Which Multidrug-Resitant Bacteria are Emerging in Patients with Hematological Malignancies?: One-Year Report

Abstract

The aim of this retrospective, observational study was to evaluate the outcomes of bacteremia attacks during neutropenic episodes caused by chemotherapy in patients with hematological cancers by assessing mortality, involved pathogens, antimicrobial therapy and treatment responses. Patients who were older than 14 years of age and developed at least one neutropenic episode after chemotherapy to treat hematological cancer between November 2011 and November 2012 were included in the study. We retrospectively collected demographic, treatment, and survival data for 68 patients with 129 neutropenic episodes. The mean age was 59.36 ± 15.22 years (range 17–80 years), and 41 cases were male. The mean Multinational Association of Supportive Care in Cancer score was 19.56 ± 9.04. A total of 37 (28 %) bacteremia attacks were recorded in 20 cases (29 %). Fatality rates were 50 % in the six cases with bacteremia caused by carbapenem-resistant Gram-negative bacteria; death occurred in two patients with carbapenem-resistant Acinetobacter baumannii and in one patient with carbapenem-resistant Pseudomonas aeruginosa. Clinical and microbiological responses were achieved using PIP-TAZ or CEP-SUL treatment in 80 % (16/20) of the cases with bacteremia caused by carbapenem-sensitive Gram-negative bacteria (CS-GNB). During 547 colonization-days in 21 (30 %) vancomycin-resistant enterococci (VRE)-colonized cases among 68 patients, vancomycin-resistant Enterococcus faecium bacteremia developed in two patients. Non-carbapenem-based therapy can cure most bacteremia attacks caused by CS-GNB in patients with hematological cancer. However, bacteremia and other infections caused by drug-resistant pathogens, such as A. baumannii, P. aeruginosa, and VRE, are a growing concern in hematological patients.

Introduction

Infection is a serious complication in patients with hematological malignancies due to immunosuppression and intensive chemotherapy. Patients who develop febrile neutropenic episodes (FNEs) due to chemotherapy for hematological malignancies should be treated as soon as possible [1]. The infectious etiology is identified in only 20–30 % of all patients with FNEs. Fever may occur and persist in 40 % of these patients despite broad spectrum antimicrobial therapy [1, 2]. The Multinational Association of Supportive Care in Cancer (MASCC) prognostic index suggests that the mortality rate is approximately 3 % if the MASCC score is >21; however, the mortality rate escalates to 36 % if the MASCC score is <15 [3]. Patients with bacteremia caused by Gram-negative bacteria were reported to have poor prognoses. Mortality rates were reported to be 18 % for patients with Gram-negative bacteremia and 5 % for patients with Gram-positive bacteremia [4]. Mortality rates are higher in cases with bacteremia caused by Staphylococcus aureus than in cases with bacteremia caused by enterococci and coagulase-negative staphylococci (CNS) [4]. Although CNS are isolated from blood cultures more frequently than other antimicrobial-resistant Gram-negative bacteria, including Pseudomonas aeruginosa, Acinetobacter baumannii, and Stenotrophomonas maltophilia, the isolation of extended-spectrum beta-lactamase (ESBL)-producing Gram-negative bacteria has steadily increased in recent years [5, 6].

The aim of this retrospective, observational study was to evaluate the outcomes of bacteremia attacks in hematological patients during neutropenic episodes by assessing mortality, involved pathogens, antimicrobial therapy and treatment responses.

Materials and Methods

Patients in the hematology ward who were older than 14 years of age and developed at least one neutropenic episode after chemotherapy to treat hematological cancer between November 2011 and November 2012 were included in the study. This study was approved by the local ethics committee. Patients who were treated for other hematological diseases (e.g., anemia, idiopathic or immune thrombocytopenic purpura, etc.) and had not been screened with rectal swab cultures for VRE colonization while residing in the hematology ward were excluded. The hematology department was equipped with 23 beds. The patient rooms were designed as single, double and four-person rooms without high-efficiency particulate air (HEPA) filters. Patients and their attendant resided in the same room and used three shared toilets in the ward. Febrile neutropenia (FN) was defined as an oral temperature >38.3 °C or two consecutive readings >38.0 °C for 2 h and an absolute neutrophil count <0.5 × 10⁹/L or a count expected to fall below 0.5 × 10⁹/L [1]. The collected data included the patients’ demographics and diagnosis, neutropenic episodes, clinical presentations and laboratory findings, antimicrobial treatments, microbiological data and outcomes. In our hospital, the management of FN was based on the clinical practice guidelines of both the Infectious Diseases Society of America (IDSA) and the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) [1, 7, 8].

