Inducible Clindamycin-Resistant Staphylococcus aureus Strains in Africa: A Systematic Review

Introduction Excessive use of clindamycin enhances the acquisition of inducible clindamycin-resistant S. aureus strains, which is a significant health problem in Africa. The main objective of this review study was to determine the prevalence of inducible clindamycin resistance and related genes among S. aureus isolates in Africa. Methods A qualitative systematic review was conducted on inducible clindamycin resistance among S. aureus isolates in Africa using electronic databases such as Google Scholar and PubMed. Articles published in English before 2021 were selected, and relevant data were extracted, collected, and analyzed. Results In our search, 22 articles met the eligibility criteria for this review study. Of 3064 total S. aureus isolates, 605 had iMLSB phenotype. The overall prevalence of inducible clindamycin resistance in S. aureus isolates was 19.8% with a range of 2.9% to 44.0%. A high number of iMLSB phenotypes were observed in MRSA isolates (3.6–77.8%) than MSSA (0–58.8%). The overall prevalence of the iMLSB phenotype in MRSA strains was 26.8% (279/1041). The maximum peak prevalence of inducible clindamycin resistance among S. aureus isolates recorded in the continent was 44.0% in Egypt, followed by 35.8% in Libya and 33.3% in Uganda in 2017, 2007, and 2013, respectively. The highest prevalence of iMLSB phenotype in MRSA strains was reported in Egypt, 77.8%, followed by Nigeria, 75.0%, and Libya, 66.2%. Among the recovered drug-resistance genes, ermA, ermC, and msrA genes were commonly detected in Egypt with 67.9%, 70.0%, and 70.0% prevalence, respectively. Conclusion This review highlights a higher inducible resistance of S. aureus, including MRSA strains to clindamycin in the continent. Regular screening of these strains, wise use of clindamycin, and molecular detection and genotyping of resistant genes are urgent.


Introduction
Staphylococcus aureus is normally found in human skin and mucous membranes. It is a common human pathogen that causes skin and soft-tissue infections, abscesses, pneumonia, osteomyelitis, endocarditis, arthritis, and sepsis in both the community and hospital environment, and the spread of methicillin-resistant Staphylococcus aureus (MRSA) through the acquisition of highly transmissible mecA/mecC genes has made treatment difficult [1,2].
Even though the global average incidence of MRSA is 40.0%, reports from African countries reveal rates ranging from 12.0 to 80.0%, with some countries exceeding 82.0% [3,4]. Antimicrobial resistance has become a severe health hazard worldwide, and its burden has increased in Africa because of a highly infectious disease burden, poor hygiene, lack of environmental sanitation, and poor infection control. e treatment of MRSA infections in African nations is problematic due to the lack of antibiotics with proven efficacy [5]. e rising prevalence of community-acquired MRSA has sparked interest in using macrolide-lincosamidestreptogramin (MLSB) antibiotics, particularly clindamycin to treat S. aureus-associated pneumonia and skin and softtissue infections [6].
Clindamycin is the chosen antibiotic because of its superior pharmacokinetics, availability in intravenous and oral formulations with 90% oral bioavailability, low cost, strong tissue penetration, accumulation in deep abscesses, and capacity to inhibit toxin generation in S. aureus [7]. Excessive use of clindamycin, on the other hand, enhanced the acquisition of inducible resistance, leading to therapeutic failure [8]. e main mechanisms of resistance in the MLSB drugs include target site alteration, efflux pump expression, and mutation [9]. e MLSB phenotype can be either constitutive (cMLSB phenotype) in which rRNA methylase is always produced) or inducible (iMLSB phenotype) in which methylase is produced only when an inducing substance like erythromycin is present. During treatment, iMLSB phenotypes can be mutated into cMLSB phenotypes [10]. Owing to this, the Clinical and Laboratory Standards Institute (CLSI) recommends using the double-disk diffusion method (D-test) for detecting inducible resistance to clindamycin among Staphylococcus aureus isolates [11].
In Africa, S. aureus is becoming increasingly resistant to clindamycin due to clinicians' rash use of antibiotics without performing D-test and lack of laboratory facilities for molecular approaches such as polymerase chain reaction (PCR) based resistance gene detection. Although there has been published research on inducible clindamycin resistance, little is known about its dissemination and clinical significance in Africa, necessitating a compilation of data from the continent. To fill this gap, this systematic review provides an updated summary and valuable data. erefore, this study mainly aimed to determine the prevalence of inducible clindamycin resistance and related resistance genes among S. aureus isolates in Africa.

