Successful treatment with cefiderocol for compassionate use in a critically ill patient with XDR Acinetobacter baumannii and KPC-producing Klebsiella pneumoniae: a case report Free

Sir,

Cefiderocol is a new siderophore cephalosporin, with potent activity against MDR Gram-negative bacilli. It is particularly active against XDR Acinetobacter baumannii (Ab) infections, for which treatment options are quite limited.^1–3

Herein, we describe a case of an adult male patient with severe H1N1 influenza complicated by ventilator-associated pneumonia (VAP) and bloodstream infection (BSI) caused by XDR Ab and carbapenemase-producing Klebsiella pneumoniae (KPC-Kp).

Nothing remarkable was in his medical history apart from favism and Aarskog–Scott syndrome.

In February 2019, due to the onset of acute respiratory failure, he was admitted to a peripheral primary ICU where he was diagnosed with H1N1 influenza virus complicated by bilateral pneumonia. The patient was immediately intubated and then transferred to our hospital where extracorporeal membrane oxygenation (ECMO) was started. Also, treatment with intravenous zanamivir and empirical antimicrobial therapy with piperacillin/tazobactam, clarithromycin and linezolid were prescribed. Zanamivir was obtained through the compassionate-use programme run by GlaxoSmithKline (Brentford, UK).

Table 1 shows SOFA scores and laboratory results during the inpatient stay.⁴

Due to both clinical and radiological improvement, ECMO was stopped after 9 days.

On day 14, the patient reported a clinical worsening (SOFA score increased by 3 points) with fever and increased levels of procalcitonin. So, antimicrobial therapy was switched to meropenem plus vancomycin, and anidulafungin. At that time, surveillance swab cultures came back positive for XDR Ab and for KPC-Kp. A broth microdilution-based antibiogram showed that the Ab was only susceptible to colistin (MIC ≤0.5 mg/L), while the Kp appeared to be susceptible to colistin (MIC ≤0.5 mg/L), gentamicin (MIC 2 mg/L) and ceftazidime/avibactam (MIC 4 mg/L).

On day 18, XDR Ab with the same pattern of resistance was also isolated from blood cultures. Considering these results, we prescribed a combination treatment with colistin, daptomycin and fosfomycin. Fever persisted after 3 days of this treatment, so fosfomycin was replaced by high-dose tigecycline, according to possible susceptibility of the Ab shown on the antibiogram (MIC 1 mg/L).⁵

On day 28, after an incomplete response to treatment, the SOFA score further worsened (by 8 points), body temperature rose to 38.5°C and procalcitonin reached 174.15 ng/mL. Moreover, onset of liver and renal failure was observed, the latter probably due to colistin treatment. Also, VAP was diagnosed following standard criteria, seen as lung infiltrates on chest X-rays.⁶

A T2Bacteria magnetic resonance assay (T2MR; T2 Biosystems Inc.) was performed on a new blood sample, still identifying Ab, but also Kp. By contrast, blood cultures performed with conventional methods came back negative. Notwithstanding the low level of agreement between the two techniques, assuming higher sensitivity of the T2MR, we decided to start a combination rescue therapy with colistin (both intravenous and aerosolized by a vibrating mesh nebulizer), ampicillin/sulbactam, ceftazidime/avibactam and rifampicin.

Meanwhile, thanks to a compassionate-use programme run by the drug manufacturer (Shionogi, Rome, Italy), we requested cefiderocol.

On day 35, despite procalcitonin levels decreasing to 6.13 ng/mL, fever persisted and lung infiltrates worsened on chest X-rays. So, on the same day, on availability, cefiderocol was initiated in association with linezolid.

Disc diffusion testing performed on Ab from blood and on Kp from surveillance swab cultures gave zones of 23 and 22 mm, respectively, against prospective 15 and 16 mm breakpoints for a 30 μg disc.

Under this treatment, patient conditions rapidly improved with resolution of fever and normalization of procalcitonin levels.

On day 44, the patient was transferred to the Unit of Infectious and Tropical Diseases where the same combination therapy was continued for an overall course of 14 days.

On day 48, chest X-rays showed complete resolution of lung infiltrates. The patient was then transferred to a rehabilitation department without antibiotic treatment and with no signs or symptoms of ongoing infections.

In the pipeline of new antibiotics, cefiderocol represents the most promising opportunity for treating infections due to XDR Ab strains.³ Colistin often remains the only drug effective in such cases, but it is limited in its clinical use by both pharmacokinetic limitations and significant toxicities.⁷

Our patient was diagnosed with a BSI secondary to VAP due to XDR Ab, so a colistin-based combination therapy was chosen, without achieving a clinical or microbiological cure.

Moreover, after about 10 days of therapy with colistin, acute renal failure emerged as a further complication. Therefore, it is reasonable to think that cefiderocol was relevant to the increased efficacy and reduced toxicity of the combined therapy, leading to complete recovery of our patient.

Also, the present case highlights the importance of the T2MR assay, because without its use we would have likely missed persistent BSI due to Ab and overlapping Kp.⁸

In conclusion, this is the first case report (to the best of our knowledge) confirming the efficacy and safety of cefiderocol monotherapy for treatment of invasive infections due to XDR Ab and KPC-Kp in a real-life setting.

Ethics

Local Ethics Committee approval was obtained both for zanamivir (22 February 2019) and cefiderocol (22 March 2019). The two drugs were administered after gaining the patient’s informed consent.

Acknowledgements

We thank the other members of the IMAGES (Integrated MAnaGEment of Sepsis) Group.

Other members of the IMAGES Group

Vincenzo Pisani, Chiara Costa, Giuseppe Greco, Valentina La Gamba, Aida Giancotti, Giorgio Settimo Barreca, Cinzia Peronace, Ovidia La Valle, Giuseppina Cimino, Paola La Torre, Antonio Gemelli, Francesco Antonio Tropea and Francesco Picicco. We thank Stefan Renstrom (Imperial College, London, UK) for kindly revising our manuscript with respect to English language and style prior to submission.

Funding

This study was carried out as part of our routine clinical practice.

Transparency declarations

None to declare.

References

Author notes