The characteristics of cefepime-resistant Pseudomonas aeruginosa bacteremia and the risk factors for mortality

Ting-Yi Su1, Jung-Jr Ye1, Po-Chang Hsu1, Hsuan-Feng Wu1, Ju-Hsin Chia2,3, Ming-Hsun Lee1*

1 Division of Infectious Diseases, Department of Internal Medicine, Chang Gung Memorial Hospital at Linkou, Chang Gung University College of Medicine, Taoyuan, Taiwan

2 Department of Laboratory Medicine, Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan

3 Department of Medical Biotechnology and Laboratory Science, Chang Gung University, Taoyuan, Taiwan

*Correspondence to Dr. Ming-Hsun Lee,MD,PhD

Division of Infectious Diseases, Department of Internal Medicine, Chang Gung Memorial Hospital at Linkou, 5 Fu-Shin Street, Gueishan 333, Taoyuan, Taiwan.

Tel: +886 3 328 1200 ext 8450

Fax: +886 3 328 9410

E-mail:

Running Title

Cefepime-resistant Pseudomonas auriginosa bacteremia

Key Words

Cefepime-resistant; Pseudomonas aeruginosa; bacteremia.

Abstract

Background/Purpose

Cefepime was an important treatment option for severe Pseudomonas aeruginosa infection, and increasing resistance of cefepime in P. aeruginosa posed a great therapeutic challenge. This study is aimed to investigate the characteristics of cefepime-resistant P. aeruginosa (FRPa) bacteremia and identify the risk factors for mortality.

Methods

A restrospective study of 78 patients with FRPa bacteremia collected from January 2006 to December 2011.

Results

Of the 78 patients, 46 (59.0%) were man, and the mean age was 72.2 years. The mean time interval between admission and FRPa bacteremia occurrence was 45.8 days. While FRPa bacteremia occurred, 46 patients (59.0%) hand ventilator use, and 45 (57.7%) stayed in intensive care unit. The mean Pittsburgh score was 5.0 and Charlson index score was 6.0. Forty-two patients (53.8%) had identified sources of bacteremia, and pneumonia was the most common identified source (28/42, 66.7%). Most FRPa blood isolate had multiple drug resistance (55.1%). The 30-day mortality rate was 65.4%. Forty-one patients (52.6%) with appropriate antibiotic therapy within 72 hours after FRPa bacteremia occurrence had a significantly higher 30-day survival rate compared to the others (48.8% versus 18.9%). In multivariate analysis, higher Pittsburgh score was an independent risk factor for the 30-day mortality, and appropriate antibiotic therapy within 72 hours was an independent protecting factor.

Conclusion

FRPa bacteremia occurred in patients with high disease severity and complicated clinical conditions. Disease severity and the timing of appropriate antimicrobial therapy initiation were associated with mortality.

Introduction

Pseudomonas aeruginosa is one of the leading causes of nosocomial infections,1,2 and infections caused by P. aeruginosa are often life threatening.3 Emergence of antibiotic resistance was increasingly reported in P. aeruginsa,4,5 and the resistant strains of P. aeruginosa were associated with increased mortality, cost, and hospital stay.1,6,7

The clinical blood isolates of P. aeruginosa from our institution had the second lowest resistant rate to cefepime (7.8%), second only to amikacin (2.3%). Therefore, cefepime became a very important empirical treatment option for patients with suspected P. aeruginosa infections in our clinical practice, particularly for the cases with high disease severity. However, Akhabue et al.8 had described the trend of increasing cefepime-resistance in P. aeruginosa, and the emergence of cefepime-resistant strains might pose a great therapeutic challenge. This study is aimed to investigate the characteristics and outcomes of cefepime-resistant P. aeruginosa (FRPa) bacteremia and identify the risk factors for mortality.

Materials and methods

Setting

This retrospective study was conducted at the Chang Gung Memorial Hospital (CGMH)-Linkou, a 3715-bed university-affiliated tertiary-care medical centre with 308 intensive care unit (ICU) beds in northern Taiwan. A central microbiology laboratory is responsible for processing all clinical specimens using computer-assisted microbiology laboratory databases. This study was approved by a research ethics committee (institutional review board) of CGMH- Linkou (Number: 101-0929B).