Literature Search Strategy in Databases.
A systematic literature search was performed on published articles for inducible clindamycin resistance among S. aureus isolates in Africa with a study period before 2021 using electronic databases such as PubMed and Google Scholar. e following keywords were used with the help of Boolean operators: "inducible clindamycin resistance" OR "macrolidelincosamide-streptogramin B resistance" OR "D-test" AND ("Staphylococcus aureus" OR "S. aureus" OR "methicillinresistant Staphylococcus aureus" OR "methicillin-sensitive Staphylococcus aureus" OR MRSA OR MSSA) AND (Africa). e references of included articles were appropriately scanned to access related articles of interest. e literature search was not limited to a specific publication or year of study. In this review, we considered all studies that described inducible clindamycin resistance in S. aureus obtained from any type of human study participant in Africa.
e procedure of eligible study selection is demonstrated in Figure 1.

Study Selection and Eligibility Criteria.
All studies from Africa reported the following information in the full-text selected: (I) Articles performed D-test for detecting the iMLSB phenotype in S. aureus according to CLSI guideline (II) All articles published with a study period until 2021 (III) Articles published and written in English (IV) Articles with well-defined objectives and methodology (V) Articles from a human source of specimen (VI) Articles including data on the number of S. aureus isolates and any source of specimen used (VII) Studies which investigated antibiotic resistance genes using PCR were also summarized Articles that lacked all or most of the above variables, such as abstract only, not in English, duplicate reports, ambiguous results, and articles with overlapping data, were excluded.

Assessment of the Study Validity.
e validity of each study was illustrated by the use of the selection and eligibility criteria described above, thereby excluding studies that have unclear results, are unrepresentative of the human population, or studies with noncomparable data. Studies vary in specimen source, and the human study population was not excluded.

Data Extraction and Collection.
Essential data were extracted from eligible studies using Excel spreadsheet format, and any discrepancies were handled by the author. e following information was extracted from the selected studies: the percentage of iMLSB phenotype detected, the number of S. aureus isolates identified, the study period, the study population, geographic area where the study was conducted, the source of the specimen, the method of detection, the coexistence of antibiotic resistance genes, and the references were all considered.

Data Synthesis and Analysis.
e data were synthesized qualitatively. Because of the relatively small number of studies used, inconsistencies between studies, and Studies with no information about study period/ population/specimen source.

2.
3. heterogeneity of the study populations between countries, we did not perform a quantitative synthesis. Data were summarized in the extraction table and analyzed manually. e overall prevalence of inducible clindamycin resistance in S. aureus or MRSA strains was calculated using the following formula:

4.
overall prevalence � sum of the iMLSB phenotypes total number of S. aureus/MRSA * 100%. (1) According to the United Nations list of 54 African countries, the map of Africa was created using the website (https://mapchart.net/). Finally, charts were created using the Excel 2019 software.

Literature Search.
In electronic database searches, 465 articles were retrieved. After removing duplicates, 361 articles were avoided based on their titles and abstracts. e full-text articles of the remaining 104 articles were reviewed in detail for eligibility. Of these, 82 articles were discarded after the full-text had been reviewed for appropriate methodology, study population, the source of the specimens, clear result, and standard microbiological technique. Finally, 22 articles were included in the synthesis of the review ( Figure 1).

