Multiresistant Bacterial Pathogens Causing Bacterial Pneumonia and Analyses of Potential Risk Factors from Northeast Ethiopia

Background Pneumonia is the most common cause of morbidity and mortality in developing countries, mostly caused by different species of bacterial pathogens. Hence, patient management needs awareness of the pathogens and antimicrobial susceptibility testing (AST). This study was aimed to assess the type of bacterial isolates and their antimicrobial susceptibility patterns among pneumonia suspected patients at Dessie Referral Hospital, Northeast Ethiopia. Potential risk factors were also assessed to apply preventive measures accordingly. Materials and Methods A cross-sectional study design was employed among pneumonia suspected patients from February to April 2020 at Dessie Referral Hospital. Sociodemographic characteristics and associated risk factors were collected using a pretested questionnaire, and clinical data were extracted by reviewing medical records. Sputum specimens were collected and inoculated into chocolate agar, blood agar, mannitol salt agar, and MacConkey agar which are then incubated at 35°C or 37°C for 24–48 hours. Bacterial species were identified based on Gram stain, colony characteristics, and biochemical techniques. The data were entered in to Epi-Info version 7.1.5 and analyzed with SPSS software version 20. p value <0.05 at 95% CI was considered as statistically significant. Results A total of 406 sputum specimens were collected and cultured, among which 157 (38.7%) were positive for different bacterial pathogens. The predominant pathogens were Klebsiella pneumoniae (28.0%), Streptococcus pneumoniae (24.8%), Staphylococcus aureus (18.5%), and Pseudomonas aeruginosa (14.0%). Majority of the isolates exhibited resistance to ampicillin with 81.5% followed by penicillin with 75.9% and amoxicillin-clavulanate with 61.2%. Multivariable logistic regression showed a significant association of culture positivity with older age (AOR = 2.43, CI: 1.12–5.28, p value = 0.025), cigarette smoking (AOR = 4.67, CI: 2.39–9.20, p value <0.001), and alcohol use (AOR = 5.58, CI: 3.14–9.92, p value <0.001). Resistance to ampicillin and penicillin was associated with repeated prescription and use. Conclusions This study found high prevalence of bacterial pneumonia in the study area, and high rate of bacterial resistance was observed in ampicillin, penicillin, and amoxicillin-clavulanate. Repeated prescriptions and use of antimicrobials were significantly independent factors of bacterial resistance. Therefore, patient management needs identification of bacteria by routine culture with antimicrobial susceptibility testing.


Introduction
Pneumonia, supported with a clinical and/or radiological consolidation of the lungs, is an inflammation (acute or chronic) of the lung parenchyma [1]. is inflammation is produced mostly by microorganisms, particularly bacteria [2,3]. e extracellular bacteria responsible for pneumonia include S. pneumoniae, H. influenzae, and Staphylococcus aureus, while Pseudomonas aeruginosa and other Gramnegative bacilli are infrequent causes of the disease. Moreover, the "atypical" intracellular bacteria responsible for this disease, without being cultured by routine culture methods, include Mycoplasma pneumoniae, Chlamydophila pneumoniae, and Legionella pneumophila [4]. Pneumonia can be transmitted by different ways, inhalation of droplets (e.g., C. pneumoniae and M. pneumoniae), environmentally In developing countries, bacterial pneumonia is treated usually empirically; by medical history and physical examinations, the etiologic agent is rarely identified [35] and results in high prevalence of multidrug-resistant (MDR) pathogens [7]. Hence, identifying the most common bacterial pathogens and their antimicrobial susceptibility patterns timely and accurately is key to reduce morbidity and mortality due to the disease [36], as the empiric treatment is based on severity of pneumonia, the prevalence, and local antimicrobial resistance patterns [22,37,38].
erefore, awareness about the etiology of pneumonia is fundamental for appropriate patient management. In Northeast Ethiopia, however, there was scarce of studies showing the real burden of bacterial pneumonia among suspected patients. In addition, there are scarce of studies about antimicrobial susceptibility patterns of the common bacterial agents for pneumonia. erefore, this study aimed to assess the prevalence of bacterial pathogens, their antimicrobial susceptibility patterns, and associated risk factors among patients suspected for bacterial pneumonia attending Dessie Referral Hospital, Northeast Ethiopia.

