Multidrug-resistant Acinetobacter baumannii outbreaks: a global problem in healthcare settings

Abstract

INTRODUCTION:

The increase in the prevalence of multidrug-resistant Acinetobacter baumannii infections in hospital settings has rapidly emerged worldwide as a serious health problem.

METHODS:

This review synthetizes the epidemiology of multidrug-resistant A. baumannii, highlighting resistance mechanisms.

CONCLUSIONS:

Understanding the genetic mechanisms of resistance as well as the associated risk factors is critical to develop and implement adequate measures to control and prevent acquisition of nosocomial infections, especially in an intensive care unit setting.

METHODS

A comprehensive search of the literature was performed using PubMed, ScienceDirect, and Web of Science. The search was restricted to original articles published in English related to risk factors, epidemiology, and multidrug-resistant A. baumannii (MDR-Ab). The key words used were (Acinetobacter baumannii OR A. baumannii) AND infection AND (multidrug-resistant OR MDR) AND (ICU), or (Acinetobacter baumannii OR A. baumannii) AND risk factors AND epidemiology. Case reports or conference abstracts were excluded. Two independent investigators searched the electronic databases using an identical method. The full texts of articles were reviewed by two independent reviewers to determine whether they met the eligibility criteria for inclusion. References in the included articles were reviewed to explore additional papers.

ACINETOBACTER BAUMANNII CONTEXT

Acinetobacter spp. is a pathogen that belongs to the Moraxellaceae family, which consists of 59 different species ^{1 , 2} . In this family, Acinetobacter spp. is the fifth most frequently isolated microorganism, distributed across five continents, among the gram-negative bacteria involved in nosocomial infections ³ . It is known that the species Acinetobacter baumannii is an opportunistic pathogen with clinical relevance ^{3 - 6} . The most frequent clinical manifestations are pneumonia associated with mechanical ventilation, bloodstream infections, urinary tract infections, and bacteremia associated with long periods of device use, meningitis, eye infections, intra-abdominal infections, surgical sites, the respiratory tract, and the gastrointestinal tract ^{7 , 8} . Nonetheless, this pathogen can survive in the intensive care unit (ICU) environment for up to four weeks due to its capacity to produce biofilms and thus contaminates patients admitted later ⁹ . Lipopolysaccharides (LPS), vesicles and proteins, polysaccharide capsules, phospholipases, proteases, outer membrane porins, and iron uptake systems are the most important factors for A. baumannii resistance ¹⁰ .

MDR-Ab is considered a hospital-acquired infection, which has been rapidly increasing worldwide due to the fitness effect of its resistance mutations ³ . The exacerbated and undue use of antibiotics associated with ineffective hospital interventions are related to the spread of MDR and consequently reduce treatment options. The World Health Organization (WHO) published in early 2017 a list of priorities for research into the development of active antibiotics against MDR and extensively resistant bacteria, which put A. baumannii first in the list of critical situations around the world ¹¹ . It was estimated that multidrug-resistant A. baumannii can cost $33,510 to $129,917 per infection ¹² . Moreover, patients with bacteremia can be related to high mortality rates due to multidrug-resistant A. baumannii (56.2%), when compared to A. baumannii strains with no multidrug resistance (4.7%) ¹³ . An average of 10.6% of patients die as a result of infections caused by MDR-Ab ¹² .

OVERVIEW OF A. BAUMANNII ANTIBIOTIC RESISTANCE

The key resistance mechanisms of A. baumannii are the low permeability of the outer membrane, alteration in antibiotic binding sites, and mutations, which can cause upregulation or downregulation of efflux system activity ^{4 , 10} . Among these mechanisms, alteration of bacterial membrane permeability by the outer membrane proteins (OMPs) is associated with the loss or reduced expression of porins ⁸ . This group is represented by OmpA, OprD, and CarO proteins ¹⁴ . The OccD1 (OprD) channel of the Pseudomonas aeruginosa species plays an important role in the uptake of molecules such as imipenem and meropenem. This OM channel is closely related to the OM family in A. baumannii and is the largest pore described amongst Occ proteins with efficient in vitro uptake responsible for transporting small molecules, presenting a huge potential for future antibiotic design ¹⁵ .