Blood cultures were inoculated into BactAlert 3D bottles (bioMérieux, Marcy-L’Etoile, France). Microbiological analysis of samples, including urine, sputum, wound, conjunctive, abscess, and catheter samples, was performed by inoculation onto 5 % sheep-blood agar (Salubris Inc., Istanbul, Turkey), chocolate agar (Salubris Inc.) and MacConkey agar (Salubris Inc.). The E test method (AB BIODISK, Solna, Sweden) and the Vitek 2 automated broth microdilution method (bioMérieux) were used to perform susceptibility testing and confirmation. Clinical and Laboratory Standards Institute (CLSI 2008) breakpoints were used. Bile-esculin-azide agar plates containing 6 μg/mL of vancomycin (Becton, Dickinson and Company, Sparks, MD, USA) were used to screen the patients for vancomycin-resistant enterococci (VRE) colonization. After inoculating rectal swabs, the plates were subsequently incubated aerobically at 35–37 °C with 5–10 % CO₂ for up to 48 h. Samples from the patients were cultured at two-week intervals. VRE colonization was defined as a positive culture from a rectal swab, and colonization continued until two consecutive rectal swab cultures were negative. The invasive procedures performed in VRE-colonized patients included the placement of chemotherapy port catheter lines or central line catheters and bone marrow biopsy during follow-up. Patients and their attendants were informed about resistant microorganisms and appropriate preventive measures by a nurse and doctor for 1 h each week while residing in the ward. Patients were encouraged to use alcohol-based hand disinfectant after contact with contaminated or probable contaminated materials and zones. Patients colonized with VRE resided in the same room. Health care workers were advised to use gloves and a gown when entering the room if contact with patient bodily fluids was possible. The hospital floors were cleaned daily with sodium hypochlorite (1,000 parts per million). The use of glycopeptide and anti-anaerobic antibiotics was restricted. Central line colonisation with Gram-positive bacteria was defined as presence of 15 (CFU) or more in a semiquantitative culture of distal segment of catheter tip (approximately 5 cm) using a simple pour-plate technique in patient who did not have clinical manifestations of sepsis and apparent source of blood stream infection (BSI) with blood cultures that did not grow the same microorganism.

The therapeutic regimens that were active against P. aeruginosa included piperacillin-tazobactam (PIP-TAZ), cefoperazone-sulbactam (CEP-SUL) and PIP-TAZ in combination with ciprofloxacin (CIP), in accordance with local antibiotic resistance status. The antibacterial treatment was changed to either imipenem or meropenem or to other antibiotics effective against bacteria grown in the sample culture if the patient exhibited a persistent fever after 2 days of empirical antibiotic therapy or had clinical, laboratory and radiological findings. Vancomycin was administered according to the previously mentioned guidelines. Patients with VRE bacteremia were treated with linezolide (2 × 600 mg/day) for at least 14 days. Patients with VSE were treated with ampicillin-sulbactam (8–12 gram/day) plus gentamycin (160–240 mg/day) for at least 14 days. Patients who had bacteremia caused by carbapenem-resistant Gram-negative bacteria (CR-GNB) were treated with colimycin. The recommended dose of this drug is 2.5–5 mg/kg of colistin base a day, which equals 6–12 mg/kg of colistimethate sodium per day. The drug was administered as a monotherapy or in combination with another active antibiotic (e.g., rifampicin, aminoglycosides, tigecycline, etc.) for at least 14 days after receiving susceptibility results. Antibiotic doses were adjusted if the patient experienced hepatic or renal failure. Antifungal and antiviral treatments were not considered in this study, as they were not relevant to the aim and content of the study.

A response to treatment was defined as defervescence in the 48–72 h following the initiation of antimicrobial therapy and recovery of all laboratory findings or clinical symptoms associated with infection. In-hospital mortality during the neutropenic phase and the clinical outcomes of FNEs were the primary outcomes that were investigated in this study.

