Carbapenem-resistance in Acinetobacter baumannii: prevalence, antibiotic resistance profile and carbapenemase genes in clinical and hospital environmental strains

Abstract

Background

Acinetobacter baumannii, a Gram-negative member of the ESKAPE pathogen group, is known to develop resistance to several antibiotics rapidly, and carbapenem-resistant A. baumannii (CRAB) is highly implicated in life-threatening infections, especially within hospital settings.

Objectives

This study detected CRAB in clinical and hospital-environmental samples, evaluated the antibiotic resistance patterns and screened for prevalent carbapenemase genes in isolates from a hospital in Southwest Nigeria.

Methods

A total of 150 clinical and hospital environmental samples were analysed using culture-dependent and molecular methods for the detection of Acinetobacter baumannii. Antibiotic susceptibility test was done using the Kirby-Bauer disk diffusion technique. Phenotypic screening for carbapenemase was via simplified carbapenem inactivation method (sCIM), and molecular detection of bla_KPC type, bla_{OXA−48−like}, bla_VIM type, bla_NDM−1, bla_IMP variants and bla_{OXA−23−like} genes by Polymerase chain reaction.

Results

Altogether, only 29.4% (42/143 isolates) of recovered isolates were identified as A. baumannii, giving a prevalence of 28.0% (42/150 samples), predominantly from sputum. All isolates had the gluconolactonase gene, while 5/42 had the bla_OXA-₅₁-_like gene. Resistance to meropenem and cefiderocol was 100.0% and 88.1%, respectively, while gentamicin was most effective in vitro (7.1%); 54.8% were multidrug-resistant, and 88.1% (37/42) had MARI ≥ 0.2. Overall, 39/42 (92.9%) isolates had ≥ one or more carbapenemase genes; 61.9% (26/42) had the bla_KPC type gene, 59.5% (25/42) had the bla_IMP variants while 45.2% had the bla_VIM type gene; no strain had the bla_NDM−1 or the bla_{OXA−23−like} gene.

Conclusion

This study reports the occurrence of MDR strains, and of bla_KPC type, bla_IMP variants and bla_VIM type carbapenemase genes in A. baumannii isolates from clinical and hospital environmental samples, contributing to the pool of existing data on their occurrence. It also highlights the need for monitoring and continued surveillance of the strains, most especially in the clinical setting.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-025-11169-x.

Introduction

Acinetobacter baumannii, a Gram-negative member of the ESKAPE (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp.) pathogen group is an opportunistic pathogen highly implicated in life-threatening infections (such as bloodstream, central nervous system, urinary tract, skin and soft tissue, digestive and respiratory tract infections) within and outside the hospital settings [1–3]. It was discovered and first reported in soil in 1911 and called Micrococcus calcoaceticus [4] but became formally acknowledged and renamed Acinetobacter in 1971 [2, 4, 5]. The genus Acinetobacter is of the order Pseudomonadales and family Moraxellaceae, and presently comprise more than 85 valid species (validly published with correct names as of 9th February 2025) [6]. It is a non-motile, coccobacillary form which develops resistance consistently and rapidly to multiple antibiotics (including drugs of last resort like the carbapenems) as the recovery of multidrug-resistant (MDR), extensively-drug resistant (XDR) and even pan-drug-resistant (PDR) strains have been reported [3, 7, 8].

Acinetobacter baumannii possesses intrinsic resistance to numerous antibiotics and can acquire additional genes and also upregulate existing resistance genes. These traits make it a formidable pathogen as viable treatment options are almost non-existent, raising mortality rates in infected patients [9, 10]. Reports of resistance to disinfectants and desiccation [11] and the ability to form biofilms have also been documented [3, 12]. Moreover, a wide array of virulence determinants has been recognized, aiding its adherence to biotic and abiotic surfaces, colonization, invasion, persistence in the environment and evasion of the host immune system [1, 2, 12–14]. The colonization of hospital equipment and devices is not left out as the persistence of A. baumannii on hospital equipment, devices [15–17] and environmental surfaces has been associated with the rapid transmission of the pathogen [11].

Carbapenems were previously first-line drugs for treating several bacterial infections as they have broad-spectrum antibacterial activity and are comparatively more able to resist the hydrolytic action of most β-lactamases. Prominent members of the class of antibiotics include imipenem, meropenem, ertapenem doripenem, and biapenem while the newer drugs include razupenem, tebipenem, tomopenem, and sanfetrinem [18]. However, resistance to carbapenems may occur by the activity of certain hydrolytic class D β-lactamases [acquired oxacillinases (OXAs)] [19–21], class B Metallo-β-lactamases (MBLs) [22], class A β-lactamases, or via overexpression of intrinsic carbapenem-resistant genes like OXA-_51−like gene [11]. Another mechanism may be the loss of porins in the cell wall of bacteria cells, which reduces the susceptibility of bacteria to carbapenems [11, 18, 23].

