Background

Staphylococcus aureus (S. aureus) is well adapted to various environments due to their metabolic versatility and pharmic resistance ability. S. aureus colonize the skin and nasopharyngeal membranes as normal microbiota in healthy individuals [1]. However, they cause myriad of detrimental infections when they invade the internal tissues or enter the bloodstream. S. aureus is an important pathogen involved in both hospital-acquired and community-acquired infections and causes many infectious diseases ranging from mild skin and soft tissue infections, bones and joint infections, infective endocarditis, cardiovascular disorders, osteomyelitis, bacteremia, and fatal pneumonia in both healthy and individuals with underlying diseases [2]. The high incidence of both community and nosocomial staphylococcal infections coincide with the emergence of multidrug resistant S. aureus which renders antibiotic treatments ineffective [3].

S. aureus has become resistant to various antibiotics over the past years especially to the beta-lactam class of antibiotics [4]. Emergence of methicillin resistant S. aureus (MRSA) and vancomycin resistant S. aureus (VRSA) constitutes a serious global public health problem. Currently, VRSA and MRSA strains are classified as very potent and dangerous agents that can potentially cause devastating damage worldwide in the absence of effective treatment options [5].

Various mechanisms of resistance utilized by S. aureus include: production of beta-lactamase enzymes to deactivate beta-lactam antibiotics, efflux pump for extruding antibiotics such as tetracyclines [6], reduced accumulation of macrolides antibiotics [7], production of aminoglycosides modifying enzymes to inactivate aminoglycoside antibiotics, alteration of DNA gyrase and topoisomerase IV expression of floroquinolones antibiotics, and expression of Mec genes which alters penicillin binding proteins [8].

In Nigeria, the prevalence of multi-drug resistant pathogens continue to be on the increase due to several factors such as drug misuse, self medication, lack of trained medical personnel, and poverty. As the world battles the persistent rise in antimicrobial resistance (AMR), it is pertinent that adequate data and information about AMR is known which can serve as the basic foundation for setting out effective interventions to contain the crisis of AMR. From the literature, no prior meta-analysis has been done on S. aureus resistance to different antibiotics routinely use in Nigeria. Due to the various infections caused by S. aureus, it is pertinent to determine the pooled prevalence of resistance of S. aureus to various routinely used antibiotics in Nigeria. This will help in improving treatment options and enlighten the populace on the menace and the possible cause of treatment failures due to the increasing rise of multidrug resistant strains. The aim of this meta-analysis was to determine the pooled prevalence of S. aureus resistance to various routinely used antibiotics in Nigeria.

Methods

Study design

Meta-analysis was adopted to evaluate the prevalence of S. aureus of resistance to various antibiotics in Nigeria using the appropriate studies that rely solely on S. aureus from the title. The prevalence of resistance of S. aureus to various routinely used antibiotics in Nigeria is a country wide study as it covers studies from the six geo-political regions of Nigeria. Meta-analysis was adopted because it is a quantitative study of pooled prevalence of resistance of S. aureus to routinely use antibiotics in Nigeria.

Search strategy

Electronic search engines including Google scholar, PubMed, ScienceDirect, and African Journal Online (AJOL) were used to search for available studies from 23rd March to May 2022. Relevant key words such as Staphylococcus, antibiotic resistance, antibacterial resistance, antimicrobial resistance, drug resistance, drug susceptibility, Nigeria were used during the search. These key words were used in different combinations (Staphylococcus OR S. aureus AND antibiotic resistance OR antibacterial resistance OR antimicrobial resistance OR drug resistance AND Nigeria) in various electronic databases using the Boolean operators. The reference lists of included articles were also check to identify studies relevant to the current study.

