Prevalence and risk factors associated with multidrug-resistant bacteria in COVID-19 patients

Abstract

Bacterial coinfection among patients with confirmed coronavirus disease 2019 (COVID-19) is a critical medical concern that increases the disease severity and mortality rate. The current study is aimed at evaluating the effects of bacterial coinfections among COVID-19 patients, especially in relation to degree of severity and mortality. A retrospective study was conducted for patients with positive COVID-19 test, admitted to a regional COVID-19 hospital in Jeddah, Saudi Arabia, between May and August 2020. A specimen (e.g., blood, urine, or sputum) was collected from patients with confirmed COVID-19, and was cultured to determine bacterial coinfection caused by multidrug resistant (MDR) bacteria. COVID-19 patients were categorized into 2 groups based on the result of bacterial coinfection culture, as COVID-19 patients with coinfection and COVID-19 patients without coinfection. Independent sample t test or Mann–Whitney U test was used to compare age and hospitalization period between these groups. In addition, binominal logistic regression was applied to identify risk factors associated with mortality and bacterial coinfection. The study included 342 patients with laboratory confirmed COVID-19. Eighty (23.3%) patients were diagnosed with bacterial coinfection, while the remaining 262 (76.6%) patients did not test positive for bacterial coinfection. Length of hospital stay was prolonged among COVID-19 patients diagnosed with bacterial coinfection (16.01 ± 11.36 days) when compared with patients without bacterial coinfection (6.5 ± 6.12 days). Likewise, the mortality rate was significantly higher among COVID-19 patients with bacterial coinfection (90%) compared to those without bacterial coinfection (49.2%). Gram-negative bacteria were predominant compared to gram-positive, as Klebsiella pneumoniae (35 [43.8%]) and Acinetobacter baumanni (32 [40%]). On the other hand, Staphylococcus aureus (4 [5%]), Enterococcus faecalis (1 [1.3%]), and Enterococcus faecium (1 [1.3%]) were identified as gram-positive bacterial species from recruited patients. The findings of the current study showed that prolong hospitalization is the main risk factor associated with bacterial coinfection and death. Thus, health care providers should minimize hospitalization as well as following a continuous monitoring for bacterial coinfection among COVID-19 patients, to control the spread of infection and reducing the severity and mortality rate among COVID-19 patients.

1. Introduction

Coronavirus disease 2019 (COVID-19), which is caused by the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), has ravaged millions of lives globally due to its severe symptoms and associated complications, especially among immunocompromised patients.^[1] On June 19, 2023, about 87,159,156 cases and 6431,700 deaths due to COVID-19 have been reported worldwide.^[2] Thus, all nations impacted by this viral infection have experienced the aftermath as a consequence. Nevertheless, the Kingdom of Saudi Arabia (KSA) will be the main focus area for this study, which till date has reported about 810,394 incidences and 9255 deaths (June 19, 2023).^[2,3] Victims of this pandemic often experience mild symptoms including fever, cough, fatigue, and loss of smell and taste.^[4] In addition, sore throat, headache, diarrhea or eye irritation or pain were also reported as clinical signs of COVID-19.^[5] However, in severe cases, patients may experience serious chest pain, loss of mobility, and shortness of breath (hypoxia).^[4] Nonetheless, a previous study in Saudi Arabia reported that old age, being a male, the presence of comorbidities, chronic respiratory diseases and cardiovascular illness are independent risk factors that heighten the severity of COVID-19 cases and its associated complications and deaths.^[6] Thus, individuals with underlying medical conditions such as obesity, diabetes, stroke, asthma, and chronic kidney diseases are at a higher risk of being seriously ill (severe symptoms and complications), hospitalized, and dead when infected by SARS-CoV-2.^[7]

Beside underlying diseases, bacterial coinfection among COVID-19 patients is a critical medical concern that increases its severity, risk of complications (e.g., respiratory failure) and mortality rate.^[8] For instance, a previous investigation in the Asir region (Asir Central Hospital, KSA), showed that microbial coinfections among COVID-19 patients doubled their hospitalization time (average = 35.2 days) compared to patients with SARS-CoV-2 only (16.2 days).^[9] The longer hospitalization for COVID-19 patients would certainly impact their health and overall economical burden since admission to intensive care units (ICUs). Additionally, the use of expensive or unsafe drugs and application of serious medical procedures as supportive therapies are all highly expected scenarios in such COVID-19 hospitalized patients,^[10] which is linked with a severe form of the illness, complicated, unsafe or/and expensive therapeutic plans.^[11] Moreover, bacterial coinfection caused by gram-negative bacteria among COVID-19 patients carries an increased risk of being multidrug resistance (MDR) bacterium, which is extremely linked with the extended hospitalization and other abovementioned complications, especially for ICU admitted patients.^[12–14] Multiple bacterial species were identified as co-pathogens among COVID-19 patients, including, Klebsiella pneumoniae and Staphylococcus aureus.^[15]