Statistical Analysis

Variables, including age and patient MASCC scores, were described as the mean ± SD and range. The overall 30-day crude mortality was calculated as death within 30 days of the development of neutropenia. The infection-related mortality rate was calculated as the proportion of all included patients who died in the neutropenic phase due to infection. The fatality rate was calculated as the proportion of patients with bacteremia caused by CR-GNB who died in the neutropenic phase due to bacteremia caused by CR-GNB.

Results

During the study period, 15 of 141 patients who were admitted to the hematology ward were excluded from the study based on the previously mentioned exclusion criteria. We retrospectively analyzed 68 patients, who experienced 129 neutropenic episodes (Table 1). The mean age was 60.01 ± 16.23 years (range 18–81 years), and 41 patients were male. The mean MASCC score was 19.46 ± 9.14 (Table 1). During 129 FNEs in 68 patients, 37 (28 %) episodes of bacteremia were recorded in 20 patients (29 %). Gram-negative bacteria caused 70 % (n: 26) of all bacteremia attacks. CR-GNB (n: 6) caused 23 and 16 % of Gram-negative bacteremia attacks and all bacteremia attacks, respectively. CR-GNB included A. baumannii (n: 4), P. aeruginosa (n: 1), and Serratia marcescens (n: 1) (Table 2). The fatality rate was 50 % among six patients with bacteremia caused by CR-GNB as a result of two cases of death associated with carbapenem-resistant A. baumannii and one case of death associated with carbapenem-resistant P. aeruginosa. The hematological cancers of those patients were acute myeloid leukemia (AML) in three cases, non-Hodgkin’s lymphoma (NHL) in one case, multiple myeloma (MM) in one case, and acute lymphoblastic leukemia (ALL) in one case. Those patients were undergoing antifungal treatment, but they were stable for invasive fungal infections at the time of death. All patients with CR-GNB were followed up and isolated in the hematology ward. Screen cultures from rectal, axillary and groin sites were negative in these patients. CR-GNB developed in these patients during carbapenem therapy to treat FNEs. Central or port line catheters were removed from patients who had bacteremia caused by CR-GNB after identification. Carbapenem-resistant isolates developed during carbapenem therapy that was administered to treat FNEs (Table 2).

Table 1.

Distribution of hematologic malignancies in the patients with febrile neutropenia (n: 68)

Hematologic malignancies	n (%)
Acute myeloblastic leukemia	37 (57)
Acute lymphocytic leukemia	17 (25)
Non-Hodgkin lymphoma	4 (6)
Multiple myeloma	1 (1)
Hairy cell leukemia	2 (3)
Mantle-cell lymphoma	1 (1)
Aplastic anemia	2 (3)
Plasma cell leukemia	2 (3)
Chronic myeloid leukemia	2 (3)
Total	68 (100)

Table 2.

Isolated bacteria that led to bacteremia in the patients with febrile neutropenia

Hematologic Malignancy	n (%)
ESBL (−) E. coli	6 (16)
Vancomycin sensitive Enterococcus faecalis	6 (16)
ESBL (−) K. pneumoniae	5 (13)
Carbapenem-resistant Acinetobacter baumannii	4 (11)
Carbapenem-sensitive Pseudomonas aeruginosa	3 (8)
ESBL (+) K. pneumoniae	2 (5)
ESBL (+) E. coli	2 (5)
Vancomycin-resistant Enterococcus faecium	2 (5)
Carbapenem-resistant S. marcescens	1 (3)
Carbapenem-resistant Pseudomonas aeruginosa	1 (3)
ESBL (−) K. oxytoca	1 (3)
Carbapenem-sensitive Serratia marcescens	1 (3)
MRSA	1 (3)
MSSA	1 (3)
MR-CNS	1 (3)
Total	37 (100)

ESBL extended spectrum β-lactamase, MRSA methicillin-resistant Staphylococcus aureus, MSSA methicillin-sensitive Staphylococcus aureus, MR-CNS methicillin-resistant coagulase negative staphylococci

Clinical and microbiological responses were achieved using either PIP-TAZ or CEP-SUL therapy in 80 % (16/20) of the cases with bacteremia caused by carbapenem-sensitive Gram-negative bacteria (CS-GNB). A treatment response was achieved by switching from non-carbapenem-based beta-lactam anti-pseudomonal antibiotics to carbapenem-based therapy in four cases with bacteremia caused by ESBL-producing K. pneumonia (n: 2), non-ESBL-producing Escherichia coli (n: 1), and ESBL-producing E. coli (n: 1).

