The Epidemiology of Bloodstream Infections and Antimicrobial Susceptibility Patterns: A Nine-Year Retrospective Study at St. Dominic Hospital, Akwatia, Ghana

Background Bloodstream infections are among the top causes of morbidity and mortality in people of all ages, especially in immunocompromised patients in sub-Saharan Africa. This study aimed at describing the epidemiology of bloodstream infections and antimicrobial susceptibility pattern over a nine-year period at St. Dominic Hospital, Akwatia, in the Eastern Region of Ghana. Method This study retrospectively analysed data from 4,489 patients who were referred to the Laboratory Department for blood culture and sensitivity testing from January 2009 to December 2017. Sociodemographic data included age, gender, and patients' department. Blood culture results were retrieved from archival records in the laboratory. The authorities of St. Dominic Hospital granted approval for the study. Results The incidence of bloodstream infection over the 9 years was 51.4 positive cultures per 100,000 hospital attendance. Staphylococcus aureus was the leading causative agent of bacteraemia for the first two scalar years (2009–2011 (38.9%) and 2012–2014 (42.2%)) while coagulase-negative staphylococcus (CoNS) (50.5%) was predominant for the last scalar year (2015–2017), followed by Staphylococcus aureus (169/587 (28.8%)). The highest incidence of bloodstream infections was recorded in the wet seasons (months of May (8.9 per 10,000 persons) and October (10.1 per 10,000 persons)). The bacterial isolates demonstrated high resistance to tetracyclines (390/531 (73.4%)), penicillins (1282/1669 (76.8%)), and sulphonamides (450/499 (90.2%)). Conclusion Bloodstream infection and antimicrobial resistance are high in patients seeking healthcare in Akwatia. This therefore calls for concerted efforts aimed at reducing the incidence in the study area.


Introduction
Bloodstream infections are the leading cause of morbidity and mortality in people of all ages [1], particularly in immunocompromised patients [2]. ese infections are frequent and present life-threatening conditions in hospital settings [3,4]. Globally, bloodstream infection affects about 30 million people leading to 6 million deaths [5], with 3 million newborns and 1.2 million children suffering from sepsis annually [6]. In Eastern African countries, the proportion of patients with bloodstream infection is reported to range from 11% to 28% [1,7,8]. In Ghana, bloodstream infection rates as published in hospital and laboratory surveillance reports are estimated at 9.3% to 11.2% [9,10]. Bloodstream infections are characterized by the presence of viable bacterial or fungal microorganisms in the bloodstream that elicit inflammatory response and often accompanied by alteration of clinical, laboratory, and haemodynamic parameters [11]. ese microorganisms may include Gram-negative bacteria such as Escherichia coli, Pseudomonas aeruginosa, Klebsiella species, Neisseria meningitidis, and Haemophilus influenzae, and Gram-positive bacteria such as coagulase-negative staphylococci (CoNS), Staphylococcus aureus, Streptococcus pneumoniae, Streptococcus pyogenes, Streptococcus agalactiae, and Enterococcus faecium [12]. e incidence of bloodstream infection is attributed to ageing of patients on admission, increasing number of patients with compromised immunity, and the acquisition of virulence factors by bloodstream pathogens [13,14] as well as factors linked to infection prevention and control measures and implementation [15]. Bloodstream infections can be categorized into three groups based on its mode of occurrence: in immunocompetent host with intact defenses, in patients at the extremes of life, and in patients affected by pathological conditions putting them at risk to the infections [11]. e symptoms associated with bloodstream infections include, but are not limited to, fever, chills, reduced vascular tone, low blood pressure, change in mental status, hyperventilation, hypothermia, excessive sweating, and likelihood of organ dysfunction [2]. In the determination of causative agents of bloodstream infection, blood cultures are the method of choice because they are highly sensitive and easier to perform [16,17]. However, this method is not ideal for uncultivable organisms or when antimicrobial treatment has commenced before blood sampling. Although there has been an improvement in public health and medical care in recent times, bacteraemia remains a major cause of sickness and death [18]. ere is also a lack of long-term monitoring of bloodstream infection in children and adult in sub-Saharan Africa [1]. Besides, the epidemiological pattern of the causative agents is not static but constantly changing over time [19], necessitating the need for frequent surveillance among the populace. Due to these reasons, this study aimed at describing the epidemiology and antimicrobial susceptibility patterns of bloodstream bacterial infections over a nine-year period at St. Dominic Hospital, Akwatia, in the Eastern Region of Ghana.

Study Design and Study Site.
is was a retrospective study of 4,489 patients suspected of bloodstream infection over a nine-year period who patronized St. Dominic Hospital from January 2009 to December 2017. e hospital is located in the Denkyembour District of the Eastern Region of Ghana with Akwatia as the district capital. e hospital has a bed capacity of 450 and serves as a referral center for other health facilities located in the district and beyond.

