Carbapenemase-producing Enterobacterales and specifically Klebsiella pneumoniae carbapenemase (KPC)-producing Klebsiella pneumoniae (KPC-Kp) are rapidly spreading worldwide. The prognosis of ventilator-associated pneumonia (VAP) caused by KPC-Kp is not well known. Our study tries to assess whether ventilator-associated pneumonia caused by a KPC-Kp strain is associated with higher all-cause mortality than that caused by carbapenem-susceptible isolates.
KEYWORDS: KPC, Klebsiella pneumoniae, ventilator-associated pneumonia, mortality
ABSTRACT
Carbapenemase-producing Enterobacterales and specifically Klebsiella pneumoniae carbapenemase (KPC)-producing Klebsiella pneumoniae (KPC-Kp) are rapidly spreading worldwide. The prognosis of ventilator-associated pneumonia (VAP) caused by KPC-Kp is not well known. Our study tries to assess whether ventilator-associated pneumonia caused by a KPC-Kp strain is associated with higher all-cause mortality than that caused by carbapenem-susceptible isolates. This is a retrospective cohort study of patients with VAP due to K. pneumoniae from a 35-bed polyvalent intensive care unit in a university hospital (>40,000 annual admissions) between January 2012 and December 2016. Adjusted multivariate analysis was used to study the association of KPC-Kp with 30-day all-cause mortality (Cox regression). We analyze 69 cases of K. pneumoniae VAP, of which 39 were produced by a KPC-Kp strain with high-level resistance to meropenem (MIC > 16 mg/ml). All-cause mortality at 30 days was 41% in the KPC-Kp group (16/39) and 33.3% in the carbapenem-susceptible cases (10/30). KPC-Kp etiology was not associated with higher mortality when controlled for confounders (adjusted hazard ratio [HR], 1.25; 95% confidence interval [CI], 0.46 to 3.41). Adequate targeted therapy (HR, 0.03; 95% CI, <0.01 to 0.23) was associated with all-cause mortality. Assuming the limitations due to the available sample size, the prognosis of VAP caused by KPC-Kp is similar to VAPs caused by carbapenem-susceptible K. pneumoniae when appropriate treatment is used.
INTRODUCTION
Pneumonia due to Klebsiella pneumoniae is not unusual in critical patients undergoing mechanical ventilation (1, 2). Therefore, antimicrobial resistance in these pathogens may impact the outcome of patients with ventilator-associated pneumonia (VAP). Specifically, the emergence and spread of strains producing different types of carbapenemase (such as Klebsiella pneumoniae carbapenemase [KPC]) are considered a global threat (3) because of the scarce therapeutic options. Prior to the commercialization of newer drugs active against KPC producers, such as ceftazidime-avibactam and meropenem-vaborbactam, and in countries where these drugs are not available, these infections were treated with “second-line” antibiotics, and the treatment was further complicated when the strain showed high-level meropenem resistance and resistance to colistin. All of these difficulties could increase the mortality of VAP caused by KPC-producing Klebsiella pneumoniae (KPC-Kp).
Our main objective was to study whether VAP caused by KPC-producing K. pneumoniae (KPC-Kp) and treated with carbapenem- and colistin-free regimens is associated with poorer outcome (30-day all-cause mortality) than infections caused by carbapenem-susceptible K. pneumoniae.
RESULTS
Participants and descriptive data.
During the study period, K. pneumoniae was isolated in respiratory samples from 186 ventilated patients. Of these, 100 patients did not meet the clinical criteria of infection, 7 patients had noncardiopulmonary resuscitation (CPR) orders, and chest X-rays or computed tomography (CT) scans were normal in 10 patients. The remaining 69 cases were included in the study (Fig. 1). The clinical characteristics of the cohort are described in Table 1. Ten cases (14.5%) were caused by extended-spectrum β-lactamase (ESBL)-producing strains, and 39 patients (56.5%) had a KPC-Kp infection. KPC-Kp cases did not show greater severity as measured by the acute physiologic score (APS) and Sequential Organ Failure Assessment (SOFA) scores than ESBL cases. Septic shock was more frequent in the KPC-Kp group (82.1% versus 53.3%; P = 0.01). In addition, patients infected with KPC-Kp had a higher percentage of malignant tumors (23.1% versus 3.3%; P = 0.03), surgery in the previous week (43.6% versus 16.7%; P = 0.02), and parenteral nutrition (56.4% versus 16.7%; P < 0.01), indicating the units affected by the outbreak of KPC-Kp. Bloodstream infection secondary to VAP was more frequent in the KPC-Kp group (61.5% versus 26.7%; P < 0.01). The empirical treatment was more frequently adequate when the strain was not KPC-Kp (76.7% versus 48.7%; P = 0.02).
FIG 1.
Flowchart of included patients with ventilator-associated pneumonia due to Klebsiella pneumoniae.
TABLE 1.
