Sunday, May 10, 2015

A look at the epidemiology of carbapenem-resistant Enterobacteriaceae

This piece is adapted from a paper I wrote recently for an emerging infectious disease epidemiology class.

Carbapenem-resistant Enterobacteriaceae: An Emerging Threat
Introduction to Enterobacteriaceae
The bacterial family Enterobacteriaceae is a diverse family that contains many of the members of the typical gut microbiota, including the potentially pathogenic Klebsiella pneumoniae and Escherichia coli, among others (1). This family has a long history of causing infectious disease when these bacteria make their way to sites other than the gastrointestinal tract. The advent of antibiotics, however, put a sharp end to that history. Since the introduction of penicillin in 1945, the incidence of these infections has decreased drastically (2). Unfortunately, in recent years there has been an emergence of carbapenem-resistant Enterobacteriaceae that are now threatening to un-do the strides that have been made in elimination of these infections.

Enterobacteriaceae is a family of gram-negative, rod-shaped bacteria. They can cause a variety of disease manifestations, depending upon the body system infected. Symptoms can include, but are not limited to, fevers, pain and/or pus from wounds, and severe pneumonia (2). These infections are typically diagnosed through laboratory analysis to isolate the organism from blood, urine, or cerebrospinal fluid (2). Typical Enterobacteriaceae infections are easy to treat with a class of antibiotics known as the carbapenems. The carbapenems include drugs such as ertapenem, meropenem, and imipenem. They are bactericidal members of the β-lactam family (3), the same family as penicillin, and they have a long track record of successful clearance of infections.

Carbapenem resistance and the impact on disease
Carbapenem resistance among Enterobacteriaceae in the United States and the world was almost non-existent before 1992. In the time from 1986 to 1990, the National Nosocomial Infection Surveillance system reported that 2.3% of the Enterobacteriaceae samples tested were deemed “not susceptible” (4). It was known during that time that many of these bacteria possessed extended-spectrum β-lactamases that were capable of degrading many other antibiotics and giving them the distinction of “not susceptible,” but they were still susceptible to the carbapenems (5). Since then, carbapenem resistance has been on the rise. Using isolates submitted for the Meropenem Yearly Susceptibility Test Information Collection Program, the incidence of resistance by different bacteria to these carbapenems has been tracked. Among K. pneumoniae alone, the incidence of resistance to meropenem increased from 0% in 1999 to 5.6% in 2008. The trend for E. coli was slightly less worrisome, with the incidence of resistance to meropenem and imipenem rising to just 0.8% and 0.2% respectively (6). With the emergence of resistance to the carbapenems, there are highly limited therapeutic options remaining to fight these infections (1). Most carbapenem-resistant isolates show resistance to all standard antibiotics. The drug of choice for these infections has typically become colistin. Alarmingly, however, cases of K. pneumoniae that were resistant to both carbapenems and colisin were found in 2009 in three different medical centers in Detroit, MI (7). Other potential treatment options, such as polymyxin and tigecycline, remain experimental at this time (8).

The increase in drug resistance among these pathogens has led to an analogous increase in disease. These pathogens are typically acquired in the hospital setting, so this is where surveillance generally takes place. Risk factors for infection include advanced age, increased illness severity, increased length of hospital or intensive care unit stay, the use of catheters, ventilators, dialysis, surgery, and prior exposure to β-lactam or other antibiotics (9). According to data from Centers for Disease Control (CDC) surveillance programs, in 2012, 3.9% of patients admitted to a surveilled hospital for acute care reported one or more infections with a carbapenem-resistant Enterobacteriaceae. The numbers are even worse for those under long-term care; 17.8% of patients were reported to have at least one infection (10). Overall, the CDC reports that in 2013, there were 9000 carbapenem-resistant Enterobacteriaceae infections, with about 85% of these being caused by K. pneumonia. These infections led to over 600 deaths (11), with the mortality approaching 50% in some high-risk populations, such as those with poor functional status or additional antibiotic exposure (12). A matched retrospective historical cohort study that examined 32 Israeli patients who were diagnosed with bacteremia from carbapenem-resistant K. pneumoniae and compared the outcomes for these patients with those of patients diagnosed with susceptible K. pneumoniae found that the attributable mortality rate for the carbapenem-resistant infection was 50% (13). Similarly, two different matched case-control studies at Mount Sinai Hospital found that being infected with a carbapenem-resistant strain was associated with a two-fold increase in mortality rate (14).

