Epidemiology of infections and colonization caused by Klebsiella pneumoniae NDM in the Mazovian Voivodeship in 2016–2017

Klebsiella pneumoniae are clinically significant microorganisms that have garnered much public health concern [1]. K. pneumonia is the most clinically significant Klebsiella species, being the cause of the majority of human infections due to this genus [2]. K. pneumoniae is a common cause of antimicrobial-resistant opportunistic infections in hospitalized patients. Because of its natural resistance to penicillin and its acquired resistance to multiple antimicrobials, K. pneumoniae is a particular threat in health care [3]. K. pneumoniae was the first identified New Delhi metallo-β-lactamase (NDM) producer [4]. NDM is a metallo-β-lactamase (MBL) which hydrolyzes almost all β-lactam antibiotics, including carbapenems. Since 2008, NDM has been found in other Enterobacteriaceae, Acinetobacter, and Pseudomonas strains [5, 6]. Nowadays, K. pneumoniae and Escherichia coli are the predominant NDM producers [7]. NDM-producing Enterobacteriaceae are involved in a broad spectrum of infections and have been isolated from urinary tract, blood, airways, peritoneum, soft tissues, devices, and implants [8].

The rapid spread of K. pneumoniae NDM is currently one of the most serious health care problems, which is a risk to patient safety and is related to increased treatment costs, prolonged hospitalization, therapeutic failures, and patient death [9, 10, 11]. In the first quarter of 2016 in Poland, 310 isolates of Enterobacteriaceae NDM, mainly K. pneumonia NDM, were reported. In the same period in 2017, this number increased to 785. A particularly dangerous epidemic situation was observed in the Mazovian Voivodeship, where 88% (n=273) of strains isolated in 2016 and 69% (n=545) in 2017 originated [9]. According to the European Centre for Disease Prevention and Control (ECDC) data, between 2014 and 2017, an increase was observed of MBL producers in Poland among K. pneumoniae from 1.4% to 6.4% [12]. The data from the Epidemiological Station in Warsaw show that in 2018, there were 23 outbreaks caused by K. pneumoniae MBL; among them, 7 (30%) were caused by NDM producers [13].

Rational antibiotic therapy, hand hygiene, and microbiological screening of carriers are common methods aimed at reducing the spread of alarm pathogens [14, 15]. Screening for asymptomatic K. pneumoniae NDM colonization and rapidly implementing contact isolation may reduce transmission between patients. Recent guidelines described infection control mechanisms and practices related to carbapenemase-producing Enterobacteriaceae (CPE) [16, 17].

Epidemiological surveillance is important for monitoring and evaluating emerging patterns and trends in microbial spread. Surveillance is crucial because it contributes to better prevention and management of infection outbreaks. Collected data may help countries set their priorities and develop targeted interventions to prevent outbreaks.

The aim of this study was the assessment of the epidemiological situation related to the spread of symptomatic infections and colonization caused by K. pneumoniae NDM in the Mazovian Voivodeship in 2016–2017.

We assessed the epidemiology of infections and colonization caused by K. pneumoniae NDM in the Mazovian Voivodeship in 2016–2017. We have performed a retrospective analysis of data sent from hospitals to the Provincial Sanitary and Epidemiological Station (PSES) in Warsaw. Data was extracted from reports on suspected epidemic outbreaks and the elimination of outbreaks and annual reports on nosocomial infections and alarm pathogens.

The analysis included the number of infections, colonization, transmissions, and risk of epidemic outbreaks caused by K. pneumoniae NDM. The results were stratified according to hospital location (hospitals in Warsaw or hospitals outside of Warsaw).

An outbreak caused by K. pneumoniae NDM was determined when 1) at least two cases were diagnosed in the ward, 2) at least one case was diagnosed by a test performed during hospitalization, and 3) patients had symptomatic infection and colonization. Further analysis characterizes two situations:

symptomatic epidemic outbreak (in which there is at least one symptomatic patient and the remaining patients are symptomatic or asymptomatic (colonized)

The NDM mechanism was detected by using routine procedures applied by participating hospitals. All K. pneumoniae NDM isolates were sent to the National Reference Center for Medicinal Sensitivity of Microorganisms to confirm.

The study included data from 168 hospitals under the supervision of the PSES in Warsaw in 2016–2017. Among them, 87 (52%) were in Warsaw (an average of 700,000 hospitalizations annually) and 81 (48%) were located outside of Warsaw (an average of 600,000 hospitalizations annually).