Clinical characteristics of patients with ventilator-associated pneumonia caused by Klebsiella pneumoniae
| Variable | Global (n = 69) | No KPC-Kp (n = 30; 43.5%) | KPC-Kp (n = 39; 56.5%) | P value |
|---|---|---|---|---|
| Age, years, median [IQRa] | 57 [46–69] | 57 [45–70] | 57 [46–48] | 0.86 |
| No. male (%) | 48 (69.6) | 20 (66.7) | 28 (71.8) | 0.65 |
| Total length of stay (days, median [IQR]) | 50 [29–68] | 38 [27–60] | 53 [30–86] | 0.12 |
| Length of stay before isolation of Klebsiella pneumoniae (days, median [IQR]) | 17 [9–25] | 10 [6–23] | 19 [12–30] | 0.51 |
| Charlson comorbidity index (median [IQR]) | 3 [1–5] | 3 [1–5] | 3 [1–6] | 0.29 |
| Comorbidities and conditions (no. [%]) | ||||
| Chronic renal failure | 4 (5.8) | 1 (3.3) | 3 (7.7) | 0.63 |
| Chronic bronchial disease | 10 (14.5) | 4 (13.3) | 6 (15.4) | 1 |
| Solid organ transplantation | 3 (4.3) | 1 (3.3) | 2 (5.1) | 1 |
| Malignant tumors | 10 (14.5) | 1 (3.3) | 9 (23.1) | 0.03 |
| Immunodeficiency | 14 (20.3) | 3 (10) | 11 (28.2) | 0.06 |
| Surgical intervention 3 previous months | 34 (43.3) | 11 (36.7) | 23 (59) | 0.07 |
| Surgical intervention previous week | 22 (31.9) | 5 (16.7) | 17 (43.6) | 0.02 |
| Parenteral nutrition | 27 (39.1) | 5 (16.7) | 22 (56.4) | <0.01 |
| Antibiotics in previous month | 68 (98.6) | 29 (96.7) | 39 (100) | 0.43 |
| Severity of illness | ||||
| Septic shock (no. [%]) | 48 (69.6) | 16 (53.3) | 32 (82.1) | 0.01 |
| APS at time of diagnosis (median [IQR]) | 9 [13–19] | 13.5 [9.75–19] | 13 [9–18] | 0,36 |
| SOFA at time of diagnosis (median [IQR]) | 6 [5–8] | 6 [4–8] | 7 [6–8] | 0.49 |
| Secondary bloodstream infection (no. [%]) | 32 (46.4) | 8 (26.7) | 24 (61.5) | <0.01 |
| Adequate empirical therapy (no. [%]) | 42 (60.9) | 23 (76.7) | 19 (48.7) | 0.02 |
| Targeted combination therapy (no. [%]) | 40 (58) | 10 (33.3) | 30 (76.9) | <0.01 |
IQR, interquartile range.
The combination treatment was more frequent in patients with KPC-Kp (76.9% versus 33.3%; P < 0.01). The antimicrobial agents indicated in our patients are shown in Table 2. Most patients infected with KPC-Kp (28 patients, 71.8%) received combination regimens of tigecycline, gentamicin, and fosfomycin depending on the sensitivity of each strain. Two patients (5.1%) received a ceftazidime-avibactam-based regimen of combination therapy. Five patients (12.8%) received tigecycline monotherapy since other tested antibiotics were not active and ceftazidime-avibactam was not available at that time; of these patients, 2/5 (40%) died. Most patients without KPC-Kp VAP received monotherapy (20 patients, 66.6%) as follows: ciprofloxacin (20%), carbapenem (16.7%), piperacillin-tazobactam (16.7%), and others. Ten patients (33.3%) received a combination regimen of meropenem or piperacillin-tazobactam or cephalosporin plus ciprofloxacin.
TABLE 2.
Targeted antibiotic therapy in the study populationa
| Therapy | Global (n = 69) | KPC-Kp (n = 39) | No KPC-Kp (n = 30) |
|---|---|---|---|
| Combination therapy (no. [%]) | 40 (58) | 30 (76.9) | 10 (33.3) |
| Tigecycline ± gentamicin ± fosfomycin | 28 (40.6) | 28 (71.8) | |
| Ceftazidime-avibactam ± gentamicin ± fosfomycin | 2 (2.9) | 2 (5.1) | |
| Meropenem + ciprofloxacin | 6 (8.7) | 6 (20) | |
| Piperacillin-tazobactam + ciprofloxacin | 2 (2.9) | 2 (6.7) | |
| Third- and fourth-generation cephalosporins + ciprofloxacin | 2 (2.9) | 2 (6.7) | |
| Monotherapy (no. [%]) | 25 (36.2) | 5 (12.8) | 20 (66.6) |
| Carbapenems (meropenem or imipenem) | 5 (7.2) | 5 (16.7) | |
| Piperacillin-tazobactam | 5 (7.2) | 5 (16.7) | |
| Tigecycline | 5 (7.2) | 5 (12.8) | |
| Ciprofloxacin | 6 (8.7) | 6 (20) | |
| Others | 4 (5.8) | 4 (13.3) | |
| No antibiotic therapy (no. [%]) | 4 (5.8) | 4 (10.3) |
All data are expressed as n (%).
Outcome of VAP due to K. pneumoniae and impact of KPC-Kp etiology.
Table 3 shows that the use of an adequate targeted treatment (adjusted hazard ratio [HR], 0.03; 95% confidence interval [CI], <0.01 to 0.23) is associated with 30-day all-cause mortality in the adjusted multivariate analysis. All-cause mortality at 30 days was 41% in the KPC-Kp group (16/39 patients) versus 33.3% in the rest (10/30 patients). A KPC-Kp etiology and secondary bacteremia could not be shown to be significantly associated with higher mortality, but the precision of the estimate was low (KPC-Kp etiology adjusted HR, 1.25; 95% CI, 0.46 to 3.41; secondary bacteremia adjusted HR, 0.69; 95% CI, 0.27 to 1.74). Figure 2 shows the survival curves of both etiological groups. Adequate empirical therapy was not associated with 30-day all-cause mortality (HR, 1.01; 95% CI, 0.46 to 2.23).
