High Burden of Extended-Spectrum β-Lactamase–Producing Escherichia coli and Klebsiella pneumoniae Bacteremia in Older Adults: A Seven-Year Study in Two Rural Thai Provinces

Abstract. Bloodstream infection surveillance conducted from 2008 to 2014 in all 20 hospitals in Sa Kaeo and Nakhon Phanom provinces, Thailand, allowed us to look at disease burden, antibiotic susceptibilities, and recurrent infections caused by extended-spectrum β-lactamase (ESBL)–producing Escherichia coli and Klebsiella pneumoniae. Of 97,832 blood specimens, 3,338 were positive for E. coli and 1,086 for K. pneumoniae. The proportion of E. coli isolates producing ESBL significantly increased from 19% to 22% in 2008–2010 to approximately 30% from 2011 to 2014 (P-value for trend = 0.02), whereas ESBL production among K. pneumoniae cases was 27.4% with no significant trend over time. Incidence of community-onset ESBL-producing E. coli increased from 5.4 per 100,000 population in 2008 to 12.8 in 2014, with the highest rates among persons aged ≥ 70 years at 79 cases per 100,000 persons in 2014. From 2008 to 2014, community-onset ESBL-producing K. pneumoniae incidence was 2.7 per 100,000, with a rate of 12.9 among those aged ≥ 70 years. Although most (93.6% of E. coli and 87.6% of K. pneumoniae) infections were community-onset, hospital-onset infections were twice as likely to be ESBL. Population-based surveillance, as described, is vital to accurately monitor emergence and trends in antimicrobial resistance, and in guiding the development of rational antimicrobial therapy recommendations.


INTRODUCTION
Escherichia coli and Klebsiella pneumoniae are two of the most common causes of bloodstream infections (BSIs) worldwide. 1,2 β-Lactam antibiotics are the preferred treatment because of their selective toxicity, broad spectrum of activity, and low cost. 3,4 However, E. coli and K. pneumoniae, and other Enterobacteriaceae, have developed extended-spectrum β-lactamase (ESBL)-producing capability and can hydrolyze penicillins, cephalosporins, and monobactams, rendering them clinically ineffective. 5 Extended-spectrum β-lactamaseproducing E. coli and K. pneumoniae are increasingly reported worldwide in hospital and community settings, resulting in prolonged and more expensive hospitalizations and increased morbidity and mortality. 6,7 Prompt identification is essential for appropriate treatment and improved patient outcomes. 8,9 Extended-spectrum β-lactamase-producing E. coli have been identified throughout Thailand, [10][11][12][13] with the first documented case in 1994, 14 and a prevalence among different clinical specimens ranging from 22% to 59%, and evidence that they are becoming more common. [10][11][12][13]15 However, representative estimates of the incidence of ESBL-producing E. coli and K. pneumoniae and longitudinal trends in antimicrobial resistance are currently limited in Southeast Asia.
