Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Klebsiella and Raoultella Species
Microbiology And Epidemiology
The genus Klebsiella , family Enterobacteriaceae, order Enterobacterales, contains at least five Klebsiella species and three Raoultella species: K. pneumoniae, K. oxytoca, K. granulomatis, K. variicola, K. singaporensis, R. terrigena, R. ornithinolytica, and R. planticola . These organisms are gram-negative, facultative anaerobic rods or coccobacilli. Like other members of the Enterobacteriaceae family, they ferment glucose and are oxidase negative. Enterobacter aerogenes is now reclassified as Klebsiella aerogenes . Despite the new designation, this organism will continue to be discussed in Chapter 140 with other members of the Enterobacter cloacae group, given the implications for identification and antimicrobial susceptibility.
The two most important human pathogens among Klebsiella spp. in terms of frequency and disease severity are K. pneumoniae and K. oxytoca . K. pneumoniae is further divided into subspecies pneumoniae , ozaenae , and rhinoscleromatis . The latter two K. pneumoniae subspecies are associated with chronic respiratory tract infections with unusual clinical presentations: ozena, a chronic atrophic rhinitis, and rhinoscleroma, a chronic granulomatous infection of the nose, respectively. These diseases occur most frequently in the tropics.
Subspecies may be difficult to separate from each other and from biochemically inactive strains of K. pneumoniae subsp. pneumoniae . K. granulomatis , formerly known as Calymmatobacterium granulomatis , is the agent of granuloma, inguinale or donovanosis, a sexually transmitted ulcerative disease of skin and subcutaneous tissues of the genital region (see Chapter 139 ).
Most Klebsiella and Raoultella spp. are ubiquitous in nature. They are found in environmental sources such as surface and drinking waters, soil, vegetation, sewage, and industrial waste, and they are associated with mucosal surfaces and diseases of humans, other mammals, birds, and reptiles. , K. variicola is a recently described species that has been isolated from plants and human sources, including the bloodstream. K. ozanae, K. rhinoscleromatis, and K. granulomatis have been recovered only from human clinical specimens. Raoultella spp. are isolated infrequently from human sources . R. ornithinolytica and R . planticola produce histidine decarboxylase and have been implicated in causing scombroid fish poisoning.
Members of the genera Klebsiella and Raoultella have properties similar to other Enterobacteriaceae with some distinctive features (see Key Points). Only R. ornithinolytica produces ornithine decarboxylase. Klebsiella spp. typically use citrate as a sole carbon source, often produce urease, and are positive on the Voges-Proskauer test. They are hydrogen sulfide and indole negative, except for K. oxytoca, which is indole positive.
K. pneumoniae subsp. pneumoniae and K. oxytoca usually grow as large, mucoid, lactose-fermenting colonies on MacConkey agar. Commercial and conventional biochemical identification systems can differentiate these two species; the lack of indole production by K. pneumoniae is the main identifying feature.
K. pneumoniae is difficult to differentiate from R. planticola and R. terrigena using basic biochemical tests and commercial identification systems. Studies conducted in Europe have suggested that a significant percentage of organisms identified as Klebsiella spp. may be R. planticola, but similar studies conducted in the US suggest that recovery of R. planticola is rare. , Prevalence of these organisms in human disease may vary by geographical location.
Matrix-assisted laser desorption/ionization–time of flight (MALDI-TOF) mass spectrometry has been used for identification of bacterial isolates, has demonstrated overall good performance in the identification of gram-negative bacilli, and can correctly identify and differentiate Klebsiella from Raoultella spp . Early reports revealed some misidentifications among the Klebsiella species and subspecies, but supplementation of databases should improve performance. , Raoultella sp. are reliably differentiated using MALDI-TOF.
