The Prevalence of Antimicrobial Resistance and Serotypes of Streptococcus pneumoniae in the Kingdom of Bahrain

Streptococcus pneumoniae (S. pneumoniae), also known as pneumococcus, is found in the upper respiratory tract of 5-70% of healthy people. It causes otitis media, pneumonia and invasive pneumococcal diseases (IPD). The group with the highest risk for pneumococcal diseases are children, elderly people and people with chronic systemic illnesses. The identification of these serotypes is crucial to monitor the prevalent serotypes, to track resistance patterns, and to evaluate the efficacy of treatment and vaccination. Our aim was to determine antibiotic resistance in S. pneumoniae, identify the most common serotypes, determine the association between highly resistant strains and most common serotypes, and select an effective vaccine against the predominant serotypes in the Kingdom of Bahrain. One hundred isolates were included in the study. Antimicrobial susceptibility was determined by agar dilution and serotypes were determined by the Quellung reaction. Antimicrobial resistance rates were highest for penicillin in children ≤ 5 years old (34%) and elderly patients (29%). For the serotypes distribution of S. pneumoniae, the predominant serotypes were 19, 6, 23, 3 and 14 in the different age groups. The predominant serotypes that showed resistance to three antimicrobial agents were 19, 6, 23 and 14. The increasing pneumococcal resistance to penicillin, erythromycin, and tetracycline indicates the need for preventive measures, antibiotics use monitoring and stewarding. In light of the expanding prevalence of S. pneumoniae resistant to multiple antimicrobial agents, the need for an effective pneumococcal vaccine has become very important.


INTRODUCTION
Streptococcus pneumoniae also known as pneumococcus, is found in the upper respiratory tract of 5-70% of healthy people. It causes otitis media, pneumonia, sinusitis and also invasive pneumococcal diseases (IPD) 1,2 .
The serotypes identification is crucial to monitor the prevalent serotypes, resistance patterns, and to evaluate the efficacy of treatment and vaccination 3 . Serotype distribution in different countries is shown in Table 1. Although no data on serotypes are available, yet the use of pneumococcal vaccines is common in the Kingdom of Bahrain. This present study was aimed to determine the serotypes and antibiotic susceptibility to penicillin, erythromycin, and tetracycline used for the treatment of pneumococcal infection.

MATERIALS AND METHODS Bacterial isolates
A total of 100 S. pneumoniae isolates from patients were evaluated after getting an ethical approval from Arabian Gulf University. There were sixty-two isolates from Salmaniya Medical Complex (SMC) and thirty-eight isolates from Bahrain Defence Force Hospital (BDF), that were isolated from inpatients and outpatients with various infections, such as pneumonia, otitis media, sinusitis, conjunctivitis, meningitis, and septicemia. Isolates from blood or cerebrospinal fluid (CSF) were considered invasive, and isolates from upper respiratory tract samples (e.g., ear swab), lower respiratory specimens (e.g., sputum, and deep tracheal aspirates), eye swab, pus, anal swab or wound fluids were considered noninvasive. The demographic data for all patients such as sex, age, and nationality, as well as clinical diagnosis and the sources of the isolates were recorded. All collected strains were taken as subculture isolates from hospital lab on sheep blood agar plates and incubated at 37°C for 18-24 hours in the presence of 5% CO 2 . Several colonies of the bacterial culture were then picked, placed in skimmed milk, and stored at -80°C until further testing. The isolates were confirmed as S. pneumoniae by positive tests for alpha-hemolysis on blood agar, optochin susceptibility, and bile solubility.

Study design
It was a cross sectional study. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were used.

