Assessment of risk factors associated with multidrug-resistant organism infections among patients admitted in a tertiary hospital - a retrospective study

Abstract

Background

Bacterial resistance has become a global health concern. To treat suspected multidrug resistant organisms (MDROs), physicians first use broad-spectrum antibiotics; however, this increases the chance of developing antimicrobial resistance. Thus, defining the risk factors for MDROs could aid in the selection of the ideal initial antimicrobial therapy and improve clinical outcomes.

Objective

This study aimed to identify the common risk factors for MDRO infection among patients admitted to King Fahad Hospital (KFH) and to analyze the comorbidity factors associated with MDRO infections.

Methods

This retrospective, observational, case-control study included adult patients $\ge$ 18 years old admitted to KFH between 1st of January to 31st of March 2021, with positive microbial culture. Pediatric patients, outpatients, or patients with only positive fungal cultures were excluded. Data were obtained from the KFH laboratory MDRO documenting database.

Results

Two hundred and seventy patients were included in this study: 136 in the study group and 134 in the control group. Among patients, 167 (61.9 %) were males and 184 (68.1%) were 18 to 65 years old. The use of drugs such as cotrimoxazole, amikacin, and imipenem (OR = 4.331, C. I. of OR:1.728, 10.855, p = 0.002) were significantly associated with MDRO infections, whereas cefazolin was associated with a lower risk of MDRO infections (OR = 0.080, C.I. of OR:0.018, 0.347, p < 0.001). The intensive care unit showed higher odds of significant association with MDRO infections than those of the surgical unit (odds ratio [OR] = 8.717, 95% C.I. of OR: 3.040, 24.998, p < 0.001). Patients who previously consumed acid-suppressive medications showed higher odds of developing MDRO infections (OR = 5.333, C.I. of OR: 2.395, 11.877, p < 0.001).

Conclusion

The most significant comorbidities were diabetes, hypertension, antibiotic use prior to hospitalization and the use of cotrimoxazole, amikacin and imipenem among other antibtiotics was mostly associated with MRDO infections. This study revealed an increasing trend of MDRO infections and a positive correlation with the incidence of strokes and mortality, which highlights the importance of understanding the risk factors for MDRO infections.

1. Introduction

Bacterial resistance is a global health threat and has intensified with the inappropriate use of antibiotics (Nohl et al., 2020). Multidrug resistant organisms (MDROs) occurs when microorganisms develop antibiotic-resistance and evolve in response to antibiotics misuse and overuse, among other factors. Specifically, isolates are considered multidrug resistant (MDR) if they are resistant to more than one class of antibiotics (Martin-Loeches et al., 2015). Recently, Gonzalez et al. (2019) reported that mortality was more prevalent among patients infected with MDROs than among those infected with non-MDROs. The treatment of bacterial infection is generally considered inappropriate if the culture is not susceptible to the first agent used and this could lead to poor clinical outcomes in patients with severe illness (González Del Castillo et al., 2020). Clinical management of bacterial infections, therefore, usually involve the use of broad-spectrum antibiotics to treat suspected MDROs. However, this may further aggravate the risk of antimicrobial resistance (Oxman et al., 2020).

Consequently, several MDROs have been identified recently, such as methicillin-resistant Staphylococcus aureus (MRSA), methicillin-susceptible Staphylococcus aureus, multidrug-resistant Pseudomonas aeruginosa, susceptible Pseudomonas aeruginosa, extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae (Escherichia coli, Klebsiella pneumoniae), carbapenemase-producing Enterobacteriaceae (Escherichia coli, Klebsiella pneumoniae), non-ESBL- or carbapenemase-producing Enterobacteriaceae (Escherichia coli, Klebsiella pneumoniae), Enterococcus faecalis, Enterococcus faecium, Acinetobacter baumannii, Haemophilus influenzae, Streptococcus pneumoniae, and carbapenemase-resistant Enterobacteriaceae (González Del Castillo et al., 2020).

MDROs are categorized into three types based on their susceptibility to antimicrobial chemotherapy. Type one MDROs are MDR with acquired non-susceptibility to at least one agent of three or more antimicrobial classes. Type two MDROs are microorganisms extensively drug-resistant and non-susceptible to at least one agent in all, except two or fewer classes of antimicrobials. The third type of MDROs are pan-drug resistant and non-susceptible to all agents in all antimicrobial categories (Martin-Loeches et al., 2015). The mechanisms underlying antimicrobial resistance include mutational alteration of the target protein, enzymatic inactivation of the drug, bypassing the target, preventing drug access to targets, and expression of multidrug efflux pumps (Nikaido 2009).

Recently, the prevalence of MDROs, including MRSA, in nosocomial and community settings has increased in Saudi Arabia (Nazeer and Al-Tawfiq, 2012, Al-Asmari et al., 2015, Baig et al., 2015, Farah et al., 2019). Vancomycin-resistant enterococci, Enterococcus faecalis and Enterococcus faecium, have developed considerable resistance to several antibiotics, such as penicillin, macrolides, sulfamethoxazole, and aminoglycosides (Khan et al., 2018). Streptococcus pneumoniae isolates are developing resistance to penicillin, while resistance to cotrimoxazole and macrolides is increasing in other species (Yezli et al., 2012, Cherazard et al., 2017). Defining the risk factors for MDROs is a key factor in the selection of the effective initial antimicrobial agents and improve clinical outcomes (Martin-Loeches et al., 2015). Therefore, the purpose of this study was to identify the common risk factors for MDROs among patients admitted at King Fahad Hospital (KFH).