Inclusion and exclusion criteria

The titles of search results of all retrieved articles were screened independently by two authors with the aim of including studies that address the research question. The articles were inserted into Zotero version 5.0.95.1 referencing application which helped in detecting duplicate articles. The title of the study which solely focused on prevalence of antimicrobial resistance of S. aureus was grouped as eligible for inclusion. S. aureus resistance in any state in Nigeria and studies only done in Nigeria represented in the title is the first criteria for inclusion. However, studies that focused on many microbial strains antimicrobial resistance were excluded.

In general, retrieved studies selected from predefined criteria were screened further using the inclusion criteria: studies that were research articles and used cross sectional design, studies that used human samples, studies that conducted antimicrobial susceptibility tests using the Clinical Laboratory Standard Institute (CLSI) guidelines, studies written in English language and studies with full text.

Exclusion criteria in this meta-analysis include: studies conducted on non-human samples, studies with isolates below 20, duplicate studies, studies that did not conduct antimicrobial susceptibility tests using the Clinical Laboratory Standard Institute (CLSI) guidelines studies not written in English, and review articles.

Data extraction

Relevant data such as name of author (s) and publication year, study design, study place, clinical sample size, isolate source, total number of Staphylococcus aureus isolates tested in each research article, and total No. of isolates resistant each antibiotics. In situations where the proportion of susceptible isolates was reported, then the No. of resistant Staphylococcus aureus isolates was calculated by subtracting the percentage susceptibility from 100 and then dividing the result by 100 and multiplying to the total number of isolates. However, in situation where the proportion of the resistant isolates was given, then the No. of resistant Staphylococcus aureus isolates was calculated by dividing the proportion of the resistant isolates by 100 and multiply with the total number of isolates. The formula is given as thus:

$${\text{Prevalence}}\;{\text{of}}\;{\text{resistance}}\left( \% \right) = \frac{{number\;of\;resistant\;isolates}}{{total\;number\;of\;isolates}} \times 100$$
(1)

To ascertain the reporting of all relevant information in this meta-analysis, we followed the Preferred reporting Items for Systematic Review and Meta-analysis (PRISMA) [9] (Additional file 1: S1) guidelines.

Statistical analysis procedures

In this meta-analysis, statistical analyses were performed using MedCalc statistical software version 20.0.1. The pooled prevalence of antibiotic resistance of S. aureus was evaluated using the meta-analysis proportion command in MedCalc. A total of 23 separate meta-analyses were carried out to evaluate the pooled prevalence of S. aureus resistance to 23 different antibiotics. Between 6 and 77 studies were included in the 23 different meta-analyses. I2 statistic command in MedCalc was used to evaluate the heterogeneity among the included studies. Random effect and fixed effect are two models used to estimate pooled prevalence in meta-analysis. In this study, due to the characteristically high heterogeneity of the included studies, the random effect model was used for meta-analysis at 95% CIs. Egger test was employed for assessing the presence of publication bias [10].

The Freeman-Tukey double arcsine transformation was used to ensure studies which report proportions near or at 0 and 1 were not being excluded. In addition, studies that report unusually high prevalence of resistance when compared to others, a sensitivity analysis was perform by removing the studies. If the point estimate of pooled prevalence after removing a study that reported unusually high prevalence of resistance lies within the 95% CI of the overall pooled estimate for all studies combined, the study is considered as having no significant influence on the overall estimate and vice versa.

Results

Characteristics of included studies

Studies search record from electronic databases yielded 40, 682 of which 35, 400, 2, 180, 1,706, and 1396 were from Google scholar, AJOL, PubMed, and Science Direct, respectively. Articles from Google Scholar gave 35,400 results comprising of many studies irrelevant or that does not fit to the study aim; hence, they were screened randomly from titles alone. Screening of the titles reduced the number of eligible articles to 134 for full text assessment. After going through the full texts, 36 articles were excluded (reported small number of isolates and isolates not from human samples). Thus, 98 studies met the inclusion criteria of the study (Fig. 1).

