KPC - 3 Klebsiella pneumoniaeST258 clone infection in postoperative abdominal surgery patients in an intensive care setting: analysis of a case series of 30 patients

Abdominal surgery carries significant morbidity and mortality, which is in turn associated with an enormous use of healthcare resources. Sepsis is a serious complication and may be diagnosed during pre-, intra-, or postoperative periods [1, 2].

Klebsiella pneumoniae (Kp) is an emerging major pathogen in surgical settings, especially after emergency abdominal surgery [3, 4].

In 2010, the first outbreak of K. pneumoniae carbapenemase (KPC)-Kp sequence type (ST)258 was reported in ICU patients in Palermo, Italy [5‐7]. Since then, colonizations or infections with KPC-Kp have become endemic and have also been described in different healthcare settings, such as in surgical wards [8].

The aim of this study was to describe the clinical aspects of surgical KPC-Kp infections in patients who had undergone emergency or elective abdominal surgery. Risk factors for mortality and the impact of a combination therapy of colistin plus recommended regimen or higher dosage of tigecycline on the patients’ clinical course were evaluated.

During the study period, we collected 30 cases of infection caused by KPC-Kp.

Patient characteristics are shown in Table 1. The mean age of the patients was 56.6 ± 15 years, and the APACHE score on admission averaged 23.4 ±1.7. Twenty of the 30 patients came from the surgical emergency unit. Fifty percent of these patients had been hospitalized in the previous two years. Four patients had chronic obstructive pulmonary disease (COPD) as underlying medical condition and two had diabetes mellitus. Smoking habit was present in the anamnesis of 16 patients, obesity in two, alcohol abuse in four and cocaine abuse in one patient. Multiple underlying conditions were observed in 6 patients.

The most common surgical indication was rectal cancer in seven patients, followed by colon cancer in six patients. Other indications were less represented (Table 1). KPC-Kp post-surgical infections were as follows: intra-abdominal abscess in 15 cases, anastomotic leakage in eight, surgical site infection (SSI) in four and peritonitis in three patients, respectively. Figure 1 shows KPC-Kp infection in a patient with rectal cancer.

As shown in Table 1, KPC-Kp was isolated from all surgical samples as well as from broncho-alveolar lavage (BAL) in 50% of the patients who developed ventilator-associated pneumonia (VAP) during their ICU stay.

Fifteen of the 30 patients under study who had a history of previous hospitalizations proved to be colonized by KPC-Kp strains on admission to the surgical emergency unit. Four patients acquired KPC-Kp colonization during their stay in the surgical unit. Ten patients had also been exposed to carbapenems before admission.

All the KPC-Kp strains were resistant to imipenem (MIC ≥16 μg/ml), meropenem (MIC ≥32 μg/ml) and ertapenem (MIC ≥8 μg/ml). They were also resistant to amikacin (MIC ≥64 μg/ml), amoxicillin- clavulanic acid (MIC ≥32 μg/ml), cefepime (MIC ≥8 μg/ml), cefotaxime (MIC ≥8 μg/ml), ceftazidime (MIC ≥64 μg/ml), ciprofloxacin (MIC ≥4 μg/ml), levofloxacin (MIC ≥8 μg/ml), piperacillin-tazobactam (MIC ≥128 μg/ml), tobramycin (MIC ≥16 μg/ml), gentamicin (MIC ≥8 μg/ml) and trimethoprim-sulfamethoxazole (MIC ≥320 μg/ml). Before starting combined antibiotic treatment, all the strains isolated from blood samples were susceptible to colistin (MIC ≤0.5 μg/ml) and tigecycline (MIC ≤0.5 μg/ml). After starting the treatment, five strains of KPC-Kp showed tigecycline MICs of 0.8 – 1 μg/ml.