2. Patients and methods

2.1. Study design and setting

This retrospective, observational, case-control study was conducted at KFH in Madinah, Kingdom of Saudi Arabia. The study included $\ge$ 18 years old adult patients admitted to KFH between the 1st of January and 31st of March 2021, whose microbial cultures tested positive. Pediatric patients, outpatients, or patients with only positive fungal cultures were excluded from the study.

For the study group (MDRO bacteria-positive culture), patients were identified using data obtained with the assistance of the laboratory department at the KFH MDRO documenting database. For the control group (non-MDRO bacterial positive culture), hospital records were used to obtain the results of all non-MDRO cultures for the same period (1st of January and 31st of March 2021). This study was approved by the Institutional Review Board of the General Directorate of Health Affairs in Madinah, Saudi Arabia (H-03-M−084). The data were used for research only and confidentiality was maintained throughout the study.

2.2. Demographic data

The collected data included age, sex, body mass index (BMI), area of care in the hospital, duration of stay, mortality (within the admission), vital signs (temperature, systolic blood pressure, diastolic blood pressure, heart rate, and respiratory rate), culture information (type of sample, type of organism, and type of acquired resistance), comorbidities (hypertension, diabetes mellitus, ischemic heart disease, congestive heart failure, chronic obstructive pulmonary disease, dementia, stroke, chronic liver disease, chronic kidney disease (CKD), peripheral vascular disease, connective tissue disease, peptic ulcer, hemiplegia, solid tumor, leukemia, and acquired immunodeficiency syndrome), Charlson comorbidity index, medical history (coagulopathy, hospital admission within the previous 90 days, whether using acid-suppressive medications or corticosteroids, and any previous surgery), use of medical equipment (nasogastric tube (NGT), urinary catheter, mechanical ventilation, or central venous line), laboratory results within 2 days of admission (neutrophils %, lymphocytes %, monocytes %, eosinophils % basophils %, hemoglobin, serum creatinine, urea, and blood glucose), and the type of antibiotic used previously and during admission (Supplementary tables).

2.3. Definitions

MDROs were identified using standardized international terminology created by the European Centre for Disease Prevention and Control and the Centers for Disease Control and Prevention (Magiorakos et al., 2012).

2.4. Statistical analyses

Data analyses were performed using IBM SPSS software for Windows, version 26. Descriptive statistics are presented as means and standard deviations for numeric variables and frequencies and percentages for categorical variables. Differences between cases and controls were tested using the independent samples t-test for numeric variables and the chi-square test for categorical variables. Multiple logistic regression analysis was used to study the association between MDROs and other variables. Variables with a p-value < 0.2 in the univariate analysis were included in the multivariable analysis. Crude and adjusted odds ratios (ORs) with 95 % confidence intervals (Cis) were reported as measures of association. Statistical significance was set at p < 0.05. A comparison of MDROs and control cases was performed using the chi-square or exact test. The Charlson comorbidity index was compared using an independent sample t-test.

3. Results

3.1. Characteristics of the study cohort

The study included 270 patients, among which, 136 patients were in the study group and 134 patients the control group, respectively. Majority of the patients were males (n = 167, 61.9%). Most of patients (n = 184, 68.1%) were 18–65 years old, with variable duration of stay in hospital, and an overall mortality of 22.2% (Table 1).

Table 1.

Demographic data and medical history of the study cohort.

Patient demographic data
		Frequency (n)	Percentage (%)
Gender	Male	167	61.9
	Female	103	38.1
Age	18–65	184	68.1
	> 65	86	31.9
Area of care	Surgical	99	36.7
	Medical	84	31.1
	ICU	87	32.2
Length of stay	1–3 days	32	11.9
	4–7 days	48	17.8
	8–14 days	53	19.6
	15–30 days	55	20.4
	31–60 days	47	17.4
	61–90 days	13	4.8
	91–180 days	17	6.3
	180–360 days	3	1.1
	> 360 days	2	0.7
Mortality	No	210	77.8
	Yes	60	22.2
Comorbidities
Disease		Frequency (n)	Percentage (%)
Hypertension		132	48.9
Diabetes Miletus		135	50
Ischemic Heart Disease		28	10.5
Congestive Heart Failure		7	2.6
Chronic Obstructive Pulmonary Disease		3	1.1
Dementia		5	1.9
Stroke		37	13.9
Chronic Liver Disease		5	1.9
Chronic Kidney Disease		46	17
Peripheral Vascular Disease		4	1.5
Connective Tissue Disease		1	0.4
Peptic Ulcer		11	4.1
Hemiplegia		8	3
Solid Tumor		14	5.2
Leukemia		2	0.7
Lymphoma		5	1.9
Acquired Immunodeficiency Syndrome (Aids)		1	0.4
Medical History
		Frequency (n)	Percentage (%)
Coagulopathy		19	7.1
Previous 3 months hospital admission		87	32.8
Acid suppressive therapy		133	49.3
Corticosteroids		41	15.3
IV antibiotics in the previous 3 months		93	35.1
Previous surgery		50	18.5
History of antibiotics use in the Previous 3 months
		Frequency (n)	Percentage (%)
Ampicillin		3	1.1
Amoxicillin		13	4.8
Metronidazole		31	11.5
Cefazolin		29	10.7
Ceftriaxone		40	14.8
Cefuroxime		16	5.9
Ceftazidime		30	11.1
Cefotaxime		7	2.6
Cefepime		1	0.4
Cotrimoxazole		2	0.7
Amikacin		2	0.7
Gentamycin		0	0
Ciprofloxacin		13	4.8
Linezolid		17	6.3
Moxifloxacin		4	1.5
Levofloxacin		2	0.7
Meropenem		15	5.6
Tigecycline		16	5.9
Vancomycin		34	12.6
Clindamycin		4	1.5
Colistin		9	3.3
Imipenem		24	8.9
Tazocin		39	14.4