TABLE 3.
Univariate and multivariate Cox regression analysis of the association between different variables and 30-day mortality in patients with ventilator-associated pneumonia due to Klebsiella pneumoniae
| Variable | Total (n = 69) |
Crude analysis |
Adjusted analysis |
|||
|---|---|---|---|---|---|---|
| Alive (n = 43) | Dead (n = 26) | HRa (95% CI) | P value | HRa (95% CI) | P value | |
| Age, years, median [IQR] | 53 [38–66] | 64 [51–72] | 1.03 (1.00–1.06) | 0.01 | 1.01 (0.97–1.05) | 0.63 |
| No. male (%) | 30 (69.8) | 18 (69.2) | 0.93 (0.40–2.14) | 0.86 | ||
| Length of stay before isolation of Klebsiella pneumoniae (days, median [IQR]) | 13 [9–25] | 18 [9–26] | 1.00 (0.98–1.02) | 0.77 | ||
| Known previous colonization by KPC-Kp (no. [%]) | 10 (23.3) | 6 (23.1) | 0.90 (0.36–2.25) | 0.83 | ||
| Charlson comorbidity index (median [IQR]) | 2 [0–4] | 4 [2–7] | 1.19 (1.06–1.33) | <0.01 | 1.12 (0.88–1.42) | 0.35 |
| Chronic renal failure (no. [%]) | 3 (7) | 1 (3.8) | 0.54 (0.07–4.03) | 0.55 | ||
| Malignant tumors (no. [%]) | 3 (7) | 7 (26.9) | 3.36 (1.41–8.02) | 0.01 | 0.79 (0.16–3.87) | 0.77 |
| Surgical intervention previous week (no. [%]) | 14 (32.6) | 8 (30.8) | 1.10 (0.48–2.53) | 0.83 | ||
| Parenteral nutrition (no. [%]) | 15 (34.9) | 12 (46.2) | 1.45 (0.67–3.15) | 0.34 | ||
| Antibiotics in previous month (no. [%]) | 42 (97.7) | 26 (100) | 1 | |||
| Severity of illness | ||||||
| Septic shock (no. [%]) | 27 (62.8) | 21 (80.8) | 2.18 (0.82–5.77) | 0.12 | ||
| APS II at time of diagnosis (median [IQR]) | 12 [9–18] | 14.5 [9–20.25] | 1.04 (0.98–1.11) | 0.19 | ||
| SOFA at time of diagnosis (median [IQR]) | 6 [4–8] | 8 [6-8] | 1.12 (0.98–1.28) | 0.09 | ||
| Bloodstream infection (no. [%]) | 22 (51.2) | 10 (38.5) | 0.65 (0.29–1.43) | 0.28 | 0.69 (0.27–1.74) | 0.69 |
| Renal failure at diagnosis of infection (no. [%]) | 8 (18.6) | 8 (30.8) | 1.88 (0.82–4.32) | 0.14 | ||
| KPC-producing K. pneumoniae | ||||||
| No. no (%) | 20 (46.5) | 10 (38.5) | 1.38 (0.62–3.03) | 0.43 | 1.25 (0.46–3.41) | 0.65 |
| No. yes (%) | 23 (53.5) | 16 (61.5) | ||||
| Adequate empirical therapy (no. [%]) | 26 (60.5) | 16 (61.5) | 1.01 (0.46–2.23) | 0.98 | ||
| Adequate targeted therapy (no. [%]) | 43 (100) | 22 (84.6) | 0.02 (0.00–0.10) | 0.00 | 0.03 (0.00–0.23) | <0.01 |
| Time to targeted therapy, n = 65 (no. [%]) | n = 43 | n = 22 | ||||
| <24 h | 28 (65.1) | 18 (81.8) | Reference | 0.33 | ||
| 24–48 h | 4 (9.3) | 2 (9.1) | 0.85 (0.20–3.66) | 0.83 | ||
| 48–72 h | 11 (25.6) | 2 (9.1) | 0.33 (0.08–1.42) | 0.14 | ||
| Time to targeted therapy, n = 65 (hours, mean [SD]) | 20.65 [28.16] | 11.91 [22.71] | 0.99 (0.97–1.01) | 0.20 | ||
| Targeted combination therapy (no. [%]) | 27 (62.8) | 13 (50) | 0.61 (0.28–1.33) | 0.22 | ||
In quantitative variables, the HR is per unit.
FIG 2.
Survival curves of ventilated patients with respiratory infection due to Klebsiella pneumoniae according to KPC production.
In the same way, mortality was analyzed at 14 days. The use of an adequate targeted treatment (adjusted HR, 0.06; 95% CI, 0.01 to 0.71) is associated with 14-day all-cause mortality in the adjusted multivariate analysis. A KPC-Kp etiology could not be shown to be significantly associated with higher mortality (adjusted HR, 0.98; 95% CI, 0.28 to 3.48).
Outcome predictors among patients with VAP due to KPC-Kp.
Table 4 shows the adjusted analysis of variables associated with 30-day all-cause mortality among the 39 patients with VAP due to KPC-Kp. The use of adequate targeted therapy was the only variable associated with mortality (adjusted HR, 0.03; 95% CI, <0.01 to 0.31). Four patients did not receive adequate antibiotic treatment because they died before the culture was available and the empirical treatment was inadequate. Five patients were treated with monotherapy upon judgment of the attending physician, and 40% (2/5 patients) died. The number of available cases prevented us from analyzing the type of targeted therapy (monotherapy versus combination therapy).
TABLE 4.