This study aimed to expand our understanding of ESBLproducing pathogen disease burden by estimating trends in population-based E. coli and K. pneumoniae bacteremia incidence, changes over time in the proportion of these cases due to ESBL-producing organisms, susceptibility profiles to other antibiotics, and the probability of recurrent infections. This study used the same well-established surveillance database that was previously used to estimate the incidence and susceptibility profiles of several other pathogens in rural Thailand. [16][17][18][19] MATERIAL AND METHODS Study setting. We conducted comprehensive BSI surveillance in two rural provinces: Sa Kaeo (SK) in eastern Thailand bordering Cambodia and Nakhon Phanom (NP) in northeast Thailand bordering Laos People's Democratic Republic. Both provinces are agrarian with estimated populations of approximately 526,000 and 734,000, respectively, from National Economic and Social Development Board (NESDB) projections based on 2010 census data. 20,21 This study included data between 2008 and 2014 from all 20 hospitals of the two provinces (16 district, two military [10-140 beds], and two provincial hospitals [225-327 beds]), and included all laboratories performing hemoculture. 22 Specimen collection and laboratory testing. As clinically indicated, hemocultures were performed using two different conditions, one optimized for aerobic growth and the other for enhanced growth of mycobacteria and other fastidious organisms as previously reported. A modification was applied in October 2011; the mycobacterial type bottles were not routinely processed but available on physician request. 23 Hemoculture bottles inoculated at district or military hospitals were kept at 15-30°C and transported to the provincial laboratory within 24 hours before placement in the BacT/ALERT 3D system (BioMérieux, Marcy-l'Étoile, France). Alarm-positive bottles were subcultured on chocolate, sheep blood, and MacConkey agars and incubated overnight at 35°C. Standard biochemical tests were performed for organism identification. 24 From 2012, Analytical Profile Index Microbial Identification Kits (BioMérieux) were used when standard testing was not definitive. Escherichia coli-and K. pneumoniae-positive cultures that grew a likely contaminant in the same bottle were excluded from our case counts. Repeat positive cultures that grew the same species within 30 days were excluded. Antibiotic susceptibility testing was performed using a modified Kirby-Bauer disk diffusion method. 25 To improve screen test sensitivity, ESBL screening used two singledisc diffusion tests: ceftazidime (CAZ) and cefotaxime (CTX) with zones of inhibition of £ 22 or £ 27 mm, respectively, were considered positive. 25 Confirmatory testing used a combination disc method. Ceftazidime/Clavulanic acid (CAZ/CLA) or CTX/CLA discs were placed on the same culture plates as the single discs. A ³ 5-mm increase in combination disc zone diameter compared with single disc alone confirmed ESBL production. Confirmation testing was routinely performed in SK. Nakhon Phanom started confirmatory testing in 2014. We defined an ESBL-producing pathogen as an isolate that was either confirmation or screen test positive when confirmatory testing was not available.
Escherichia coli and K. pneumoniae cases were classified as hospital onset (HO) when the hemoculture was drawn after day 3 of hospital stay, with day of admission being day 1. 27 Community onset (CO) cases were described by a positive hemoculture collected on or before day 3 of hospital stay. Preculture antibiotic use was any antibiotic received within 72 hours of culture determined by patient report and review of medical records. Recurrent infections were defined as positive hemoculture for the same pathogen more than 30 days after the initial positive culture. Mortality outcome was based on information available at the time of hospital discharge and, therefore, reflects solely in-hospital mortality.
The CO incidence rate of E. coli and K. pneumoniae bacteremia was calculated using age-specific population estimates from the NESDB. 20 For the period 2007-2009, the 2010 NESDB age distribution was applied to the 2007-2009 NESDB overall provincial population estimates, 20 as official intercensal estimates were not available. 21 To calculate HO incidence cases, we used the number of hospitalizations from 2008 to 2014 from hospital registration databases (S. Naorat and B. Piralam, personal communication). In SK, four district hospitals were missing 2008 and 2009 data. Together, these hospitals account for £ 15% of annual hospitalizations in our surveillance system, so we imputed hospitalization data from 2010 data for these four hospitals. Military hospitals were also excluded from HO incidence calculations, as hospitalization data were not available. Military hospitals account for < 1% of hospitalizations in each year of surveillance.
The chi-squared or Fisher exact tests were used as appropriate. Ninety-five percent CIs on incidence estimates were calculated from the normal approximation to the Poisson distribution.
Ethical considerations. The CDC Human Subjects Review Office determined that this protocol constituted routine public health activities and did not involve human subject research (CGH 2014-273).

Blood culturing practices and bacteremia incidence.