During 547 colonization-days in 21 (30 %) colonized cases among 68 patients, vancomycin-resistant Enterococcus faecium-associated bacteremia developed in two patients (Table 2). Of the two patients with VRE bacteremia, a male patient with NHL survived and a female patient with AML died of VRE bacteremia. Vancomycin-sensitive Enterococcus faecalis (VSE) bacteremia developed in six patients, and VSE bacteremia developed in one patient (Table 2). Of those seven patients, four were male and the median age was 44 years (range 25–73). The hematological malignancies present in those patients were AML (n: 3), ALL (n: 1), MM (n: 1), NHL (n: 1), and hairy cell leukemia (n: 1). Only two VRE-colonized patients who exhibited persistent fever accompanied with distinctive clinical findings (e.g., cough, pain in the anal region, ulcerations in oral mucosa, etc.) without confirmed VRE bacteremia were cured with linezolid therapy. The placement of chemotherapy port catheter lines and bone marrow biopsy were only invasive procedures that were performed for patients colonized with VRE during follow-up. No patient underwent intra-abdominal or urinary tract implementations.

The overall 30-day crude mortality rate was 21 % (17/68), and the distribution of hematological malignancies in patients who died included AML (n: 12), ALL (n: 4), and non-Hodgkin lymphoma (n: 1). A total of 11 patients (10 %) died of infections. Probable invasive fungal infection (n = 1) and primary bloodstream infections associated with carbapenem-resistant A. baumannii (n = 2), carbapenem-resistant P. aeruginosa (n = 1), non-ESBL producing E. coli (n = 2), vancomycin-sensitive Enterococcus faecalis (n = 2), vancomycin-resistant Enterococcus faecium (n = 1), C. albicans (n = 1) and C. parapsilosis (n = 1) were recorded in the patients who died.

Discussion

Bacteremia caused by Gram-negative bacteria was observed in approximately one-third of our cases during associated neutropenic episodes. Gram-negative bacteria, which colonize the gastrointestinal tract, can cause infection after passing through impaired mucosal barriers in patients undergoing chemotherapy [9]. The frequency of bacteremia attacks among neutropenic patients was reported to be between 11 and 38 % [10]. Gram-positive bacteria were reported to predominate in bloodstream infections of neutropenic patients during the 20 years after the 1960s and 1970s [5]. Gram-negative bacteria cause 24 of 30 (80 %) bacteremia attacks of our hematological patients as being consistent with reported rates [11]. Although bacteremia caused by CR-GNB was not reported in that study, carbapenem resistance rates of up to 23 % were observed among our isolates; thus, carbapenem resistance emerged more frequently than ESBL production (15 %) in our setting. The observed carbapenem resistance rates of 23 % (6/26) were notably higher than the ESBL production rates of 15 % that were observed in our hematology ward. Carbapenem therapy is likely to cause to the selection of carbapenem-resistant Gram-negative bacteria during antibacterial treatment of hematological patients [12]. The rates of carbapenem resistance observed among our bacteremia isolates were similar to the rate of 22 % that was reported in a study from Turkey [13]. Colistin monotherapy achieved clinical responses in half of patients (3/6) with CR-GNB. Colistin was used as a monotherapy due to the presence of other drugs that can cause renal impairment. In our setting, colistin and tigecyline are not used empirically to prevent selection until CR-GNB-associated infection is documented [1]. Colistin is the drug of last resort for CR-GNB-associated bacteremia and other severe infections. Hematological patients’ long-term port catheters should be removed to prevent port pocket site infection, endocarditis, septic thrombosis, sepsis, tunnel infection, and persistence of bloodstream infection, even after 72 h of appropriate antibiotic treatment [1]. The observed fatality rates of 50 % demonstrate the severity of the situation. Most of the fatal cases (n: 12) and half of the cases with CR-GNB occurred in patients with AML in our study. The 30-day mortality rates associated with bloodstream infections were reported as 44.8 % for patients with carbapenem-susceptible P. aeruginosa bacteremia, 54.2 % for patients with carbapenem-resistant P. aeruginosa bacteremia, and 41 % for patients with carbapenem-resistant A. baumannii. [14, 15]. The severity of organ failure and leukocytosis at bacteremia onset, age, bacteremia, inadequate empiric treatment, and inadequate definitive treatment were reported to be independently associated with mortality in patients with carbapenem-resistant A. baumannii-related bacteremia [16, 17]. Carbapenem resistance is increasing worldwide among Gram-negative bacteria, leading to more fatal outcomes in hematological patients and other immunosuppressed and comorbid patients. Catheters were removed in those patients because catheter removal is recommended in case of port pocket site infection, endocarditis, septic thrombosis, sepsis, tunnel infection, or persistence of BSI, even after 72 h of appropriate antimicrobial treatment [1].