Study Design and Setting.
A hospital-based crosssectional study was conducted from February to April 2020 in Dessie Referral Hospital, Northeast Ethiopia. e hospital had around 500 beds with annual ambulatory cases of around 14400; hence, it was purposefully selected to conduct this study. All pneumonia suspected patients visiting the hospital were used as the source population, and all patients aged ≥5 years who were clinically suspected for bacterial pneumonia were included in the study. Patients who were under antimicrobial treatment within the last 14 days during data collection were excluded from the study. Pneumonia prevalence of 40.3% from the previous study was used to determine the number of study participants [1].

Sample Size and Sampling Technique.
A total of 406 study participants were proposed and systematically recruited.

Data Collection.
Data related to sociodemography and risk factors for bacterial pneumonia were collected by pretested and structured questionnaire through face-to-face interview. Gene Xpert results, initial diagnosis, and nutritional status of study participants were collected by reviewing medical records. Pneumonia suspected patients were clinically selected based on chest pain, shortness of breath, cough with sputum production, fever, night sweats, shaking chills, hemoptysis, and altered mental state (confusion). Sputum specimens were collected in a sterile, disposable, leak proof, and wide-mouthed container with tight-fitting lid. To reduce the number of commensals, the purulent parts of the sputum specimens were washed in about 5 ml of sterile physiological saline. To keep pathogens (such as S. pneumoniae and H. influenzae) alive, the washed sputum specimens were inserted to a cotton-wool swab which then inserted in containers of Amies transport medium; then, all specimens were transported with cold box to Amhara Public Health Institute (APHI), Dessie Branch Bacteriology Laboratory for culture and antimicrobial susceptibility testing.

Bacterial Isolation and Identification.
All specimens received for culture were evaluated macroscopically followed by microscopic inspection by Gram stain before culture analysis began.
us, sputum specimens with at least 25 polymorph-nuclear leukocytes and <10 epithelial cells per low power field and >10 bacteria per high-powered field were processed for culture [39][40][41].
e sputum specimens appropriate for culture were inoculated into blood agar plate (BAP), MacConkey agar plate (MAC), mannitol salt agar (MSA), and chocolate agar plate (CHO). Subsequently, BAP, MAC, and MSA were incubated at 37°C for 18-24 hours, while CHO (in humid, 5% CO 2 atmosphere) was incubated for 18-24 hours at 35°C-37°C. All the plates were examined for growth after 24 hours; the plates without growth were further incubated for up to 48 hrs. e colonies were subcultured on BAP and MAC for further identification. Bacterial species were identified based on Gram stain, colony characteristics (such as size, shape, pigmentation, and color) zones of hemolysis, and other biochemical characteristics. Streptococcus pneumoniae was identified by catalase and optochin (5 μg) sensitivity tests, while S. aureus isolates were confirmed by catalase, coagulase, and the mannitol fermentation tests. Chocolate agar, enriched with factor V (NAD) and factor X (hemi), was used to enhance the growth of H. influenzae. Gram-negative isolates were inoculated onto different biochemical tests such as motility, indole, urea, lysine decarboxylase, triple sugar iron agar, and citrate utilization tests for identification [42].
A young culture growth of bacterial suspensions was prepared by picking parts of similar colonies with a sterile wire loop in which these suspensions were adjusted to McFarland 0.5 turbidity standard. e bacterial suspensions in a sterile broth were incubated up to 2 hours to allow the bacteria to reach their log-phase in growth. en, inoculums were swabbed on to Muller-Hinton agar. After drying the agar for 3-5 minutes, the antimicrobial impregnated disks were placed with sterile forceps on the agar surface in such a way that each disk was placed at least 24 mm away from each other to avoid the overlapping zone of inhibition. After placing the discs, the plates were allowed to stand for 30 minutes to help the antimicrobial to be dissolved in the media. Following inverting and incubating for 24 hours at 37°C, the plates were read for the diameter of zone of inhibition. e susceptibility patterns were graded as sensitive, intermediate, and resistant. Muller-Hinton agar (MHA) supplied with 5% sheep blood was used for S. pneumoniae, while Muller-Hinton agar containing 1.0% hemoglobin and 1.0% IsoVitaleX supplement (CHOC-MHA) was used for H. influenzae [42].

Laboratory Quality Control.
Manufacturer instructions and bacteriological standard procedures were followed strictly throughout the whole technical processes including culture media preparation, inoculation, and AST testing. e sterility of culture media was checked by incubating 5% of the batch at 35-37°C overnight and was evaluated for possible contamination. e standard reference bacterial strains such as S. aureus (ATCC ® 25923), H. influenzae (ATCC ® 49247), E. coli (ATCC ® 35218), and S. pneumoniae (ATCC ® 49619) were used as a quality control [43].