Fig. 1
figure 1

PRISMA flowchart for the selection and screening of eligible studies

Full size image

About 46, 640 S. aureus isolates were tested against different antibiotics and 23,048 isolates were resistant to various antibiotics. Isolates sources include: nasal, blood, vaginal, ear, wound, urine, throat, pimples, hand, and mixed samples were collected from both symptomatic patients [61] and asymptomatic people [37]. Eighty six studies used primary data while twelve used records from hospitals. The characteristics of each study included is summarized Table 1.

Table 1 Characteristics of included studies
Full size table

Heterogeneity survey and publication bias

The included studies were conducted in the six geo-political zones of Nigeria; a total of 98 studies comprising of 26 from South South, 23 South West, 20 South East, 18 North West, 8 North Central and 3 North East. Quality assessment (risk of bias) was done in line with the following criteria: studies which used CLSI guideline for antibiotic resistant assessment, studies that used more than 20 S. aureus isolates and studies that used adequate sample representative of the region where testing was done. Agar diffusion based method was used to determine the resistance level of S. aureus isolates in all included studies. High heterogeneity was observed for each of the meta-analyses performed with I2 ranging from 79.36 to 98.60%; at p-values ≤ 0.01). This is due to vast difference in sample sizes; some studies used 20 isolates while some used 400 isolates which impacted on the resistance profile of each antibiotic. Also, number of clinical samples and recovered S. aureus isolates differ in all studies and these disparities resulted in high heterogeneity. More studies were conducted in the Southern (South South, South West, and South East) part of Nigeria giving rise to high heterogeneity. Studies were done in different hospitals within these regions with different prevalence estimates. Random sampling was used in most of the studies and different clinical samples were collected. More than one clinical sample per patient was collected in 51 studies while one clinical sample was collected per patient in 47 studies. Egger’s test for a regression intercept gave a p-value range of 0.06 to 0.99, indicating no evidence of publication bias (Additional file 2: S2) following Eggers’ test rule which state that ‘P-value less than 0.05 indicates the presence of publication bias’.

Prevalence of S. aureus resistance to different antimicrobial agents

In this meta-analysis, the pooled prevalence of S. aureus resistance to twenty-three different antibiotics and the number of studies included in each meta-analysis is summarized in Table 2. Prevalence of resistance of S. aureus to each antibiotic based on pharmacological classification is given below for antibiotics routinely used in Nigeria.

Table 2 Pooled prevalence of S. aureus resistance to different antibiotics in Nigeria
Full size table

Prevalence of resistance S. aureus to rifamycins (rifampicins)

Seven studies involving the prevalence of resistance to rifampicin was analyzed. The pooled prevalence of resistance of S. aureus to rifampicin in Nigeria is 24% (95% confidence interval [CI] 6%, 48%). The forest plot (rifampicin) is presented in Fig. 2.

Fig. 2
figure 2

Forest plot of the prevalence of S. aureus resistance to rifampicin

Full size image

Prevalence of resistance of S. aureus to glycopeptides (vancomycin)

The pooled prevalence of S. aureus resistance to vancomycin is 13% (95% CI 7%, 21%) and the forest plot is presented in Fig. 3. Sensitivity results after exclusion of four studies [20, 22, 27, 36] that reported high prevalence of S. aureus resistant to vancomycin is 7% (95% CI 3.3%, 12%). Hence, there was significant decrease in poled prevalence.