Although, bacterial coinfection among COVID-19 patients is well-known to reduce immunity of the host, which is linked with long hospitalization (ICU) admission, treatment failure, development of serious complications and death, comprehensive studies discussing the correlation between bacterial coinfections and abovementioned adverse outcomes among COVID-19 are lacking or very limited worldwide. For instance, both health and economical burdens upon Saudi Arabia residents, who were diagnosed with COVID-19 and secondary infections (coinfection) and/or superinfections (infection by MDR) have not been studied yet. These aspects imply clear and critical gaps of research that need to be addressed. Thus, the main focus of this study is to comprehensively and analytically evaluate and elucidate the effects of bacterial coinfections among COVID-19 patients in Jeddah, Saudi Arabia, especially in relation to the health burdens (severity and mortality).

2. Materials and methods

2.1. Study design and data collection

This is a retrospective study that was conducted at a regional COVID-19 hospital in Jeddah, Saudi Arabia, between May and August 2020. Nasopharyngeal swab specimens of suspected patients with COVID-19, underwent a real-time reverse transcriptase polymerase chain reaction (RT-PCR) TaqMan 2019-nCoV Assay Kit v1 (Thermo Fisher Scientific, USA), and only patients with positive RT-PCR results were included in this study. In addition, based on the clinical manifestations, a specimen (e.g., blood, urine, wound, or sputum), was collected from patients with confirmed COVID-19, and was cultured to determine bacterial coinfection caused by MDR bacteria. MDR bacteria were identified as microorganisms that become resistant to 1 antibiotic drug of 3 or more different classes.^[16] Then, COVID-19 patients were categorized into 2 groups based on the results of bacterial coinfection culture as COVID-19 patients with coinfection caused by MDR (COV-MDR) and COVID-19 patients without coinfection (COV). Next, age, gender, date of admission/discharge, patients’ outcome (recovered, died), type of specimen tested for bacterial coinfection and name of isolated MDR bacteria, application of central line or catheters were all retrieved and analyzed from targeted patients.

2.2. Ethical approval

The study was approved by the Jeddah Health Affaires ethics board at the Ministry of Health in Saudi Arabia (H-02-J-002-01285). Consent form was waived as this was a retrospective study. Data was collected anonymously.

2.3. Statistical analysis

Data was coded in Excel version 16.57, and all statistical analyses were done using SPSS version 27. The following variables were collected for the participants, age (continuous variable), gender (male, female), mortality (died, not died), duration (continuous variable), MDR infection (yes, no), bacteria types (Acinetobacter baumannii, Escherichia coli, K pneumoniae, Enterococcus faecalis, Enterococcus faecium, S aureus, Serratia marcescens, Pseudomonas aeruginosa), specimen type (blood, sputum, urine, wound), utilization of central lines/catheters (yes, no). Descriptive statistical analysis (numbers and percentages) was carried out on demographic characteristics. The categories of each variable (present/absent) were compared between patients. Chi-square test was used when all the parameters had >5 expected counts; otherwise, Fisher exact test was used. Independent sample t test was used to compare age and hospitalization period between groups (bacterial coinfection, no bacterial coinfection) when assumptions were met (no outlier and data were normally distributed and homogeneity of variances) otherwise Mann–Whitney U test was used. Binominal logistic regression was used to identify risk factors for increased likelihood of mortality and bacterial coinfection. Statistically significant cutoff was defined at <.05.

3. Results

3.1. Demographics and clinical characteristics

During the study period, 342 nasopharyngeal swabs were tested positive for SARS-CoV-2 (COVID-19) using RT-PCR. Out of the total number (342) of patients confirmed with COVID-19, 266 (77.7%) were male and 76 (22.2%) were female. In addition, around 80 (23.3%; 65 = males [81%] and 15 females [18.8%]) out of the total patients (342) were diagnosed with bacterial coinfection, while the remaining patients 262 (76.6%; 201 = males [76.7%] and 61 females [23.3]) showed negative bacterial culture (Table 1). The average age of recruited patients with and without bacterial coinfections are similar (P = .127), but the hospitalization stay was found to be higher among COVID-19 patients diagnosed with bacterial coinfection (16.01 ± 11.36 days) when compared with patients of the same illness but without bacterial coinfection (6.5 ± 6.12 days; P < .01).