Study design and patients

From January 2006 to December 2011, all the hospitalized patients with an age ≥ 18 years who had at least one blood culture positive for FRPa accompanied with symptoms and signs suggestive of systemic infection were enrolled. The symptoms and signs of infection included at least two of the following clinical characteristics: (1) body temperature less than 36°C or greater than 38°C; (2) heart rate greater than 90 beats per minute; (3) tachypnea greater than 20 breaths per minute; or, an arterial partial pressure of carbon dioxide less than 32 mmHg; (4) white blood cell count less than 4000 cells/mm³ or greater than 12,000 cells/mm³; or the presence of greater than 10% immature neutrophils. If a single patient had two or more episodes of FRPa bacteremia, only the first episode was included.

Microbiology

Blood cultures were processed in the clinical microbiology laboratory, using the automated blood culture system (BACTEC 9240 system; Becton Dickinson Diagnostic Instrument Systems, Sparks, MD, USA). The isolated organisms that grew on culture were identified according to routine bacteriological procedures. Antibiotic susceptibility testing was determined by the disk diffusion method of the Clinical and Laboratory Standards Institute (CLSI). The antibiotic disks (BD Microbiology Systems, Cockeysville, MD, USA) for P. aeruginosa included amikacin, gentamicin, ceftazidime, cefepime, piperacillin, piperacillin-tazobactam, imipenem, meropenem, and ciprofloxacin. Cefepime non-susceptibility of P. aeruginosa was initially screened with disk diffusion testing, and was then confirmed by the E-test method according to the manufacturer’s instructions (AB Biodisk, Solna, Sweden). The minimal inhibitory concentration (MIC) value for cefepime resistance was 16 μg/mL (intermediate resistance) or ≥ 32 μg/mL (resistance). Multidrug-resistant (MDR) FRPa was defined as FRPa with resistance to three or more of the following four classes of agents: antipseudomonal carbapenems, antipseudomonal beta-lactams (penicillins and cephalosporins), aminoglycosides, and fluoroquinolones.

Data collection and definition

Data on age, sex, concomitant diseases, and clinical characteristics of the patients with FRPa bactremia were gathered by reviewing the inpatient medical records. The concomitant diseases included hypertension, end stage renal disease, diabetes mellitus, cerebral vascular accident, liver cirrhosis, chronic pulmonary disease, and malignancy. Central venous catheter (CVC) and ventilator use, intensive care unit (ICU) stay, and the time interval between admission and occurrence of FRPa bacteremia were recorded.

On the day when FRPa bacteremia occurred, Pittsburgh bacteremia score was calculated as follows: (1) ear temperature: 2 points for a temperature of ≦35°C or ≥40°C, 1 point for a temperature of 35.1–36.0°C or 39.0–39.9°C, and 0 point for a temperature of 36.1–38.9°C; (2) hypotension: 2 points for an acute hypotensive event with decreases in systolic and diastolic blood pressure of >30 and >20 mm Hg, respectively, use of intravenous vasopressor agents, or systolic blood pressure <90 mm Hg; (3) receipt of mechanical ventilation: 2 points; (4) cardiac arrest: 4 points; and (5) mental status: alert, 0 point; disoriented, 1 point; stupor, 2 points; and comatose, 4 points.

On the day when FRPa bacteremia occurred, Charlson comorbidity index was calculated as follows: (1) myocardial infarct, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic pulmonary disease, connective tissue disease, ulcer disease, mild liver disease, diabetes; 1 point each; (2) hemiplegia, moderate or severe renal disease, diabetes with end-organ damage, any tumor, leukemia, lymphoma; 2 points each; (3) moderate or severe liver disease: 3 points; (4) metastatic solid tumor and AIDS; 6 points each.