Between 2008 and 2014, 97,832 hemocultures were performed, with 12,813 (13.1%) positive for a pathogen. This excludes 3,859 cultures that grew a likely contaminant and pathogen in the same culture bottle. A total of 3,338 E. coli infections (1,595 from SK and 1,743 from NP) and 1,086 K. pneumoniae infections (529 from SK and 557 from NP) were identified. Annual E. coli bacteremia case counts increased from 373 in 2008 to 593 in 2014, with corresponding increases in annual incidence from 32.9 (2008) to 51.6 (2014) per 100,000 persons. Likewise, K. pneumoniae case counts increased from 116 in 2008 to 178 in 2014, with corresponding increases in incidence from 10.2 (2008) to 15.5 (2014) per 100,000 persons.
Overall proportion and trend of ESBL-producing E. coli and K. pneumoniae bacteremia. Extended-spectrum β-lactamase-producing capability among E. coli bacteremia cases ranged from 23.8% to 31.9% by age group; however, ESBL-producing K. pneumoniae was significantly more common among children aged 0-4 years compared with older persons (P < 0.01, Table 1). Prevalence of ESBL in E. coli increased over time: from 19% to 22% in 2008-2010 to approximately 30% from 2011 to 2014 (P-value for trend = 0.02). Prevalence of ESBL among K. pneumoniae cases had no apparent trend over time (P = 0.09). There were no statistically significant differences between the two provinces for the percentage of E. coli and K. pneumoniae that were ESBL producing (Table 1).
Hospitalization in the past year was significantly more frequent among patients with ESBL-producing E. coli and K. pneumoniae (63.7% and 71.6%, respectively) than in patients with non-ESBL-producing organisms (36.3% and 28.4%, respectively). The percentage of ESBL E. coli did not differ between provincial and district level hospitals (27.1% versus 25.7%, P = 0.37), but ESBL K. pneumoniae was significantly more common in provincial level hospitals than district hospitals (30.6% versus 22.6%, P = 0.01).
Trends in hospital-onset ESBL-producing E. coli and K. pneumoniae incidence. The incidence of HO ESBLproducing E. coli during the 7-year period varied between 0.13 Antimicrobial usage and microbiological characteristics of ESBL and non-ESBL-producing infections. Most E. coli (88.8%; 2,500/2,815) and K. pneumoniae (86.0%; 772/ 898) case patients did not receive antibiotics in the 72 hours before blood draw; however, pre-culture antibiotic use was three to four times more common among ESBL-producing case patients than in non-ESBL-producing case patients ( Table 2). The most common pre-culture antibiotics used were cephalosporins, which was significantly higher among ESBLproducing cases ( Table 2).
There were significantly different antibiotic susceptibility profiles between ESBL and non-ESBL CO E. coli infections. Extended-spectrum β-lactamase CO E. coli infections had significantly lower susceptibility to amoxicillin ( (Figure 3).
Of 864 ESBL-producing E. coli, two were carbapenemresistant Enterobacteriaceae (CRE) (one in 2012 encoding genes other than five tested for, and one in 2014 carrying bla NDM-1 gene). Of 277 ESBL-producing K. pneumoniae, three were CRE (two cases with bla NDM-1 in 2013 and 2014, and one case in 2012 encoding genes not tested for). Of 2,400 non-ESBLproducing E. coli, two were CRE (one case in 2011 encoding bla IMP and another in 2014 encoding genes not covered). There were no non-ESBL-producing K. pneumoniae CRE cases.

DISCUSSION
This is one of the largest population-based studies to estimate the incidence of ESBL-producing organisms. It includes almost 100,000 specimens, ordered for clinical purposes from all hospitals in two rural Thai provinces over a 7-year period, providing large-scale, longitudinal antibiotic usage and resistance data, and documenting increasing incidence of ESBL production and the development of CRE in these areas.