The emergence of carbapenem-resistant K. pneumoniae (CRKP) and hypervirulent K. pneumoniae (hvKP) strains have contributed to the increase importance of K. pneumoniae as a pathogen. The spread of the bla KPC gene, by a variety of mechanisms, is of great epidemiological significance leading to an increase in CRKP infections. The worldwide spread of KPC-producing K. pneumoniae is primarily associated with a single multilocus sequence type (ST), ST258, and its related variants. The infections with hvKP are often community acquired and in previously healthy individuals. They are more often seen in the Asian Pacific rim but are spreading globally. These infections tend to present at multiple sites or spread rapidly. hvKP also has an increased ability to infect the central nervous system (CNS) and cause endophthalmitis. Like classic K. pneumoniae , hvKP strains are becoming increasingly resistant to antimicrobials. Note, the clinical microbiology lab has no current testing available to distinguish classic K. pneumoniae from hvKP. There is ongoing research suggesting that several biomarkers and quantitative siderophore production may be useful in identification.
The role of specific virulence factors in the pathogenesis of infection due to Klebsiella spp. is still being investigated, and the importance of different factors may vary by site of infection. Identified virulence factors include cell surface adhesins that may mediate attachment to mucosal surfaces, cell surface capsular polysaccharides and lipopolysaccharide that may inhibit phagocytosis and serum bactericidal activity, and other factors such as siderophores (i.e., iron-chelating compounds), cytotoxins, enterotoxins, and hemolysins. , The capsule is an established virulence factor for K. pneumoniae , but hvKN is able to produce increase amounts of capsular polysaccharide using genes encoded on a large virulence plasmid. Strains of K. pneumoniae , and K. oxytoca that produce enterotoxins and cytotoxins have biological effects in various animal models of gastrointestinal pathology. Overgrowth of these organisms in the gut of patients receiving antibiotics may lead to gastrointestinal syndromes, including antibiotic-associated colitis.
K. pneumoniae subsp. pneumoniae and K. oxytoca may be carried in the nasopharynx (1%–6%) and in the intestinal tract (5%–40%) of the general population. Carriage in hospitalized patients and healthcare personnel can be dramatically higher, particularly in the respiratory tract. Airway colonization with Klebsiella spp. and other gram-negative bacilli in the mechanically ventilated neonate is a moderate risk factor for bloodstream infections (BSIs) and is associated with severe bronchopulmonary dysplasia.
Clinical Manifestations
Klebsiella spp . are important opportunistic pathogens of children that cause a wide variety of infections, including healthcare- and community-associated BSIs; infections of the CNS, respiratory tract, urinary tract, musculoskeletal system, and peritoneal cavity; and deep abscesses and postsurgical, trauma, and burn wound infections. Approximately 75% of BSIs are healthcare associated. Outbreaks in neonatal units are a major concern. Neonatal infections include BSI, meningitis, brain abscess, conjunctivitis, hepatic abscess, endocarditis, pneumonia, necrotizing fasciitis, arthritis, osteomyelitis, urinary tract infection (UTI), and necrotizing enterocolitis (NEC). The clinical manifestations of infection are no different from those of other gram-negative bacilli. Although Escherichia coli is the most frequent cause of UTI, Klebsiella and Proteus spp . are also common members of the order Enterobacterales associated with UTI.
Gram-negative organisms account for >50% of BSIs in children. In a study of over 8000 pediatric bacteremia cases over an 11-year period at a pediatric tertiary care hospital in Philadelphia, Klebsiella accounted for 4.4% (363 isolates) of all infections. Among gram-negative pathogens, Klebsiella spp. were the most common organisms isolated. During a 6-year prospective study of gram-negative BSIs in a pediatric tertiary care medical center in Israel, K. pneumoniae was the most commonly identified pathogen, comprising 26% (109 of 419) of isolates. In a review of 57 cases of pediatric BSI caused by K. pneumoniae , 67% were younger than 12 months of age, and most had at least one underlying condition, including gastrointestinal tract abnormality in 56%, central venous catheter in 35%, neutropenia in 25%, and urinary tract abnormality in 16%.