The sources of bacteremia were determined according to the medical records, image studies, surgical findings and microbiological evidence. The sources of bacteremia were categorized into lower respiratory tract, urinary tract, skin and skin structure, central catheter associated bloodstream infection (CABSI) and intra-abdominal infection. FRPa bacteremia with multiple sources was defined as FRPa bacteremia with at least two identified sources. If there was no identified source, it was categorized as primary FRPa bacteremia. Polymicrobial bacteremia was defined as bacteremia with blood cultures growing pathogens in addition to FRPa.

Treatment and outcome

Appropriate antibiotic therapy was defined as the antibiotic therapy subsequently proved to be effective in-vitro against the FRPa blood isolates. The time interval between occurrence of FRPa bacteremia and appropriate antibiotic therapy was recorded. Combination therapy for FRPa bacteremia was defined as concurrent use of at least two classes of antipseudomonal antimicrobial agents. Finally, 30-day mortality was recorded.

Statistical method

All statistical analyses were performed using the Statistical Package for the Social Sciences for Windows (Version 18.0; SPSS Inc., Chicago, IL, USA). Categorical variables were compared using χ2 test or Fisher exact test, as appropriate. Continuous variables were compared by Mann-Whitney U test. Variables with a p value < 0.1 in univariate analysis were included in a multiple logistic regression model using backward stepwise method to identify the risk factors for the 30-day mortality. Adjusted odd ratios (ORs) and 95% confidence interval (CI) were calculated. The survival curve was plotted by means of the Kaplan-Meier method, and the log rank test was used to compare univariate survival distribution. All tests were two-tailed, and a p value of < 0.05 was considered significant.

Results

Enrolled patients and clinical characteristics

Between January, 2006 and December 2011, 1323 blood isolates of P. aeruginosa were identified, including 103 FRPa blood isolates. Finally, 78 adult patients with FRPa bacteremia were enrolled in the study, and their demographic and clinical characteristics were shown in Table 1. Of the 78 patients, 46 patients (59.0%) were men, and the mean age was 72.2 years. The most common concomitant disease was hypertension (52.6%), followed by end stage renal disease (50%), and solid organ malignancy (38.5%). The time intervals between admission and occurrence of FRPa bacteremia ranged from 0 to 227 days with the mean of 45.8 days. In 58 patients (74.4%), FRPa bacteremia occurred more than one week after admission. Among the other 20 patients with FRPa bacteremia occurring within one week after admission, 17 (85%) and 2 (10%) had prior hospitalizations in our hospital in one and two months before the hospitalization episodes with FRPa bacteremia respectively, and 1 (5%) was transferred from a local hospital.

Sixty-five patients (83.3%) had CVC use, 46 patients (59%) had ventilator use, and 45 patients (57.7%) stayed in ICU when FRPa bacteremia occurred. The mean Pittsburgh score was 5.0, and mean Charlson comorbidity index was 6.0. Thirty-six patients (46.2%) hand primary bacteremia, and the other 42 patients (53.8%) had indentified sources for FRPa bacteremia. The most common source of bacteremia was lower respiratory tract infection (28/42, 66.7%), followed by urinary tract infection (11/42, 26.2%). Ten patients (12.8%) had FRPa bacteremia with multiple sources, and 28 patients (35.9%) had polymicrobial bacteremia. Forty-three patients (55.1%) had MDR-FRPa bacteremia.

Treatment and outcomes

Eighteen (23.1%), 28 (35.9%), and 41 (52.6%) patients had appropriate antibiotic therapy within 24, 48, and 72 hours after occurrence of FRPa bacteremia, respectively. Floroquinolones were the most common regimens used as appropriate antibiotic therapy (23.1%), followed by amyloglycosides (20.5%) and imipenem-cilastatin (20.5%). Thirteen patients (16.7%) had combination therapy. The most common co-administered antimicrobial agents were amyloglycosides (8/13, 61.5%). Finally, the 30- day mortality was 65.4% (Table 1).

Co-resistance of FRPa blood isolates

The co-resistance to anti-pseudomonal agents other than cefepime was shown in Figure 1. The FRPa blood isolates in this study had the lowest co-resistance rate to amikacin (14.1%), and the co-resistance rates were high for ceftazidime (78.2%) and piperacillin (87.2%). About half of the isolates were simultaneously resistant to ciprofloxacin (55.1%) and imipenem-cilastatin (50%).