Our study shows more than 25 percent of E. coli and K. pneumoniae isolates causing bacteremia in hospitalized patients from 2008 to 2014 were ESBL producers. The proportion of these isolates increased over time leading to a notable upsurge in the incidence of ESBL E. coli bacteremia, particularly among persons ³ 50 years old. Our findings agree with increases in ESBL incidence and other antibiotic resistance in E. coli over the past 15-20 years seen in such diverse locations as Hong Kong, 28 Canada, 29 and Malawi. 30 However, this was not a universal finding as shown by stable levels of resistance in New Zealand from 2005 to 2011. 31 Variations in overall incidence and trends in antimicrobial resistance are likely driven in part by antibiotic usage patterns in different locations. In our study, patients with ESBL-producing  infections were more likely to have received antibiotics, most frequently cephalosporins, in the preceding 72 hours similar to that previously documented in a Korean study. 32 It is difficult to discern whether cephalosporin use led to the development of ESBL production or whether it resulted in clearance of non-ESBL-producing infections. Extendedspectrum β-lactamase-producing infections were associated with increased mortality in our study, which is consistent with previous reports from tertiary-level hospitals in Thailand 10 and elsewhere. 6,7 Our findings that ESBL-producing E. coli and K. pneumoniae are causing both CO and HO bacteremia agree with other studies in Thailand, 33,34 and the Thai studies that looked at temporal trends found ESBL-producing E. coli isolates are increasing. 34 Two factors explain the increase in incidence: a growing total case count and an increasing percentage of E. coli infections caused by ESBL-producing organisms. As in other studies, the increase in ESBL E. coli infections appears to add to the total burden of BSIs rather than represent a replacement of ESBL-negative with ESBL-positive infections. 28 Our observed prevalence of ESBL-producing capability among E. coli (22.0%) and K. pneumoniae (27.7%) isolates are higher than 2010 estimates from Kanoksil et al. 34 They reported an increase in the percentage of CO bacteremia due to ESBLproducing E. coli and K. pneumoniae in Northeast Thailand between 2004 and 2010 (2.9-18.0% and 10.0-16.4%, respectively). Part of this difference is likely due to variations in antibiotic susceptibility testing methodologies between the multiple provincial hospitals included in the Kanoksil study.
There are very few studies which provide population-based estimates of the incidence of ESBL-producing E. coli and K. pneumoniae globally. In Canada, an overall incidence of CO infections from ESBL-producing E. coli from all specimens of 7.7-13.3 per 100,000 population between 2004 and 2006 was reported, with blood accounting for only 10% of specimens. In our study, the incidence of CO ESBL-producing E. coli BSIs alone was 5.4-13.3 per 100,000 population. This demonstrates an incidence of ESBL-producing E. coli BSIs 8-10 times higher than in Canada. 35 Despite this, the age-related patterns of infection were similar with the highest incidence in persons aged > 70 years and an increasing incidence in people aged older than 50 years. Similar age patterns have been seen in multiple geographic locations for overall E. coli and all etiology BSI incidence. 30,31,36 According to projections for Thailand 2010-2040 by NESDB, 20 the population of persons over 50 years old is rising rapidly. If disease incidence remained at 2014 levels, by the year 2020, the number of CO ESBL-producing E. coli bacteremia cases will increase 19% in NP and SK because of the aging population alone. If we factor in the increasing incidence over time (1.2 cases/year), by 2020 case counts will increase 63% compared with that in 2014. Community onset ESBL-producing K. pneumoniae bacteremia incidence was stable, but application of the 2008-2014 K. pneumoniae incidence estimate to the projected 2020 population age structure results in an estimated 20% increase in cases compared with 2014.