The rising prevalence of Carbapenem-resistant Acinetobacter baumannii (CRAB) across continents has been reported by various authors. In two previous studies, 97.0% and 95.0% prevalence of carbapenem-resistant A. baumannii were reported jointly in Greece, Italy and Spain [7] and also in Lebanon [17] respectively; in comparison, 64.6% in one of the two studies was PDR or XDR [7]. A recent study reported a prevalence rate of 62.8% for CRAB in Nigeria [24]. However, a pooled prevalence of 20.0% has been recorded for CRAB in the Sub-Saharan African region [25]. CRAB is the third pathogen in the priority category “critical” generated by the World Health Organization [26]. It is classified as a strain that poses a great hazard to public health due to restricted therapeutic choices and has been placed in the class of organisms for which new antibiotics are urgently required [26]. Furthermore, according to the Centre for Disease Control (CDC) in its updated factsheet “Antimicrobial Resistance Threats in the United States, 2021–2022” [27], CRAB is the second pathogen in the “urgent” category, and the rate of hospital-onset infections caused by it increased significantly in 2022 when compared with 2019. Carbapenem-resistant A. baumannii is reported to belong to the top five global pathogens based on high mortality, antibiotic resistance, and is presumed to be the foremost pathogen in Oceania, Eastern and Southeast Asia [27–30]. In the USA, CRAB was implicated in 700 fatalities arising from infections in 8,500 in-patients in 2017 alone [11].

Previously, data from the African continent on this pathogen were minimal. In a systematic review conducted in 2019 [31], out of 24 studies analysed, only two were from the African continent– namely Benin and Ethiopia. More recently, in the African continent, within Nigeria and particularly southwest Nigeria, reports of CRAB strains carrying various carbapenemase genes are beginning to emerge [24, 25, 32]. The possibility of under-reporting, however, strongly exists. As such, there is a necessity for continued surveillance of A. baumannii prevalence and carbapenem resistance in A. baumannii in Nigeria. Odih et al. [24] also reported the frequent transmission of carbapenem resistance genes within strains, emphasizing the need for more autochthonous studies to characterize the currently circulating CRAB strains within Nigeria. In light of the above, the present study was designed to characterize Carbapenem-Resistant Acinetobacter baumannii in clinical and hospital environmental samples, determine the antibiotic resistance profile and screen for the prevalent carbapenemase genes in the recovered strains from a private hospital in Iwo, Southwest Nigeria.

Materials and methods

Study design and study population

This cross-sectional study analysed clinical and hospital environmental samples. The sample size was calculated using the formula as previously described [33]. In this case, 9.2% prevalence was used [34], with a 10% attrition value summing up to a minimum of 141 samples. Participants (90 in all) were male and female in- and out-patients from a hospital at Iwo, Osun State, Nigeria (7º 25’ 35.2524’’ N and 3º 50’ 30.5298’’ E), included in the study based on individual or parental consent. Ethical approval for the study was obtained from the Osun State Ministry of Health (via the Ethics Committee of the Health Planning, Research, and Statistics Department with approval number OSHREC/PRS/569T/482). During sample collection, a structured questionnaire detailing demographic information, history of antibiotic use, and other relevant information was used to collate data from the study participants (Supplementary File 1). Participation in the study was completely voluntary.

Sample collection and processing

Between January and March 2024, 150 samples, comprising 90 clinical samples from an equal number of participants (viz. skin swabs, sputum, urine, and wound samples) and 60 swabs from the hospital environment (vis bedpans, bed rails, doorknobs, intravenous fluid (IV) stands, sinks, stair rails, and ward sinks), were collected via standard laboratory techniques. All swabs were collected using sterile swab sticks dampened with sterile Ringer solution, while mid-stream urine samples were collected into sterile universal bottles, immediately transported in cool boxes to the Microbiology Laboratory of Osun State University and processed within 4 h of collection. The swabs were inoculated overnight in Tryptone Soy Broth (TSB) at 35 ± 2℃, and tubes with visible growth were afterwards streaked out on MacConkey agar at 35 ± 2℃ for 24 h. Additionally, a loopful of each urine sample was streaked out on MacConkey agar as well as Cystine-Lactose-Electrolyte-Deficient (CLED) agar [35] and incubated for 24 to 48 h at 35 ± 2 °C. Discrete, light-pink and cream-coloured colonies on MacConkey and CLED agar were subcultured for purity and analyzed via standard Gram staining, morphological, and biochemical tests, including catalase, oxidase, citrate utilization, motility, and growth at 37 °C and 42 °C [36]. Bacterial cultures were preserved in freshly prepared TSB supplemented with 15% glycerol and stored at -20℃ for further processing.