Univariate and multivariate Cox regression analysis of the association between different variables and 30-day mortality in patients with ventilator-associated pneumonia due to KPC-Kp
| Variable | KPC-Kp (n = 39) |
Crude analysis |
Adjusted analysis |
|||
|---|---|---|---|---|---|---|
| Alive (n = 23) | Dead (n = 16) | HRa (95% CI) | P value | HRa (95% CI) | P value | |
| Age, years, median [IQR] | 51 [38–66] | 64 [47–72] | 1.03 (0.99–1.07) | 0.10 | ||
| No. male (%) | 15 (65.2) | 13 (81–3) | 0.44 (0.13–1.56) | 0.20 | ||
| Length of stay before isolation of Klebsiella pneumoniae (days, median [IQR]) | 19 [13–37] | 22 [10–26] | 0.98 (0.94–1.02) | 0.41 | ||
| Known previous colonization by KPC (no. [%]) | 10 (43.5) | 5 (31.3) | 0.61 (0.21–1.76) | 0.36 | ||
| Charlson comorbidity index (median [IQR]) | 2 [1–4] | 4 [1–8] | 1.16 (1.01–1.32) | 0.03 | 1.13 (0.91–1.4) | 0.28 |
| Chronic renal failure (no. [%]) | 2 (8.7) | 1 (6.3) | 0.66 (0.09–5.02) | 0.69 | ||
| Surgical intervention previous week (no. [%]) | 10 (43.5) | 7 (43.8) | 1.21 (0.45–3.25) | 1.21 | ||
| Parenteral nutrition (no. [%]) | 12 (52.2) | 10 (62.5) | 1.39 (0.50–1.39) | 0.52 | ||
| Malignant tumors (no. [%]) | 3 (13) | 6 (37.5) | 2.98 (1.08–8.27) | 0.03 | 0.51 (0.07–3.84) | 0.51 |
| Antibiotics in previous month (no. [%]) | 23 (100) | 16 (100) | 1 | |||
| Severity of illness | ||||||
| Septic shock (no. [%]) | 17 (73.9) | 15 (93.8) | 4.32 (0.57–32.77) | 0.16 | ||
| APS at time of diagnosis (median [IQR]) | 11 [8–16] | 15 [9.25–19.75] | 1.05 (0.97–1.14) | 0.17 | ||
| SOFA at time of diagnosis (median [IQR]) | 6 [4–8] | 8 [7–9] | 1.28 (1.03–1.60) | 0.03 | 1.32 (1.01–1.73) | 0.04 |
| Bloodstream infection (no. [%]) | 16 (69.6) | 8 (50) | 0.51 (0.19–1.35) | 0.17 | 0.5 (0.15–1.68) | 0.26 |
| Renal failure at diagnosis of infection (no. [%]) | 7 (30.4) | 9 (56.3) | 2,43 (0.90–6.57) | 0.08 | ||
| Adequate empirical therapy (no. [%]) | 12 (52.2) | 7 (43.8) | 0.76 (0.28–2.05) | 0.59 | ||
| Adequate targeted therapy (no. [%]) | 23 (100) | 12 (75) | 0.03 (0.00–0.18) | 0.00 | 0.03 (0.00–0.31) | <0.01 |
| Time to adequate therapy (n = 35) | n = 23 | n = 12 | ||||
| <24 h (no. [%]) | 14 (60.9) | 8 (66.7) | Reference | 0.78 | ||
| 24–48 h (no. [%]) | 3 (13) | 2 (16.7) | 1.15 (0.24–5.41) | 0.86 | ||
| 48–72 h (no. [%]) | 6 (26.1) | 2 (16.7) | 0.60 (0.13–2.84) | 0.52 | ||
| Time to adequate therapy, n = 35 (h, mean [SD]) | 23.43 (28.73) | 20.33 (28.06) | 1.00 (0.97–1.02) | 0.68 | ||
In quantitative variables, the HR is per unit.
We calculated the associations excluding the 2 ceftazidime-avibactam-treated patients from analysis, and the results did not change. The only variable associated with mortality at day 14 and 30 was “adequate targeted treatment.”
DISCUSSION
The published information on VAP caused by KPC-Kp is scarce and limited to case reports (4, 5), outbreaks (6), or small observational studies (7). With a limited number of available cases of KPC-Kp, our series permits us to analyze the impact of KPC-producing strains on the prognosis of VAP due to K. pneumoniae. Previous studies have observed that not only carbapenemase-producing Enterobacterales infections (8) but also colonization (9, 10) is associated with higher all-cause mortality in intensive care unit (ICU) patients. The main conclusion of our study is that when VAP caused by KPC-Kp is treated adequately, it may have a similar prognosis to that produced by carbapenem-susceptible K. pneumoniae. The survival curve shows an early separation and then roughly parallel curves, implying early appropriate antibiotics are critical. It should be noted that this conclusion is drawn from a cohort of KPC-Kp-infected patients that were not treated with either carbapenem or colistin at a time when the use of ceftazidime-avibactam was not generalized (it was used only in 2 cases; see Table 2), as previously described in patients with bacteremia (11). To be helpful for centers that cannot afford or have no access to this drug, we calculated the associations excluding the 2 ceftazidime-avibactam-treated patients from analysis, and the results did not change. The only variable associated with mortality at day 14 and 30 was “adequate targeted treatment.”