Risk factors for the 30-day mortality of FRPa bacteremia

The cumulative survival rate at 30 days revealed the patients with appropriate antibiotic therapy within 72 hours had a significantly higher survival rate than those with delayed initiation of appropriate antibiotic therapy , and the survival curves showed significant difference (48.8% versus 18.9%, p = 0.011) (Figure 2). In urivariate analysis (Table 2), the patients with 30-day mortality had higher rates of renal insufficiency (60.8% versus 29.6%, p = 0.009) and ICU stay (68.6% versus 37.0%, p = 0.007), higher Pittsburgh score (mean ± standard deviation: 5.7 ± 3.5 versus 3.6 ± 2.9, p = 0.007), and a lower rate of appropriate antibiotic therapy within 72 hours (41.2% versus 74.1%, p = 0.006) while comparing to the survivals. In multivariate analysis, higher Pittsburgh score was an independent risk factor for the 30-day mortality (adjusted OR, 1.261; 95% CI, 1.507-1.504; p = 0.010), and appropriate antibiotic therapy within 72 hours was an independent protecting factor. (adjusted OR, 0.252; 95% CI, 0.082-0.779; p = 0.017).

Discussion

Our enrolled patients were aged and had complicated co-morbidities. The FRPa bacteremia episodes were health care associated or hospital acquired. The disease severity was high. Ventilator use and ICU stay was common. Akhabue E. et al.8 had recognized that the emergence of FRPa was associated with ICU stay, transfer from another facility, and multiple concurrent illness. Prior studies also reported that MDR-P. aeruginosa bacteremia was usually nosocomial-onset9, and associated with longer hospitalization,9,10 mechanical ventilator use and ICU stay.11,12

In addition to cefepime, most FRPa blood isolates had multiple drug resistance to most currently available anti-pseudomonal agents. The condition increased the difficulty on empiric antibiotic choice. Lodise et al.13 had approved that the number of drug resistance to P. aeruginosa bacteremia were significantly related to delayed appropriate antibiotics use, and only about half of our patients had appropriate antibiotic therapy in the first three day of FRPa bacteremia. Our study showed that the timing of appropriate antibiotic initiation was independently related to mortality. In Gram-negative bacilli bacteremia, inappropriate initial antimicrobial therapy was approved independently associated with increased mortality.14 Micek ST.15 also revealed that hospital mortality was statistically greater for patients receiving inappropriate initial antimicrobial treatment in P. aeruginosa bacteremia. However, delay appropriate antibiotics were not approved independently associated with mortality in prior studies with MDR-P. aeruginosa bacteremia, possibly due to these authors put both multidrug resistance and appropriate empirical therapy into the same multivariate analysis model and these two factors had strong correlation.9,10,16Above studies could not clarify the influence of early appropriate antibiotic therapy to the mortality among highly-resistance P. aeruginosa bacteremia patients. Our study had clearly pointed out the timing of appropriate antibiotic therapy had strongly related to FRPa bacteremia patients’ outcome.

The other independent risk factor for the 30-day mortality was Pittsburgh score, which was reflecting the high disease severity in our patient group. Our study reported a high 30-day mortality rate of 65.4%, which might be associated with the high disease severity in addition to delayed initiation of appropriate antibiotic therapy. In previous studies, the mortality rates of P. aeruginosa bacteremia ranged from 25.2 to 58.8%.17-21 and disease severity or intensive care were associated with unfavorable outcomes.18,19,22 In studies of MDR-P. aeruginosa bacteremia, critical illness and ICU admission were usually associated with increased mortality rates.10-12 However, in our study, mortality did not change by the factors including concomitant diseases and primary infection sources. These results were probably due to our FRPa bacteremia patients had similar background with multiple concomitant conditions, complicated infections and prolonged hospitalization. Though, after eliminating the compounding factors, the importance of disease severity and early initiating effective antimicrobial agents were protruded.