Fig. 3
figure 3

Forest plot of the prevalence of S. aureus resistance to vancomycin

Full size image

Prevalence of resistance of S. aureus to beta-lactams antibiotics

Estimation of the pooled prevalence of S. aureus resistance to penicillin antibiotics (penicillin G, methicillin, amoxicillin, cloxacillin, ampicillin, and amoxacilin/caluvanic acid are here presented. Resistance to penicillin G, amoxicillin, cloxacillin, ampicillin, and augmentin were estimated based on 15, 40, 22, 28 and 20 studies respectively. Pooled prevalence resistance rates were highest in penicillin G at 82% (95% CI 61%, 96%). Resistance to cloxacillin [77% (95% CI 64%, 88%)], to amoxicillin [74% (95% CI 66%, 81%)], to ampicillin [68% (95% CI 53%, 81%)] and to amoxacilin/caluvanic [62% (95% CI 50%, 73%)]. However, resistance rate was moderate for methicillin [46% (95% CI 37%, 56%)]. Forest plots for antibiotics (methicillin and penicillin G) resistance are shown in Fig. 4 and 5, respectively while the forest plots for amoxicillin, ampicillin, amoxicillin/clavulanic acid and cloxacillin resistance are presented in Additional file 3: S3, Additional file 4: S4, Additional file 5: S5 and Additional file 6: S6 respectively.

Fig. 4
figure 4

Forest plot of the prevalence of S. aureus resistance to methicillin

Full size image
Fig. 5
figure 5

Forest plot of the prevalence of S. aureus resistance to penicillin G

Full size image

Higher prevalence of resistance among cephalosporin antibiotic was observed in cefuroxime 69% (95% CI 51%, 85%) followed by ceftazidime 61% (95% CI 46%, 75%). Resistance to ceftriaxone is 44% (95% CI 34%, 54%) and to cefoxitine is 43% (95% CI 31%, 546%). The forest plot for ceftriaxone resistance is presented in Fig. 6 while the forest plots for cefuroxime and cefoxitine resistance are presented respectively in Additional file 7: S7 and Additional file 8: S8.

Fig. 6
figure 6

Forest plot of the prevalence of S. aureus resistance to ceftriaxone

Full size image

Prevalence of resistance of S. aureus to floroquinolones

Three antibiotics (ciprofloxacin, ofloxacin, and norfloxacilin) from floroquinolones were included in the study. For ciprofloxacin, 44 studies were used to estimate the pooled resistance, 25 were used for ofloxacin and 9 studies were used for norfloxacilin. The pooled prevalence of resistance of S. aureus to ciprofloxacin [31% (95% CI 24%, 38%)], ofloxacin [24% (95% CI 18%, 31%)], and to norfloxacillin [33% (95% CI 17%, 52%)]. The forest plot for ofloxacin resistance is presented in Fig. 7 while the forest plot for ciprofloxacin and norfloxacilin included in Additional file 9: S9 and Additional file 10: S10.

Fig. 7
figure 7

Forest plot of the prevalence of S. aureus resistance to ofloxacin

Full size image

Prevalence of resistance of S. aureus to protein synthesis inhibitors

Tetracycline a reversible protein synthesis inhibitor showed the highest resistance rate [65% 995% CI 56%, 76%)] followed by erythromycin (macrolides) [47% (95% CI 40%, 53%)] and chloramphenicol [47% (95% CI 37%, 56%)], respectively. Aminoglycosides (gentamycin and streptomycin) and lincosamides (clindamycin) showed relatively lower level of resistance. The pooled prevalence of resistance to streptomycin [45% (95% CI 34%, 57%)], to clindamycin [35% (95% CI 23%, 49%)] and to gentamycin [31% (95% CI 25%, 37%)]. The forest plot for chloramphenicol resistance is presented in Fig. 8 while the forest plots for tetracycline, erythromycin, gentamycin, streptomycin, and clindamycin resistance are presented in Additional file 11: S11, Additional file 12: S12, Additional file 13: S13, Additional file 14: S14, and Additional file 15: S15 respectively.

Fig. 8
figure 8

Forest plot of the prevalence of S. aureus resistance to chloramphenicol

Full size image

Prevalence of resistance of S. aureus to antimetabolites

High resistance was observed among the antimetabolites antibiotics. Pooled prevalence of S. aureus resistance to cotrimoxazole was found to be 66% (95% CI 55%, 76%) and to trimethoprim is 55% (95% CI 35%, 74%). The forest plot for cotrimoxazole resistance is presented in Fig. 9 while the forest plot for trimethoprim is presented in Additional file 16: S16.