General Characteristics of the Studies Included in is Systematic Review.
e main characteristics of the 22 studies from 8 African countries included in this systematic review have been summarized in Table 1. All the studies used D-test for detecting inducible clindamycin resistance in S. aureus isolates, but some of them also used PCR for confirming the MLSB resistance genes. Studies used various specimens from human sources including swabs such as nasal, vaginal, cervical, urethral, wound, throat, ear, eye, and palm, respiratory specimens such as sputum, bronchoalveolar lavages, and tracheal aspirates, pus, urine, catheter, blood, semen, cerebrospinal fluid, pleural fluid, ascetic fluid, synovial fluid, and others. In this systematic review, most of the studies were conducted in clinical patients admitted to hospitals or outpatients, and the rest were conducted in healthy individuals with S. aureus carriers. Out of six studies that performed PCR for detecting MLSB resistance genes, 5 (83.3%) were conducted in Egypt and 1 (16.7%) was in Uganda.
e detection rate of inducible clindamycin resistance in MRSA and MSSA was not reported in Ethiopian studies and a study from Uganda. Additionally, the prevalence of the iMLSB phenotype was relatively higher in children and burn patients (Table 1).

4.1.
e Prevalence of Inducible Clindamycin Resistance among S. aureus Isolates. In our review, we assessed the overall prevalence of inducible clindamycin resistance among S. aureus isolates by adding all iMLSB phenotypes and dividing it by the total number of S. aureus isolates. e total number of S. aureus isolates in the review was found to be 3064. Among the total S. aureus isolates, 605 had an iMLSB phenotype. us, the overall prevalence of inducible clindamycin resistance in this review was found to be 19.8%. Similarly, the overall prevalence of the iMLSB phenotype in MRSA was calculated by adding all the number of iMLSB phenotypes and dividing it by the total number of MRSA isolates, which was 26.8% (279/1041). Studies were conducted between 2007 and 2021 from different areas of the country .
e prevalence of inducible clindamycin resistance among S. aureus isolates varies from place to place due to the difference in local clindamycin resistance. Inducible clindamycin resistance was first reported in 2007 in Libya among burn patients [12]. e prevalence range of inducible clindamycin resistance among the S. aureus isolates was 2.9-44%   (Figure 3). e highest peak prevalence of inducible clindamycin resistance among S. aureus isolates documented on the continent was 44.0% in 2017 in Egypt [21] and the minimum prevalence was 2.9% from Côte d'Ivoire [24]. In 2007 and 2013, respectively, the second (35.8%) [12] and third (33.3%) [17] highest peaks of inducible clindamycin resistance prevalence were reported. Despite a 33.3% record of inducible clindamycin resistance in 2013, there has been a progressive drop in the prevalence of inducible clindamycin resistance since 2007 [17]. A similar 10.0% prevalence was observed in Egypt during 2017 and 2018 [23,27]. Generally, there were heterogeneous distribution and prevalence rate of inducible clindamycin resistance among S. aureus isolates in Africa according to the reviewed studies (Figure 3).

Resistance Genes Related to MLSB Resistance.
Despite the lack of complete information on the resistance genes, investigations have shown that many S. aureus isolates coproduced resistance genes such as erm (A, B, C, E) genes, msrA genes, mphC genes, and lnuA genes. Among the recovered erm genes in Egypt, the ermC gene was a highly detected resistance gene, at 70.0% [23], followed by the ermA gene, with a 67.9% detection rate [21]. e msrA gene was also detected among S. aureus isolates, with a 70.0% high detection rate in Egypt [23]. A study in Uganda also revealed ermC genes (32.7%) [17]. Additionally, the mphC genes (40.0%), and lnuA genes (20.0%) were also detected from a study in Egypt [25] (Table 1).