Klebsiella spp. are common pathogens causing healthcare-associated infections. The Surveillance and Control of Pathogens of Epidemiologic Importance project based at Virginia Commonwealth University in Richmond has provided important data for identifying the predominant pathogens responsible for healthcare-associated BSIs in children. Although gram-positive organisms, especially coagulase-negative Staphylococcus spp., accounted for more than 50% of isolates, Candida spp . accounted for 9.3%, and Klebsiella spp . , Enterobacter spp., and E. coli accounted for 5.8%, 5%, and 5% of isolates, respectively. Amongst pediatric HAIs reported to NHSN, Klebsiella spp. account for 8.7% of overall infections and are the fourth most commonly reported pathogen behind Staphylococcus aureus , coagulase-negative Staphylococcus spp., and E. coli.
Other conditions associated with Klebsiella spp. BSI in childhood include prematurity, prior antibiotic exposure, preceding rotavirus gastroenteritis, Kasai procedure for biliary atresia, malignancy, postoperative status, burns, multiple trauma, acquired immunodeficiency syndrome, granulocyte disorders, homozygous sickle cell disease, solid-organ transplantation, the postoperative period after splenectomy, systemic lupus erythematosus, and interleukin-12 receptor β1 chain deficiency. , , Contaminated intravenous solutions also have been identified as a source of Klebsiella spp. BSI. ,
The overall mortality rate for Klebsiella spp. BSI is approximately 11%, with a mortality rate of 14.5% for healthcare-associated BSI. , An increased mortality rate is significantly associated with acute leukemia, neutropenia, healthcare-associated infections (HAIs), and previous corticosteroid therapy.
Cytotoxin-producing strains of K. oxytoca have been implicated in antibiotic-associated hemorrhagic colitis. , , The clinical manifestations of abdominal pain and bloody diarrhea are usually preceded by antibiotic treatment with a β-lactam agent. Colitis is usually segmental and located predominantly in the right colon. Cessation of the offending antimicrobial agent and supportive care typically result in clinical resolution. Enterotoxigenic and enteroaggregative strains of K. pneumoniae also have been implicated as causes of childhood gastroenteritis.
Raoultella spp. have been reported inconsistently as colonizing bacteria of the gastrointestinal tract of hospitalized neonates and as agents of BSI and adult pneumonia. , R. planticola has been isolated from adult urine; wounds, including an intra-abdominal abscess associated with pancreatitis; and contaminated infant formulas. R. ornithinolytica has been recovered from the blood of a neonate with visceral heterotaxy, functional asplenia, and complex congenital heart disease who had NEC; from one adult with a giant renal cyst; and from another with an enteric fever–like syndrome.
R. ornithinolytica and R. planticola appear to cause scombroid (histamine) fish poisoning. This condition follows ingestion of improperly stored (>20°C) scombroid fish (e.g., tuna, mahi-mahi, bonito, mackerel, sardine) and results from the breakdown of high concentrations of histidine into histamine by the action of bacterial histidine decarboxylase. , The incubation period can be 1 minute to 3 hours. Clinical manifestations can include marked skin flushing (especially of the face, upper trunk, and arms), headache, dizziness, abdominal cramps, vomiting, diarrhea, burning of the mouth, urticaria, generalized pruritus, and (rarely) hypotension and bronchospasm.
Treatment
K. pneumoniae and K oxytoca frequently are susceptible to the aminoglycosides, carbapenems, third- and fourth-generation cephalosporins, extended-spectrum penicillin/β-lactamase (ESBL) inhibitor combinations, trimethoprim-sulfamethoxazole, and fluoroquinolones; K. pneumoniae commonly is susceptible to cefazolin. Both organisms are routinely resistant to ampicillin due to plasmid-mediated β-lactamases such as TEM-1, TEM-2, and SHV-1.
In 2021, the Infectious Diseases Society of America published guidance for management of infections due to ESBL-producing Enterobacterales (newly termed ESBL-E) and carbapenem-resistant Enterobacterales (CRE). Because resistance and management are dynamic, the practitioner should consult the latest version of such guidance in managing these difficult infections: idsociety.org/practice-guideline/amr-guidance.