Antibiotic sensitivity testing
The dilution of antimicrobial agents was made in dH 2 O. Two ml of the previous antimicrobial concentration was poured into the Petri dishes. Then 18 ml of Mueller Hinton agar (MHA) with sheep blood was poured into plates. After that, plates to be used were left at room temperature overnight to dry. The colonies were emulsified in Mueller Hinton broth where the suspension optical density was adjusted to 0.5 of McFarland standards. Then 2 µl of the colony suspension was inoculated as a spot onto the surface of each agar plate. The S. pneumoniae ATCC 49619 strain was used as quality control for testing. The plates were incubated at 37°C in a CO 2 incubator for 18-24 hours. The strains and sensitivity testing (MIC) were performed according to Clinical and Laboratory Standards Institute guidelines (CLSI), see Table 2

Statistical analysis
The chi-square test was used to identify differences in prevalence of antimicrobial resistance. A p-value of < 0.05 was considered significant.

RESULTS
The demographic data for patients is shown in Table 3. The male-to-female ratio was 1.4:1. Many of the patients were children ≤ 5 years old (44%), the next biggest group was of patients 16-64 years of age (34%). There was no statically significant difference between the age groups in this study.    Penicillin resistance was higher among upper respiratory tract than among blood isolates. The 100 pneumococcal isolates were first tested for antimicrobial susceptibility. The sources of the isolates, and their antimicrobial resistance pattern are shown in Table 3. The prevalence of penicillin-resistant isolates in this study was high in children ≤ 5 years old (34%) and elderly patients (29%). For erythromycin resistance, high rates were found in patients between 6-15 years old (60%) with only five isolates which is a very small group and in children ≤ 5 years old (39%). For tetracycline resistance, high prevalence rate was found in the patients between 6-15 years old (60%) again with only five isolates which is a very small group and in patients between 16-64 years old (35%). There was no significant difference in resistance for any of the antibiotics between the age groups. The prevalence of penicillin resistance among the upper respiratory tract isolates (8 of 16) was significantly higher than that among the blood isolates (4 of 21) (p=0.046). Among the 100 isolates, 39 were sensitive to all the three antibiotics and 26 isolates showed resistance to more than one antibiotic. No differences were found in antibiotics resistance between upper and lower respiratory tract infections. The antibiotics resistance between invasive and noninvasive isolates was also not significantly different.
Of the isolates, 95 (95%) were serotyped and 5 (5%) were unidentifiable. The predominant serotypes among all patients were the serotypes 19 (n=17), 6 (n=16), 23 (n=11), 3 (n=10), and 14 (n=8) is shown in Table 4. The distribution of these serotypes were not different between the age groups. The strains with serotypes 19, 23, 6, and 14 were most often resistant to antibiotics, while of the five most common strains the one with serotype 3 was never found to be resistant to antibiotics.

All serotypes of invasive infections in children are included in the vaccines
The prevalence of S. pneumoniae serotypes in different infections as per the age group in children is shown in Table 5. The PPV23 vaccine given to individuals of ages >2 and <65 years in high risk groups includes the serotypes 1, 2, 3, 4, 5, 6B, 7F, 8, 9N, 9V, 10A, 11A, 12F, 14, 15B, 17F, 18C, 19F, 19A, 20, 22F, 23F, and 33F. The serotypes of the isolates of invasive infections observed in children > 2 years old were covered by the PPV23 vaccine. Also, in the groups of at 6-15 years old (60%) and in children ≤ 5 years old (39%). This is similar to the result from US 20 but different from results reported in Thailand (2005), where more resistant isolates were seen in the age group < 6 years (37.9%) and between 15-60 years old (27.5%) 21 . For tetracycline resistance, the highest prevalence rates were found in the patients between 6-15 years old (60%) and in patients between 16-64 years old (35%). In Spain, a similar finding was reported for the adults 22 .
According to the site of infection, in this study, the highest prevalence rate for penicillinand erythromycin-resistant strains were found in upper respiratory tract isolates, similar result for penicillin were reported in Kuwait 9 and US 18 . Another study from Saudi Arabia in 2004 14 , Iran 23 , and Taiwan 24 showed highest resistance rates in blood and CSF isolates. But those studies were different from the current one because highest prevalence was found in blood and CSF isolates. For tetracycline resistance, the highest prevalence was found in blood, which is different from UAE 25 where the highest rate was seen in lower respiratory tract isolates. This study also found that invasive infections were found mainly in the group of children ≤ 5 years old (34%). Similar findings were reported in the Kingdom of Saudi Arabia 8 , Kuwait 9 , and Spain 22 .
When looking at antibiotic resistant strains, resistance was most commonly found for the three antibiotics in the serotypes 19, 6, 23 and 14. In Kuwait, a study was conducted in which similar results were obtained for penicillinresistant serotypes (23F, 19F, 6B and 14) 9 . On the other hand, in the Kingdom of Saudi Arabia, a study showed that the main serotypes among the multidrug-resistant isolates were 19F, 23F and 6A 14 . In Turkey 11 and South Africa 26 , also comparable researches were performed. The most frequently encountered serotypes were 6, 19, 1, 23, and 14 in Turkey and 6, 19, 14, and 23 in South Africa.