The efflux system expels toxic compounds to the extracellular environment. Within it, five families of systems have been described in A. baumannii, such as the major facilitator super family (MFS), ATP binding cassette (ABC), resistance nodulation division (RND), small multidrug resistance family 1 (SMR), multidrug and toxic compound extrusion (MATE), and drug/metabolite transporter (DMT) ¹⁶ . The RND family is well characterized and is represented by the AdeABC, AdeIJK, and AdeFGH efflux system ¹⁷ . Mutations can influence the expression of the efflux system, resulting in increased cases of clinical infections. A study highlighted resistance to aminoglicosides, tetracyclines, chloramphenicol, fluoroquinolones, some beta-lactams, and tigecycline related to mutations on the chromosome or plasmids ¹⁸ . The efflux systems CraA, AmvA/AedF, Tet(A), and Tet(B) of the MFS system are known to have a drug-specific substrate profile, and are involved in chloramphenicol, erythromycin, chlorhexidine, and tetracycline resistance ^{19 , 20} . The expression of Acel protein is strictly related to chlorhexidine transportation and the AbeM gene (a member of the MATE family), which confers resistance to fluoroquinolones through the H⁺ antiport ^{20 , 21} . Quinolone resistance can be related to the AbaQ gene, which belongs to the MFS transporter and has its N- and C- ends located in the cytoplasm, which confers its characteristic as a drug H⁺ antiporter-1 (DHA1). AbaQ knockout in A. baumannii confirmed its involvement with quinolone susceptibility, resulting in decreased susceptibility caused by active efflux transportation ²² .

It is known that the fluoroquinolone resistance mechanism is mainly encoded by mutations in DNA gyrase (gyrA, gyrB genes) and topoisomerase IV (parC, parE ), with gyrB and parE mutated at a lower frequency. These mutations are sequential, as primary mutations in gyrA81 are followed by mutations in parC88 and parC84 in A. baumannii. However, a study described strains carrying mutations in only the parC gene, revealing the involvement of other resistance mechanisms for fluoroquinolone ^{23 - 24} .

One of the main mechanisms of resistance to beta-lactam antibiotics is associated with changes in the structure or expression profile of penicillin binding proteins (PBPs) ²⁵ . PBPs are transglycosylases, transpeptidases, and carboxypeptidases, enzymes located in the plasma membrane, and are involved in the synthesis of peptidoglycan, an essential component of the bacterial cell wall. Once a PBP is acylated by a beta-lactam antibiotic, it is unable to catalyze hydrolysis of the covalent acyl-enzyme intermediate and is inactivated. Peptidoglycan transpeptidation cannot occur; thus, the cell wall is weakened ²⁵ .

PBPs are divided into high molecular mass (HMM) and low molecular mass (LMM). The first is responsible for insertion into the cell wall, which, depending on the structure and catalytic activity of the N-terminal domain, can be classified as class A or B ²⁶ . Therefore, changes in PBP expression lead to decreased susceptibility to these antimicrobial agents, favoring the occurrence of beta-lactam-resistant strains ²⁷ . Due to the lack of interaction that occurs in the connection between beta-lactams and PBPs, the susceptibility of A. baumannii strains to beta-lactams has been observed ^{27 - 29} .

Mutations can occur and modify the binding of antibiotics, inactivating some lipids, such as lipid A ³⁰ . Polymyxins interact with lipid A through the addition of phosphoethanolamine (PEtn), resulting in displacement of cations Mg²⁺ and Ca ²⁺, which destabilizes the membrane. These molecules are mediated by the pmrCAB operon ^{31 - 33} . Alterations in the pmrA-pmrB two-component system, which is also involved in lipid A biosynthesis, upregulate pmrC, influencing the synthesis of PEtn. It is known that LPS is synthesized through the lpx pathway; mutations in lpxA, lpxC, and lpxD genes lead to deficiency in LPS production and its complete loss, conferring the colistin resistance phenotype ^{34 - 35} . Colistin resistance can be chromosomal or plasmid-encoded, carrying the mcr gene (mcr-1 to mcr-5) ^{36 - 37} .

Carbapenemases, belonging to class A of Ambler (1980) and to group 2 of Bush and Jacob (2010) are considered one of the most versatile enzymatic families among β-lactamases, since they are able to hydrolyze most β-lactam antibiotics, such as carbapenems, penicillins, cephalosporins, and monobactams, in addition to being resistant against some commercial β-lactamase inhibitors ^{35 - 38} . Enzymes such as KPC-2, KPC-3, KPC-4, and KPC-10 ³⁹ , as well as GES-11, GES-12, and GES-14 ⁴⁰ , have already been described in A. baumannii ³⁸ .

Metallo-β-lactamases belong to class B of Ambler (1980) and group 3 of Bush and Jacoby (2010). They confer resistance against penicillins, cephalosporins, and carbapenems, and are inhibited by β-lactamase inhibitors (clavulanic acid, sulbactam, and tazobactam). The enzymes representing this family are VIM-1 and NDM-1, commonly related to penicillin hydrolysis ^{39 - 43} . Class C of Ambler (1980), group 1 of Bush and Jacob (2010), is represented by chromosomal cephalosporinases (AmpC), which hydrolyze penicillins, and cephalosporins at a low level. When the insertion element ISAba1 or ISAba125 is inserted upstream of the bla _AmpC gene, it is overexpressed, resulting in resistance to extended-spectrum cephalosporins as upstream ISAba induces strong promoter sequences ^{44 - 45} .