Results
e crude incidence of bacteraemia among the population was 77.7 and 72.2 per 10,000 persons for male and female residents, respectively. e age-standardized bacteraemia rates recorded were 77.3 per 10,000 persons for the male population and 75.1 per 10,000 persons for the female population. After age standardization, the overall case density was highest among children under 5 years irrespective of gender. Decreasing incidence of bacteraemia was recorded with increasing age group among both males and females (Table 1).
Within the years under review, 51.4 cases per every 100,000 were diagnosed of bacteraemia. e incidence over the 9-year period was 574.4 and 2.8 per 100,000 per hospital attendance for inpatients and outpatients, respectively. Although not statistically significant, the rate of bacteraemia was observed to have increased from the first 3 years of the review to the last, and this trend was also seen in both inpatients and outpatients ( Table 2).
We found male gender to be more susceptible to bloodstream infections in the present study.
us, more males (77.7 cases per 10, 000 persons) recorded bacteraemia compared to their female counterparts (72.2 cases per 10,000 persons) ( Table 1). Our results add to a growing body of knowledge where male preponderance to bloodstream infections has been reported in the previous studies [7,23,[28][29][30]. Some reasons proposed to explain the male gender vulnerability include less frequent hand hygiene practice which could potentially provide enabling environment for large reservoirs of common pathogens responsible for causing bloodstream infections [31,32], biological makeup of women where oestrogen suppresses the expression of virulence factors of some microorganisms especially Pseudomonas aeruginosa [31], and the onset of urinary tract infection in men which often goes undetected, providing fertile environment for the organisms to make their way into the bloodstream [33]. e infectious agents responsible for bloodstream infections vary from country to country with unique geographical peculiarities [3,23,34]. e research works of Bouza et al. [25] in Spain, Koupetori et al. [35] in Greece, and Musicha et al. [1] in Malawi attributed bloodstream infections to the predominance of Gram-negative bacteria. However, those results were challenged in the works of Kolonitsiou et al. [36] in Greece, Bassetti et al. [37] in Italy, and Wasihun et al. [8] in Ethiopia, as well as Chiduo et al. [7] in Tanzania who found Gram-positive bacteria to be predominantly responsible for bloodstream infections. In our study, the first two scalar years of the review (2009-2011, 37/95 (38.9%); 2012-2014, 78/185 (42.2%)) revealed Staphylococcus aureus as the leading causative agent of bacteraemia (Figures 1(a) and 1(b)) while coagulase-negative staphylococcus (CoNS) (155/307 (50.5%)) was observed to be predominantly responsible for bacteraemia in the last scalar year (2015-2017), followed by Staphylococcus aureus (54/307 (17.6%)) ( Figure 1(c)). In contrast, however, our results contradict the findings of Labi et al. [20], Obeng-Nkrumah et al. [9], and Opota et al. [17] who reported Escherichia coli as the leading cause for bloodstream infections. Staphylococcal predominance may be due to the rise in methicillin-resistant staphylococcus and catheterrelated infections, whereas Escherichia coli-associated bloodstream infection is usually secondary to hepatobiliary sepsis, abdominal, urinary tract, and surgical tract infections [19].
After age standardization, bacteraemia was highest among children under 5 years irrespective of gender (male vs. female: 50.9 vs. 45.2 per 10,000 persons) compared to the older age groups. e highest age category (<5 years) had the highest incidence (male vs. female: 372.5 vs. 330.9 per 10,000 persons). us, an inverse relationship between age and the incidence of bloodstream infection was established in both male and female subpopulations (Table 1). e finding suggests that children were at a greater risk of acquiring bloodstream infections compared to the older age groups. Postulated factors attributed to higher bloodstream infection rate in young patients particularly neonates include immature immune system, poor Year-on-year trends skin integrity, frequent exposure to healthcare environments, and low socioeconomic status of parents, as well as poor hygiene practices, bottle feeding, and high incidence of delivery at home [12,[38][39][40]. According to Newman [41], hospital environments abound with lots of nosocomial organisms and this situation could render neonates more susceptible to bloodstream infections. Seasonal variation is an important determinant of bloodstream infection burden [42,43]. In Ghana, the months of December-January and May and October coincide with the dry and wet seasons, respectively [44]. In the current study, the epidemiological pattern of bloodstream infection based on the crude incidence revealed an upward trend during the rainy season and a dip in the dry season, with the highest bacteraemia cases recorded in the months of May (8.9 per 10,000 persons) and October (10.1 per 10,000 persons) (Figure 3(a)). ough it was previously reported that higher rates of bloodstream infection directly correlate with increasing temperature [43,45], our finding seems to contradict those reports. A plausible reason may be due to climatic changes that appear to underlie individuals' susceptibility to bacterial pathogens.
Microbial agents that are associated with bloodstream infections may be modified via antibiotic administration and other factors specific to the patients such as surgical procedures, trauma, or underlying conditions, or by the quality of specimen collection, transport, and culture [46]. Reports of antimicrobial susceptibility patterns of bacterial isolates in body fluids are well documented in the literature worldwide and in the African settings but with varying outcomes [46,47]. In the present study, the bacterial isolates demonstrated varying resistance patterns to    (Tables 3 and 4). Similar results of high resistance to ampicillin (94.4%), cefuroxime (79.0%), and cefotaxime (71.3%) were reported by Agyepong and his colleagues in Ashanti Region [26] and in other parts of the country [48,49]. e high level of bacterial resistance to antimicrobials in sub-Saharan Africa, including Ghana, is believed to be due to easy availability; thus, they are routinely used in many healthcare facilities and low cost [26]. Moreover, the increasing resistance to the secondand third-generation cephalosporins including cefuroxime and cefotaxime is shown to be associated with the expression or production of extended-spectrum betalactamase enzyme [50].
We have identified some limitations worth mentioning in the present study that could potentially influence the interpretations of our findings. Data were retrospectively obtained, and this did not allow for random selection of  cases. e study fell short of describing factors that may be associated with bloodstream infections due to unavailability of information, hence limiting our understanding of potential risk factors. Although a considerable number of staphylococcus isolates were included in this study, no data on methicillin resistance test were available. e study was also based on a single-site analysis; hence, findings cannot be generalized. Potentially, changes in hygiene and infection prevention and control practices among the healthcare givers over the period could also affect the findings of this study.

Conclusion
Bloodstream infection and antimicrobial resistance are high in patients seeking healthcare in Akwatia. is therefore calls for concerted efforts aimed at reducing the burden in the study area.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they do not have any conflicts of interest.