Data Quality Control.
Training was given for the data collector about data collection procedures and interview techniques. To assure the quality of the data, a pretested, structured questionnaire was used for data collection. e questionnaire was objective-based and logically sequenced. It was checked daily by the principal investigator for its completeness.

Data
Analysis. Data were checked for completeness, cleaned, coded, and entered in to Epi-Info version 7.1 software and then exported to SPSS version 20 for analysis. Frequency, proportions, and summary statistics were used to describe study participants in relation to relevant variables. Bivariate and multivariate logistic regression analyses were carried out to identify the association between bacterial pathogens, antimicrobial resistance, and possible risk factors. Odds ratio and p value were used to assess the presence and degree of association. p value <0.05 at 95% CI was considered as statistically significant.

Ethical Approval and Consent to Participate.
Ethical clearance was obtained from University of Gondar, College of Medicine and Health Sciences Institutional Review Board. Informed written consent was obtained from the study participants after explaining the purpose and objective of the study. e laboratory results from the study participants were communicated to their physicians for appropriate patient management. Data from all patients were kept confidential.

Limitations of the Study.
is study did not consider the "atypical" intracellular bacteria, such as Mycoplasma pneumoniae, Chlamydophila pneumoniae, and Legionella International Journal of Microbiology pneumophila, since they cannot be cultured by routine culture methods. is study did not also consider the anaerobic bacteria (Prevotella spp., Fusobacterium spp., and Clostridium spp.) as routine culture methods are mostly aerobic. Hence, it underestimates the actual prevalence in the study area.
is study did not include serotyping techniques to Haemophilus influenzae and Streptococcus pneumoniae. Characterization of methicillin-resistant Staphylococcus aureus (MRSA) was not performed.

Sociodemographic and Clinical Data Characteristics.
is study was conducted among 406 pneumonia suspected patients, of which 221 (54.4%) were males and 249 (61.3%) were urban dwellers with 158 (38.9%) participants unable to read and write (Table 1). e median age of study participants was 45.0 with a range of 10-95 years. Among study participants, 131 (32.3%) were smokers and 158 (38.9%) were alcohol consumers.
Among all participants, 43 (10.6%) were HIV positives and 39 (9.6%) had active TB cases at the time of data collection. As indicated in Table 1, asthma, diabetes, and hypertension comorbidities accounted 6%, 5%, and 3%, respectively. e nutritional status of children (n � 30) was assessed by mid-upper arm circumferences (MUAC) and body mass index (BMI), all of which (n � 30, 100%) showed normal nutritional status.  (Table 2). Additionally, all study participants were screened for TB, by Gene Xpert and Mycobacterium tuberculosis was detected in 39 (9.6%) of the participants. Among TB-positive participants, rifampicin-resistant was detected in 5 (12.8%) of the cases.
International Journal of Microbiology