Oxacillinases belong to class D of Ambler (1980) and group 2 of Bush and Jacob (2010) and are encoded by the bla _OXA genes. These proteins hydrolyze carbapenems and penicillins at a low level and has weak hydrolysis of second and third generation cephalosporins ⁴⁴ . Oxacillinases have been reported in clinical isolates of A. baumannii associated with hospital outbreaks ⁴⁶ . Six subgroups of Class D carbapenem-hydrolyzing enzymes (CHDLs), including OXA-23, OXA-24, OXA-51, OXA-58, OXA-143, and OXA-235, were identified ⁴⁷ . These enzymatic groups hydrolyze penicillins at a high level and carbapenems at a low level. However, the presence of insertion sequence (IS) is considered a strong promoter for the increase of oxacillin expression and dissemination ⁴⁸ . It was reported that the ISAba1/bla _OXA-23 or ISAba1/bla _OXA-51 combination amplified resistance to carbapenems ⁴⁹ .

Aminoglycosides bind to 16S rRNA in the 30S ribosomal subunits and inhibit protein synthesis. Resistance is mediated by aminoglycoside-modifying enzymes (AMEs), such as acetyltransferases (AAC), adenyltransferases (ANT), and phosphotransferases (APH), which are found on mobile elements such as transposons and plasmids. AAC enzymes are responsible for modifying amino groups, while the ANT and APH enzymes act on hydroxyl groups, breaking bonds and inactivating the antibiotic molecule ¹⁰ . Methylase production (armA, rmtA, rmtB, rmtC, rmtD) decreases the affinity of the aminoglycosides for 30S ribosomal subunits ⁵⁰ . A study with carbapenem-resistant (CR) A. baumannii identified 97.2% of the isolates carrying the aph(3´)-VI gene, with the majority found in 4 different clusters (A, B, C, and E), conferring resistance to amikacin, and group D, harboring AME genes (aac(6´)-Ib, aac(3)-Ia, and aph(3´)-Ia), responsible for gentamicin resistance and intermediate resistance to amikacin ^{51 , 52} . The presence of methylase armA coexisting with bla _OXA-23 in MDR A. baumannii has been previously described and identified in quinolone-resistant A. baumannii ^{53 , 54} .

In addition to the multiple mechanisms of resistance, A. baumannii can acquire resistance genes through mobile genetic elements. Mobile elements, such as IS, transposons, genomic islands, integrons, and plasmids, are related to variations in the insertion site and carry strong transcriptional promoters that are abundantly synthesized ^{55 , 56} . Multiple A. baumannii plasmids have been reported: pA297-1, carrying gentamicin, kanamycin, and tobramycin resistance genes; pA297-3, carrying sulfonamide and streptomycin resistance genes; and pAb-G7-2, carrying an amikacin resistance gene ^{57 , 58} .

Transposons, such as Tn2006, Tn2007, and Tn2008, increase the spread of resistance genes and may present integrons, which were captured and express exogenous resistance genes ^{40 , 48 , 59} . Thus, integrons are composed of gene cassettes, and classes 1 and 2 are commonly found in A. baumannii clinical isolates ^{60 - 62} . As previously stated, insertion sequences act as strong promoters that increase the resistance levels of OXA carbapenemases in A. baumannii isolates ^{47 , 59 , 63} . Insertion sequence Acinetobacter baumannii (ISAba) can be located upstream of the resistant gene, overexpressing genes such as AmpC and OXA-51, which increases cephalosporin resistance ^{64 , 65} . Resistance to colistin in A. baumannii clinical isolates was related to the presence of the ISAba125 at the 3' end of the hns gene, disrupting the normal expression of a transcriptional gene regulator ⁶⁶ .

RISK FACTORS RELATED TO A. BAUMANNII

Risk factors are directly related to increased susceptibility in hospitalized patients who develop some type of infectious disease involving bacterial resistance, consequently resulting in mortality in nosocomial environments. Investigation of the risk factors associated with A. baumannii infection/colonization contributes to the prevention and control of bacterial resistance, reducing the impact of A. baumannii isolates ⁶⁷ (Table 1 and Table 2). The prevalence of A. baumannii infection and colonization is higher in ICUs, since patients with severe clinical conditions are hospitalized in such wards. In addition, these patients have compromised immune systems due to the presence of comorbidities, altered nutritional status, prolonged hospitalization, invasive procedures, immunosuppressive drugs, and broad-spectrum antibiotics ^{67 , 68} .

TABLE 1: Risk factors associated with infection and colonization caused by A. baumannii in adult ICUs.