The most frequently reported comorbidities among patients were diabetes mellitus (n = 135, 50 %), followed by hypertension (n = 132, 48.9%) (Table 1). Other comorbidities were the presences of coagulopathy, admission to hospital in the previous 3 months, the use of acid suppressive therapy or corticosteroids or IV antibiotics in the previous 3 months (yes/no) and history of previous surgeries (Table 1). The intravenous (IV) administered antibiotics in the last 3 months prior to hospitalization were also investigated as a possible comorbidity. The most administered antibiotics were ceftriaxone (14.8%), piperacillin/tazobactam (14.4%), metronidazole (11.5%), ceftazidime (11.1%), and cefazolin 29 (10.7%), among others (Table 1). Charlson comorbidity index (CCI) was used to assign patients’ groups accordingly (Supplementary tables). Finally, samples were withdrawn from patients (Table 2) and microbiological analysis was carried out to identify a range of microorganisms. The most common infective microorganisms among patients were Klebsiella Pneumoniae (21.5%) followed by Staphylococcus aureus (14.8%), Acinetobacter baunmanni (12.6%), Escherichia coli (14.4%), and Klebsiella pneumoniae (21.5%) (Table 2). Classification of identified MDRO indicated that the majority belonged to the carbapenem-resistant enterobacterales, CRE, group (24.3%), while vancomycin resistant enterococcus, VRE, group was the least identified group, in only 2 patients (1.5%) (Table 2).

Table 2.

Microbiological findings and identified microorganisms.

Sample withdrawn from patients
	Frequency (n)		Percentage (%)
Blood	46		17
Sputum	66		24.4
Urine	61		22.6
Wound	50		18.5
Biopsy	10		3.7
Abscess	27		10
Other	10		3.7
Identified microorganisms
	Frequency (n)		Percentage (%)
S. Aureus	40		14.8
S. Marcescens	5		1.9
P. Stuartii	2		0.7
A. Baunmanni	34		12.6
E. Coli	39		14.4
E. Faecium	6		2.2
Enterobacter Aerogenes	1		0.4
K. Pneumonia	58		21.5
P. Aeruginosa	17		6.3
P. Mirabilis	11		4.1
Providencia Rettgeri	1		0.4
Bacillus Sp.	2		0.7
E. Faecalis	7		2.6
Enterobacter Cloacae	5		1.9
gram-negative rods	2		0.7
Micrococcus Luteus	1		0.4
Morganella Morganii	3		1.1
Sphingomonas Paucimobilis	1		0.4
Staphylococcus Epidermidis	2		0.7
Staphylococcus Lugdunensis	2		0.7
Streptococcus Anginosus	4		1.5
Streptococcus Sp.	1		0.4
Streptococcus Viridans	1		0.4
Serratia Rubidaea	1		0.4
Aeromonas Caviae	1		0.4
C. Amalonaticus	1		0.4
Corynebacterium Jeikeium	1		0.4
Enterococcus Raffinosus	1		0.4
S. Agalactiae	4		1.5
Salmonella Sp.	1		0.4
Staphylococcus Hominis	1		0.4
Stenotrophomonas Maltophilia	2		0.7
C. Freundi	1		0.4
Citrobacter Koseri (diversus)	1		0.4
Escherichia Hermannii	1		0.4
Klebsiella Ozaenae	1		0.4
Proteus Vulgaris	1		0.4
Streptococcus Constellatus	1		0.4
Streptococcus Dysgalactiae	1		0.4
Streptococcus Intermedius	1		0.4
Streptococcus Mitis	1		0.4
Streptococcus Oralis	2		0.7
Streptococcus Porcinus	1		0.4
GRouping of MDRO
	Frequency (n)		Percentage (%)
Methicillin resistant staphylococcus aureus (MRSA)	26		19.1
Multidrug resistant (MDR)	21		15.4
carbapenem-resistant enterobacterales (CRE)	33		24.3
extended spectrum beta lactamase (ESBL)	26		19.1
Extensively drug resistant (XDR)	28		20.6
Vancomycin resistant enterococcus (VRE)	2		1.5
MDRO Vs Control
	MDRO		Control
	N	%	N	%
S. Aureus	26	19.1	14	10.4
S. Marcescens	1	0.7	4	3
P. Stuartii	2	1.5	0	0
A. Baunmanni	28	20.6	6	4.5
E. Coli	18	13.2	21	15.7
E. Faecium	3	2.2	3	2.2
Enterobacter Aerogenes	1	0.7	0	0
K. Pneumonia	42	30.9	16	11.9
P. Aeruginosa	2	1.5	15	11.2
P. Mirabilis	8	5.9	3	2.2
Providencia Rettgeri	1	0.7	0	0
Bacillus Sp.	0	0	2	1.5
E. Faecalis	0	0	7	5.2
Enterobacter Cloacae	2	1.5	3	2.2
gram-negative rods	0	0	2	1.5
Micrococcus Luteus	0	0	1	0.7
Morganella Morganii	0	0	3	2.2
Sphingomonas Paucimobilis	0	0	1	0.7
Staphylococcus Epidermidis	0	0	2	1.5
Staphylococcus Lugdunensis	0	0	2	1.5
Streptococcus Anginosus	0	0	4	3
Streptococcus Sp.	0	0	1	0.7
Streptococcus Viridans	0	0	1	0.7
Serratia Rubidaea	1	0.7	0	0
Aeromonas Caviae	0	0	1	0.7
C. Amalonaticus	0	0	1	0.7
Corynebacterium Jeikeium	0	0	1	0.7
Enterococcus Raffinosus	0	0	1	0.7
S. Agalactiae	0	0	4	3
Salmonella Sp.	0	0	1	0.7
Staphylococcus Hominis	0	0	1	0.7
Stenotrophomonas Maltophilia	0	0	2	1.5
C. Freundi	1	0.7	0	0
Citrobacter Koseri (diversus)	0	0	1	0.7
Escherichia Hermannii	0	0	1	0.7
Klebsiella Ozaenae	0	0	1	0.7
Proteus Vulgaris	0	0	1	0.7
Streptococcus Constellatus	0	0	1	0.7
Streptococcus Dysgalactiae	0	0	1	0.7
Streptococcus Intermedius	0	0	1	0.7
Streptococcus Mitis	0	0	1	0.7
Streptococcus Oralis	0	0	2	1.5

