Resistant	2.5	–	50	Resistant	30	2.8	Resistant
Staphylococcus Aureus	Resistant	Resistant	Resistant	Resistant	–	100	Resistant	60	Resistant	Resistant
Enterococcus sp.	25	Resistant	Resistant	–	Resistant	100	50	Resistant	23.07	21.4

OPEN IN VIEWER

Table 3.

Antimicrobial sensitivity pattern of Gram-positive bacteria to various antibiotics with sensitivity percentage.

Bacteria	Gentamicin	Erythromycin	Tetracycline	Doxycycline	Daptomycin	Tigecycline	Vancomycin	Teicoplanin	Linezolid
CONS	27.5	25	75	Resistant	100	100	100	100	100
Staphlococcus Aureus	80	25	100	100	100	100	100	100	100
Enterococcus sp.	(High genta) 33.3	Resistant	37.5	Resistant	100	100	76.9	84.6	100

Among the Gram-negative bacteria, E. coli and Acinetobacter sp. were completely sensitive to colistin and tigecycline. Klebsiella sp. showed 100% sensitivity only towards colistin, and Pseudomonas sp. was 100% susceptible to colistin, minocycline and tigecycline (Tables 4 and 5).

OPEN IN VIEWER

Table 4.

Antimicrobial sensitivity pattern of Gram-negative bacteria (in %).

Bacteria	Ampicillin	Ceftazidime	Cefepime	Cefuroxime	Cefotaxime	Cefazolin	Ceftriaxone	Amoxicillin- clavulnic acid	Piperacillin- tazobactam	Ticarcillin- clavulanate	Cefoperazone- sulbactam	Co- triamoxazole
E. coli	Resistant	Resistant	14.28	10	Resistant	Resistant	13.3	13.3	53	Resistant	60	29.4
Klebsiella sp.	Resistant	33.3	28.5	22.2	Resistant	Resistant	18.2	33.3	35.7	Resistant	46.2	28.5
Aceinatobacter sp.	50	13.3	–	Resistant	–	–	Resistant	Resistant	Resistant	26.6	58.8	23.5
Pseudomonas Sp.	–	43.8	43.8	–	–	–	–	–	71.4	48.2	71.4	Resistant

OPEN IN VIEWER

Table 5.

Antimicrobial sensitivity pattern of Gram-negative bacteria (in %).

Bacteria	Ciprofloxacin	Levofloxacin	Gentamicin	Amikacin	Doxycycline	Tetracycline	Minocycline	Imipenam	Meropenam	Ertapenam	Tigecycline	Colistin
E. coli	Resistant	5.8	50	80	30	60	60	76.4	23.2	73.3	100	100
Klebsiella sp.	50	50	61.5	66.6	Resistant	–	70	53.8	50	50	87.5	100
Aceinatobacter sp.	33.3	33.3	5.8	Resistant	–	–	87.5	12.5	Resistant	29	100	100
Pseudomonas Sp.	50	50	50	42.8	–	–	100	50	50	–	100	100

Besides the bacterial isolates, there were 10.90% fungal isolates which belonged to Candida tropicalis and Candida parapsilosis predominantly. All these isolates were found to be 100% susceptible to caspofungin, flucytosine and amphotericin B while they showed 79.4% susceptibility towards azoles.

Discussion

Blood culture positivity was seen in 23.2% of samples in our study. The blood culture positivity rates depend on various factors such as volume of the blood taken from patient for culture, number of blood culture samples taken for study, the difference in blood culture systems, geographical location and epidemiological difference of aetiological agents. Furthermore, several patients already had undergone some kind of primary antibiotic treatment at peripheral health centres before reaching our tertiary care hospital, thus affecting the positivity of the blood culture. S. haemolyticus found in our study is a true pathogen and not a contaminant, as we received two blood culture samples at 30 min apart from two different sites from the same patient after using proper aseptic precaution. For CONS including S. haemolyticus, we included only those isolates which were positive from both the bottles which adds to authenticity of our observation.

There was a high incidence of methicillin-resistant staphylococcus (MRSA) (97.5%), showing its alarming high prevalence in the population. Resistance to vancomycin was high in our population, thus raising a red flag for vancomycin resistance which is developing gradually in ICUs and thus highlighting the need for strict antibiotic stewardship and antibiotic usage protocols in each ICU.

All Gram-negative isolates showed no resistance to colistin and were highly susceptible to tigecycline, but these drugs need to be used as reserve drugs with strict protocols and used only after discussion with microbiologists. Klebsiella and Pseudomonas sp. were only moderately sensitive, while Aceinetobacter was poorly sensitive, thus showing growing resistance of these pathogens to carbapenems, drugs which were considered as very effective only a decade ago.

We suspect an indiscriminate and continuous use of sub-therapeutic doses of commonly available antimicrobials.

Conclusion

S. haemolyticus and E. coli emerged as the leading pathogens causing septicemia. For Gram-positive isolates, linezolid and vancomycin were the drugs of choice, while colistin and tigecycline were the most effective drug for Gram negatives. A unit-based microbiological surveillance, timely and repeated investigation of bacterial flora of BSI, regular PDSA cycle for improvement in hospital infection control measures, monitoring of antibiotic susceptibility patterns, reinforcement of infection control procedures (especially hand washing) and appropriate isolation and barrier precautions for patients infected or colonised with resistant organisms are mandatory to preserve good antibiotic stewardship.