Table 1.

Demographic characteristics of diagnosed patients with coronavirus disease 2019 (COVID-19), with and without bacterial coinfection, collected from a regional COVID-19 hospital in Jeddah, Saudi Arabia.

		MDR coinfection (COV-MDR)	No MDR coinfection (COV)	Total	P-value
Gender	Male	65 (81.3)	201 (76.7)	266	.394
	Female	15 (18.8)	61 (23.3)	76
Mortality	Discharged or transferred	8 (10)	133 (50.8)	141	<.01
	Died	72 (90)	129 (49.2)	201	<.01
Age (yr)		56.15 ± 12.68	53.55 ± 15.02		.127
Hospitalization (d)		16.01 ± 11.36	6.5 ± 6.12		<.01

Date are presented in number and (%) and  ±  indicates the standard deviation.

COVID-19 = coronavirus disease 2019, MDR = multidrug resistance.

Different types of specimens were collected from patients with confirmed COVID-19 cases to identify bacterial coinfection. About 69% of tested blood specimens among COVID1-19 patients have been reported with coinfection, followed sputum, urine and wound (19%, 11%, and 1%, respectively; Table 2). Based on the application of central line or/and urinary catheter patients with positive coinfection (total = 80), were divided into 2 groups as patients with (28.8%) and without (71.2%) central lines/catheters. Additionally, the results of this study showed a high rate of mortality among patients with bacterial coinfection 72 (90%) compared to patients without coinfection 129 (49.2%) which was statistically significant (P < .01; Table 1). COVID-19 patients who tested positive for bacterial coinfections from urine and blood specimens were associated with the use of these medical indwelling devices (Table 2). Calculated mortality rates of COVID-19 patients with positive bacterial coinfection across all tested specimens was highest among patients who gave wound and blood samples followed by others (Table 2).

Table 2.

Types of requested clinical specimens from patients diagnosed with coronavirus disease 2019 (COVID-19) and bacterial coinfections.

Specimen type	Distribution	Gender		Age (average (yr) ± SD)	Duration (average (d) ± SD)	Distribution central line/catheter	Mortality
		Males	Females
Blood	61 (69%)	51 (83.6%)	10 (16.4%)	55.3 ± 12.87	14.95 ± 7.6	(37.7)	56 (91.8%)
Sputum	17 (19%)	14 (82.4%)	3 (17.6%)	52.4 ± 10.58	21.27 ± 19.50	(5.8)	14 (82.4%)
Urine	9 (11%)	8 (88.9%)	1 (11.1%)	61.8 ± 13.73	19.11 ± 9.60	(66.6)	8 (88.9%)
Wound swabs	1 (1%)	0 (0%)	1 (100%)	80	19	(0)	1 (100%)

Date are presented in number and (%) and ±  indicates the standard deviation (SD).

COVID-19 = coronavirus disease 2019, SD = standard deviation.

3.2. Prevalence of MDR coinfection among COVID-19 patients

With respect to the type of bacteria that were identified from COVID-19 patients, generally gram-negative bacteria were predominant including K pneumoniae (35 [43.8%]) and A baumanni (32 [40%]) followed by E coli (5 [6.3%]), S marcescens (1 [1.3%]), and P aeruginosa (1 [1.3%]). On the other hand, S. aureus (4 [5%]), E faecalis (1 [1.3%]), and E faecium (1 [1.3%]) were identified as gram-positive bacterial species from recruited patients (Table 3). K pneumoniae were the predominant isolated bacteria in blood, sputum and urine while E coli was the only bacteria isolated from wound (Table 3). Regression analysis revealed that mortality was positively associated with age, hospital duration and strongly linked with coinfection (Table 4). The model was significant (P <.1), explained 35.4% (Nagelkerke R²) of variation in the mortality and correctly classified 78.1%. The sensitivity is 83.1% and specificity is 70.9%, positive predictive value 80.29% and negative predicted value was 74.63%. Regression analysis revealed that bacterial coinfection COV-MDR was positively associated with mortality and duration. The model was significant (P < .1), explained 38.8% (Nagelkerke R²) of the variation in the mortality and correctly classified 81.6%. The sensitivity is 37.5% and specificity is 95%, positive predictive value 69.77% was and negative predicted value was 83.28% (Table 4).

Table 3.

Distribution of isolated causative bacterial species for coinfection from various clinical specimens among (COVID-19) patients.