Fig. 9
figure 9

Forest plot of the prevalence of S. aureus resistance to cotrimoxazole

Full size image

Comparison of the prevalence of S. aureus resistance to different antibiotics

The trend of prevalence of S. aureus resistance to different antibiotics addressed in this meta-analysis is shown in Fig. 10. From observation, the prevalence of resistance of S. aureus to the different antibiotics in this study ranges from 13 (vancomycin) to 82% (penicillin G).

Fig. 10
figure 10

Comparison of the prevalence of S. aureus resistance to different antibiotics in Nigeria

Full size image

The order of resistance in increasing order based on the pooled prevalence of S.aureus resistance to different antibiotics was observed to be vancomycin, ofloxacin, rifampicin, ciprofloxacilin, gentamycin, norfloxacillin, clindamycin, cefoxitine, ceftriaxone, streptomycin, methicillin, chloramphenicol, erythromycin, trimethoprim, ceftazidim, amoxicillin-clavulanic acid, tetracycline, cotrimoxazole, ampicillin, cefuroxime, amoxacilin, cloxacillin, and pencilin G.

Discussion

Antimicrobial resistance continues to be on the rise which constitutes a serious public health problem globally. Many microbes have developed resistance to many different antimicrobial agents over time. This meta-analysis estimated the pooled prevalence of resistance of Staphylococcus aureus to 23 different antibiotics routinely used in Nigeria. Ninety eight studies [98] were included in this meta-analysis study with variation in the number of studies included in each meta-analysis which ranged from 6 to 77. In general, the 98 studies evaluated the rate of S. aureus resistance to different antibiotics based on 46,640 isolates of which 23, 048 were resistant to various antibiotics. Prevalence of resistance of S. aureus to different antibiotics ranges from 13 to 82%. Results from the meta-analysis showed that resistance of S. aureus to routinely used antibiotics in Nigeria was alarmingly high. From the studies, it was found that 82% S. aureus were resistant to penicillin G. However, it was observed from the studies that 24% of S. aureus were resistant to ofloxacin and rifampicin. In general, clinical samples (nasal, urine, wound, pimple, ear, blood, and vaginal swab) were collected from both symptomatic patients [61] and asymptomatic people [37].

High heterogeneity was observed for each of the meta-analyses performed with I2 ranging from 79.36 to 98.90% at p-values ≤ 0.01). This is because many studies used varying number of isolates/sample sizes. Some studies used 20 isolates while some used 400 isolates which impacted on the resistance profile of each antibiotic. This can better be illustrated in the prevalence of resistance of S. aureus to vancomycin. Sensitivity test was carried out to by removing studies that reported very high prevalence of S. aureus to vancomycin and the overall pooled prevalence reduced from 13 to 7%. This showed the degree of heterogeneity among studies. Possible cause of heterogeneity is due to different number of clinical samples and number of isolates recovered which were subjected to antibiotic sensitivity tests. Also random sampling of clinical samples can also be the possible cause. Publication bias was evaluated for all meta-analysis of the 23 antibiotics and publication bias was not found. Egger test is use to estimate asymmetry of data using funnel plots. p-value less than 0.05 using Egger criteria indicate no presence of publication bias even though erythromycin had p-value of 0.017 which is below 0.05. This is because a p-value of 0.017 for the Egger test means that the results found have a 1.7% chance to occur when there is no 'small sample bias.