3.1

Incidence of K. pneumoniae NDM infections

Based on the number of hospitalizations in the Mazovian Voivodeship, the incidence of K. pneumoniae NDM (infection and colonization) was 0.96/1,000 hospitalizations in 2016 and 2.04/1,000 hospitalizations in 2017. The highest incidence (per 10,000 citizens) in 2016 was observed in Płońsk (5.11), Pruszków (4.22), and Ciechanów County (3.55). In 2017, the highest incidence was observed in Płońsk (14.81), Siedlce (12.16), and Ciechanów County (10.21). Detailed data are presented in Figure 1.

Fig. 1

3.2

K. pneumoniae NDM transmission

In 2016, hospitals in the Mazovian Voivodeship reported 50 transmissions of K. pneumoniae NDM. In 2017, this value increased to 74. The number of K. pneumoniae NDM transmissions reported in hospitals in Warsaw in 2017 was slightly lower than in 2016 (34 vs. 30 cases), while in hospitals outside of Warsaw, the opposite outcomes were observed, with an increased number of transmissions from 16 to 44 cases (Fig. 2).

Fig. 2

In 2016, among 50 transmissions, 31 and 19 of them caused colonization and symptomatic infections, respectively. In 2017, 43 (58%) transmissions were colonization and 31 (42%) were symptomatic infection transmissions. Detailed data are presented in Table 1.

Table 1

Transmission of K. pneumoniae NDM in the Mazovian Voivodeship in 2016–2017

K. pneumoniae NDM transmissions and number of affected patients	2016	2017
Transmissions	50 (100%)	74 (100%)
Symptomatic transmissions	19 (38%)	31 (42%)
Colonization transmissions	31 (62%)	43 (58%)
Patients with transmission of infection	360 (100%)	602 (100%)
Symptomatic patients	79 (22%)	138 (23%)
Patients with transmission of colonization	281 (78%)	464 (77%)

During the study period, the most common transmissions of symptomatic infections caused by K. pneumoniae NDM were reported in internal medicine departments, neurology, and general surgery. In internal medicine and surgical departments, colonization transmissions were prevalent, while in neurological departments, symptomatic infections were dominant. The number of symptomatic infection transmissions on intensive care units increased from 0 in 2016 to 13 in 2017 (Table 2).

Table 2

The occurrence of K. pneumoniae NDM transmissions by hospital department in 2016–2017

	2016		2017
Department	Number of symptomatic transmissions	Number of colonization transmissions	Number of symptomatic transmissions	Number of colonization transmissions
Total	19 (100%)	31 (100%)	31 (100%)	43 (100%)
Surgery	3 (15.8%)	8 (25.8%)	3 (9.68%)	7 (16.3%)
Neurology	7 (36.8%)	1 (3.2%)	4 (12.90%)	5 (11.6%)
Neonatology	0 (0%)	1 (3.2%)	0 (0%)	1 (2.3%)
Intensive care unit	0 (0%)	5 (16.1%)	13 (41.9%)	9 (20.9%)
Psychiatry	1 (5.3%)	2 (6.4%)	1 (3.2%)	1 (2.3%)
Internal medicine	8 (42.1%)	14 (45.2%)	10 (32.3%)	20 (46.5%)

3.3

Duration of symptomatic outbreaks and colonization caused by K. pneumoniae NDM

The duration (from diagnosis to extinction) of symptomatic outbreaks caused by K. pneumoniae NDM in hospitals in the Mazovian Voivodeship in 2016 ranged from 1 to 168 days (average 49 days ± 25 days), and in 2017, from 3 to 246 days (mean 59 days ± 24 days). In most cases, epidemic outbreaks were extinguished within more than 30 days (58% of outbreaks in 2016 and 68% in 2017). The duration of colonization in 2017 ranged from 1 to 146 days (mean 22 days ± 14 days) and in 2016 it ranged from 1 to 237 days (mean 52 days ± 37 days). In most cases, colonization lasted up to 60 days (68% in 2016 and 71% in 2017). It was also found that the mean duration of a symptomatic outbreak and colonization did not differ significantly in 2016, but in 2017, the mean duration of a symptomatic outbreak was significantly longer (p<0.01) (Fig. 3).