Although carbapenem-resistant Enterobacteriaceae has so far been limited to the hospital setting, Enterobacteriaceae in general can cause community infections as well. The threat of carbapenem resistance spreading outside the hospital setting and contributing to community infections is worrisome (1).

Major mechanism of carbapenem resistance: the KPC enzymes
The largest proportion of carbapenem-resistant Enterobacteriaceae infections are caused by K. pneumoniae. This is likely due to the fact that it appears the emergence of a large proportion of the resistance can be traced to this organism. Original resistant isolates in 1997 were found to contain the AmpC β-lactamase chromosomally, as well as not contain one of the outer membrane porins (15). The AmpC β-lactamase is a cephalosporinase, not a carbapenemase, which gives bacteria resistance to cephamycins, but not to carbapenems. However, in combination with the loss of a 42 kDa outer membrane protein, resistance to the carbapenems is achieved. The identity and function of this outer membrane protein is not well understood (15). In 2001, a K. pneumoniae carbapenemase-producing Enterobacteriaceae was reported from a clinical isolate in North Carolina. This novel carbapenemase, termed K. pneumoniae carbapenemase-1 (KPC-1), demonstrated broad β-lactamase activity, and was capable of inactivating both imipenem and meropenem, therefore allowing the pathogen to survive in spite of the antibiotic treatment. Additionally, it was shown that this isolate was resistant to other extended-spectrum drugs. The gene encoding this carbapenemase was found to reside on a plasmid (16), which facilitates its ability to spread from organism to organism. The spread of carbapenem resistance has been traced from this original clinical isolate to subsequent outbreaks. In 2002-2003, surveillance in New York City identified a rising number of K. pneumoniae isolates that were carbapenem-resistant. The study also found that all isolates from two separate hospital outbreaks contained the same KPC enzyme, KPC-2 (17), which is genetically identical to the KPC-1 from the original North Carolina isolate (5). After these initial outbreaks in New York City, KPC-containing bacteria could be found endemically in hospitals throughout the New York and New Jersey areas (17).

From 2001 to 2005, the carbapenem resistance stayed within the eastern United States. Since then, KPC-producing organisms have spread to other countries and across the United States. The first intercontinental transfer was reported from the United States to France. There, in 2005, an 80-year old man was found to have an infection with a carbapenem-resistant K. pneumoniae. The man had had a recent visit and short hospital stay in New York City, where it is likely he acquired the bacteria (18). Shortly after the case in France, the first outbreak outside of the United States occurred in Tel Aviv, Israel. During a three year study from 2004-2006 in Tel Aviv, the proportion of isolates showing resistance to carbapenems increased dramatically. The first two years of the study showed rates of just 0.4% and 0.07%, while in 2006 3.1% of isolates were resistant. Of the 2004-2005 resistant isolates, none of them contained a KPC, whereas the majority of the 2006 resistant isolates did. It was found that 75% of these resistant isolates were clonal, highlighting the transmissibility of these pathogens (19). When the pathogens isolated from the Israeli outbreak were compared to those from the United States, 35% of the isolates were found to be genetically identical or highly similar (20). This suggests that the Israeli outbreak was precipitated by a pathogen that originated in the United States.

Bacteria containing the KPC enzyme can now be found endemically in the United States, Israel, and Greece. Additionally, these bacteria have been reported around the globe in Brazil, China, Colombia, Norway, the United Kingdom, India, Sweden, Italy, Finland (5), and Canada (3). It is likely these bacteria also exist in other countries where surveillance for them has not yet taken place. Molecularly, it has been found that 70% of the isolates that have been logged in the CDC database from 18 states and from Israel and India can be linked to a single strain, multilocus sequence type 258 (5).

Testing for the infections caused by carbapenem-resistant Enterobacteriaceae has been difficult due to the presence of the KPC enzyme. The KPC enzymes are not always detected by the routine microbiological susceptibility testing (5). Reports estimate that the automated testing systems will label anywhere from 7-87% of the KPC-producing bacteria as susceptible to the carbapenems (21). This is due to the fact that the KPC-producing bacteria have highly variable minimal inhibitory concentrations depending upon exactly which carbapenem is used in the test. Ertapenem has been shown to be the most reliable indicator of the presence of the KPC enzyme (22).