The pooled prevalence of S. aureus resistance to Beta-lactams class of antibiotics was extremely high especially for penicillins. S. aureus showed highest resistance to penicillin G (82%) and 69% resistance to cefuroxime (cephalosporin). The pooled estimate of S. aureus resistance to penicillin G is comparable with the reported estimation of worldwide resistance of 90–95% [108]. This is not surprising due to the fact that penicillin G is the first antibiotic to be discovered. Bacteria are able to develop resistance to antibiotics due to selective pressure from antibiotics. Selective pressure from penicillin led to the production of beta-lactamase to conuter the effect of beta-lactam antibiotics. Consequently, semi-synthetic beta-lactam antibiotics such as ampicillin, Amoxicillin/clavulanic acid and amoxicillin with different side chains were developed to counter such bacteria strains. However, S. aureus resistance to amoxicillin and ampicillin is relatively high from our results. Lower rate of resistance was observed among beta-lactamase-resistant antibiotics (methicillin, ceftriaxone, cefoxitine). Also, lower rate of resistance to clindamycin might be attributed to infrequent use of the antibiotic. Amoxicillin-clavulanic acid was developed as a combination of an antibiotic (amoxicillin) and non-antibiotic (clavulanic acid). Clavulanic acid inhibit beta-lactamase enzyme which prolong the antibacterial activity of amoxicillin component; however, results from the meta-analysis showed high resistance of S. aureus to amoxicillin/clavulanic acid.

Another semi-synthetic penicillin resistant antibiotic called methicillin was developed which is resistant to hydrolysis of beta-lactamase was developed. The term Methicillin Resistant Staphylococcus aurues (MRSA) is synonymous with multi-drug resistance (MDR) because MRSA are invariably resistant to different antibiotics. Acquisition of mec A gene that encodes penicillin binding protein confers resistance to S. aureu [109]. The pooled prevalence of S. aureus to methicillin (46% [95% CI 37%, 56%]) in Nigeria is similar to 2014 global surveillance reports of the world health organization (WHO) [110] 2014. Which depicted MRSA prevalence ranged 33–95% in Africa. Similarly, the pooled estimate of 46% in our study is also in agreement with the pooled prevalence estimate of MRSA in continents such as North America, Asia, and Europe which ranges from 23.1 to 47.4% [109]. The high pooled prevalence in our study might be due to certain factors and variables such as the inclusion of nosocomial and community acquired infections in the original studies analyzed. Generally, nosocomial infection causing pathogens are believed to possess higher resistance rate due to prolonged and higher exposure to different antimicrobial agents and exchange of genetic materials. Thus, there is greater transmission of resistant genes through various means within the hospital settings [111]. The implication of infections cause by MRSA is difficulty in treatment which often requires alternative antimicrobial agents which are most times very expensive.

The pooled prevalence of S. aureus resistance to vancomycin (13% at 95% CI [0.7%, 21%]) in this meta-analysis is high and a cause for concern when compared to global prevalence estimate [4]. The prevalence of vancomycin resistant S. aureus (VRSA) in Africa was reported to be 2.5% [4]. This is quite low when compared to the result from this study which is very high (13%). With this increased resistance, the use of vancomycin to treat MRSA is becoming problematic and poses serious health challenge. The rise in VRSA might be due to the indiscriminate use of vancomycin in Nigeria. By the way, Four studies [20, 22, 28, 36] reported a very high prevalence of VRSA; however, sensitivity analysis showed that they had high significant influence on the overall pooled prevalence estimate. Removing the three studies reduces the pooled prevalence of S. aureus resistance to vancomycin from 13 to 7%. Analyzing studies that depicted high prevalence of resistance of S. aureus to vancomycin showed that the same author conducted and published the three studies in peer reviewed journals. Urine samples were mainly used for S. aureus isolation by the author in the three studies of which [20, 22] were from symptomatic urinary tract infection patients who visited the hospitals and [27] from healthy volunteers. Urinary tract infection is a common infection and a reason for antibioticl use; consequently, resistant microbial strains have emerged. This reason might be attributed to the high prevalence of S. aureus resistant to vancomycin in the three studies. Exposure to resistant strains especially in hospital settings might have resulted in the increased resistance to vancomycin in the three studies [112, 113]. This is because in Nigeria, expired or waste antibiotics are not properly discharged. This could result in selective pressure on inhabitant microorganisms which results in development of various resistant mechanisms.