	A baumannii	E coli	K pneumoniae	E faecalis	E faecium	S aureus	S marcescens	P aeruginosa
Blood	27 (44.3%)	2 (3.3%)	28 (45.9%)	1 (1.6%)	1 (1.6%)	1 (1.6%)	0 (0%)	1 (1.6%)
Sputum	6 (35.3%)	1 (5.9%)	7 (41.2%)	0 (0%)	0 (0%)	2 (11.8%)	1 (5.9%)	0 (0%)
Urine	1 (11.1%)	2 (22.2%)	6 (66.7%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Wound	0 (0%)	1 (100%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)	0 (0%)

A baumannii = Acinetobacter baumannii, E coli = Escherichia coli, K pneumoniae = Klebsiella pneumoniae, E faecalis = Enterococcus faecalis, E faecium = Enterococcus faecium, S aureus = Staphylococcus aureus, S marcescen = Serratia marcescens, P aeruginosa = Pseudomonas aeruginosa.

COVID-19 = coronavirus disease 2019.

Table 4.

Multiple linear regression analyzing the associations between risk factors and mortality and coinfection among COVID-19 patients.

	β	SE	P-value	95% Confidence interval
				Upper	Lower
Mortality
Age	1.048	0.010	<.001*	1.028	1.069
Gender (male)	1.315	0.316	.387	0.708	2.442
Duration	1.104	0.025	<.001*	1.052	1.159
COV-MDR	4.672	0.433	<.001*	1.999	10.920
Constant	0.034	0.644	<.001*
COV-MDR
Age	0.987	0.012	.256	0.964	1.010
Gender (male)	1.281	0.393	.529	0.593	2.770
Mortality (yes)	5.864	0.471	<.001*	2.330	14.754
Duration	1.155	0.025	<.001*	1.101	1.212
Constant	0.031	0.842	<.001*

β = slope of the regression, COV-MDR = COVID-19 patients with coinfection caused by MDR, COVID-19 = coronavirus disease 2019, MDR = multidrug resistance, SE = standard error.

^*indicate significant change.

4. Discussion

COVID-19 is a serious viral infection that can lead to high mortality rate, especially among patients with incompetent immune system, including patients with chronic diseases (e.g., diabetes).^[17] Dysregulation of immune responses together with increased inflammatory conditions contribute to adverse patient outcomes and patient mortality.^[18] In addition, bacterial coinfections among COVID-19 patients were claimed as a major risk factor causing severe form of SARS-COV-2 infection and high rate of death.^[18,19] Thus, the current study focuses on the impact of different bacterial coinfections among COVID-19 patients. Above all, the current data supports the suggestion that bacterial coinfection is a major risk factor for severe form of COVID-19, serious complications and high mortality rate. Prevalence of bacterial coinfection among COVID-19 patients might be strongly linked with the hospitalization period and the use of medical intervention devices (e.g., ventilators), which are used to treat associated severe symptoms (e.g., shortness of breath) and other clinical complications (e.g., respiratory failure), leading to higher probability of death (90%).^[10] In contrast, a lower rate of mortality (6.1%) was claimed for COVID-19 patients without bacterial coinfection.^[19] The high mortality rate of COVID-19 patients with bacterial coinfections should indicate the urgent need for modern medicine to invest significant amounts of funds into countering the threat in question.^[20] Firstly, the issue is vital since most COVID-19 guidelines focus on traditional case scenarios without considering the possibility of bacterial coinfection. Although clinicians are fully aware of following local up-to-date antibiogram for empirical antibiotic prescription, availability of these data, especially among COVID-19 patients, is very limited and not sufficiently addressed and discussed. Namely, each hospital must establish its own guidelines/recommendation in relation to antibiotic empirical prescription based on the analysis of recent antibiotic surveillance program. Secondly, a significant factor is the possibility of nosocomial coinfections in critically ill patients during COVID-19 related which results in prolonged hospitalization.^[21] To preclude this situation, targeted plans to improve the long-term infection control measures within the relevant facilities are essential.

Another finding that may require scientific interpretation is that the high prevalence of males (77.7%) in recruited patients compared to females. This data is in full accordance with previous report as Mourosi and his group that have claimed male gender as major risk factor among COVID-19 patients.^[22] Studies indicate that males had a higher risk of hospitalization with COVID-19 related complications due to adaptive and innate immune responses to SARS-CoV-2 in males.^[23] Additionally, factors such as higher expression of angiotensin-converting enzyme-2 receptors which is target of the coronavirus in males may also contribute to this trend.^[24] This clear association between male and higher incidence of COVID-19 should be of interest to staff at the infection control unit within hospitals, to develop a modified approach to control dissemination of COVID-19 more effectively. For example, application of strict precaution when treating male cases such as wearing mask, gloves and ensuring that patients follow the guidelines. Nevertheless, as reported previously it could be attributed to the fact that males simply tend not to follow the proper COVID-19 preventive measures (e.g., social distancing and washing hands).^[25] This may be due to cultural views, as males believe that their immunity could be more robust than women.^[26] This unscientific belief should be corrected among people in question by different means, namely through intensive guidance and education by public health and information campaigns.