3.2. The effect of comorbidities and risk factors in MDRO and control groups

Variables and risk factors were compared in MDRO and non-MDRO infected patients. Regarding demographic parameters, age and gender did not show any significant differences between the study group and the control group. However, significantly higher number of cases were reported in the intensive care unit (ICU) than in the surgical unit (Table 3). Regarding body mass index (BMI), 44.8% of the patients had a BMI in the range of 18.5–24.9, and 34.4% reported a BMI in the range of 25–29.9, while differences were not significant (results not shown). Significantly higher mortality, however, was found in the study group showing that MDRO infections can contribute to increased mortality (Table 3). Additionally, data were collected regarding vital signs on admission and routine laboratory investigations showed significantly lower hemoglobin (<12.5 g/dL) and higher urea levels (>8.3 mmol/L) in the study group as compared to the control group (p < 0.001 and p = 0.007, respectively) (Supplementary tables).

Table 3.

Risk factors and comorbidities in MDRO compared to the control group.

Demographic data and mortality in MDRO and control
		MDRO		Control		P-value
		N	%	N	%
Gender	Male	90	66.2	77	57.5	0.141
	Female	46	33.8	57	42.5
Age	18–65	94	69.1	90	67.2	0.730
	> 65	42	30.9	44	32.8
Area	Surgical	33	24.3	66	49.3	< 0.001
	Medical	31	22.8	53	39.6
	ICU	72	52.9	15	11.2
Length of stay	1–7 days	34	25.0	67	50.0	< 0.001
	8–14 days	31	22.8	24	17.9
	15–30 days	34	25.0	13	9.7
	31–60 days	11	8.1	2	1.5
	> 60 days	26	19.1	28	20.9
Mortality	No	89	65.4	121	90.3	< 0.001
	Yes	47	34.6	13	9.7
Comorbidities
		Average MDRO (SD)		Average control (SD)		P-value
		N	%	N	%
Hypertension		69	50.7	63	47.0	0.541
Diabetes Miletus (DM)		63	46.3	72	53.7	0.224
Ischemic heart disease		15	11.3	13	9.7	0.674
Congestive heart failure		5	3.7	2	1.5	0.447
Chronic obstructive pulmonary disease		2	1.5	1	0.7	> 0.999
Dementia		2	1.5	3	2.2	0.683
Stroke		24	18.2	13	9.7	0.046
Chronic liver disease		3	2.2	2	1.5	> 0.999
Chronic kidney disease		29	21.3	17	12.7	0.059
Peripheral vascular disease		1	0.7	3	2.2	0.369
Connective tissue disease		0	0	1	0.7	0.496
Peptic ulcer		5	3.7	6	4.5	0.739
Hemiplegia		8	5.9	0	0	0.007
Solid tumor		5	3.7	9	6.7	0.266
Leukemia		1	0.7	1	0.7	> 0.999
Lymphoma		1	0.7	4	3.0	0.212
Acquired immunodeficiency syndrome (AIDS)		0	0	1	0.7	0.496
History of coagulopathy		13	9.6	6	4.5	0.099
Previous hospital admission in 3 months		52	39.7	35	26.1	0.019
Acid suppressive medications		94	69.1	39	29.1	< 0.001
Corticosteroids		30	22.4	11	8.2	0.001
Previous IV antibiotics in 3 months		68	51.9	25	18.7	< 0.001
Previous surgery		34	25.0	16	11.9	0.006
Use of external medical support equipment
		MDRO		Control		P-value
		N	%	N	%
NGT or oGt		70	51.5	27	20.1	< 0.001
Urinary catheter		96	70.6	55	41.0	< 0.001
Mechanical ventilation		54	39.7	23	17.2	< 0.001
Central venous line		56	41.2	25	18.7	< 0.001