Molecular identification of Acinetobacter baumannii isolates

Molecular identification of bacterial strains was performed as follows: chromosomal DNA was extracted using the thermal cell lysis method with slight modifications [37] and quantified for concentration and purity using the Nanodrop-One Spectrophotometer (Thermo-Fisher Scientific, USA). The extracted DNA was subjected to polymerase chain reaction (PCR), a multiplex reaction to detect bla_OXA−51−like and gluconolactonase genes specific for A. baumannii [38]. Amplification of extracted DNA templates was carried out in 20 µL solutions, comprising 4 µL of 5x Master-Mix (Genewiz, USA), 1 µL each of the two 10 µM forward and reverse primers (Genewiz, USA), and 5 µL of each DNA template made up to 20 µL with 7 µL of DNAse/RNAse-free sterile water (BioConcept, USA) using the Master Cycler Nexus Gradient 230 (Eppendorf, Germany). Acinetobacter baumannii ATCC 19606 served as the positive control, while DNAse/RNAse-free sterile water was used as a template in the negative control setup. All PCR products (10µL each) were run on a 1.0% agarose gel stained with SafeView-Classic at 80 V for 60 min and visualized under a UV trans-illuminator E-BOX-CX5 TS imaging system (Vilber, France). A 100 bp DNA ladder (Biolabs, England) was used as the DNA molecular weight standard. The details of the primer sequences and PCR protocols are highlighted in Tables 1 and 2, respectively.

Table 1.

Primer sequences of selected genes for Acinetobacter baumannii characterization

	Name	Sequences (5’ − 3’)	Product size	Reference
Identification genes	bla OXA−51−like	F: CTAATAATTGATCTACTCAAGTTAC R: GAATACTCCATTTGAACCARTGG	988	[38]
	gluconolactonase	F: TTGGAGAATGCCCAACTTGG R: CCCGTCTTCGAGCGCAAC	185	[38]
Carbapenem resistance genes	bla KPC	F: TGTTGCTGAAGGAGTTGGGC R: ACGACGGCATAGTCATTTGC	340	[39]
	blaVIM type	F: CGCGGAGATTGARAAGCAAA R: CGCAGCACCRGGATAGAARA	247	[39]
	bla OXA−48−like	F: AACGGGCGAACCAAGCATTTT R: TGAGCACTTCTTTTGTGATGGCT	585/597	[39]
	bla NDM−1	F: TAAAATACCTTGAGCGGGC R: AAATGGAAACTGGCGACC	439	[39]
	bla OXA−23−like	F: GTGGTTGCTTCTCTTTTTCT R: ATTTCTGACCGCATTTCCAT	736	[39]
	blaIMP variants except IMP-3, IMP-16, IMP-27, IMP-31, IMP-34 and IMP-35	F: GAGTGGCTTAATTCTCRATC R: CCAAACYACTASGTTATCT	183	[39, 40]

Note: R = A or G; S = G or C; Y = C or T.

Table 2.

PCR protocols for the amplification of selected genes for Acinetobacter baumannii characterization

Genes	Initial Denaturation	No. of cycles	Denaturation Temp (℃)	Annealing Temp (℃)	Extension Temp (℃)	Final Extension
Identification
bla OXA−51−like gluconolactonase	95 ℃ for 5 min	30	95 ℃ for 1 min	56.5 ℃ for 45 s	72 ℃ for 1 min	72 ℃ for 10 min
Resistance genes
blaKPC type bla OXA−48−like blaVIM type	95 ℃ for 5 min	35	95 ℃ for 1 min	56℃ for 1 min	72 ℃ for 1 min	72 ℃ for 5 min
bla NDM−1 blaOXA−23−like blaIMP variants except IMP-3, IMP-16, IMP-27, IMP-31, IMP-34 and IMP-35	95 ℃ for 5 min	35	95 ℃ for 1 min	52 ℃ for 1 min	72 ℃ for 1 min	72 ℃ for 5 min

Antibiotic susceptibility testing (AST)