The average duration of treatment with a combination of tigecycline and intravenous colistin was 18 ± 6.5 days. Twelve patients were administered a combined treatment for more than 21 days. Two surviving patients were treated with prolonged courses of combined antibiotic treatment, with more than twenty-day intervals between courses. Twelve patients received a combination treatment of high-dose tigecycline and intravenous colistin. They included the five cases with KPC-Kp isolates exhibiting tigecycline MICs >0.5 μg/ml.

Overall crude ICU mortality rate was 40% (12 out of 30 patients). Risk factors for mortality in the ICU were assessed (Table 2). Using univariate analysis, a better outcome was associated with the presence of a surgical drainage, and a worse one with a higher APACHE II score and VAP. Treatment with high doses of tigecycline was associated with lower mortality (P = 0.005). Klapan-Meier curve showed that patients treated with high doses of tigecycline had a significant favorable outcome (log-rank test, p = 0.0035) (Figure 2).

The results of our study draw attention to two important issues: the impact of KPC-Kp bloodstream infection in ICU patients who have had abdominal surgery, and the use of standard or high doses of tigecycline in combination with polymyxins to treat infection due to KPC-Kp in patients with serious surgical complications, such as intra-abdominal abscesses [1, 2, 7]. Moreover, our study highlights additional postoperative infectious complications, such as colorectal anastomotic leakage, which can be responsible for severe sepsis [13].

Some results appear to be relevant. Firstly, in all patients, sources other than blood samples yielded KPC-Kp strains and all clinical isolates proved to belong to the pandemic clone KPC-Kp ST258. This confirms our previous reports suggesting that this strain may have a possible selective advantage over other MDR pathogens in the hospital setting [5, 6]. Moreover, about 50% of our surgical patients were colonized by KPC-Kp at admission and had a history of previous hospitalization. Ten patients had also received carbapenem therapy in the past. This is consistent with data from surveillance studies that report patients colonized with MDRGN before hospital admission [5, 8] and with previous literature data describing the prolonged carriage of multidrug resistant bacterial strains following hospital discharge [8, 14, 15].

Secondly, half of our patients had monomicrobial intra-abdominal abscess due to KPC-Kp. As in our previous report [8], this observation is consistent with those of other authors, such as Hirsch [16], who describe even higher clinical success rates when using polymyxins in combination.

In our study, we report on the use of high-dose tigecycline in patients with intra-abdominal abscess, especially in those with a less favorable clinical course and with KPC-Kp isolates showing MICs of tigecycline near the upper limit of the EUCAST susceptibility range http://​www.​eucast.​org). Since tigecycline is one of the few antibiotics which is generally active against KPC-Kp ST258 epidemic strain, optimal dosing is critical for maximal therapeutic effectiveness. There is still some debate over appropriate dosage and the most favorable pharmacokinetic/pharmacodynamic profiles in some kinds of infection, such as MDR K. pneumoniae /Acinetobacter baumannii urosepsis/bacteremia [17, 18]. Some preliminary experiences with loading doses of 200 mg and 400 mg, then decreased by 50% to achieve a maintenance dose (100 mg or 200 mg every 24 hours respectively) seem to guarantee serum concentrations (3 and 6 mg/L respectively) and urine concentrations (0.6 and 1.2 mg/L respectively) that are two and four times higher than those attained with the usual doses [18].

Tigecycline has been approved for the treatment of intra-abdominal and skin and soft tissue infections (and for community-acquired pneumonia in the USA), although a warning was recently issued by the FDA (http://​www.​fda.​gov) and the European Medicines Agency (EMA, http://​www.​ema.​europa.​eu), after clinical trials indicated a higher mortality rate in patients treated with tigecycline compared to control groups. A recent meta-analysis has confirmed increased mortality and has also indicated that, compared to other antibiotics, tigecycline has higher rates of clinical failure, septic shock, superinfections and other adverse events that can lead to therapy with the drug being discontinued. [19]. Yahav et al. advise against using tigecycline as monotherapy, especially in the treatment of serious infections [19, 20].