Discussion
e phenotypic analysis of the inducible resistance in S. aureus to clindamycin was demonstrated through the Dtest across African countries. In this review study, the prevalence of inducible clindamycin resistance among the S. aureus isolates was found to be 19.8%, ranging from 2.9 to 44.0% . Various studies have reported a comparable proportion of inducible clindamycin resistance among S. aureus isolates obtained from human sources such as studies in Odisha state, eastern India, 22.0% [7]; Nepal, 23.4% [34]; Malaysia, 22.1% [35]; and Israel, 20.0% to 25.0% [36]. is finding is significantly higher than the findings of the iMLSB phenotype from the systematic review and metaanalysis in Iran with 10.4% overall prevalence [37]. In Indian studies, inducible clindamycin resistance rate was 5.2% in Kashmir valley [38], 7% in Assam [39], 15.2% in Chennai [40], 13.71% in the sub-Himalayan region [41], and 14.8% in central India [42]. In Nepal, lower inducible clindamycin resistance rate was indicated across different areas with a prevalence of 14.9% [43], 11.48% [8], 15.2% [44], and 12.1% [45]. While studies have also reported a high proportion of inducible clindamycin resistance as the prevalence in West Bengal, India, 41.3% [46]; Nepal (39.7% [47]; 34.8% [48]); Jordan, 76.7% [49]; and Tokyo, Japan, 91.0% [50]. e different iMLSB phenotypes observed in various studies are because of the variation in the study population, geographic region, the source of the specimen, methicillin susceptibility, usage of MLSB antibiotics in the community and hospital settings, and drug-resistant clones. e emergence of inducible clindamycin resistance was comparatively higher among the clinical MRSA isolates as up to 77.8% recovered from cancer patients with febrile neutropenia [29]. In this study, the overall prevalence of inducible clindamycin resistance in MRSA strains was 26.8%. Comparable findings were reported in Nepal, 24.5% [44], and India (28.0% [41]; 25.0% [42]). Studies in Nepal (34.3% [47]; 76.4% [34]); India, 37.5% [51]; Malaysia, 46.7% [35]; and Jordan, 76.7% [49] demonstrated higher prevalence. However, this finding is higher than a study conducted in India (7.5% [39]; 18.7% [52]). e cMLSB phenotype prevalence in MRSA strain (28.9%) was in agreement with the study conducted in India, 29.26% [41], higher than studies conducted in Nepal (5.7% [47]; 11.2% [44]), India   (16.6% [51]; 16.9% [39]), and Malaysia, 11.1% [35], but lower than a study from India, 64.8% [42]. is shows that clindamycin treatment proved effective against MSSA infections, but it can lead to treatment failure in MRSA infections, and iMLSB phenotypes can be mutated into cMLSB phenotype. e most common mechanism for MLSB resistance in S. aureus is the target site modification of 23S ribosomal RNA mediated by erm genes and strains exhibiting the iMLSB phenotype having a high frequency of spontaneous constitutive resistance mutations. Regarding the genotypic confirmation of MLSB resistance genes, our findings demonstrated that erm (A, B, C, E) genes and msrA genes were commonly detected genes. is finding is supported by other studies [34,49,[53][54][55][56][57][58]. is indicates that the high spread and transmission of these genes significantly contribute to the increasing acquiring clindamycin resistance in S. aureus strains. As a limitation, the inclusion of studies with lower sample size results in a bias in the finding. Most studies studied only the prevalence of inducible clindamycin resistance; only a few studies reported resistance genes.

Conclusion
e current review study demonstrated a high prevalence of inducible clindamycin resistance S. aureus isolates with varying proportions throughout the country. A relatively higher number of iMLSB phenotypes was observed in MRSA than in MSSA isolates and a high figure was reported in Egypt, 77.8%, and Nigeria, 75.0%. Additionally, these strains are closely related to resistance genes such as the ermA, ermC, and msrA genes. Hence, there is an urgent need for ongoing studies to further assess iMLSB-positive S. aureus strains especially MRSA and in the revision of clindamycin prescription. Genotypic detection of resistance genes is mandatory to minimize treatment failure.

Data Availability
All extracted data are freely available in the article and Supplementary Materials.

Conflicts of Interest
e authors declare no conflicts of interest.

Authors' Contributions
MA conceived the review idea and content of the manuscript. MA also involved in data extraction, collection, and analysis and wrote the manuscript and was responsible for the final approval and submission of the manuscript.  International Journal of Microbiology 7