It is generally accepted that KPC-Kp infections should be treated with combination therapy. This concept, which has been reported in observational studies on bacteremia (12–15), may be applicable to any serious infection (14–18). VAP is an example of a high-inoculum infection that affects very vulnerable patients and is associated with high mortality (17), which explains why >75% was treated with combination therapy. The previously reported increase in mortality associated with severe KPC-Kp infections (8) and the particular susceptibility profile of our strain, which prevents the use of meropenem and colistin, justifies that we expected higher mortality in patients infected with KPC-Kp. With this phenotype, it is not always easy to design an effective therapeutic regimen. We must resort to nephrotoxic drugs (gentamicin), drugs of doubtful efficacy (gentamicin, tigecycline), or forgotten drugs (fosfomycin) (14). To achieve efficacy, it is necessary to use very high doses (tigecycline, fosfomycin) that are frequently poorly tolerated (14). However, this expected increase in mortality has not been confirmed in our study. This could be due to the following reasons: (i) the infection occurs anyway in patients with a high basal risk of mortality, which prevents finding differences between different strains or treatments, and (ii) KPC-Kp infections occur in the context of outbreaks or endemicity in which colonization is systematically detected, thus facilitating early adequate treatment. The precocity of the appropriate targeted treatment has been related to a better prognosis of respiratory infection in ventilated patients (19).
Our study is limited by its design (a single-center, observational, and retrospective study, case-selection). As we have previously discussed, patients infected with KPC-Kp preferably received combination therapy. However, it was not possible to include a propensity score for using combination therapy because the only variable associated with the use of combination therapy was infection with a KPC-Kp strain. Another important limitation is the sample size, which depends on the number of available cases. For all these reasons, we consider it necessary to confirm our data in further studies. Considering these limitations, we can conclude that, in our cohort, the prognosis of VAP caused by KPC-Kp is similar to that of VAPs caused by carbapenem-susceptible K. pneumoniae when appropriate treatment is used.
MATERIALS AND METHODS
Study design and population.
This is a retrospective cohort study performed at a university hospital (>40,000 annual admissions) during a nosocomial outbreak and subsequent endemicity of KPC-Kp (2012 to 2016). A subset of isolates at the beginning of the KPC-Kp outbreak was determined to produce KPC-3 and correspond to sequence type 512 (ST512) (20). Phenotypic resistance profiles of subsequent isolates indicated that the strain was maintained over time in our institution. Patients were identified through a search of the microbiology database. All episodes of VAP due to K. pneumoniae in the same source population (carbapenem-resistant, cephalosporin-resistant, and sensitive) and diagnosed between January 2012 and December 2016 were eligible. The following inclusion criteria were used: (i) patients aged ≥18 years and (ii) first episode of VAP was due to K. pneumoniae. Patients with nonresuscitation orders were excluded from the study. The day of onset of the infection (day 0) was defined as the day of VAP diagnostic sample collection in which K. pneumoniae was isolated.
This report was made following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations (21). The study was approved by the Spanish Agency for Medicine and Health Products (AEMPS, code FRE-MER-2018-01) and by the local ethics committee (code 2495), which exempted the need to seek written informed consent due to the observational nature of the study. We used an anonymized database designed for this study.
Outcome variable.
The primary outcome variable was 30-day all-cause mortality. Survival status on day 30 was ascertained through a phone call to patients discharged before day 30.
Exposure variables.
The main exposure of interest was KPC-Kp etiology (for microbiological characteristics of the strain, see below). Any antimicrobial treatment that was initiated or maintained after receiving the susceptibility testing results was considered targeted therapy. The time “zero” used as a reference for the therapies was the moment when microbiological cultures were taken. A targeted antibiotic treatment regimen was considered active when it included antibiotics to which the isolate was susceptible in vitro (for gentamicin, intermediate isolates were also considered as susceptible; see below). As this was an observational study, the antibiotic treatment regimen and other treatment decisions were at the discretion of the treating physician. Patients treated with a single in vitro active drug were considered to be undergoing monotherapy. Patients treated with two or more in vitro active drugs were considered to be undergoing combination therapy.
To classify patients as receiving a specific regimen, the regimen should have been maintained for at least 50% of the duration of treatment (or at least 48 h if the patient died before) in order to guarantee a minimum exposure time to the regimen. Patients who died before the culture results were received and who did not receive adequate empirical treatment were classified as inadequate therapy.
In patients with KPC-Kp infection, gentamicin was given intravenously as a single daily dose of 5 mg/kg of body weight/day and adjusted according to blood level concentrations. Fosfomycin was administered at an intravenous dose of 4 g every 6 h and adjusted according to renal function. Tigecycline was administered in a double dose (200 mg loading dose followed by 100 mg intravenously every 12 h). When indicated (MIC < 16 mg/ml), meropenem was administered at a dose of 2 g every 8 h in extended perfusion (2 h). The duration of treatment ranged from 10 to 14 days according to the judgment of the attending physician. Ceftazidime-avibactam was used at the approved dose (2 g/500 mg in a 2-h intravenous infusion every 8 h, adjusted according to renal function).
Other study variables.
The study variables were age, gender, number of hospitalization days in the previous 6 months, previous colonization by KPC-Kp, Charlson comorbidity index (22), comorbidities (chronic renal failure, chronic bronchial disease, solid organ transplantation, malignant tumors, immunodeficiency), surgical intervention and parenteral nutrition in the previous week, antibiotics in the previous month, presentation with septic shock (23), the acute physiologic score (APS) component of acute physiology and chronic health evaluation II (APACHE-II) (24) and SOFA (25) scores on the day of index culture, secondary bloodstream infection, microbiological data, empirical therapy, time to targeted therapy, and targeted therapy.
Definitions.