Fig. 3

Considering the number of hospitalizations, the risk of symptomatic infection was higher in hospitals outside of Warsaw than in hospitals in Warsaw, while the risk of colonization was higher in hospitals in Warsaw. The risk of symptomatic infection doubled in 2017 in hospitals in Warsaw (OR 1.8; 95% CI, 1.2–3.5) and tripled in hospitals outside of Warsaw (OR 2.9; 95% CI, 1.8–4.2) compared to 2016. During the analyzed period, there was a downward trend in the incidence of symptomatic and colonization outbreaks caused by K. pneumoniae NDM in Warsaw hospitals, while in hospitals outside of Warsaw, a reverse trend was observed.

K. pneumoniae is a common, opportunistic Gram-negative bacterium affecting patients with compromised immune response. Antimicrobial drug abuse, mainly of third-generation cephalosporins and carbapenems, resulted in a rapid increase in drug resistance. K. pneumoniae NDM infections represent an urgent public health problem with significant economic burden [18, 19].

Based on our results, reported infections caused by K. pneumoniae NDM in hospitals in the Mazovian Voivodeship increased from 0.96/1,000 hospitalizations in 2016 to 2.04/1,000 hospitalizations in 2017. Also, the number of symptomatic transmissions increased from 19 in 2016 to 31 in 2017. Interestingly, the number of symptomatic transmissions in hospitals located in Warsaw decreased from 11 to 7, while in hospitals located outside of Warsaw, it increased from 8 in 2016 to 24 in 2017. An upward trend in the number of symptomatic transmissions in Poland, including K. pneumoniae, is observed by other authors [20, 21]. In the Lublin Voivodeship in 2017, three outbreaks of K. pneumoniae CPE were reported to PSES, all caused by K. pneumoniae NDM. In 2018, this number increased to 24, including 11 outbreaks caused by K. pneumoniae NDM. In the first 10 months of 2019, 37 outbreaks were reported, among them, 9 were caused by K. pneumoniae NDM [22]. The spread of bacterial infections in Poland, including K. pneumoniae NDM, may be the result of irrational antibiotic therapy, as well as insufficient sanitary, hygienic, and epidemiological supervision and an insufficient number of microbiological screening tests. According to the Supreme Chamber of Control in Poland and ECDC data, only 10% of hospitals in Poland have properly equipped isolation rooms (a separate room with full sanitary facilities) [23, 24].

In our study, geographic differentiation within districts was observed in the frequency of determining the presence of K. pneumoniae NDM, with the worst situation in both analyzed years taking place in the districts outside of Warsaw. This may seem surprising since the largest number of multi-profile hospitals are based in Warsaw. Reasons for such a situation can be seen in the underestimation of the problem in smaller centers and the misconception that colonization and infections caused by K. pneumoniae NDM occur in patients of highly specialized departments. It may also be due to financial reasons that limit microbiological screening.

In both analyzed years, hospitals reported more colonization transmission than symptomatic transmission (62% vs. 38% in 2016 and 58% vs. 42% in 2017). However, the duration of the symptomatic outbreak was significantly longer compared to the duration of colonization. This can be explained by the need for longer hospitalization periods for patients with symptomatic infections who require long and expensive therapy due to their condition. The problem of the high cost of treating diseases caused by multi–drug resistant pathogens is well described [25, 26]. For infections caused by CPE, these costs can be significantly higher than for other resistant bacteria because the disease is more difficult to treat and requires wider prophylaxis and proper control. On the other hand, the role of the carriers in the transmission of the pathogen in the population should not be minimized. It was found that the risk of developing a symptomatic infection in the case of a K. pneumoniae CPE carrier is 16.5%, which was demonstrated in a meta-analysis by Tischendorf et al. [27].

The risk of developing a symptomatic infection in the case of early colonization increases in patients with leukemia during chemotherapy, after organ transplantation, and during treatment in the intensive care unit [28]. Our results demonstrated an increased number of colonization transmissions (from 5 in 2016 to 9 in 2017) and symptomatic infection transmission (from 0 in 2016 to 13 in 2017). Such a situation may negatively affect treatment outcomes and lead to patient death.

The main limitation of this study is that not all hospitals submitted data concerning nosocomial infections to PSES. Healthcare facilities should obligatorily report data on microbiological screening tests for alarm pathogens, including K. pneumoniae NDM. Another limitation is the presentation of data obtained between 2016 and 2017, but despite this, the results obtained may be relevant in retrospective analyses.

In the Masovian Voivodeship, the epidemiological situation related to infections and colonization caused by K. pneumoniae NDM is dynamic and disadvantageous, which implies the necessity to monitor anti-epidemic measures. The epidemic situation in hospitals outside of Warsaw seems to be worse compared to hospitals in Warsaw, with a higher risk of symptomatic infection caused by K. pneumoniae NDM.