About 3.8 percent of E. coli bacteremia were recurrent infections. This has been shown in other population-based studies to be strongly associated with HO infections and an independent predictor for death. 37 Our recurrence rate was slightly higher than the 2.4% documented in a 7-year population-based Canadian study. 29 The antibiotic treatment choices for infections caused by ESBL-producing pathogens are limited. We found a high prevalence of resistance to third-generation cephalosporins among ESBL-producing E. coli and K. pneumoniae, which is consistent with previous reports from Thailand. 38 A 2016 report 39 examined treatment options for children (0-15 years) with E. coli and K. pneumoniae infections visiting a tertiary care center in Bangkok. They also found the vast majority of ESBL-producing isolates were susceptible to amikacin but noted that poor tissue penetration and potential nephrotoxicity limited the use of amikacin and other aminoglycosides outside of urinary tract infections. 40 Another potential treatment option is piperacillin/tazobactam, but we found substantial non-susceptibility (> 25%) to this combination among ESBL-producing K. pneumoniae isolates, similar to Sethaphanich, 39 who found only 62.6% of K. pneumoniae isolates were susceptible to piperacillin/tazobactam. This makes it a poor choice for empiric treatment. For now, carbapenems remain a reliable treatment option, but it is troubling that we identified carbapenem-resistant, ESBL-producing isolates from five patients (two E. coli and three K. pneumoniae), especially in light of the recent identification of colistin-resistant E. coli from clinical isolates, healthy adults, and the environment in Thailand. 41,42 If we cannot prevent and control the spread of these types of multidrug-resistant bacteria, few options will remain for patients infected with this organism in the near future.
Our study had several limitations, including missing data to determine whether CO infections were health care associated or not. The retrospective nature and limited scope of data collection did not allow us to identify risk factors (apart from recent antibiotic use) for the presence of an ESBL-producing infection or potential primary foci of infection. Outcome data were only available for the period of hospital admission, which likely led to an underestimated case fatality rate, as previous work in rural Thailand show that many deaths among patients with bacteremia occur following hospital discharge. 34 The methods used for ESBL testing have changed over time, as have data collection practices, and data were not always available on antibiotic use before blood draw, treatment regimen after diagnosis/culture, length of stay, other indicators of disease severity, and patient outcome after treatment. Our incidence rate calculations are underestimated, as our data only cover hospitalized patients. Exclusion of cases that grew a contaminant in the same hemoculture bottle as a pathogen led to slight underestimation (overall ESBL-producing E. coli incidence differed by 1.4% [42.1 versus 41.5 per 100,000 population] and K. pneumonia differed by 0.3% [13.6 versus 13.3]).
Our study also had several important strengths. The inclusion of smaller district-level hospitals and all laboratories with hemoculture capability helped provide a comprehensive estimate of ESBL incidence at the community level, and the use of automated hemoculture systems enhanced the sensitivity of organism detection. 22 Furthermore, the use of this comprehensive data source allowed us to identify repeat cultures on the same patient and describe recurrent infections. Rigorous efforts were also made to identify and remove duplicate isolates to avoid multiple counting; failure to do this has been shown to overestimate incidence and resistance rates. 43 Population-based studies such as this are considered the gold standard for accurately estimating the incidence of BSIs and other infectious diseases and comparing incidence rates over time and place. 44  However, such studies are vanishingly rare in most parts of the world. A 2012 review of community-acquired BSIs did not identify any truly population-based incidence estimates from South or Southeast Asia, although one study extrapolated an incidence rate from an estimated capture rate for the catchment population. 45 Data collected from other studies can result in biased estimates which frequently underestimate the total burden of disease but overestimate the frequency of antimicrobial resistance. 43,44 Common issues leading to increased rates of drug resistance in many bacteria are the use of antibiotics without proper examination and diagnostic testing, for conditions not requiring antibiotics, the extensive use of antibiotics in the animal sector, and the use of overly broad-spectrum antibiotics. Surveillance systems are vital to monitoring emergence and trends in antimicrobial resistance. They are helpful for the development of rational therapy recommendations and policy for limiting the use of certain classes of antibiotics, especially in outpatient settings, and for enhancing efforts to prevent hospital-acquired infections which are more likely to be antibiotic resistant. These policies are especially important as an aging population will put more and more individuals in Thailand at risk of contracting difficult or impossible to treat bacterial infections. Laboratory-enhanced surveillance in Thailand is vitally important, given the high incidence of invasive ESBLproducing E. coli and K. pneumoniae bacteremia, together with the predicted increases in case counts for persons 50 years and older, and the emergence of carbapenemase resistance.