The definitions were established prior to the data collection and statistical analysis. Ventilator-associated pneumonia was defined as follows: (i) presence of at least two of the following clinical criteria: fever (>38°C) or hypothermia (<36°C), leukocytosis (>12,000/mm3) or leukopenia (<3,500/mm3), presence of purulent respiratory secretions; (ii) microbiological confirmation in quantitative culture of valid respiratory samples as follows: tracheal aspirate (≥106 CFU/ml), protected bronchial aspirate (≥103 CFU/ml), or bronchoalveolar lavage (≥104 CFU/ml); (iii) appearance of symptoms and signs 48 h after the start of mechanical ventilation; or (iv) the chest radiograph or CT scan showed a pulmonary infiltrate.
Septic shock was defined according to the latest consensus criteria (23). The Cockcroft-Gault formula was used to calculate creatinine clearance (<60 ml/min was considered renal failure). Empirical therapy was defined as the administration of treatment within the first 24 h following the collection of cultures and prior to determining the susceptibility of the isolate. Empirical therapy was considered appropriate when the isolate was susceptible in vitro to at least one of the prescribed antibiotics.
Patients who had one of the following conditions were considered immunocompromised: neutropenia (<500/μl), solid or active hematologic malignancy, prolonged corticosteroid treatment (>1 mg/kg/day for >1 month), HIV infection (CD4 cells < 500/μl in the previous 6 months), and solid organ or hematopoietic progenitors transplantation.
Microbiological studies.
Identification and susceptibility testing of all isolates were performed using the wider panel MIC/UD Gram-negative Uricult (Siemens Healthcare Diagnostics, Inc., West Sacramento, CA, USA) and/or Microscan panel NC-54 (Siemens Healthcare Diagnostics, Inc., West Sacramento, CA). Identification was confirmed with matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) (Bruker Daltonics GmbH, Bremen, Germany). Susceptibility testing of ceftazidime-avibactam was performed for selected strains using a gradient diffusion method (Etest, bioMérieux). Colistin resistance, as defined with commercial panels, was confirmed by broth microdilution following the recommendations of the joint CLSI-EUCAST Polymyxin Breakpoints Working Group (http://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/General_documents/Recommendations_for_MIC_determination_of_colistin_March_2016.pdf) in 22 of the 24 available isolates (MICs of colistin were 2 and 1 mg/liter and 0.25 and 1 mg/liter in duplicate assays for the other two isolates). MICs were interpreted following the breakpoint of the Clinical and Laboratory Standards Institute’s (CLSI) standard criteria for broth microdilution (26). The presence of mcr-1 and mcr-2 genes was investigated in selected colistin-resistant strains using PCR with specific primers. Meropenem was not considered active against any isolate because the MIC was ≥64 mg/liter in all cases. Gentamicin and fosfomycin were considered active when the MIC was ≤8 mg/liter and ≤64 mg/liter, respectively. Ceftazidime-avibactam was considered active when the MIC was ≤8 and 4 mg/liter (27).
An additional confirmatory test for extended-spectrum β-lactamase (ESBL) production was performed using ceftazidime and cefotaxime discs with and without clavulanic acid on Mueller-Hinton agar plates according to CLSI recommendations (26).
The KPC-Kp index isolates in the outbreak were characterized as belonging to the ST512 clone and producing KPC-3 by the reference laboratory of the Virgen Macarena University Hospital of Seville, Spain. The characteristics of the strain have been previously reported (20).
Statistical analysis.
Continuous variables were compared by the Mann-Whitney U test; categorical variables were compared by the chi-square test or Fisher’s exact test as appropriate. Multivariate analyses were performed using a Cox regression after assessing the proportional hazard assumption. Interactions between selected covariates were explored. The predictive ability validity of each multivariate model was studied by calculating the area under the receiver operating characteristic (ROC) curve (AUC). The analyses were performed using the SPSS software program (SSPS 15.0, IBM Corp, Armonk, New York, USA).
ACKNOWLEDGMENTS
Supported by Plan Nacional de I+D+i 2013 to 2016 and Instituto de Salud Carlos III, Subdirección General de Redes y Centros de Investigación Cooperativa, Ministerio de Economía, Industria y Competitividad, Spanish Network for Research in Infectious Diseases (REIPI RD16/0016/0001; RD16/0016/0008). Cofinanced by the European Regional Development Regional Fund “A way to achieve Europe,” Operational Programme Smart Growth 2014 to 2020. B.G.-G. is a recipient of the “Río Hortega” grant by the Spanish Ministry of Economy and Competitiveness, Instituto de Salud Carlos III.
Isabel Machuca, Rocio Tejero, Ángela Cano, Belén Gutiérrez-Gutiérrez, Jesús Rodriguez-Baño, Luis Martínez-Martínez, and Julian Torre-Cisneros are members of the Spanish Network for Research in Infectious Diseases (REIPI).
Conflicts of interest are as follows: J.R.-B. served as scientific advisor for a research project for AstraZeneca and InfectoPharm and as a speaker in unrestricted accredited educational activities funded by Merck. J.T.-C. served as a scientific advisor for research/consensus projects for Pfizer and as an expert in a consensus document for InfectoPharm. He has received payment for lectures including service on speaker bureaus and for the development of educational presentations for Pfizer, AstraZeneca, and Merck. L.M.-M. has been a consultant for MSD and Shionogi; served as a speaker for Merck, Pfizer, AbbVie, and AstraZeneca; and received research support from Janssen-Cilag and Pfizer. A.C. has received honoraria for the development of educational presentations for Pfizer in 2019. The remaining authors have disclosed that they do not have any conflicts of interest.
