Clinical outcomes after combination treatment with ceftazidime/avibactam and aztreonam for NDM-1/OXA-48/CTX-M-15-producing Klebsiella pneumoniae infection

Abstract

Sir, The growing spread of bacteria producing carbapenemases, such as the New Delhi MBL (NDM),1 has created an urgent need to identify effective therapeutic options that can treat serious infections caused by these XDR bacteria. Ceftazidime/avibactam has been successfully used to treat infection caused by carbapenem-resistant Enterobacteriaceae;2,3 however, the combination lacks activity against strains producing NDM. These carbapenemases remain susceptible to aztreonam, although most MBL-producing isolates also harbour ESBLs or other β-lactamases that confer resistance to aztreonam.4 The combination of aztreonam and avibactam has demonstrated potent in vitro activity against MBL-producing Enterobacteriaceae including those isolates that also carry other β-lactamases.5 However, this combination is currently in clinical development and unavailable for clinical use. In our hospital, we experienced an outbreak of Klebsiella pneumoniae ST147 in late 2015 caused by an XDR strain producing NDM-1 + OXA-48 + CTX-M-15 (KP-HUB-ST147).6 Herein, we retrospectively review the outcomes of 10 patients treated with ceftazidime/avibactam plus aztreonam between January 2016 and June 2017 at Bellvitge University Hospital, Barcelona, Spain. All patients were attended by an infectious diseases physician after informing the patient about the need for using the combination therapy and obtaining oral consent. As ceftazidime/avibactam was still not approved by the Spanish Medicines Agency, this antibiotic was obtained on a compassionate basis. This study had the approval of the local Ethics Committee (reference: EPA033/17). Clinical success was defined as survival and the absence of recurrence at day 30 after the onset of the infection, with resolution of the signs and symptoms of infection, and sterilization of site-specific cultures within 7 days of starting combination therapy.7 To check for long-term recurrence, patients were followed up for 90 days after the onset of infection. Recurrence was defined as a new positive culture with the reappearance of clinical signs and symptoms of infection after clinical success. The MICs of antibiotics were determined via microdilution, using a commercial NC53 panel by the MicroScan system (Beckman Coulter, Inc.) and a DKMGN panel by the Sensititre™ system (TREK Diagnostic Systems Ltd). The synergistic activity of ceftazidime/avibactam and aztreonam was tested in vitro, using disc diffusion [ceftazidime/avibactam (30/20 μg) and aztreonam (30 μg)]. All isolates showed synergistic activity. MIC clinical breakpoints were defined according to EUCAST.8 Ten patients received ceftazidime/avibactam and aztreonam; five of them (50%) were receiving immunosuppressant therapy (three were liver recipients, one was a kidney recipient and one was receiving long-term corticosteroid therapy). In total, five (50%) infections were bacteraemic (four were secondary and one was primary). All infections were caused by a KP-HUB-ST147, which was resistant to β-lactam antibiotics, including aztreonam (MIC >32 mg/L), ceftazidime/avibactam (MIC >16/4 mg/L), meropenem (MIC >16 mg/L), imipenem (MIC >16 mg/L), aminoglycosides, fluoroquinolones and trimethoprim/sulfamethoxazole, and only showed intermediate susceptibility to tigecycline (MIC = 2 mg/L). In four patients, the strain was resistant to colistin (MIC >8 mg/L). Clinical success was achieved in 6 of the 10 patients (60%). Failures were due to death (n = 3) and recurrence (n = 1). Two of the six patients with clinical success (33.3%) had a recurrence within 90 days, but these were patients with structural anomalies at the site of infection. One infection was a bacteraemic cholangitis in a liver recipient and the other was a febrile urinary tract infection in a patient with radical cystoprostatectomy with Studer reconstruction. Recurrences occurred 30 days and 12 days after the antibiotic therapy had been withdrawn. The 30 day mortality rate was 30% (3/10) and the median number of days to death was 20 days (range, 17–21 days). None of the deaths was considered related to the infection. Moreover, there were no adverse events related to the combination therapy during follow-up. Table 1 summarizes patient characteristics. Characteristics and outcomes of patients treated with ceftazidime/avibactam and aztreonam M, male; F, female; cUTI, complicated urinary tract infection; CI, continuous infusion; EI, extended infusion. SOFA was recorded at onset of the infection. They included targeted therapies to treat the infection. Patient required surgical drainage. Patient required a ureteral stent placement (JJ stent). Patient received concomitantly inhaled colistin. Patient received only 3 days of therapy because he had improved after the catheter removal. Blood cultures at the end of therapy were negative. Patient received concomitantly intravenous tigecycline and inhaled colistin. Characteristics and outcomes of patients treated with ceftazidime/avibactam and aztreonam M, mal