Another enzymatic player
While much of the Enterobacteriaceae resistance to carbapenems starting in 2001 can be traced with the KPC enzymes, a second novel class of β-lactamases has also contributed to the spread in more recent years. In December of 2007, while traveling to India, a 59-year old Swedish man was hospitalized in New Delhi. Upon his return to Sweden in January 2008, a K. pneumoniae clinical isolate that was resistant to carbapenems was found. After further evaluation of the isolate, it was found that this isolate produced a novel metallo-β-lactamase, termed New Delhi Metallo-1 (NDM-1) (23). The United Kingdom saw an influx of this enzyme around the same time period. The enzyme was most often found on a plasmid (24). A strong link to India and Pakistan was found for this particular enzyme, as 9 of the 19 affected patients in the U.K. had been recently hospitalized in India or Pakistan (8).

Factors contributing to the spread of resistance

There are a number of major factors that have played a role in the rapid spread of these carbapenem-resistant Enterobacteriaceae. One is the presence of the resistance elements within plasmids, which facilitates the transfer from one bacterium to the next. Another is the migration of humans through the ease of air travel, which has allowed the original U.S. isolates to spread to other countries (24). An additional factor that serves as a double-edged sword is the high use of antibiotics. The use of carbapenems against Enterobacteriaceae was originally necessitated by the emergence of extended-spectrum β-lactamases, which were selected for by the widespread use of the β-lactamase antibiotics. Recent studies now suggest that 70-90% of Enterobacteriaceae possess these extended-spectrum β-lactamases, leaving carbapenems as the only class of antibiotics left. In light of increased use of the carbapenems, which were originally reserved as a drug of last resort, there has been strong selective pressure on the bacteria for carbapenem resistance to develop (25). Also, it has been shown that there are environmental reservoirs of these pathogens that are difficult to deplete. A study in the ICU of Dandenong Hospital in Melbourne, Australia from 2009 to 2012 identified the grating and drain of sinks in the unit as a persistent reservoir of the pathogens; even after six decontamination attempts, the bacteria could still be isolated from these areas. They also clonally linked the isolates from the drain area to isolates found in patients, confirming that this environmental reservoir contributed to the caseload (26).

Recent outbreaks and control strategies

In spite of added precautions and monitoring to prevent the resistance from spreading, outbreaks of these bacteria still occur. The most recent notable outbreak of carbapenem-resistant Enterobacteriaceae occurred at the UCLA Medical Center in late 2014. In this outbreak, 7 patients became infected and 2 died following exposure to the bacteria during tests involving a duodenoscope. Although the scopes were cleaned following FDA guidelines, the contaminating bacteria were still present and able to be transmitted to the patients. This outbreak comes on the tail of multiple other similar outbreaks affecting 150 patients in Illinois, Pennsylvania, and Washington (27). In Illinois, hospitals have put a stop to this transmission by instating a new method of cleaning the duodenoscope (28), which has now also been adopted in California (27)

Studies have shown that appropriate interventions can play a crucial role in containing these pathogens. One study in Israel that focused on a 2006 outbreak highlighted the effectiveness of an appropriate country-wide containment strategy. During this outbreak, the government took a very active role in surveillance and case monitoring, allowing them to reduce the incidence of carbapenem-resistant infections from 41.9 cases per 100,000 patient days to just 11.7 cases per 100,000 patient days in one year, nearly an 80% decrease. The study found a direct correlation between compliance with the national guidance and reduction in incidence (29).

Control strategies to limit the spread of infections and surveillance programs to allow early identification are likely the best current strategies for containment of outbreaks with carbapenem-resistant Enterobacteriaceae. As far as a long-term strategy to deal with these infections, however, discovery of new classes of antibiotics is likely the strongest approach. Until earlier this year when teixobactin was discovered (30), new classes of antibiotics had not been discovered since the golden age of antibiotic discovery in the 1930s-1960s (31). Even the most recent teixobactin does not work well against gram-negative bacteria, such as the Enterobacteriaceae (30). For continued control of these and other emerging pathogens, it is likely new antibiotic discovery routes will need to be pursued. While there is hope to prevent spread and combat these infections with novel antibiotics, the never-ending arms race between humans and bacteria is sure to continue.