Study	Place of Study	Study Period	No. of Patients	Cases	Controls	Risk Factors	P-value
JANG et al., 2009	Taiwan	1997-2006	154	77 patients with AB bloodstream infection.	77 patients with bloodstream infection without AB.	Use of central venous catheter, mechanical ventilation, colonization by AB, respiratory failure, cardiovascular failure.	P < 0.05
YE et al., 2010	Germany	2001-2005	209	49 patients with MDRAB.	160 patients with CSAB.	Previous use of antibiotics, use of mechanical ventilation, > 60 years, length of hospital stay.	P < 0.05
ROCHA et al., 2008	Brazil	2005-2006	275	84 patients with PAVM.	191 patients without PAVM.	Stay > 7 days in hospital, use of corticoids, invasive procedures, use of central venous catheter, and tracheostomy.	P < 0.05
BROTFAIN et al, 2016	Israel	2005-2011	129	46 patients with pneumonia and positive sputum culture for MDRAB 72 h after MV onset and bacteremia.	83 patients with pneumonia and positive sputum culture for MDRAB 72 h after the onset of MV, without developing bacteremia.	Hospitalization > 3 days in the ICU, advanced age, and recent bacteremia.	P < 0.05
BLANCO et al., 2017	United States	2005-2009	101	90 patients with MDRAB.	11 patients with CSAB.	Advanced age, previous hospitalization, heart failure, paralysis, HIV-AIDS, and rheumatoid arthritis.	P < 0.05
ELLIS et al., 2015	United States	2006-2012	671	302 patients with infection caused by MDRAB.	369 patients with infection caused by CSAB.	Length of hospital stay, transfer from another hospital, previous use of antibiotics	P < 0.25
HENIG et al., 2015	Israel	2007-2012	2380	1190 patients with CRAB.	1190 patients without AB.	Chemotherapy, organ transplant, chronic diseases, invasive procedures, recent bacteremia, tumor, hematological diseases, and recurrent hospitalizations.	P < 0.05
JUNG et al., 2010	South Korea	2008-2009	200	108 patients with bacteremia caused by AB.	92 patients without bacteremia.	Respiratory failure, mechanical ventilation, tracheal tube, central venous catheter, bacteremia caused by other microorganisms, previous use of antibiotics.	P < 0.05
NUTMAN et al., 2014	Israel	2008-2011	172	83 patients with bacteremia who died within 14 days.	89 patients with bacteremia who survived after 14 days.	Disease severity and surgical procedure.	P ≤ 0.10
CHUSRI et al., 2015	Thailand	2010-2011	394	139 patients with CRAB.	197 patients without AB and 58 patients with CSAB.	Use of fluoroquinolones, broad spectrum cephalosporins, and carbapenems > 3 days.	P < 0.05
MOGHNIEH et al., 2016	Lebanon	2012-2013	257	40 patients with AB.	217 patients without AB.	Use of urinary catheter, ICU contact pressure, gastrectomy tube, and carbapenem use.	P < 0.05
GUO et al., 2016	China	2012-2015	87	64 patients with bloodstream infection by MDRAB.	23 patients with bloodstream infection by CSAB.	Pneumonia, drain use, ICU stay> 7 days, and use of mechanical ventilation.	P < 0.05

AB: A. baumannii; MDRAB: multidrug-resistant A. baumannii; PAVM: pneumonia associated with mechanical ventilation; MV: mechanical ventilation; CRAB: carbapenem-resistant A. baumannii; CSAB: carbapenem-susceptible A. baumannii; ICU: intensive care unit.

TABLE 2: Risk factors associated with infection and colonization caused by A. baumannii in pediatric and neonatal ICUs.