Prolonged plasma half-life (24–48 hours) and pharmacodynamic index (AUC/MIC) suggest, however, that higher doses of tigecycline should be used. Nevertheless, the off-label use of tigecycline and higher doses should be reserved for the treatment of multidrug resistant Gram negative infections in patients with serious intra-abdominal infections, such as abscesses.

As we recently reported, we observed KPC-Kp isolates, which developed increasing tigecycline MICs after treatment [21]. It is therefore important to carefully evaluate the benefit of a higher dose of tigecycline or other treatment options in this complex cohort of patients.

Thirdly, our mortality rate was 40%, which is higher than reported by other studies [22]. Nonetheless, our patients were particularly complicated and showed at least one additional site of KPC-Kp infection, such as severe intra-abdominal infection. In 16 cases, a KPC-Kp VAP was also present. Although numerous studies have shown that VAP frequently complicates the ICU patient’s course, the issue of whether or not critically-ill patients are at an increased risk of death because of the acquired pneumonia remains controversial [23]. In a large study analysis, VAP-attributed mortality was higher in surgical patients and patients with intermediate (but not high) Simplified Acute Physiologic Score II values on ICU admission [24]. In our study, mean APACHE II score was higher in non-survivors than in survivors. It is postulated that both derangements in acute physiology and severity of underlying diseases at the time of admission could be responsible for mortality, rather than the underlying conditions. Despite improvements in our knowledge of the physiopathology of severe infection and the evolution of diagnostic methods, antibiotic therapy, postoperative care and surgical techniques, a substantial number of patients develop severe intra-abdominal infection and advanced stages of septic insult requiring admission to the ICU. The success of treatment of these conditions is multifactorial and the best antibiotic protocol may not suffice, unless adequate control of the focus of infection is achieved.

Fourthly, in our study underlying conditions, such as COPD, diabetes and lifestyle risk factors such as smoking, were often present. This is in line with the analysis of Huttenen [25], who reported a high prevalence of obesity and smoking and their association with poor prognosis in patients with bacteraemia.

Our study is not without limitations. It includes a small number of patients with very complex and heterogeneous conditions, as well as the bias associated with non-random enrollment, as the choice of treatment was based on the patients’ medical conditions.

Additionally, our results are not supported by studies comparing different protocols. Moreover, antibiotic therapy pattern and quality of spread of multidrug resistant pathogens in the ICU under study could not be generalized. A further limitation was that data on all patients with KPC-Kp bacteraemia admitted to our university hospital during the study period was not available.

Surveillance of healthcare associated infections has become an integral part of infection control and quality assurance of healthcare in many countries. Gastmeier et al. reported that effective surveillance could reduce the NI rate by about 20–30% on average [26]. Surveillance programs provide data about the organisms causing specific infections and their antibacterial drug resistance patterns. Moreover, such programs can guide clinical practices and infection prevention and control efforts in different geographic regions and clinical settings [6, 26, 27].

At present, hand hygiene and contact precautions have been improved, including dedicated patient-care equipment, disposable gloves and aprons, and environmental cleaning. The ICU has been thoroughly cleaned and respiratory equipment disinfected. Because of insufficient bed capacity, isolation is not feasible, but colonized or infected patients are spatially segregated. Compliance with hand hygiene and environmental cleaning procedures are monitored. Staff adherence is strongly promoted by weekly meetings between healthcare workers and the hospital infection control team.

In summary, present-day surgical intensive care units continue to evolve with the latest technological advances. The modern ICU team provides continuous daily care to the critically-ill surgical patient in close cooperation with the surgical team. A vital role of the intensive care specialist is to establish and enforce protocols, guidelines and patient care pathways that ensure continued quality performance improvement. Despite the evolution of sophisticated adjuncts to healthcare and the improvement of structured critical care systems, the surgical critical care specialist continues to play a key role in improving patient outcomes.