DISCUSSION
This is the first study of the kind in the Kingdom of Bahrain to determine S. pneumoniae serotypes. The most common serotypes in all age groups were 19, 6, 23, 3 and 14 which is shown in Table 6.
This study is different from ones conducted in the Kingdom of Saudi Arabia 14 and Kuwait 9 , where the most common serotypes were 6, 19, 1 and 3. Similar results to the current study were obtained in Burkina Faso 15 as well as in England and Wales 16 for the serotypes 6, 19, 1 and 3. The same study in England and Wales gave a different result for the serotypes (14, 9, 4, 23, 3, 12 and 18). Serotypes 19, 23, 3, 6 and 14 were the predominant serotypes among the noninvasive isolates in all age groups in our study. Similar results were found for some serotypes (3 and 6A) in the Kingdom of Saudi Arabia 14 , Kuwait (23F, 19F, 6A, 6B and 14) 9 and a study done in Thailand (2007) reported similar serotypes (19, 6 and 23) 17 .
The prevalence of penicillin-resistant isolates in this study was high in children ≤ 5 years old (34%) and in elderly patients (29%). Similar results were found in studies from Kuwait 9 , US 18 , and Caledonia 19 . For erythromycin resistance, the highest rates were found in patients between polysaccharide vaccine (PPV23) is available in the Kingdom of Bahrain for high-risk groups such as HIV positive individuals, sickle cell disease patients and elderly above 65 years old. This study demonstrated that PPV23 had an excellent coverage (91.4%) for all strains, with even 100% coverage for invasive, erythromycin resistant, and tetracycline resistant isolates and high coverage (95.2%) for penicillin-resistant isolates for children ≥ 2 years old. Our results demonstrated higher coverage rates by PPV23 for invasive and penicillinresistant strains than in Kuwait (76 to 82%) 7 . Also, high coverage was reported in South Africa (97.4%) 26 and Uruguay (96.1%) 27 .
For PCV13 coverage for children < 2 years old, the vaccine showed good coverage rates for all serotypes (66.7%), penicillin-resistant (77.8%), erythromycin-resistant (88.9%), tetracyclineresistant (85.7%) and excellent coverage rate for invasive isolates (100%). High coverage rate demonstrated in the study from Chile which showed coverage rate of (80%) for all serotypes 28 . On the other hand, this present study reported higher coverage rate for invasive isolates than another study in Spain (77.9%) 29 and Thailand (85.7%) 30 . For children < 5 years old, this study showed coverage rates of (69%) for all serotypes, (100%) for invasive isolates and (78.6%), (87.5%), (84.6%) for penicillin-, erythromycin-, and tetracycline-resistant isolates, respectively. On the other hand, high coverage rate was demonstrated in a study from China (87.8%) for all serotypes 31 . In Thailand 30 , lower coverage rates was reported for IPD, it was (87.8%).
In conclusion, in light of the increasing prevalence of S. pneumoniae resistant to multiple antimicrobial agents, the local policies and the immunization should consider these changes and adapt accordingly.