Study	Place of study	Study period	No. of patients	Cases	Controls	Risk factors	P-value
BRITO et al., 2010	Brazil	2001-2002	33	11 patients with infectious conditions caused by AB.	22 patients without infectious conditions caused by AB.	Birth weight <2500 grams, respiratory syndromes, parental feeding, re-intubation, carbapenem use, and mechanical ventilation.	P < 0.05
DENG et al., 2011	China	2002-2008	349	117 patients with PAVM caused by AB.	232 patients without PAVM caused by AB.	Use of mechanical ventilation> 7 days.	P < 0.01
HSU et al., 2014	Taiwan	2004-2010	248	37 patients with bacteremia caused by AB.	74 patients without bacteremia and 137 patients with bacteremia caused by Escherichia coli or Klebsiella spp.	Cholestasis, gestational age < 29 weeks.	P < 0.05
LEE et al., 2017	China	2004-2014	40	37 patients with AB susceptible to imipenem	3 patients with AB resistant to imipenem	Prematurity, low birth weight (70% < 1500 g), prolonged intubation, percutaneous use of central venous catheter, inappropriate initial therapy, infection within the first 10 days of life, use of imipenem for up to 5 days, and high frequency oscillation ventilation.	P < 0.05
PUNPANICH et al., 2012	Thailand	2005-2010	176	91 patients with bacteremia caused by CRAB.	85 patients with bacteremia caused by CSAB.	Prematurity, use of mechanical ventilation, previous exposure to carbapenems.	P < 0.05
HOSOGLU et al., 2012	Turkey	2006-2007	192	64 patients with AB sepsis.	128 patients with blood samples without AB.	Stay in the ICU> 7 days, re-intubation.	P < 0.001
De OLIVEIRA COSTA et al., 2015	Brazil	2009-2012	101	47 patients with infection caused by BGN.	54 patients without infection caused by BGN.	Hematologic diseases, neutropenia > 3 days, previous use of antibiotics, previous hospitalization, stay in the ICU > 3 days.	P < 0.05
THATRIMONTRICHAI et al., 2013	Thailand	2009-2014	101	63 patients with CRAB pneumonia and 13 patients with CSAB.	25 patients with pneumonia without bacterial growth or caused by other microorganisms.	Weight of newborns, previous use of cephalosporins, surfactant replacement therapy, re-intubation, umbilical artery catheterization.	P < 0.05
REDDY et al., 2015	South Africa	2010	388	194 patients with blood culture or respiratory sample positive for AB.	194 patients with blood culture or negative respiratory sample for AB.	Mechanical ventilation and traumatic brain injury.	P < 0.05
ZARRILLI et al., 2012	Italy	2010-2011	161	22 patients with AB.	139 patients without AB in the first 48 h.	Use of mechanical ventilation and central venous catheter.
TRAN et al., 2015	Vietnam	2010-2011	2555	69 patients with sepsis caused by AB.	2486 patients without sepsis caused by AB.	Maternal infection, gestational age, central catheter, surgical procedure, and blood transfusion.	P < 0.05
KUMAR et al., 2014	India	2010-2012	65	33 patients with CRAB bloodstream infection.	32 patients without CSAB bloodstream infection.	Previous use of antibiotics, hospitalization > 7 days, use of mechanical ventilation > 7 days.	P < 0.05
WEI et al., 2014	Taiwan	2010-2013	59	12 deaths due to sepsis caused by MDRAB.	47 deaths due to sepsis caused by other microorganisms.	Prolonged intubation, mechanical ventilation, peripheral central venous catheter, umbilical catheter, total parental nutrition, ICU stay > 7 days, surgical procedure, and bronchopulmonary dysplasia.	P < 0.05
MACIEL et al., 2017	Brazil	2013-2015	21	21 patients with AB colonization without clinical manifestation.	17 patients without sepsis.	Low birth weight, prematurity, hospitalization time, previous exposure to beta-lactams, use of peripheral access, and respiratory syndromes.	P < 0.05

AB: A. baumannii; PAVM: pneumonia associated with mechanical ventilation; CRAB: carbapenem-resistant A. baumannii; CSAB: carbapenem-susceptible A. baumannii; BGN: gram-negative bacillus; MDRAB: multidrug-resistant A. baumannii; ICU: intensive care unit.

Skin colonization, length of hospital stays > 7 days, use of corticosteroids, and invasive procedures such as central venous catheter or tracheostomy, were the main risk factors related to the development of pneumonia associated with mechanical ventilation by MDR A. baumannii in hospitalized patients (Table 1) ^{69 , 70} . Risk factors such as use of urinary catheters for more than 6 days, ICU contact pressure > 4 days, presence of gastrectomy tubes, chemotherapy, organ transplantation, chronic diseases, invasive procedures, recent bacteremia, tumors, hematological diseases, recurrent hospitalizations, hospitalization time > 7 days, transfer from another hospital, and previous use of carbapenems or broad-spectrum cephalosporins were related to acquisition of MDR A. baumannii infection in adult patients hospitalized in the ICU ^{69 , 71} . Isolation of MDR A. baumannii after medical ICU (MICU) admission was related to a greater likelihood of the patient being older ⁷² . Previous hospitalization was associated with the isolation of A. baumannii after admission to the surgical ICU (SICU). Positive colonization in SICU was strongly correlated with heart failure, paralysis, human immunodeficiency virus infection and acquired immune deficiency syndrome (HIV-AIDS), and rheumatoid arthritis ⁷³ .

Bloodstream infections by A. baumannii are frequent in ICUs and have been associated with central venous catheters, mechanical ventilation, pneumonia, drain use, and respiratory and cardiovascular failure ⁷⁴ . The risk of bacteremia caused by A. baumannii was associated with respiratory failure, mechanical ventilation, endotracheal tubes, central venous catheters, surgical procedures, and previous use of antibiotics ^{75 , 76} .