IV antibiotics 3 Months before admission
	MDRO		CONTROL		P-VALUE
	N	%	N	%
Ampicillin	2	1.5	1	0.7	> 0.999
Amoxicillin	7	5.1	6	4.5	0.797
Metronidazole	21	15.4	10	7.5	0.040
Cefazolin	21	15.4	8	6	0.012
Ceftriaxone	32	23.5	8	6	< 0.001
Cefuroxime	13	9.6	3	2.2	0.011
Ceftazidime	27	19.9	3	2.2	< 0.001
Cefotaxime	6	4.4	1	0.7	0.120
Cefepime	1	0.7	0	0	> 0.999
Cotrimoxazole	1	0.7	1	0.7	> 0.999
Amikacin	2	1.5	0	0	0.498
Gentamycin	4	2.9	5	3.7	0.748
Ciprofloxacin	10	7.4	3	2.2	0.050
Linezolid	15	11	2	1.5	0.001
Moxifloxacin	2	1.5	2	1.5	> 0.999
Levofloxacin	2	1.5	0	0	0.498
Meropenem	14	10.3	1	0.7	0.001
Tigecycline	15	11	1	0.7	< 0.001
Vancomycin	30	22.1	4	3	< 0.001
Clindamycin	4	2.9	0	0	0.122
Colistin	9	6.6	0	0	0.003
Imipenem	18	13.2	6	4.5	0.011
Tazocin	34	25	5	3.7	< 0.001

antibiotics during admission
	MDRO		Control		P-value
	N	%	N	%
Ampicillin	5	3.7	4	3	> 0.999
Amoxicillin	14	10.3	20	14.9	0.251
Metronidazole	26	19.1	41	30.6	0.029
Cefazolin	9	6.6	17	12.7	0.091
Ceftriaxone	34	25	60	44.8	0.001
Cefuroxime	7	5.1	8	6	0.768
Ceftazidime	23	16.9	28	20.9	0.403
Cefotaxime	2	1.5	2	1.5	> 0.999
Cefepime	1	0.7	2	1.5	0.621
Cotrimoxazole	15	11	4	3	0.010
Amikacin	12	8.8	2	1.5	0.007
Gentamycin	4	2.9	5	3.7	0.748
Ciprofloxacin	12	8.8	11	8.2	0.856
Linezolid	32	23.5	18	13.4	0.033
Moxifloxacin	2	1.5	2	1.5	> 0.999
Levofloxacin	3	2.2	0	0	0.247
Meropenem	66	48.5	13	9.7	< 0.001
Tigecycline	46	33.8	0	0.0	< 0.001
Vancomycin	64	47.1	21	15.7	< 0.001
Clindamycin	9	6.6	8	6.0	0.827
Colistin	16	11.8	2	1.5	0.001
Imipenem	57	41.9	16	11.9	< 0.001
Tazocin	36	26.5	39	29.1	0.629

Similarly, the duration of stay and mortality were significantly different, wherein a larger number of patients had a shorter hospital stay (1–7 days); furthermore, higher mortality rates were reported in the study group than those in controls (p < 0.001). The comparison between the study and control groups is presented in Table 3.

Among various pathogens, Klebsiella pneumoniae was the most common pathogen, accounting for 30.9% in the study group and 11.9% in the control group, followed by Acinetobacter baunmanni (20.6% in the study group and 4.5% in the control group), then Staphylococcus aureus (19.1% in the study group and 10.4% in the control group) (Table 2).

Patients of both the groups had no significant differences in pathological comorbidities, however the incidence of stroke (p = 0.046) and hemiplegia (p = 0.007) were significantly higher in the study group than in controls. Additionally, admission to hospital in the previous 3 months, the use of acid suppressive therapy or corticosteroids or IV antibiotics in the previous 3 months and history of previous surgeries previous hospitalization in the last 3 months or history of previous surgeries were all significantly higher in the study group than in the controls (Table 3). Similarly, the use of NGTs, urinary catheters, mechanical ventilation, or central venous lines were significantly higher in the study group c compared to the control group (Table 3).