Current data should gain the attention of researchers and clinicians in relation to the high (23%) incidence of MDR coinfections among COVID-19 patients. This is because previous reports have claimed limited or absence of association between COVIDD-19 and bacterial coinfection, especially among cases reported in 2020.^[10] Although, COVID-19 might be considered less lethal as the overall management approach is currently directed to the treatment of only serious associated complications, the consideration of bacterial coinfections and development of a comprehensive therapeutic plan is still required. Bacterial coinfection among recruited patients was commonly caused by gram-negative bacterial species, including K pneumoniae and A baumannii as evidenced in other similar studies.^[11,27] These bacterial species are frequently linked with multidrug resistance (MDR) and treatment failure. Thus, coinfection caused by K pneumoniae or/and A baumannii would be more serious and challenging due to the higher rate of treatment failure, severe complications and death.^[8,9] Consequently, a comprehensive antibiotic scheme for bacterial coinfection among COVID-19 patients should be presented, especially for MDR bacteria. This is an urgent and critical requirement since recruited patients who tested positive for bacterial coinfection showed higher rates of mortality (90%) as compared to patients without bacterial coinfection (49%). Nevertheless, age and long hospitalization were positively linked to high mortality among COVID-19 patients. High mortality rate in old patients could be attributed to decreased immunity, slow response to treatment, and comorbidities such as cardiovascular disease, hypertension, and diabetes which hamper the recovery and associated with worse outcomes in those patients.^[28]

Although the highlighted correlation between MDR coinfection and various factors among COVID-19 patients, restricted access to the patients’ files resulted in several limitations, including unavailable information regarding the type of other bacteria (non-MDR) and fungi, antibiogram data, and prescription of antibiotics. This information is useful to the analysis of prevalence of bacterial coinfection, bacterial resistance patterns, and prescribed antibiotics.

5. Conclusions

The current investigation indicates that MDR coinfection, old age, and prolong hospitalization are main risk factors associated with mortality in COVID-19 patients. The predominant MDR bacteria recovered from these patients were gram-negative including K pneumoniae, A baumanni, and E coli. The existing infection control measures at hospitals could be a major factor for bacterial-coinfection among COVID-19 patients. Therefore, reevaluation and strict application of approved control measures are needed to decrease the rate of both COVID-19 and bacterial coinfection. In addition, targeted investments and research for novel antibiotics and/or alternative agents/approaches that can facilitate control of the existing resistance of the relevant bacterial species, especially among COVID-19 patients. All abovementioned measures are useful and applicable to COVID-19 pandemic and any possible infectious pandemics, which are inevitable in the near future.

Acknowledgments

We would like to thank our patients for their participation and health care providers who assisted in data collection.

Author contributions

Conceptualization: Abdu Aldarhami, Naif K. Binsaleh, Subuhi Sherwani.

Data curation: Abdu Aldarhami, Ahmed A. Punjabi, Naif K. Binsaleh, Subuhi Sherwani.

Formal analysis: Ahmed A. Punjabi, Abdulrahman S. Bazaid, Omar W. Althomali.

Funding acquisition: Ali A. Almishaal, Abdu Aldarhami, Ahmed A. Punjabi, Abdulrahman S. Bazaid, Naif K. Binsaleh, Subuhi Sherwani.

Investigation: Abdu Aldarhami, Ahmed A. Punjabi, Naif K. Binsaleh, Omar Hafiz.

Methodology: Abdu Aldarhami, Abdulrahman S. Bazaid, Naif K. Binsaleh, Omar Hafiz.

Project administration: Ali A. Almishaal, Abdulrahman S. Bazaid, Naif K. Binsaleh, Subuhi Sherwani.

Resources: Omar W. Althomali, Subuhi Sherwani.

Software: Omar W. Althomali.

Supervision: Subuhi Sherwani.

Validation: Ali A. Almishaal, Abdulrahman S. Bazaid, Naif K. Binsaleh.

Visualization: Ahmed A. Punjabi, Omar Hafiz.

Writing – original draft: Ali A. Almishaal, Abdu Aldarhami, Naif K. Binsaleh, Subuhi Sherwani.