Newborns are considered susceptible to A. baumannii colonization and infections, since they have immature immune systems. The risk is greater for newborns if they are also preterm (< 28 weeks) and underweight (< 2,500 g) ^{76 , 77} . Birth weight < 2500 grams, respiratory syndromes, parental feeding, re-intubation, carbapenem use, mechanical ventilation, hematologic diseases, neutropenia > 3 days, previous use of broad-spectrum antibiotics, use of invasive devices, immunosuppressants, corticosteroids, previous hospitalization, and ICU stay > 3 days were considered risk factors for the acquisition of A. baumannii infections in the neonatal ICU (Table 2) ^{78 - 80} .

Bloodstream infections caused by A. baumannii in neonates were related to the use of mechanical ventilation, and additionally to the presence of traumatic brain injury, previous use of antibiotics, hospitalization > 7 days, and use of mechanical ventilation > 7 days ^{81 - 83} . The weight of newborns (1000-1499 g), previous use of cephalosporins, surfactant replacement therapy, re-intubation, and umbilical artery catheterization were also indicated as risk factors for the development of neonatal pneumonia caused by carbapenem-resistant A. baumannii ⁸⁴ . Maternal infection, gestational age among 26 to 36 weeks, use of central venous catheters, surgical procedures, blood transfusions, prolonged intubation, use of mechanical ventilation, central peripheral venous catheters, umbilical catheters, total parental nutrition, ICU stay > 7 days, surgical procedures, and bronchopulmonary dysplasia were described as risk factors for sepsis by A. baumannii ^{77 , 85} . Cholestasis, gestational age < 29 weeks, prematurity, low birth weight (70% < 1500 g), prolonged intubation, central venous catheters, use of imipenem for up to 5 days, mechanical ventilation, and prior carbapenem exposure are related to A. baumannii bacteremia in neonates ^{10 , 86 , 87} . Similar results were reported for colonization in neonates ⁸⁸ . These studies pinpoint persistent endemic isolates in hospitals, highlighting the need to implement efficient control measures and prevent outbreaks.

Seasonality of A. baumannii infection is another risk factor that should be taken into consideration. A systematic review compiled studies showing 57.1% (12/21) of A. baumannii infections occurred in warmer seasons. The hypothesis for this was that it was due to enhanced lipid A moiety regulation, which was responsible for the virulence; it was also reported there was biofilm formation and a higher flow of people entering the hospital facility (carriers, patients, healthcare workers, and sanitation workers) in warmer months. This study highlights the importance of correlating different factors of A. baumannii adaptability in the ambient environment to implement preventive measures for seasonal peaks of infection ⁸⁹ .

Information related to colonization pressure (CP) is important for mediating risk factors. CP is a tool to measure the proportion of A. baumannii reservoirs within a health care facility. For A. baumannii surveillance, CP can help enhance patient screening and determine infection control measures ^{90 , 91} .

MOLECULAR EPIDEMIOLOGY OF A. BAUMANNII IN BRAZIL

In Brazil, the first outbreak associated with OXA-23-producing A. baumannii isolates was in 1999 ⁹² . Subsequently, different outbreaks were reported ⁹³ . A. baumannii dissemination in different Brazilian hospitals was associated with bla _OXA-51 and bla _OXA-23 genes and highlighted the prevalence of ISAba1/OXA-23 and ISAba1/OXA-51 genetic profiles ⁹⁴ . Isolates carrying the bla _OXA-51, bla _OXA-58, and bla _OXA-23 genes, and ISAba1 upstream of OXA-51 and OXA-23 were found in different ICUs, indicating an outbreak of cross-contamination among patients, equipment, or medical staff ⁹⁴ . The bla _OXA-58 and bla _OXA-65 genes with the upstream ISAba1 sequence for both genes have been reported. The bla _OXA-58 gene is prevalent in Argentina, indicating a possible spread from the border with Rio Grande do Sul ⁹⁵ . In addition, two genotypes of OXA-23-producing A. baumannii were present at 8 hospitals in the same city, suggesting the spread of isolates in these environments ⁹³ . The sequence type (ST) 156, ST25, and ST160 were identified in a Brazilian hospital ⁹⁶ . Cephalosporin-resistant A. baumannii and producers of extended-spectrum beta-lactamases (ESBL) were identified in a neonatal intensive care unit (NICU), causing septicemia in hospitalized neonates (Table 3) ⁵ . A study in neonates described most isolates as belonging to ST1 and had ISAba1 upstream of the bla _OXA-51 and bla _OXA-23 genes ⁸⁸ .

TABLE 3: Outbreaks of Acinetobacter baumannii in Brazil.