REFERENCES
- 1.American Thoracic Society, Infectious Diseases Society of America. 2005. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 171:388–416. doi: 10.1164/rccm.200405-644ST. [DOI] [PubMed] [Google Scholar]
- 2.Hunter JD. 2012. Ventilator associated pneumonia. BMJ 344:e3325. doi: 10.1136/bmj.e3325. [DOI] [PubMed] [Google Scholar]
- 3.Logan LK, Weinstein RA. 2017. The epidemiology of carbapenem-resistant Enterobacteriaceae: the impact and evolution of a global menace. J Infect Dis 215:S28–S36. doi: 10.1093/infdis/jiw282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yan Q, Zhou M, Zou M, Liu WE. 2016. Hypervirulent Klebsiella pneumoniae induced ventilator-associated pneumonia in mechanically ventilated patients in China. Eur J Clin Microbiol Infect Dis 35:387–396. doi: 10.1007/s10096-015-2551-2. [DOI] [PubMed] [Google Scholar]
- 5.Sbrana F, Malacarne P, Bassetti M, Tascini C, Vegnuti L, Della Siega P, Ripoli A, Ansaldi F, Menichetti F. 2016. Risk factors for ventilator associated pneumonia due to carbapenemase-producing Klebsiella pneumoniae in mechanically ventilated patients with tracheal and rectal colonization. Minerva Anestesiol 82:635–640. [PubMed] [Google Scholar]
- 6.Hu L, Liu Y, Deng L, Zhong Q, Hang Y, Wang Z, Zhan L, Wang L, Yu F. 2016. Outbreak by ventilator-associated ST11 K. pneumoniae with co-production of CTX-M-24 and KPC-2 in a SICU of a tertiary teaching hospital in Central China. Front Microbiol 7:1190. doi: 10.3389/fmicb.2016.01190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kontopidou F, Giamarellou H, Katerelos P, Maragos A, Kioumis I, Trikka-Graphakos E, Valakis C, Maltezou HC, Group for the Study of KPC-producing Klebsiella pneumoniae infections in intensive care units . 2014. Infections caused by carbapenem-resistant Klebsiella pneumoniae among patients in intensive care units in Greece: a multi-centre study on clinical outcome and therapeutic options. Clin Microbiol Infect 20:O117–O123. doi: 10.1111/1469-0691.12341. [DOI] [PubMed] [Google Scholar]
- 8.Falagas ME, Tansarli GS, Karageorgopoulos DE, Vardakas KZ. 2014. Deaths attributable to carbapenem-resistant Enterobacteriaceae infections. Emerg Infect Dis 20:1170–1175. doi: 10.3201/eid2007.121004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Papadimitriou-Olivgeris M, Marangos M, Fligou F, Christofidou M, Sklavou C, Vamvakopoulou S, Anastassiou ED, Filos KS. 2013. KPC-producing Klebsiella pneumoniae enteric colonization acquired during intensive care unit stay: the significance of risk factors for its development and its impact on mortality. Diagn Microbiol Infect Dis 77:169–173. doi: 10.1016/j.diagmicrobio.2013.06.007. [DOI] [PubMed] [Google Scholar]
- 10.Dautzenberg MJ, Wekesa AN, Gniadkowski M, Antoniadou A, Giamarellou H, Petrikkos GL, Skiada A, Brun-Buisson C, Bonten MJ, Derde LP, Mastering Hospital Antimicrobial Resistance in Europe Work Package 3 Study Team . 2015. The association between colonization with carbapenemase-producing Enterobacteriaceae and overall ICU mortality: an observational cohort study. Crit Care Med 43:1170–1177. doi: 10.1097/CCM.0000000000001028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Machuca I, Gutierrez-Gutierrez B, Gracia-Ahufinger I, Rivera Espinar F, Cano A, Guzman-Puche J, Perez-Nadales E, Natera C, Rodriguez M, Leon R, Caston JJ, Rodriguez-Lopez F, Rodriguez-Bano J, Torre-Cisneros J. 2017. Mortality associated with bacteremia due to colistin-resistant Klebsiella pneumoniae with high-level meropenem resistance: importance of combination therapy without colistin and carbapenems. Antimicrob Agents Chemother 61:e00406-17. doi: 10.1128/AAC.00406-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gutiérrez-Gutiérrez B, Salamanca E, de Cueto M, Hsueh P-R, Viale P, Paño-Pardo JR, Venditti M, Tumbarello M, Daikos G, Pintado V, Doi Y, Tuon FF, Karaiskos I, Machuca I, Schwaber MJ, Azap ÖK, Souli M, Roilides E, Pournaras S, Akova M, Pérez F, Bermejo J, Oliver A, Almela M, Lowman W, Almirante B, Bonomo RA, Carmeli Y, Paterson DL, Pascual A, Rodríguez-Baño J, Investigators from the REIPI/ESGBIS/INCREMENT Group . 2016. A predictive model of mortality in patients with bloodstream infections due to carbapenemase-producing Enterobacteriaceae. Mayo Clin Proc 91:1362–1371. doi: 10.1016/j.mayocp.2016.06.024. [DOI] [PubMed] [Google Scholar]