The rate of ESBL production among our bacteremia isolates (20 %) was lower than those reported in other studies from Turkey, which reported infection rates of 33 and 25 % [11, 13]. While 80 % of our patients with bacteremia attacks caused by CS- GNB were treated with non-carbapenem-based beta-lactam anti-pseudomonal antibiotics, three of four (75 %) bacteremia attacks caused by ESBL-producing GNB were treated with carbapenem-based therapy. The use of beta-lactam antibiotics in combination with beta-lactamase inhibitors has been controversial in the treatment of bacteremia caused by ESBL-producing GNB yet. In a study published by Burgess et al. [18], clinical cure rates were only 55 % with piperacillin-tazobactam-based therapy. Mortality is more likely in patients with comorbidities, patients over the age of 75 years, patients residing in the ICU at bacteremia onset, patients with a non-urinary tract infection bacteremia focus and patients admitted with severe sepsis [19].

VRE colonization rates among our patients (30 %) were higher than those reported among patients with hematopoietic stem cell transplantation (HSCT) (27.5 %) and among patients with cancer (4.7 %) [20, 21]. The unfavorable conditions in our ward (e.g., shared toilets, housing of patients with their attendants, etc.) are likely to be associated with high VRE colonization rates. The observed VRE bacteremia attack rates of 9 % (2/21) among VRE-colonized patients were lower than previously reported rates, with VRE infection rates of 30 and 33 % reported in the previously mentioned studies [22]. Less invasive procedures and mild-moderate mucositis were likely to reduce the development of VRE bacteremia. Invasive devices, broad spectrum antibiotic use (i.e., vancomycin, cephalosporins, and anti-anaerobic agents) and the severity of illness were reported as independent risk factors for VRE infection [23]. Moreover, mucositis and increasing mucositis were reported to be independent risk factors for VRE-related bloodstream infection (BSI) [24]. Mortality rates were 33 % (2/6) and 50 % (1/2) in VSE and VRE bacteremia attacks, respectively. The mortality attributable to VRE bacteremia was reported to be 37 %, and the risk ratio for death was reported to be 2.3 [25]. Among our patients with VRE and VSE bacteremia, higher mortality rates were more likely to be associated with severe sepsis and septic shock compared to reported rates [25]. Antibacterial agents that act against VRE should be initiated in hematological patients who are colonized with VRE and exhibit persistent fever and hemodynamic instability, in accordance with recommended guidelines [1, 8].

Methicillin-resistant S. aureus (MRSA), methicillin-sensitive S. aureus (MRSA) and methicillin-resistant CNS bacteremia attacks were achieved with antibacterial therapy. MRSA infections were reported to increase in patients with FN [26, 27]. Long-term catheters that are implemented to administer chemotherapy predispose patients to MRSA bacteremia and catheter infections. Patients who are admitted from health care facilities or nursing homes should be screened for MRSA colonization [28, 29].

Non-carbapenem-based therapy can cure most bacteremia attacks in hematological patients that are caused by CS-GNB. However, bacteremia and other infections caused by drug-resistant pathogens, such as A. baumannii, P. aeruginosa, and VRE, are a growing concern in hematological patients. The outcomes of this study suggest that infection control measures should be taken not only in hematology wards but also in all health care facilities with antimicrobial stewardship.