The antibiotic susceptibility testing (AST) was done using the agar disk diffusion technique by Kirby-Bauer [41, 42]. Commonly prescribed commercially available antibiotics (Mast, UK) were selected based on the CLSI breakpoint guideline [43], and these included carbapenems [imipenem (10 µg), meropenem (10 µg)], fluoroquinolones [ciprofloxacin (5 µg), levofloxacin (5 µg)], aminoglycosides [amikacin (30 µg), gentamicin (10 µg), tobramycin (10 µg)] and cephems [cefepime (30 µg), cefiderocol (30 µg)]. For each test isolate, 3–5 colonies of an overnight culture on Tryptone Soy Agar (TSA) were inoculated into a sterile Ringer solution (5 mL) and adjusted to 0.5 McFarland standard. This was inoculated as a lawn on sterile Mueller-Hinton agar (MHA) plates, and the antibiotic discs were placed aseptically on the agar surface using a disc dispenser (Oxoid). The dispenser was wiped with ethanol before use, and measures were taken to ensure that the dispenser did not touch the surface of the inoculated agar. The plates were incubated overnight at 35 ± 2 ℃, and the resultant zones of inhibition (if any) were recorded to the nearest millimetre, and the values were interpreted using the CLSI breakpoint guideline [43]. Multidrug resistance (MDR) was recorded for isolates resistant to three or more classes of antibiotics, while Multiple antibiotic resistance index (MARI) was calculated with the formulae below:

Phenotypic screening for carbapenem resistance

The bacterial isolates recovered and confirmed to be A. baumannii in this study were phenotypically screened for carbapenemase using the simplified carbapenem inactivation method (sCIM). This was done as described by Jing et al. [44]. For this test, Escherichia coli reference strain ATCC 25922 standardized to 0.5 McFarland standard turbidity was diluted in the ratio 1:10 in sterile Ringer solution, inoculated as a lawn on sterile MHA, and the plates were left to dry. About 2–4 colonies of each test isolate were daubed evenly onto one side of a 10 µg imipenem disk (Oxoid, UK), and the test disc was immediately placed on the MHA plate inoculated with E. coli ATCC 25922. The side of the disc smeared with the bacteria was placed in direct contact with the surface of the agar. The MHA plates were incubated at 35 ± 2 ℃ for 16 to 18 h. Carbapenemase-producing bacterial strains capable of hydrolyzing imipenem denature the drug and allow the reference strain to grow uninhibited. The presence of a zone of inhibition measuring 6–20 mm, or colonies within an inhibition zone ≥ 22 mm, denoted a positive result for carbapenemase production; a zone ≤ 26 mm indicated a negative result, while a clear zone ≤ 23 but ≥ 25 mm was taken to be indeterminate. A clean 10 µg imipenem disk served as the positive control.

Molecular detection of carbapenem resistance genes

Carbapenem-resistant genes were screened using two multiplex reactions: the first multiplex reaction was targeted at detecting bla_KPC type (340 bp)/bla_OXA−48−like (597 bp)/bla_VIM type (247 bp) carbapenemase genes. In contrast, the second multiplex reaction was to screen for bla_NDM−1 (439 bp)/bla_IMP variants (183 bp)/bla_OXA−23−like (736 bp) genes. PCR mix for DNA amplification comprised 5x Master-Mix (4 µL), 10µM of the three sets of forward and reverse primers (1 µL each), the DNA template (5 µL), and DNAse/RNAse-free sterile water (5 µL), adding up to 20 µL solutions. Amplicons were run on a 1.5% agarose gel containing SafeView-Classic and visualized as described above. A 100 bp DNA ladder served as a marker for size.

Results

Socio-demographic data of the participants

The study population comprised 90 participants [34 (37.8%) males and 56 (62.2%) females], their ages ranged between 20 and 68 years (averaging 36.9 years). More than half of the female participants (73.2%) were in the age range of 21–40 years, while 79.4% of male participants were within the 31–50-year age bracket. Half of the population (50.0%) were traders, 83.3% were married, while 90.0% of the study population were of the Yoruba ethnic group, in keeping with the part of the country from where the study was conducted (Fig. 1).

Fig. 1.