Study	Place of study	Year of outbreak	Place of outbreak	No. of patients	Antibiotic Resistance	Reported genes
DALIA-COSTA et al., 2003	Curitiba	1999	Ward	8	IPM, MEM, CIP, and AMG	bla OXA-23
BRITO et al., 2005	Uberlândia	2005	NICU	11	GEN, CIP, CAZ, FEP, and ATM	-
TAKAGI et al., 2009	São Paulo	2005-2006	ICU	8	PIP, TZP, CAZ, CTX, ATM, IPM, MEM, CIP, AMK, GEN, and SXT	bla OXA-51
MARTINS et al., 2009	Porto Alegre	2007	DHW	53	CIP, GEN, TZP, and SXT	bla OXA-51, bla OXA-23
GUSATTI et al., 2012	Porto Alegre	2007	Ward	74	IPM, MEM, AMK, CIP, GEN, CET, AMA, SXT, and TIM	bla OXA-51, bla OXA-58, bla OXA-65, ISAba1/OXA-51
PAGANO et al., 2015	Porto Alegre	2011	DHW	122	FEP, CIP, CAZ, AMA, AMK, PMB, IMP and MEM	bla OXA-23
CASTILHO et al., 2017	Goiás	2010	ICU	64	AMA, FEP, AMK, PMB, and TGC	ISAba1/OXA-23 and ISAba1/OXA-51, bla OXA-51, bla OXA-23, bla OXA-58
MACIEL et al., 2017	Dourados	2013-2015	NICU	21	AMA, TZP, CAZ, CRO, FEP, GEN, AMK, CIP, and TGC.	ISAba1/OXA-23 and ISAba1/OXA-51

ICU: intensive care unit; NICU: neonatal intensive care unit; DHW: different hospital wards; IPM: imipenem; MEM: meropenem; CIP: ciprofloxacin; AMG: aminoglycoside; GEN: gentamicin; CAZ: ceftazidime; FEP: cefepime; ATM: aztreonam; TZP: piperacillin/tazobactam; AMK: amikacin; SXT: trimethoprim/sulfamethoxazole; CET: cephalothin; TIM: ticarcillin/clavulanic acid; PMB: polymyxin B; TGC: tigecycline; CRO: ceftriaxone; AMA: ampicillin/sulbactam; CTX: cefotaxime; PIP: piperacillin.

A study in Recife, Brazil described isolates belonging to ST1, ST15, ST25, ST79, ST113, and ST881 (related to ST1). Among them, ST79 and ST113 were found to be more virulent and presented resistance genes. ST113 and ST15 were commonly found in all 5 hospitals of the study, while ST79 was found in 4 hospitals and ST1 in 3 hospitals. Among the CCs circulating between hospitals, Leal et al. described CC1, CC15, and CC113, which are globally spread types, and CC79, which is found in South America, North America, and Europe ⁹⁷ .

A study carried out in nine hospitals in South America identified A. baumannii clinical isolates presenting bla _OXA-51, bla _OXA-23, bla _OXA-72, bla _OXA-132, bla _OXA-65, bla _OXA-69, and bla _OXA-64 genes. Multilocus sequence type (MLST) analysis identified ST79, ST25, and ST15 ⁹⁸ . The two major clonal complexes (CC) found in bla _OXA-23 multidrug-resistant A. baumannii are CC15 and CC79, and CC15 has already been described in 9 Brazilian states ⁷⁷ . In addition, ST15 was described in other countries, such as Argentina and Turkey, and ST79 was described in the United States, Canada, and Spain ⁹⁹ . Of the clonal profiles identified, ST15 and ST79 were described in several countries, indicating their spread among hospitals around the world and high mortality rates ¹⁰⁰ .

The Antimicrobial Surveillance Program (SENTRY) evaluated the prevalence of Acinetobacter spp. and other gram-negative bacilli isolated from Latin American (Argentina, Brazil, Chile, and Mexico) medical centers from 2008 to 2010. In this period, 5,704 gram-negative bacilli were isolated and 845 (17.7%) were classified as Acinetobacter spp. This microorganism was responsible for 7.2% of the 6,035 bloodstream infections, 7% of the 1,442 pneumonia cases, and 9.9% of the 1,531 skin and soft tissue infections. The oxacillinases found in this study were OXA-23 and OXA-24 in Argentina, OXA-23 in Brazil, OXA-58 in Chile, and OXA-24 in Mexico ¹⁰¹ . Figure 1 shows a map representing the description of the resistant gene OXA in the last eight years ^{3 , 102 - 145} .

FIGURE 1: Geographic distribution of OXA enzymes in the last seven years.