- 13.Gutiérrez-Gutiérrez B, REIPI/ESGBIS/INCREMENT Investigators, Salamanca E, de Cueto M, Hsueh P-R, Viale P, Paño-Pardo JR, Venditti M, Tumbarello M, Daikos G, Cantón R, Doi Y, Tuon FF, Karaiskos I, Pérez-Nadales E, Schwaber MJ, Azap ÖK, Souli M, Roilides E, Pournaras S, Akova M, Pérez F, Bermejo J, Oliver A, Almela M, Lowman W, Almirante B, Bonomo RA, Carmeli Y, Paterson DL, Pascual A, Rodríguez-Baño J. 2017. Effect of appropriate combination therapy on mortality of patients with bloodstream infections due to carbapenemase-producing Enterobacteriaceae (INCREMENT): a retrospective cohort study. Lancet Infect Dis 17:726–734. doi: 10.1016/S1473-3099(17)30228-1. [DOI] [PubMed] [Google Scholar]
- 14.Rodriguez-Bano J, Gutierrez-Gutierrez B, Machuca I, Pascual A. 2018. Treatment of infections caused by extended-spectrum-beta-lactamase-, AmpC-, and carbapenemase-producing Enterobacteriaceae. Clin Microbiol Rev 31:e00079-17. doi: 10.1128/CMR.00079-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tumbarello M, Viale P, Viscoli C, Trecarichi EM, Tumietto F, Marchese A, Spanu T, Ambretti S, Ginocchio F, Cristini F, Losito AR, Tedeschi S, Cauda R, Bassetti M. 2012. Predictors of mortality in bloodstream infections caused by Klebsiella pneumoniae carbapenemase-producing K. pneumoniae: importance of combination therapy. Clin Infect Dis 55:943–950. doi: 10.1093/cid/cis588. [DOI] [PubMed] [Google Scholar]
- 16.Gonzalez-Padilla M, Torre-Cisneros J, Rivera-Espinar F, Pontes-Moreno A, López-Cerero L, Pascual A, Natera C, Rodríguez M, Salcedo I, Rodríguez-López F, Rivero A, Rodríguez-Baño J. 2015. Gentamicin therapy for sepsis due to carbapenem-resistant and colistin-resistant Klebsiella pneumoniae. J Antimicrob Chemother 70:905–913. doi: 10.1093/jac/dku432. [DOI] [PubMed] [Google Scholar]
- 17.Falagas ME, Lourida P, Poulikakos P, Rafailidis PI, Tansarli GS. 2014. Antibiotic treatment of infections due to carbapenem-resistant Enterobacteriaceae: systematic evaluation of the available evidence. Antimicrob Agents Chemother 58:654–663. doi: 10.1128/AAC.01222-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cano A, Gutiérrez-Gutiérrez B, Machuca I, Gracia-Ahufinger I, Pérez-Nadales E, Causse M, Castón JJ, Guzman-Puche J, Torre-Giménez J, Kindelán L, Martínez-Martinez L, Rodriguez-Baño J, Torre-Cisneros J. 2018. Risks of infection and mortality among patients colonized with Klebsiella pneumoniae carbapenemase-producing K. pneumoniae: validation of scores and proposal for management. Clin Infect Dis 66:1204–1210. doi: 10.1093/cid/cix991. [DOI] [PubMed] [Google Scholar]
- 19.Iregui M, Ward S, Sherman G, Fraser VJ, Kollef MH. 2002. Clinical importance of delays in the initiation of appropriate antibiotic treatment for ventilator-associated pneumonia. Chest 122:262–268. doi: 10.1378/chest.122.1.262. [DOI] [PubMed] [Google Scholar]
- 20.López-Cerero L, Egea P, Gracia-Ahufinger I, González-Padilla M, Rodríguez-López F, Rodríguez-Baño J, Pascual A. 2014. Characterisation of the first ongoing outbreak due to KPC-3-producing Klebsiella pneumoniae (ST512) in Spain. Int J Antimicrob Agents 44:538–540. doi: 10.1016/j.ijantimicag.2014.08.006. [DOI] [PubMed] [Google Scholar]
- 21.von Elm E, STROBE Initiative, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, Initiative S. 2007. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 370:1453–1457. doi: 10.1016/S0140-6736(07)61602-X. [DOI] [PubMed] [Google Scholar]
- 22.Charlson ME, Pompei P, Ales KL, MacKenzie CR. 1987. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40:373–383. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 23.Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, Bellomo R, Bernard GR, Chiche JD, Coopersmith CM, Hotchkiss RS, Levy MM, Marshall JC, Martin GS, Opal SM, Rubenfeld GD, van der Poll T, Vincent JL, Angus DC. 2016. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA 315:801–810. doi: 10.1001/jama.2016.0287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Knaus WA, Draper EA, Wagner DP, Zimmerman JE. 1985. APACHE II: a severity of disease classification system. Crit Care Med 13:818–829. doi: 10.1097/00003246-198510000-00009. [DOI] [PubMed] [Google Scholar]
- 25.Vincent JL, de Mendonca A, Cantraine F, Moreno R, Takala J, Suter PM, Sprung CL, Colardyn F, Blecher S. 1998. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Crit Care Med 26:1793–1800. doi: 10.1097/00003246-199811000-00016. [DOI] [PubMed] [Google Scholar]
- 26.Clinical and Laboratory Standards Institute. 2017. Performance standards for antimicrobial susceptibility testing; 27th informational supplement. CLSI M100. Clinical and Laboratory Standards Institute, Wayne PA. [Google Scholar]
- 27.European Committee for Antimicrobial Susceptibility Testing. 2019. Breakpoint tables for interpretation of MICs and zone diameters, version 9.0, 2019. http://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/Breakpoint_tables/v_9.0_Breakpoint_Tables.pdf.