Demographic details of the study participants

Frequency of occurrence of A. baumannii from clinical and hospital-environmental samples

Altogether, 150 samples comprising 90 human and 60 environmental samples from the hospital setting were collected. A total of 106 (70.7%) samples had growth [71 (67.0%) from clinical and 35 (33.0%) from environmental samples], while 44 (29.3%) samples had no growth. The one hundred and six positive samples yielded 143 Gram-negative bacterial isolates; 69 samples (46 from clinical and 23 from the environment) yielded single isolates, while the remaining 37 samples (25 from clinical and 12 from the environment) were polymicrobial, all giving 2 isolates each. Nevertheless, only 29.4% (42/143) were conventionally and genotypically identified as Acinetobacter baumannii, generating an overall prevalence of 28.0% (42/150). The genotypic identification revealed the presence of gluconolactonase in 42 strains, 5 of which also possessed the bla_OXA-₅₁-_like gene. The highest proportion of A. baumannii in clinical samples was obtained from sputum at 65.0% (13/20) while intravenous (IV) fluid stands had the highest proportion for environmental samples at 35.0% (7/20). Overall, the highest number of isolates were from sputum, at 31.0% (13/42), followed by urine and IV stands, at 26.2% (11/42) and 16.7% (7/42), respectively. The distribution of bacterial isolates across the various sample types is detailed in Fig. 2, while the gel image of A. baumannii identification is shown in Figs. 3a and b.

Fig. 2.

The distribution of bacterial isolates across the various sample types

Figs. 3.

3a and 3b: The gel image of the molecular identification of A. baumannii isolates showing bands for (a) the gluconolactonase gene at 185 bp; and (b) the bla_OXA-₅₁-like gene at 988 bp (L– 100 bp DNA molecular weight standard)

Antibiotic resistance pattern of A. baumannii from clinical and hospital-environmental samples

All A. baumannii isolates were tested against nine antibiotics in four classes. The isolates revealed the highest rate of resistance to meropenem– all 42 (100.0%) isolates were resistant to it, trailed by cefiderocol at 88.1% (37/42). However, gentamicin proved most effective in vitro as bacterial isolates exhibited the least resistance to it (7.1%), closely followed by imipenem (9.5%) and cefepime (14.3%) (Table 3). None of the isolates were resistant to all the antibiotics, nevertheless, the most resistant isolate, obtained from an IV stand in one of the hospital wards, exhibited resistance to 7 out of nine antibiotics from all 4 classes tested. Three other isolates (1 from sputum and two from urine samples) were sensitive to 7 out of 9 antibiotics. Several of the A. baumannii isolates displayed co-resistance to at least 4 of the antibiotics– 11 strains were co-resistant to meropenem, ciprofloxacin, levofloxacin, and cefiderocol, 5 of them in addition to two other fluoroquinolones. More than half of the isolates (64.3%) had intermediate resistance to ciprofloxacin, while none of the isolates were susceptible to ciprofloxacin. Four isolates were resistant to both carbapenems (imipenem and meropenem), two of them in addition to resistance to aminoglycosides and cephems. About half of the isolates (54.8%) were resistant to ≥ 3 classes of antibiotics (MDR) (Table 4), while 88.1% (37/42) had MARI of x ≥ 0.2 (Fig. 4).

Table 3.

The antibiotic resistance patterns of the A. baumannii isolates from clinical and hospital environmental samples

Antibiotic class	Antibiotics	Clinical isolates (30)			Environmental isolates (12)			Total
		Resistant	Intermediate	Sensitive	Resistant	Intermediate	Sensitive
Carbapenem	Imipenem	1	5	24	3	4	5	42
	Meropenem	30	0	0	12	0	0	42
Fluoroquinolone	Ciprofloxacin	10	20	0	5	7	0	42
	Levofloxacin	8	3	19	5	1	6	42
Aminoglycoside	Amikacin	1	0	29	8	0	4	42
	Gentamicin	3	0	27	0	0	12	42
	Tobramycin	4	1	25	7	0	5	42
Cephem	Cefepime	5	0	25	1	0	11	42
	Cefiderocol	25	0	5	12	0	0	42

Table 4.

Details of the multiple drug resistance patterns of the A. baumannii strains

SAMPLE TYPES		Number of antibiotic classes against which isolates are resistant					No of isolates
		0	1	2	3	4
ENVIRONMENTAL SAMPLES	Bedpans	0	0	0	1	0	1
	Bed rails	0	0	0	0	0	0
	Doorknobs	0	0	0	2	1	3
	IV stands	0	0	1	3	3	7
	Stair rails	0	0	0	0	0	0
	Sink	0	0	1	0	0	1
CLINICAL SAMPLES	Skin	0	0	4	1	0	5
	Sputum	0	2	5	4	2	13
	Urine	0	3	3	3	2	11
	Wound	0	0	0	1	0	1
TOTAL		0	5	14	15	8	42

Fig. 4.