MOLECULAR EPIDEMIOLOGY OF A. BAUMANNII IN THE WORLD

In France, 110 A. baumannii clinical strains were isolated between 2010 and 2011. Of these, 90 isolates harbored bla _OXA-23, 12 bla _OXA-24, and 8 bla _OXA-58. One of the isolates simultaneously displayed bla _OXA-23 and bla _PER-1, and 2 isolates possessed bla _OXA-23 and bla _OXA-58. Pulsed-field gel electrophoresis (PFGE) analysis showed 30 clusters and MLST revealed 11 STs (ST115, ST1, ST2, ST10, ST20, ST25, ST79, ST85, ST107, ST108, and ST125) ⁴³ . A study conducted in China evaluated 57 clinical isolates of carbapenem-resistant A. baumannii that were positive for the bla _OXA-23/ISAba1 and bla _OXA-51 genes, harboring ST75 and ST137 ¹⁴⁵ . In addition, a Chinese hospital identified transposons Tn2006, Tn2007, and Tn2008 in 59 clinical isolates of OXA-23-producing A. baumannii ¹⁴⁶ .

In Saudi Arabia, 107 A. baumannii clinical isolates were identified, of which 75 harbored the genes bla _TEM and bla _CTX-M (n = 86), bla _OXA-51 (n = 100), and bla _OXA-23 (n = 97). MLST analysis identified ST195, ST557, ST208, ST499, ST218, ST231, ST222, and ST286, all belonging to CC2, except ST231 ¹⁴⁷ . In the United States, in 2008 and 2009, 65 A. baumannii clinical isolates producing bla _OXA-51/ISAba1 were found in different hospitals, harboring bla _OXA-23 (65/65) and bla _OXA-40 genes (09/65). PFGE analysis indicated 24 clusters, whereas MLST identified ST1, ST2, ST77, ST79, ST123, ST124, CC1, and CC2 ¹⁴⁸ . A total of 149 clinical isolates of A. baumannii, containing bla _OXA-58 (n = 31), bla _OXA-58/ISAba3 (n = 14), and bla _OXA-72 (n = 18) were isolated from different hospitals in Egypt. These presented as 54 clusters by PFGE and ST763, ST777, ST369, ST762, and ST229 were identified ¹⁴⁹ .

In South Africa, 94 clinical isolates of A. baumannii were found in different hospitals; 93 carried the bla _OXA-51 gene and 72 the bla _OXA-23. PFGE analysis grouped the isolates into 4 clusters with 5 STs (ST106, ST258, ST339, ST502, ST758, ST848), in which ST258 and ST758 corresponded to the international clone I, and ST502 and ST848 to the international clone II ¹⁵⁰ . In India, 100 A. baumannii strains showed high genetic variability. MLST identified ST110, ST108, ST194, ST14, ST146, ST69, ST188, ST386, ST387, ST388, ST389, ST390, and ST391 ¹⁵¹ . A total of 160 A. baumannii clinical isolates were identified in Vietnam, of which 119 were MDR or extensively resistant, presenting a high level of resistance against third- and fourth-generation cephalosporins. Of these, 128 isolates harbored the bla _OXA-51 and bla _OXA-23 genes associated with the ISAba1 element. MLST analysis identified 16 STs from 23 isolates, confirmed new STs, and some isolates belonged to ST136 ¹⁵² .

In Malaysia, 162 clinical isolates of MDR A. baumannii were identified, of which 128 were resistant to carbapenems. The bla _OXA-23, bla _OXA-IMP, and bla _OXA-ADC genes were identified, and ISAba1, upstream of the bla _OXA-23 and bla _OXA-ADC genes, was also found. Point mutations in gyrA (Ser83Leu) and parC (Ser80Leu), which provide resistance to ciprofloxacin, were also identified in the isolates. MLST identified two predominant STs (ST195 and ST208) ¹⁰⁴ .

Molecular typing of A. baumannii provides a better understanding of the epidemiology of outbreaks and identification of cross-transmission, as well as assisting in the monitoring and control of nosocomial infections ^{17 , 153} . Thus, several methods have been used to study the molecular epidemiology of A. baumannii and analyze the mechanisms involved in the resistance of this microorganism.

CONCLUSION

The increase in healthcare-associated infection (HAI) rates connected to A. baumannii antimicrobial resistance has become a major public health challenge worldwide. A. baumannii possesses several resistance mechanisms. However, hydrolysis by OXA-type carbapenemases and metallo-β-lactamases are considered the most prevalent mechanisms conferring resistance to most beta-lactam antibiotics and reduce therapeutic options. This study highlights the occurrence of outbreaks in hospital settings, especially in ICUs, which are commonly related to prolonged hospital stays and invasive procedures. Thus, epidemiological studies are important for monitoring the occurrence of A. baumannii clinical isolates and may assist in the implementation of appropriate measures, contributing to the control of hospital infections.