High-Prevalence of CTX-M-15-Producing Klebsiella pneumoniae among Inpatients and Outpatients with Urinary Tract Infection in Southern India

Sir,

The population of India of over one billion represents a potentially vast reservoir of antimicrobial resistance genes including those encoding ESBLs. Occurrence of ESBL-producing Enterobacteriaceae has been described from India mostly utilizing phenotypic detection methods alone. The present study was conducted to identify the occurrence of ESBL genes in urinary isolates of Klebsiella pneumoniae collected in Southern India.

The study was conducted at two hospitals, one in Gulbarga and the other in Raichur between July 2005 and March 2006. The two cities are approximately 150 km away from each other and have populations of ca. 440,000 and 160,000, respectively. A total of 1288 non-duplicate urine specimens were collected from 642 inpatients and 646 outpatients. K. pneumoniae was identified by standard biochemical tests including citrate, urease, indole, Voges-Proskauer, glucose and inositol. As a result, K. pneumoniae was recovered from 270 of 1288 urinary specimens. Among them, 115 (43%) were from inpatients and 155 (57%) were from outpatients. ESBL production was confirmed in 260 (96%) isolates using the disk method defined by the Clinical Laboratory Standards Institute (CLSI, 2007). All of these isolates produced ≥ 5 mm increase in zone diameters with either cefotaxime or ceftazidime disks when clavulanic acid was added. They represented 95% and 97% of inpatient and outpatient isolates, respectively. Of those, 255 isolates were resistant to cefotaxime and were assigned to phenotypic group I, which was further classified into subgroups depending on susceptibility to ceftazidime and cefepime (Table). Another 5 isolates with resistance to ceftazidime but not cefotaxime were assigned to phenotypic group II. The remaining 10 non-ESBL-producing isolates susceptible to both cefotaxime and ceftazidime were assigned to phenotypic group III. For non-β-lactam antimicrobials, 92% and 95% of inpatient and outpatient isolates were resistant to ciprofloxacin, respectively. High rates of resistance to gentamicin (79% and 83% in inpatient and outpatient isolates, respectively) and amikacin (66% and 69% in inpatient and outpatient isolates, respectively) were also observed. Resistance to chloramphenicol, which is still commonly prescribed for infections caused by K. pneumoniae in India, was observed in 90% and 92% of inpatient and outpatient isolates, respectively. All the study isolates were susceptible to ertapenem except for one isolate which was intermediately resistant. No resistance to imipenem was identified.

Table.

Phenotypic and genotypic characteristics of the K. pneumoniae clinical isolates

Phenotypic		No. of isolates (n=270)	Antimicrobial agents a										ESBLs	Pulsotypes
Group	Subgroup		CAZ	CTX	FEP	FOX	ETP	IPM	CIP	CHL	GEN	AMK
I	a	155	R	R	R	S	S	S	R	R	S/I/R	S/I/R	CTX-M-15	A
	b	71	R	R	I	S	S	S	R	R	S/I/R	S/I/R	CTX-M-15	A
	c	18	I	R	S	S	S	S	R	S/I/R	S/R	S/I/R	CTX-M-15	B
	d	7	S	R	S	S/R	S	S	I/R	I/R	R/S	S	CTX-M-15	C
	e	4	R	R	S	S	S	S	R	R	R/S	R/S	CTX-M-15	A, B ,K , L
II	-	5	R	S	S	S	S	S	S	I/S	S	S	SHV-12	D
III	-	10	S	S	S	S	S	S	S	I/R/S	S	S	None	miscellaneous

^aCAZ, ceftazidime; CTX, cefotaxime; FEP, cefepime; FOX, cefoxitin; ETP, ertapenem; IPM, imipenem; CIP, ciprofloxacin; CHL, chloramphenicol; GEN, gentamicin; AMK, amikacin. Susceptibility results were interpreted according to the CLSI criteria: R, resistant; I, intermediate; S, susceptible.

A total of 35 isolates from all phenotypic groups and subgroups were subjected to pulsed-field gel electrophoresis (PFGE) analysis using restriction enzyme XbaI (New England Biolabs, Ipswich, MA, USA) and CHEF III DR electrophoresis system (Bio-Rad, Hercules, CA, USA). Forty-five percent of isolates in the phenotypic group I that were examined by PFGE belonged to pulsotype A (Table). This pulsotype was commonly observed regardless of the source hospitals or the patient locations. The isolates in the phenotypic group II belonged to pulsotype D. A total of 60 isolates from various pulsotypes were then subjected to PCR analysis for detection of bla_TEM, bla_SHV and bla_CTX-M as previously described [1, 2]. All of the isolates belonging to the phenotypic group I yielded amplicons consistent with bla_CTX-M. The gene was confirmed as bla_CTX-M-15 in representative isolates by sequencing of the entire open reading frame using the following external primers: CTX-M-15-SF, 5′-CACACGTGGAATTTAGGGACT-3′ and CTX-M-15-SR, 5′-GCCGTCTAAGGCGATAAACA-3′. These results were consistent with the findings in a previous study reporting the predominance of bla_CTX-M-15 in a small number of isolates from Southern India [3]. Some of the group I isolates were positive for bla_TEM or bla_SHV. The sequences of representative amplicons were consistent with bla_TEM-1 and either bla_SHV-1 or bla_SHV-28, respectively, all of which are non-ESBL genes. All five isolates in group II were positive for bla_SHV-12, an ESBL gene, and negative for bla_TEM or bla_CTX-M.

The prevalence of ESBL producers in this study was strikingly high, particularly given the fact that more than half of the isolates were obtained from outpatients. One of the reasons contributing to the high prevalence of ESBL may be the crowded hospital conditions precluding implementation of optimal hygiene practices. The epidemic in the community is then likely fueled by unrestricted use of antimicrobials that may be purchased without prescription [4]. A recent study reported that ESBL-producing Enterobacteriaceae, including K. pneumoniae, was responsible for community-onset infections in India [5]. Dissemination of ESBL-producing K. pneumoniae in the community has a serious implication in empiric management of complicated urinary tract infections caused by this organism, given their tendency for co-resistance to non-β-lactam classes of antimicrobials, as was observed in the present study. CTX-M-15 is known to be an ESBL that has peculiar association with community-onset E. coli infections [6]. It may therefore be speculated that E. coli producing CTX-M-15 is already established in the community in this area and that the responsible gene is being exchanged between E. coli and K. pneumoniae by means of conjugal transfer, both in the healthcare and community environments.