Administration of IV antibiotics in the last 3 months before hospital admission was also investigated as a possible risk factor for acquiring MDRO infection. Ceftriaxone, ceftazidime, linezolid, meropenem, tigecycline, vancomycin, and tazocin were the IV antibiotics administered in the preceding 3 months and their correlation to the study group (MDRO) was highly significant (p < 0.001) when comparing to the control group (Table 3). Analyzing the effect of antibiotic administrated during the hospitalization period, the use of meropenem, imipenem, tigecycline, vancomycin, or colistin was significantly higher in the study group as compared to the control group (p < 0.001). Additionally, significant differences were also noted with metronidazole, ceftriaxone, cotrimoxazole, amikacin and linezolid (Table 3).

3.3. Simple and multiple regression analysis for MDRO-associated factors

To further confirm or otherwise disprove previous findings, simple regression analysis was used to compare the two groups (Table 4). Afterwards, variables with a p-value < 0.2 in the univariate analysis were included in the multivariable analysis (Table 5). Simple logistic regression analysis revealed that factors such as the area of care, the duration of hospital stay, stroke, previous hospital admission within the last 3 months, acid-suppressive medications, corticosteroids, administration of IV antibiotics within the last 3 months, and previous surgery were substantially linked to infection with MDROs.

Table 4.

Univariate regression analysis.

Univariate regression analysis of variables
Demographic data and clinical parameters	Univariate regression
	Crude OR	95% C.I. for OR		P-Value
		Lower	Upper
Gender (female)	0.69	0.42	1.13	0.141
Age ( > 65)	0.91	0.55	1.53	0.731
Area of care
Surgical	1
Medical	1.170	0.636	2.151	0.614
ICU	9.600	4.788	19.250	< 0.001
Length Of Stay
1–7 days	1
8–14 days	2.25	1.30	4.99	0.007
15–30 days	5.15	2.41	11.03	< 0.001
31–60 days	10.84	2.27	51.69	0.003
> 60 days	1.83	0.93	3.59	0.079
Comorbidities
Stroke	2.07	1.00	4.26	0.049
Previous hospital admission in 3 months	1.86	1.11	3.13	0.019
Acid suppressive meds	5.45	3.24	9.18	< 0.001
Corticosteroids	3.23	1.54	6.75	0.002
Previous iv antibiotics in 3 months	4.71	2.71	8.19	< 0.001
Previous surgery	2.46	1.28	4.71	0.007
Use of external medical support equipment
NGT or OGT	4.20	2.45	7.21	< 0.001
Urinary catheter	3.45	2.08	5.71	< 0.001
Mechanical ventilation	3.18	1.81	5.59	< 0.001
Central venous line	3.05	1.76	5.30	< 0.001
Previous 3 months of antibiotics
Ampicillin	1.99	0.18	22.16	0.577
Amoxicillin	1.16	0.38	3.54	0.797
Metronidazole	2.26	1.02	5.01	0.044
Cefazolin	2.88	1.23	6.75	0.015
Ceftriaxone	4.85	2.14	10.97	< 0.001
Cefuroxime	4.62	1.28	16.59	0.019
Ceftazidime	10.82	3.19	36.62	< 0.001
Cefotaxime	6.14	0.73	51.69	0.095
Cotrimoxazole	0.99	0.06	15.91	0.992
Ciprofloxacin	3.47	0.93	12.89	0.064
Linezolid	8.18	1.83	36.52	0.006
Moxifloxacin	0.99	0.14	7.10	0.988
Meropenem	15.26	1.98	117.80	0.009
Tigecycline	16.49	2.15	126.69	0.007
Vancomycin	9.20	3.14	26.93	< 0.001
Imipenem	3.25	1.25	8.48	0.016
Tazocin	8.60	3.25	22.78	< 0.001
Antibiotics during admission
Ampicillin	1.24	0.33	4.72	0.752
Amoxicillin	0.65	0.32	1.36	0.254
Metronidazole	0.54	0.31	0.94	0.030
Cefazolin	0.49	0.21	1.14	0.096
Ceftriaxone	0.41	0.25	0.69	0.001
Cefuroxime	0.85	0.30	2.43	0.768
Ceftazidime	0.77	0.42	1.42	0.404
Cefotaxime	0.99	0.14	7.10	0.988
Cefepime	0.49	0.04	5.46	0.561
Cotrimoxazole	4.03	1.30	12.48	0.016
Amikacin	6.39	1.40	29.11	0.017
Gentamycin	0.78	0.21	2.98	0.718
Ciprofloxacin	1.08	0.46	2.55	0.856
Linezolid	1.98	1.05	3.74	0.035
Moxifloxacin	0.99	0.14	7.10	0.988
Levofloxacin	8.78	4.52	17.04	< 0.001
Meropenem	4.78	2.69	8.50	< 0.001
Tigecycline	1.12	0.42	2.99	0.827
Vancomycin	8.80	1.98	39.07	0.004
Clindamycin	5.32	2.85	9.93	< 0.001
Colistin	0.88	0.51	1.49	0.629

Table 5.

Statistically significant multivariate regression analysis of variables.