Extended Spectrum β-Lactam Resistance
HAIs due to ESBL–producing isolates of K. pneumoniae and K. oxytoca are increasingly common . , , , The ESBLs are a group of diverse enzymes derived primarily from genes coding for TEM- or SHV-type enzymes by single or multiple amino acid substitutions near the active site of the enzyme. , , ESBLs can hydrolyze third-, fourth-, and fifth-generation cephalosporins and monobactams, rendering organisms harboring the enzymes clinically resistant to these agents. However, in vitro ESBL activity against different β-lactam agents varies, and resistance is not always detected.
β-Lactamase inhibitor combination antibiotics (e.g., piperacillin-tazobactam) can retain some activity in vitro due to inhibition of ESBL by the β-lactamase inhibitor, but reduced susceptibility is typical. Cefepime, a fourth-generation cephalosporin, and piperacillin-tazobactam demonstrate an inoculum effect in vitro, with diminished susceptibility as the inoculum increase from 10 5 to 10 7 organisms. Susceptibility to the cephamycins (i.e., cefoxitin and cefotetan) and carbapenems is maintained.
For Raoultella , most species are broadly susceptible to antibiotics. They do have intrinsic resistance to ampicillin, and like Klebsiella spp., there are reported cases of ampC—and ESBL- producing isolates. The first case of carbapenemase- producing Raoultella ornithinolytica was reported in an adult with a postoperative perineal infection and others have since been reported. ,
Genes coding for ESBLs are usually located on transferable plasmids that can also harbor resistance determinants to other antimicrobial agents, including the fluoroquinolones and aminoglycosides. , The incidence of ESBL-producing Klebsiella spp. has increased. The SENTRY Antimicrobial Surveillance Program reported ESBL phenotype in 7% of Klebseilla spp. isolates from 1997 to 2000. , In another US surveillance network study conducted from 1996 to 2000, the ESBL rates for K. pneumoniae and K. oxytoca was 9% and 6%, respectively. Bacterial isolates collected between 2011 and 2013 from the US as part of the International Network for Optimal Resistance Monitoring, however, demonstrated ESBL phenotype in 15% of the Klebsiella species. Fortunately, ESBL prevalence among organisms isolated from children is considerably lower. The SENTRY Antimicrobial Surveillance Program reported that 1.1%–3.2% of Klebsiella spp. produced ESBLs. Although TEM-type ESBLs have been prevalent in the US, CTX-M-type ESBLs have emerged as the predominant ESBL type in many regions of the world and are common in the US. , These enzymes preferentially hydrolyze cefotaxime over ceftazidime.
The incidence of ESBLs in Klebsiella spp. in the US varies by institution, mainly because of the isolation of ESBL-positive K. pneumoniae in intensive care units and in immunocompromised hosts in tertiary care institutions. , Among 296 pediatric BSI isolates of E. coli and Klebsiella spp . from hospitalized US children in 1999–2003, 35 (12%) produced ESBLs: 17 K. pneumoniae, 10 K. oxytoca, and 8 E. coli isolates. ESBL-producing Klebsiella spp. represented 18% of all BSI isolates of Klebsiella spp. In this population, risk factors for BSI due to these ESBL-producing organisms were exposure to third-generation cephalosporins in the 30 days before infection, female sex, infection with a Klebsiella sp ., and corticosteroid use in the 2 weeks before infection. Children with BSI due to these organisms tended to have higher mortality rates and longer lengths of hospitalization compared with children with non–ESBL-producing isolates.
Bacterial strains producing ESBLs can be under detected because they usually do not demonstrate overt resistance to the third- and fourth-generation cephalosporins and monobactams by standard in vitro laboratory test methods. Detection can be optimized by using screening and confirmatory tests recommended by the Clinical Laboratory and Standards Institute (CLSI). In 2010, CLSI also established lower breakpoints for the cephalosporins (rather than phenotypic screening among the Enterobacteriaceae) as the primary method for detecting and reporting possible ESBL-producing organisms. For several antibiotics (such as, cefotaxime, ceftriaxone, ceftazidime, cefepime, aztreonam), the revised or lower minimal inhibitory concentrations (MICs) and larger disk diffusion breakpoints required to categorize an isolate as susceptible allow reporting of susceptibility test results for these agents, obviating the need for screening and confirmatory tests for ESBLs. An additional advantage of the revised breakpoint approach is that it is applicable for detection of ESBLs in all Enterobacteriaceae organisms compared with the traditional screening and confirmation test approach, which is applicable only to Klebsiella spp., E. coli , and Proteus mirabilis .
One of the difficulties in adopting the newer approach is that US Food and Drug Administration (FDA)–cleared commercial antimicrobial test systems currently use the traditional breakpoints. Laboratories desiring to adopt the newer breakpoints must verify the newer breakpoints and devise approaches to reinterpret MIC results from the test systems before reporting. Some commercial systems do not have adequate low-end antimicrobial concentrations to allow interpretation with the revised lower-susceptibility breakpoints. Whatever system is used, it is imperative that laboratories communicate clearly and that physicians interpret susceptibility test results as reported.
While in vitro screening tests are no longer recommended for routine use by CLSI, they may be appropriate for epidemiologic purposes and the option remains for routine screening based on local considerations. The phenotypic testing is designed to detect possible ESBL-producing strains of K. pneumoniae, K. oxytoca, E. coli, and P. mirabilis based on an elevated MIC or a reduced zone of inhibition by disk diffusion testing to select cephalosporins while still falling within traditional interpretative categories of susceptibility. Multiple antimicrobial agents are used in screening tests because in vitro activities of ESBL enzymes vary among β-lactam agents. Failure to identify these strains may lead to a report of false susceptibility to third- and fourth-generation cephalosporins, selection of inappropriate therapy, and treatment failure.
If a screening test suggests the presence of an ESBL, confirmation depends on demonstrating that the activity of one or more antimicrobial agents (commonly ceftazidime and cefotaxime) is enhanced by the addition of clavulanic acid. At least an eightfold decrease in the MIC or a minimum of a 5-mm increase in the zone diameter of inhibition by disk diffusion confirms the presence of an ESBL. If confirmed, the isolate is reported as resistant to all penicillin, cephalosporin, and monobactam agents, with a comment that the organism has been confirmed to produce an ESBL. Although the fourth-generation cephalosporins have greater in vitro activity against these isolates, their clinical efficacy is unclear, and the compounds also should be reported as resistant for ESBL positive organisms. , , Although the CLSI recommendation is to report the activity of ESBL inhibitor combination agents, some authorities consider the use of these agents in treatment of ESBL-positive organisms as ill advised. Until these controversies are resolved, it is prudent not to use these agents as monotherapy for serious infections due to ESBL-producing organisms.
Aminoglycosides, fluoroquinolones, cephamycins, and carbapenems usually maintain activity against ESBL-producing Enterobacteriaceae organisms. , , , Amikacin may be the most active aminoglycoside, with susceptibility of 89% of ESBL-producing E. coli and Klebsiella spp. Fluoroquinolone resistance among Klebsiella spp. causing UTI ranges from 4.2%–7.8%, with 80% of ESBL-producing E. coli and Klebsiella spp . susceptible to ciprofloxacin. , Resistance has developed after the use of cephamycins due to loss of porin channels that are responsible for cephamycin entry into the cell, and there is little clinical experience demonstrating their efficacy. The carbapenems retain full activity against ESBL-producing isolates and should be used for treatment of serious infections. ,
Klebsiella spp. can harbor other plasmid-mediated β-lactamases, including AmpC β-lactamases, similar to the chromosomal enzymes commonly found in Enterobacter spp., inhibitor-resistant TEMs, and carbapenemases. The AmpC β-lactamases are not inhibited by clavulanic acid and other β-lactamase inhibitors, and ESBL inhibitor combination antibiotics and the cephamycins usually are ineffective. The fourth-generation cephalosporins and carbapenems retain activity. , , , , ,
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here