Generally, the global pattern of antimicrobial resistance varies among different geographical locations and socioeconomic level [114, 115]. Variations in studies can be attributed to design, time, and population involved. Heterogeneity tests at p ≤ 0.01 showed significant variation among included studies in this meta-analysis. Therefore, it is reasonable to assert that the study population might be infected with the same strains of S. aureus within the same location at a specified period. This is because most of the studies were conducted within a specified period of time and area.

Mechanisms of resistance of S. aureus include: production of beta-lactamase enzymes to deactivate beta-lactam sensitive antibiotics, efflux pump for extruding antibiotics such as tetracyclines [6], reduced accumulation of macrolides antibiotics [7], production of aminoglycoside modifying enzymes to inactivate aminoglycosides antibiotics, alteration of DNA gyrase and topoisomerase IV expression for floroquinolones antibiotics, and expression of Mec genes which alters penicillin binding proteins. From the results and mechanism of resistance of S. aureus, it can be said that S. aureus found in Nigeria are highly resistant to the beta-lactam class of antibiotics.

The pooled prevalence of S. aureus resistance to the floroquinolones class of antibiotics such as ciprofloxacin, ofloxacin, and norfloxacin was lower especially for ciprofloxacin which is commonly used within Nigeria. However, high pooled prevalence of S. aureus resistance to antimetabolites class of antibiotics (cotrimoxazole and trimethoprim) was observed.

From the meta-analysis, S. aureus mediated infection in Nigeria can be treated using vancomycin, floroquinolones, and aminoglycosides. MRSA has been a concern in Nigeria especially with the incidence of VRSA. Newer alternative antibiotics such as linezolid, telavancin, ceftaroline, tigecycline and daptomycin are rarely used in Nigeria. Various factors such as lack of infection prevention which lead to reoccurrence of infection, inappropriate use of antibiotics, poor hospital facilities, lack of routine susceptibility test before antibiotic administration, and self medication contributes to the rapid emergence and re-emergence of AMR. Tackling this factors, will go a long way in the fight against the continue rise of MDR pathogens in general.

Study limitations

Most of the included studies share similar characteristics. The search was limited to only titles that deal with antibiotic resistance. Selection was done randomly especially in Google Scholar with had 35, 400 studies results from the search. The meta-analysis was done once for each antibiotics and sub-grouping to reduce high heterogeneity and publication bias was not done due to too many meta-analysis already done. The included studies used in-vitro antimicrobial assays which has limitations such as difficulties in interpreting data, variability of testing media (differences in cation content, acidic or alkaline), and difficulty in knowing the pharmacokinetics of an antibiotic or post effect of an antibiotic (a situation where bacteria growth is inhibited even when the antibiotic concentration falls below the MIC). Most of the studies were done in teaching hospitals and tertiary institutions in big cities; hence both symptomatic and asymptomatic individuals are involved. For symptomatic individuals, most of the studies were done in teaching hospitals were patients with chronic and recurrent infections are treated; resistance level could be overestimated.

Conclusion

The results of this meta-analysis showed that S. aureus is resistant to many routinely used antibiotics in Nigeria. It is highly resistant to beta-lactams, tetracyclines, and antimetabolites antibiotics. Resistance of S. aureus to vancomycin remains a serious health problem due to limited treatment options. There is a lot of variation in resistance estimates between studies. High heterogeneity was observed in each meta-analysis for each antibiotic which was attributed to various factors such as different clinical sample and recovered isolates sizes, random sampling and method used for resistance investigation. Hence it is imperative to develop programs to promote rational use of antimicrobial agents, infection prevention and control to reduce the incidence of AMR. In addition, furthers researches focusing on identifying the dynamics promoting microbial resistance, infectious microbial strains and molecular/genetic basis of resistance should be encouraged.