Comorbidities
ICU	8.717	3.040	24.998	< 0.001
CKD	3.017	1.096	8.301	0.033
Previous hospital admission in 3 months	0.338	0.130	0.878	0.026
Acid suppressive meds	5.333	2.395	11.877	< 0.001
Previous surgery	4.054	1.565	10.500	0.004
Administration of Tazocin in the previous 3 months	14.593	3.335	63.850	< 0.001
Antibiotics administered in hospital
Cefazolin	0.080	0.018	0.347	< 0.001
Cotrimoxazole	6.229	1.425	27.229	0.015
Amikacin	13.359	1.230	145.044	0.033
Imipenem	4.331	1.728	10.855	0.002

The stepwise backward method was used to generate the final model. It was noted that the area of care was significantly associated with MDROs, with higher odds in the ICU than those in the surgical unit (OR = 8.717, 95% CI of OR: 3.040, 24.998, p < 0.001).

Furthermore, patients with CKD had a significantly greater risk of infection with MDROs than those without CKD (OR = 3.017, 95% CI of OR: 1.096, 8.301, p = 0.033). Patients who were hospitalized in the preceding 3 months had a lower OR than those who were not hospitalized during this period (OR = 0.338, C.I. of OR: 0.130, 0.878, p = 0.026). Patients who had previously taken acid-suppressive medications were at a significantly higher risk of infection with MDROs than those who had never taken any acid-suppressive medications (OR = 5.333, CI of OR: 2.395, 11.877, p < 0.001).

Consistently, previous surgery was significantly associated with MDRO infection, as patients who had undergone surgery showed a higher OR than those who had no previous surgery (OR = 4.054, CI of OR: 1.565, 10.500, p = 0.004). Regarding antibiotic use in the last 3 months, tazocin use was significantly associated with MDRO infection, as patients who took tazocin had higher odds of developing MDRO infection than those who did not (OR = 14.593, CI of OR:3.335, 63.850, p < 0.001). Regarding antibiotic administration during admission, the following antibiotics showed statistically significant associations with MDRO infections: cotrimoxazole (OR = 6.229, CI of OR: 1.425, 27.229, p = 0.015), amikacin (OR = 13.359, CI of OR: 1.230, 145.044, p = 0.033), and imipenem (OR = 4.331, CI of OR: 1.728, 10.855, p = 0.002). In contrast, cefazolin use was associated with a lower risk of MDRO infection (OR = 0.080, CI of OR: 0.018, 0.347, p < 0.001) and may exert a protective effect against MDRO infection.

4. Discussion

In this study, risk factors associated with MDRO infection and the impact of antimicrobial resistance and appropriate empirical antibiotics on patient outcomes, including mortality, were identified. The findings in this study revealed that age and sex were not associated with drug resistance. In contrast, factors including previous incidence of strokes, use of medical equipment, area of care, length of stay, previous hospitalization in the last 3 months, use of acid-suppressive medication, corticosteroid use, IV antibiotics in the last 3 months, and previous surgery were significantly associated with MDRO infections. Both groups presented similar comorbidities except for stroke and hemiplegia, whose incidences were significantly higher in the study group than in controls. The use of medical equipment, including NGTs, urinary catheters, mechanical ventilation, and central venous lines, was significantly higher in the study group than those in the controls (p < 0.001). These results are consistent with previous studies, which also reported that stroke and the use of clinical equipment, such as mechanical ventilation, urinary catheters, feeding tubes, and central venous lines, are risk factors for infection with MDROs (Vazquez-Guillamet et al., 2014, Kalluru et al., 2018).

The findings in this study revealed that the area of care had a statistically significant association with MDRO infections, with the ICU having higher odds for developing MDRO infection than the surgical unit. Another recently published multicenter study assessed the incidence, impact, and risk factors for MDRO infections in patients with major trauma and reported that being male, injury severity score (ISS), administration of packed red blood cells, ICU stay > 48 h, and mechanical ventilation were the main risk factors for MDRO infection; moreover, MDRO infection was associated with worse patient outcome, Gram-negative bacteria outnumber gram-positive bacteria in patients with numerous injuries (Nohl et al., 2020).In the same previous study found a link between injury severity and MDRO detection, with higher the ISS, the greater the likelihood of positive MDRO detection. Notably, in this study, a low hemoglobin level was linked to infection with MDROs.