The multiple antibiotic resistance Indices (MARI) of the A. baumannii isolates from clinical and hospital environmental samples

Characterization of isolates for carbapenem resistance

The resistance to carbapenems via phenotypic screening using the simplified Carbapenem Inactivating Method (sCIM) revealed that all screened isolates were negative for carbapenemase enzymes. Molecular detection of the carbapenemase genes (bla_OXA−48−like, bla_KPC type, bla_VIM type, bla_NDM−1, bla_IMP variants, and bla_OXA−23−like) in the 42 recovered A. baumannii isolates disclosed the presence of one or more genes in 39 (92.9%) isolates (Fig. 5a and b). In all, 61.9% (26/42) isolates had the bla_KPC type gene, 59.5% (25/42) had the bla_IMP variants, and 45.2% had the bla_VIM type gene (Fig. 6). Twelve (30.8%) of them had only one of the genes screened for (7 isolates had bla_VIM type, 3 had bla_KPC type, and the last two had only bla_IMP variants); 22 isolates had 2 genes each while the last five isolates carried 3 genes each– all of them had bla_KPC type, bla_VIM type, and bla_IMP variants. The details of the co-carriage of carbapenemase genes are given in Table 5. Only one strain (recovered from an IV stand, and resistant to 7 out of 9 antibiotics) had the bla_{OXA−48−like} gene, co-carried with bla_KPC. None of the isolates carried the bla_NDM−1 or the bla_{OXA−23−like} gene.

Figs. 5.

5a and 5b: The gel image of the carbapenemase resistance genes in A. baumannii isolates showing bands for (a) the bla_IMP variants gene at 183 bp; and (b) bla_VIM type (247 bp) and bla_KPC type (340 bp) (L– 100 bp DNA molecular weight standard)

Fig. 6.

The distribution of the identification and carbapenem resistance genes carried by A. baumannii isolates from clinical and hospital environmental samples

Table 5.

The details of distribution and co-carriage of screened carbapenem resistance genes in clinical and environmental A. baumannii isolates

Gene Function	Gene Distribution	Clinical isolates (30)	Environmental isolates (12)	Total
Identification genes	Gluconolactonase only	28	9	37
	Gluconolactonase/blaOXA−51−like	2	3	5
	Total	30	12	42
Resistance genes	No gene	3	0	3
	blaKPC type only	2	1	3
	blaVIM type only	7	0	7
	blaIMP variants only	1	1	2
	blaOXA−48−like/blaKPC type	0	1	1
	blaKPC type/blaVIM type	2	1	3
	blaVIM type/blaIMP variants	3	1	4
	blaKPC type/blaIMP variants	10	4	14
	blaKPC type/blaVIM type/blaIMP variants	2	3	5
	Total	30	12	42

Discussion

This study used culture-dependent and molecular techniques to assess the occurrence of carbapenem-resistant A. baumannii isolates obtained from clinical specimens and the hospital environment in a privately owned healthcare facility in Southwest Nigeria. Reports of the occurrence of A. baumannii vary widely depending on several factors, which include the study area, type of sample/patients, hospitals or communities from which samples were collected among others. Nonetheless, the rising incidence of outbreaks linked to A. baumannii in the healthcare setting within the Sub-Saharan Africa region has been reported [31]. An overall prevalence rate of 28.0% (42/150) was observed in this present study, 20.0% (30/150) from clinical and 8.0% (12/150) from environmental samples. This rate is much lower than that recently reported in Thailand by Kitti et al. [45] where 33.97% of A. baumannii isolates were recovered from hospital environmental samples. In another study, however, A. baumannii strains constituted 95.6% of ESKAPE pathogens recovered, with > 95% being carbapenem-resistant [46].

This study recorded the highest proportion of isolates from sputum (65.0%), while the proportion from IV stands was the highest for environmental samples (35.0%). The overall prevalence of isolates from sputum was 31.0% (13/42), in line with similar studies by Hafiz et al. [47] and Itani et al. [17], where 27.8% and 24.3% of A. baumannii isolates were obtained from sputum, respectively, corroborating this observation in our study. A major site of A. baumannii infections has been reported to be the lower respiratory tract [17], and the occurrence of A. baumannii in the respiratory tract of humans predisposes the host to respiratory tract infections and possible complications. Many studies have observed that a link exists between the contamination of the air, environmental/abiotic surfaces and pathogen infection, particularly in the respiratory tract [48, 49] within the hospital environment and in the wards [50, 51]. Acinetobacter baumannii has also been reported to prefer moist and wet environments, hence its predilection for moisture-laden samples such as sputum, explaining its predominance in sputum in the present study [52, 53]. Kitti et al. [45] recorded a 13.3% recovery of A. baumannii isolates from IV stands, a slightly lower rate than in this study, as the prevalence of A baumannii isolates from IV stands in the present study was 16.7% (7/42). However, the recovery from bedrails (66.7%) in their research did not correlate with ours as we recorded 0.0% A. baumannii recovery from the bedrails in our study. Similarly, Mohammed et al. [54] also obtained one isolate from the hospital sink, the same as in this study. The three (3) isolates recovered from doorknobs and bedpans were multidrug-resistant, likewise, a high proportion of MDR strains were observed in IV stands. Like-strains in the hospital setting pose a major threat to the in-patients and even hospital staff as the risk of infection becomes magnified, especially when immune defences in the host are compromised, or control measures against transmission are not strictly complied with. A. baumannii can persist for long periods in the environment, as well as on furniture, medical devices, gloves [55], and blood pressure cuffs [56] and this has been observed to be a major cause of nosocomial infections [51].

The detection of the gluconolactonase gene in 29.7% (42/143) of isolates is a good indicator of A. baumannii species identity as it is a recently developed and tested genomic identification marker specific to A. baumannii (using both in silico and in vitro assays) [38]. The gene was found to be absent in non-A. baumannii Acinetobacter isolates, however, a few multi-locus sequence types (MLST) of A. baumannii (A. baumannii ST79 and related strains) did not harbour the gene. Surprisingly, only five (5) isolates revealed the bla_OXA-₅₁-like gene. This greatly contrasts with previous studies that affirmed its presence in A. baumannii as it is an intrinsic gene used as an identification marker [36, 45, 57, 58]. The reasons for this, although not quite apparent may be due to a complex interaction of several factors which may affect PCR amplification of the targeted gene [38]. Screening for bla_OXA-₅₁-_like gene singly is reported to be undependable as the possibility of insertion elements (ISs) disrupting the gene sequence exists [59], creating situations of non-detection as the expected outcomes will not be seen after PCR. In addition, too much expression of the oxa-enzymes is frequently linked with the ISs [60]. Furthermore, Kitti et al. [45] also reported bla_OXA-₅₁-_like negative A. baumannii isolates in their study.

All 42 isolates in our study were resistant to meropenem. This trend is in line with a study from Northern Nigeria in which 100.0% of the A. baumannii isolates in their study were also resistant to meropenem [54]. Other studies have also reported high resistance rates to meropenem [61, 62]. However, resistance to imipenem was quite low in this study − 9.5%, this is in line with a study by Ike et al. [63] who also reported a higher resistance rate to meropenem (78.7%) than imipenem (57.4%); but contrary to findings by Islam et al. [62] who reported high resistance to imipenem, meropenem (carbapenems), as well as amikacin, gentamycin, tobramycin (aminoglycosides), and ciprofloxacin (a fluoroquinolone) in A. baumannii KBN10P05679 strains.

Resistance to meropenem has been linked to other mechanisms beyond carbapenem hydrolyzing enzymes. These include structural changes or non-synonymous mutations in trans-membrane efflux pump protein AdeB (AdeB G288S or AdeB F136L), whose upregulation is associated with carbapenem resistance [64, 65]. AdeB is a component of the AdeABC efflux pump, and these mutations were reported to slightly subvert the protein structure, opening up the binding site and pore, and leading to more effective exportation of meropenem. Additional mutations in the FtsI (A515V), a PBP, combined with the aforementioned AdeB mutations result in high-level resistance to meropenem via more effective binding to meropenem [65].

Ajoseh et al. [34] in a review article observed variations in the resistance rates of A. baumannii to most antibiotics between 2014 and 2021. A steady increase in the resistance of A. baumannii to tetracycline (91.7% in 2014, 93.4% in 2017 and 100.0% between 2018 and 2021); and imipenem (56.7% in 2014 and 100.0% in 2021) was observed and reported. Increased resistance to carbapenems has been attributed to its large-scale replacement of traditional first-choice drugs [34]. Most isolates have been observed to exhibit resistance, displaying high minimum inhibitory concentration (MIC) values, probably via decreased membrane permeability and/or carbapenem-hydrolyzing enzymes [66].

Resistance to cefiderocol in A. baumannii has been recently reported [66–68]. Cefiderocol, a siderophore cephalosporin, has been reported to exhibit potent activity against a host of multidrug-resistant (MDR) Gram-negative pathogens, including CRAB [68, 69]. Its action has been affiliated with the decreased expression of the pirA gene, a siderophore receptor [69], and it is known to exploit iron transport mechanisms in bacterial cells to gain entry into cells [66]. It can resist inactivation by beta-lactamases, removal by efflux pumps and bind to penicillin-binding proteins (PBPs), effectively truncating cell wall synthesis, leading to bacterial cell lysis. None of the environmental isolates in this study were susceptible to cefiderocol, and a large proportion was resistant at 88.1% (37/42). This is quite worrisome as the development of resistance to this previously effective antibiotic further limits therapeutic options against CRAB.

Gentamicin was observed to be most effective in vitro at 7.1% resistance. This study recorded moderate to minimal resistance to aminoglycosides– 26.2% to tobramycin, 21.4% to amikacin, and 7.1% to gentamicin respectively, at variance with a previous study that reported 97.9%, 89.5% and 87.5% resistance to amikacin, gentamicin and tobramycin, respectively [70]. Another related study [71] reported 50.0% resistance to amikacin and gentamicin, a value higher than the observed values in this study. Our finding suggests the possibility of gentamicin as a potential treatment option, preferably in combination with another class of drug for enhanced activity for the strains of A. baumannii in this environment. Multidrug resistance in isolates (54.8%), as well as high MARI values (88.1%) as observed in this study, is also of immense concern, as these findings portend serious implications in hospitalized patients who are exposed, colonized or already infected with these strains within the hospital setting as this may also adversely impact therapy.

None of the isolates revealed the presence of carbapenemases phenotypically. However, 92.9% of isolates harboured one or more carbapenemase genes, with co-carriage of two or more genes in 64.3% of the recovered isolates. This could be related to the fact that all our isolates were resistant to meropenem, a carbapenemase, while only 9.5% were resistant to imipenem, while the sCIM test was performed with the imipenem disc (as a routine). A high occurrence of bla_KPC type (61.9%) and bla_IMP variants (59.5%) was detected in the isolates. A study conducted in Tehran, Iran [72] contrasted greatly with our findings as they reported that 93.33% of their clinical isolates revealed the bla_OXA-₂₃-like gene, whereas we did not detect the gene in any isolate in this study. Likewise, they reported only 1.7% of bla_IMP in their isolates. However, in line with our findings, none of their isolates also had the bla_NDM−1 gene. This result also conflicts with the study of Odih et al. [24], which reported bla_OXA−23 prevalence in 34.9% and bla_NDM in 27.9% of A. baumannii strains collated in Nigerian hospitals between 2016 and 2020. Again, a meta-analysis of pooled data in Sub-Saharan Africa [25] reported the high predominance of bla_OXA−23 and bla_VIM in isolates. Our inability to detect the bla_OXA−23−like gene in our isolates is in line with a previous report [73]. This could be directly related to the failure of our isolates to express the carbapenemase enzyme during phenotypic screening as bla_{OXA−23−like} gene has been reported to be notably associated with high levels of carbapenem resistance [74]. The absence of the bla_{OXA−23 like} gene in our isolates highlights its role in conferring resistance to carbapenems in A. baumannii [75–77].

Conclusion - This study reports the detection of A. baumannii isolates in clinical and hospital environmental samples in a hospital in Southwest Nigeria. It also reports the occurrence of MDR strains, and of bla_KPC type, bla_IMP variants and bla_VIM type carbapenemase genes in the recovered strains. Acinetobacter baumannii especially carbapenem-resistant and multidrug-resistant strains pose major threats to the healthcare sector. Multidrug resistance in A. baumannii isolates, particularly carbapenem resistance gravely portends severe public health crises. There is therefore a constant need for monitoring and continued surveillance of the strains most especially in the clinical setting where prompt therapeutic responses are of the essence.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

FMA did the conceptualization, funding acquisition, formal analysis, investigation, methodology, obtained resources, supervision, and project administration and wrote the original manuscript, review and editing. EAA participated in funding acquisition, resources, data curation, methodology, formal analysis and investigation. NAY-O, OHA, APD, HSI and AOU did the data curation, methodology, formal analysis and investigation. AAW and OOO participated in project administration, resource acquisition, formal analysis and writing - review and editing.

Funding

The authors did not receive support from any organization for the submitted work.

Data availability

All data collected and analysed during this study are included in this published article {and its supplementary information files].

Declarations

Ethics approval and consent to participate

Ethical approval for the study was obtained from the Ethics Committee of Health Planning, Research, and Statistics Department of the Osun State Ministry of Health, Osogbo (OSHREC/PRS/569T/482), and all procedure in the study complied with the Helsinki Declaration of 1975 (in its most recently amended version). Informed consent was obtained from each participant, and parental informed consent was sought from the parents/guardians of children below 18 years.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.