Discussion
e increased bacterial infection causing pneumonia and antimicrobial resistance become a serious public health concern. In our study, the overall prevalence of bacterial isolates was 38.7%, consistent with studies from Ethiopia reported by Temesgen et al. [1], Adhanom et al. [12], Regasa et al. [31,32], and findings from Nigeria reported by Salami et al. [44]. However, the finding in this study was higher than the findings from Brazil, China, and the USA reported by Assunção et al. [5], Lin et al. [45], and Carugati et al. [46], respectively, while it was lower than the findings in other parts of Ethiopia and elsewhere [10,21,33,34,[47][48][49]. e inconsistency might be explained as real prevalence variations or methodological differences.
Although pneumonia can be caused by a variety of bacterial species, some bacteria are frequent causes due to a variety of reasons. In the present study, the predominant bacterial isolates were Klebsiella pneumoniae (28.0%) and Streptococcus pneumoniae (24.8%). ese two pathogens were reported as predominant in different studies [1,10,12,50].
is predominance may be due to their capsular nature and the emergence of strains from both species that can acquire additional genetic traits [51,52]. e prevalence of Staphylococcus aureus and methicillin-resistant Staphylococcus aureus (MRSA) in our study was 18.5% and 34.5%, respectively.
is indicates that MRSA is becoming an important pneumonia-causing pathogen in the study area and supported by other studies [12,53,54].
In line with previously documented results, our finding revealed an increased prevalence of drug-resistant isolates that could possess a significant health risk. Klebsiella pneumoniae was resistant and sensitive to tetracycline and ciprofloxacin in 95.5% and 97.7%, respectively. It was comparable to the study conducted in Ethiopia by Temesgen et al. with 100% resistance and 96.7% sensitivity [1]. Moreover, the sensitivity of this pathogen to ciprofloxacin was comparable with a study conducted in China (91.7%) [47]. Streptococcus pneumoniae, the second most frequent species in our study, showed resistant to oxacillin in 56.4%, while only 10.3% of the isolates were resistant to clindamycin.
is was comparable with studies conducted in Ethiopia [1,34]. Moreover, Staphylococcus aureus was resistant to penicillin in 75.9%. Other studies reported comparable findings [1,5,12]. Pseudomonas aeruginosa was resistant to gentamicin in 59.1%, compared with other studies [1,3,5,10,34]. e possible explanation of drug resistance variations might be due to difference in distribution of resistance strains in different localities. e decreasing susceptibility might be due to increasing trend of using antibiotics. e overall prevalence of multidrug-resistant (MDR) isolates in this study was 63.1% which was in agreement with findings reported from Ethiopia by Regasa et al. (56.7%) [34] and 54.8% [32]. In contrast, a study reported by Temesgen et al. identified higher prevalence (76.0%) [1], and lower results were reported from Mekelle, Ethiopia, by Adhanom et al. (17.9%) [12] and from China by Luan et al. (24.5%) [3] which could be due to different reasons [1,12] among which poor drug quality, antibiotic prescribing differences such as misuse and/or incomplete treatment courses of antibiotics, overprescription due to a poor diagnostic set-up, and irrational drug use can be mentioned. In support of this, in this study, almost all cases (99%) were due to communityacquired pneumonia (CAP) and only 1% of them were due to hospital-acquired pneumonia (HAP).
Several studies reported aging as a risk factor for bacterial pneumonia. In the present study, the age group of >64 years was 2.4 times more likely to have bacterial pneumonia compared to the age group of 5-15 years [55]. Similar findings were reported from Spain [56,57], Pakistan [49], Japan [58], and the USA [59]. e decline of the immune status in the older age may be the possible reason. On the other hand, other studies reported young age as a risk factor  [12] and Ghana [60]. is suggests that occurrence of high transmission due to crowded-living or presence of undernourishment weakens the immune system of the young population. In this study, smoking increases the risk of bacterial pneumonia 4.7 times compared to those who were nonsmokers. Likewise, cigarette smoking, as a risk factor for pneumonia, was reported from Kenya [61] and Spain [55,57]. is may be due to the fact that smoking decreases the number and, at the same time, the action of cilia facilitating the entry of microorganisms to the respiratory tract [61,62].
Several studies also showed that alcohol consumption increases the risk of bacterial pneumonia. In our study, alcohol consumers (although this study did not identify the level of consumption) were 5.6 times more likely to have bacterial pneumonia compared to those who were nonconsumers. Similarly, other studies conducted in Ethiopia [1,12], China [47], Spain [57], England [63], and Europe [62] identified alcohol consumption as a risk factor for acquiring bacterial pneumonia. is may be due to the fact that the sedative properties of alcohol minimize oropharyngeal tone that results in a high risk of aspiration of pathogens from the upper respiratory tract. Moreover, high levels of alcohol consumption can alter the alveolar macrophage function, hence withdrawing pulmonary defense against infection. Alcohol depresses cough, decreases endothelial adherence, lowers chemotaxis, and suppresses B cell and T cell spreading out which contributes to reduced clearance mechanism of lung cells [1,12,62,63].

Conclusions
e overall prevalence of bacterial isolates, in this study, was 38.7%. is high prevalence needs expanding routine bacterial culture and antimicrobial susceptibility testing in the study area. e predominant isolates were Klebsiella pneumoniae (28.0%) and Streptococcus pneumoniae (24.8%).
is indicates that there is an urgent need of strengthening vaccination practices for Streptococcus pneumoniae. e prevalence of multidrug-resistant (MDR) pathogens was 63.1%. e predominant isolate, Klebsiella pneumonia, was highly resistant to tetracycline in 95.5%, followed by penicillin and amoxicillin/clavulanic acid in 81.5% and 75.9%, respectively. ese high figures, in general, recommend avoiding misuse, incomplete treatment courses, overprescription, and irrational use of antibiotics. Aging, cigarette smoking, and alcohol use were factors associated with culture positivity. erefore, preventive measures to minimize the risk of the disease should include life-style factors such as smoking and alcohol use. Moreover, strengthening regular surveillance systems are essential for assessing predominant pathogens and antibiotic resistance patterns.