Some studies have categorized the risk factors according to the isolate type. However, structural disease are one of the potential risk factors apart from nursing home stay, prior use of corticosteroids, antibiotic use within the last 90 days, and poor nutrition. The risk factors of Pseudomonas aeruginosa infection including alcoholism, diabetes mellitus, and chronic lung disease, in addition to prolonged hospitalization (>14 days), functional disability, or medical device use are risk factors associated with infections caused by Acinetobacter baumannii isolates. Klebsiella pneumoniae shares the same risk factors as Acinetobacter baumannii in addition to the female sex. MDR Klebsiella pneumoniae is a virulent pathogen causing severe community-acquired pneumonia, which is frequently associated with septic shock, respiratory failure, and bacteremia (Lin et al., 2010, Morgan and Glossop, 2016). Other potential risk factors include age > 65 years, male sex, chronic respiratory disease, CKD, altered mental status, and temperature > 37.88 °C upon arrival at the first evaluation. A different factor accounts for each type of MDR infection: ESBL was associated with previous antibiotic use, while MRSA was related to diabetes mellitus, CKD, and altered mental status. PES pathogens (Pseudomonas aeruginosa, Enterobacteriaceae ESBL-positive, and MRSA) are associated with longer hospital stay and 30-day mortality (Prina et al., 2015). In this study, the duration of hospital stay showed a statistically significant association as the number of cases were higher at 1–7 days than those at 31–60 days. Among the sample types, sputum samples were more frequently analyzed in controls (p = 0.012) than that in the study group. This may indicate a higher risk of MDRO infection-acquired pneumonia. Another study revealed that respiratory infections are the most frequent infection associated with MDRO sepsis, followed by urinary infections (Prado et al., 2022). Previous studies in Saudi Arabia investigated the association of risk factors and MDRO, however, mostly in a group of organisms such as gram negative bacteria (Al Hamdan et al., 2022) or Acinetobacter species (Al Bshabshe et al., 2016) or in a single organism such as acinetobacter baumanni (Almaghrabi et al., 2018) or investigating bacteria resistant to a certain antibiotic (Al Mayahi et al., 2019, Alqasim, 2021). Of note, studies covering all MRDOs types in Saudi Arabia were dated back to 2014 and 2015 (Mwanri and AlSaleh, 2014, Baig et al., 2015). More recent studies focused on ICU patients in the eastwrn area of Saudi Arabia (Al Hamdan et al., 2022), and antibiotics prescription in relation to MDR in Riyadh region in 2020, but it reported data from 2016 to 2017 and did not investigate risk factors (Aldawsari et al., 2020). Additionally, none of these studies used the Charlson comorbidity index, CCI, to account for the aggregate effect of a number of risk factors.

In our study, Klebsiella pneumoniae was the most abundantly present pathogen, with a prevalence of 30.9% in the study group compared to 11.9% in the control group. This was followed by Acinetobacter baumannii, with a prevalence of 20.6% in the study group compared to 4.5% in the control group, and Staphylococcus aureus with 19.1% in the study group compared to 10.4% in the control group. Acinetobacter baumannii isolates from the ICU are considered MDR that show resistance to imipenem, piperacillin-tazobactam, and other commonly used antibiotics, including cephalosporins, aztreonam, and amikacin (El-Ageery and Al-Hazmi 2014). These findings were confirmed by another study based on 48 carbapenem-resistant Acinetobacter baumannii strains isolated from patients with different types of infections, who were either admitted or attending outpatient clinics at the same hospital; however, there were varying patterns of resistance to amoxicillin/clavulanic acid, piperacillin, gentamycin, amikacin, and aztreonam (El-Ageery and Al-Hazmi 2014).

Patients with a history of using acid-suppressive drugs and corticosteroids had higher odds of developing MDRO infections than those who did not use any of them. Incidence of hospital admission in the last 3 months was higher in the study group than in the controls, as patients hospitalized 3 months ago had a lower OR for MDRO infection than those who were not. IV antibiotics use at 3 months differed significantly between the study group and controls (p < 0.001). The use of tazocin demonstrated a statistically significant association with MDRO infection, with patients receiving tazocin having a greater risk of infection than those who did not.

Regarding antibiotic administration upon admission, use of cotrimoxazole, amikacin, and imipenem showed a statistically significant association with MDRO infection. In contrast, cefazolin was associated with a decreased incidence MDRO infection and likely exerted a protective effect against MDROs. Consistently, El-Ageery et al. (2014) reported that cefazolin is linked to a decreased incidence of MDRO infections. In the present scenario, bacterial resistance is rapidly evolving compared to the rate of discovery of antibacterial drugs, and the inappropriate use of broad-spectrum antibiotics can further exacerbate the problem of MDRO infections (Lat et al., 2019).

Chronic renal illness is a risk factor for respiratory MDRO infections. In a previous study, the proportion of patients with MDRO infections was 3.9% in those without risk factors, 12.6% in those with one risk factor, and 53.6% in those with two risk factors (Menéndez et al., 2017). CKD has a statistically significant association with MDRO infections, and patients with CKD are at a higher risk of developing MDRO infection than those without CKD. Mortality too showed a significant difference between the study and control groups, with patients with MDRO infections dying at a faster rate than in the control group (p = 0.001). Early screening for MDROs may aid in appropriate antibiotic selection and boost patient survival rate, whereas delayed screening increases mortality (Yiang et al., 2021).

There are some limitations to the present study. First, data related to vital signs were not available for several patients on the hospital record system, which could be an essential aspect of MDRO infections. Second, the data were collected from a single Saudi Arabian facility, making them unrepresentative of a larger population and internationally non-generalizable. Nevertheless, those findings emphasize that more focus should be placed on enhancing the awareness of improper antibiotic prescription and use among health care professionals and the general public to slow down the spread of antibiotic resistance. This study reveals an increasing trend of MDRO infections, implying that recognizing the risk factors for MDRO infections and appropriate administration of antibiotics may help prevent high patient mortality.

5. Conclusion

In summary, the area of care, duration of stay, type of sample, stroke, previous hospital admission, history of using acid-suppressive drugs and corticosteroids, previous IV antibiotics use, and previous surgery were strongly associated with MDRO infections. By identifying the clinical factors and administering timely and appropriate empirical antibiotic therapy, the mortality rate associated with MDRO infections can be reduced. Furthermore, frequent observations of incidence and tailored actions may slow down the spread of antibiotic resistance.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

The following are the Supplementary data to this article: