Nasopharyngeal Staphylococcus aureus colonization among HIV-infected children in Addis Ababa, Ethiopia: antimicrobial susceptibility pattern and association with Streptococcus pneumoniae colonization

Abstract

Background.

Staphylococcus aureus and Streptococcus pneumoniae are common inhabitants of the nasopharynx of children. HIV-infected children have higher risk of invasive diseases caused by these pathogens. With widespread use of pneumococcal conjugate vaccines and the emergence of methicillin-resistant S. aureus , the interaction between S. aureus and S. pneumoniae is of a particular significance. We sought to determine the magnitude of colonization by methicillin-sensitive and -resistant S. aureus and colonization by S. pneumoniae ; associated risk factors and antimicrobial susceptibility pattern among HIV-infected children in Addis Ababa, Ethiopia.

Method.

A prospective observational study was conducted among 183 HIV-infected children at ALERT hospital Addis Ababa, Ethiopia from September 2016 to August 2018. S. aureus and S. pneumoniae were identified using standard bacteriological techniques, antimicrobial susceptibility testing was performed on S. aureus and screening for methicillin resistance was carried out by amplifying the mecA gene. Risk factors were analysed by using binary logistic regression.

Results.

The prevalence of nasopharyngeal S. aureus , MRSA and S. pneumoniae colonization were 27.3, 2.7 and 43.2 %, respectively. Multivariable analysis indicated an inverse association between S. aureus and S. pneumoniae nasopharyngeal colonization [aOR, 0.49; CI, (0.24, 0.99); P=0.046]. The highest level of resistance in both methicillin-sensitive S. aureus (MSSA) and MRSA was observed against tetracycline.

Conclusions

. We found an inverse association between S. aureus and S. pneumoniae colonization among HIV-infected children. Continued assessment of the impact of pneumococcal conjugate vaccines and antiretroviral therapy on nasopharyngeal bacterial ecology is warranted.

Data Summary

All the data used in the study is available within the manuscript.

Introduction

Staphylococcus aureus is a human commensal and also causes clinically important community and hospital-acquired infections such as endocarditis, bacteremia, osteomyelitis and skin and soft tissue infections [1]. S. aureus colonizes multiple body sites in humans including anterior nares, skin, perineum and pharynx. However, the most frequent carriage site for S. aureus , with a 27 % carriage rate among adults, is the nose and a causal relationship exists between S. aureus colonization and infection [2]. Compared to adults, persistent carriage is mostly seen in children and carriage rates vary between 45 % in the first 8 weeks and decline to 21 % by 6 months of age [3].

Methicillin-resistant Staphylococcus aureus (MRSA) was discovered in 1961, soon after the introduction of methicillin in clinical isolates and then in the 1980s in the community. It has since spread globally and its prevalence is increasing resulting in increased health-care costs, morbidity and mortality [1, 4]. HIV-infected patients, due to their weakened immune system, are at increased risk for several infections, including those caused by S. aureus and MRSA [5]. The morbidity and mortality associated with MRSA in patients with HIV infection [6] makes the study of MRSA colonization among the HIV-infected patients of particular interest. In addition, colonization with MRSA has been associated with a fourfold increase in the risk of infection [7].

Studies indicate that compared to healthy individuals, the prevalence of MRSA nasal colonization is higher in HIV-positive people [8, 9]. The risk of MRSA colonization often mirrors that of infection and additional people with higher risk of colonization/infection are, children, those in prisons, military recruits, those in poor neighbourhoods, livestock workers, individuals with prior MRSA infection and hospitalization and those with cystic fibrosis [1]. The main risk factors for colonization and infection among HIV-infected patients on the other hand are use of antimicrobials, previous hospitalization and low CD4 +T lymphocyte counts [10].

MRSA is now considered an urgent threat to public health and among the priority list of pathogens for which new antibiotics are required [11]. In order to make effective treatment of patients with disease due to MRSA and to prevent further transmission, accurate detection of methicillin resistance is of the utmost importance [12].

The mecA gene is highly conserved in staphylococcal strains and thus is a useful marker and is considered as the gold standard for identifying methicillin resistance in S. aureus isolates. The mecA gene is located on the staphylococcal chromosome cassette mec and encodes penicillin binding protein 2 a (PBP2a) [12]. Other chromosomal factors such as femA and femB are also associated with the expression of methicillin resistance [13].

Pneumococcal conjugate vaccines (PCVs) began being introduced in infant immunization schedules from the year 2000 and studies indicate that administration of PCVs to infants modifies not only the carriage of Streptococcus pneumoniae but also that of S. aureus and there is an inverse relationship between the carriage of the two bacteria [14, 15]. There was also an earlier clinical trial that indicated an increase in acute otitis media due to S. aureus after PCV vaccination [16]. We have recently been able to show that 5–6 years after introduction of PCV10 in Ethiopia, S. aureus was the main cause of bacteremic community-acquired pneumonia among children [17].

Earlier studies in HIV-infected children indicated a lack of association between S. pneumoniae and S. aureus carriage and cite suboptimal adaptive immunity as a possible reason [14] but there are suggestions that S. pneumoniae – S. aureus interference is CD4⁺ T cell-mediated and returns to normal following antiretroviral therapy (ART) [18].

In Ethiopia, PCV10 was introduced in October 2011 as a three-dose primary series (3P+0) without any booster dose and a decision has now been made to switch to PCV13. A previous study among HIV-infected children in Northern Ethiopia indicated a pharngeal carriage prevalence of 29 % for S. aureus and 12.3 % for S. pneumoniae [19]. There is however a scarcity of data in Ethiopia on the relationship between the nasopharyngeal carriage of S. pneumoniae and S. aureus and the possible impact of PCV introduction especially among HIV-infected children. The aim of this study was therefore to determine the magnitude of methicillin-sensitive and -resistant S. aureus nasopharyngeal colonization and their association with S. pneumoniae colonization among HIV-infected children in Addis Ababa, Ethiopia.

Methods

Study design and setting

A prospective observational study was conducted at the All Africa Leprosy Rehabilitation and Training Hospital (ALERT), a governmental referral hospital, located in Addis Ababa, Ethiopia. The target population of this study was HIV-positive children aged 0–15 years coming for ART follow-up to the ALERT paediatric HIV clinic from 1 September 2016 to 31 August 2018.

Patient enrollment and data collection

Trained research nurses approached the parents or guardians of the HIV-infected children who came for an initiation or followup visit at the ART clinic and briefed them on the information that has been provided in the patient information sheet. Children of parents who gave consent were then included in the study. Study participants' demographic and clinical data including information on risk factors for colonization and disease were obtained and transferred to the questionnaire prepared for the study.

The dependent variable assessed was colonization with S. aureus . The independent variables assessed were gender, age, number of household members, number of rooms in the house, presence of siblings <5 years of age, parents' level of education, monthly income, exposure to cigarette smoke, reasons for hospital visit, antiretroviral treatment status, PCV vaccination, S. pneumoniae colonization, upper respiratory tract infection, lower respiratory tract infection within the last 3 months, CD4⁺ T cell count, and previous hospitalization within the last 3 months.

Specimen collection

Nasopharyngeal (NP) swabs were collected from 183 HIV-infected consecutive children using flocked mini tip rayon swabs (Copan, Brescia, Italy). Briefly, tipping the head of the children slightly backward, the swab was directly inserted parallel to the base of the NP passage until reaching the nasopharynx, which is located about one-half to two-thirds the distance from the nostril to the ear lobe. Once the swab was fully passed into the nasopharynx, it was rotated 180^o and left in place for 5 s to saturate the swab tip. The swab was then removed slowly and immediately transferred into cryovials containing 1 ml of STGG (skim-milk, tryptone, glucose, and glycerol) transport medium. The inoculated STGG was initially stored at −20 °C in the hospital ward. Samples were then transported to the Armauer Hansen Research Institute (AHRI) laboratory using ice and were frozen at −80 ˚C until further analysis.

Culture and identification

The samples were initially thawed and 100 -µl of the NP swab sample was plated on a blood agar plate (BAP) with and without gentamicin by streaking the sample using a sterile loop. BAP plates were then incubated at 37 °C with ~5 % CO₂ for 18–24 h. S. aureus was identified using colony morphology on sheep blood agar, gram staining, catalase activity, production of coagulase and growth on mannitol salt agar. Identification of S. pneumoniae was made on the basis of colony morphology, optochin susceptibility and bile solubility.

Antimicrobial susceptibility testing

The susceptibility of S. aureus isolates to chloramphenicol (30 µg), ciprofloxacin (5 µg), clindamycin (2 µg), erythromycin (15 µg), gentamicin (10 µg), tetracycline (30 µg) and trimethoprim/sulfamethoxazole (1.25 µg/23.75 µg) was determined by the disc diffusion method on Mueller–Hinton agar (Oxoid, Basingstoke, UK). Test results were interpreted according to the Clinical and Laboratory Standard Institute criteria. Multidrug resistance (MDR) was defined as resistance to three or more antimicrobial classes. S. aureus (ATCC 25923) was used as a quality control organism.

Detection of mecA genes

DNA extraction was performed by boiling S. aureus pellets in 300 µl of TE buffer as described previously [20]. Detection of 310 bp fragment of mecA was performed using primer pairs: 5′-GTAGAAATGACTGAACGTCCGATAA-3′ and

5′ CCAATTCCACATTGTTTCGGTCTAA −3′ as described previously [21]. The reaction mixture contained 12.5 µl of hot star master mix (Qiagen, Hilden, Germany), 0.5 µl each of the forward and reverse primers, 9 µl of molecular grade water and 2.5 µl of the template with a final volume of 25 µl. Amplification was carried out with 40 cycles of initial heat activation at 95 °C for 15 min, denaturation at 94 °C for 30 s, followed by annealing at 52 °C for 45 s, extension at 72^°C for 1 min, and final extension at 72^°C for 10 min. The PCR products were analysed by electrophoresis on a 2 % agarose gel.

Statistical analysis

Data were initially entered in ReDCap (Vanderbilt University, Nashville, TN, USA), exported into Excel and analysed using PASW Statistics 20 software (SPSS, Chicago, IL, USA). Sociodemographic, environmental and clinical characteristics were analysed using descriptive statistics. Bivariate analysis, using binary logistic regression, was initially performed in order to determine factors associated with S. aureus colonization. Multivariate logistic regression was then used to assess independent associations for variables significant at P< 0.1. Variables at P<0.05 were then considered statistically significant in the multivariable analysis.

Ethical considerations

The study procedures were in accordance with the Helsinki Declaration. The study protocol was approved by the AHRI/ALERT Ethical Review Committee (AAERC) (PO/017/2015). An 0fficial permission letter was obtained from the ALERT hospital. A written informed assent and consent was obtained from study participants, and parents or guardians of children, respectively, before including them in the study. The study participants’ right to refuse or not give nasopharyngeal samples without affecting their routine medical services was granted. Samples were coded to keep the confidentiality of the study participants' personal information.

Results

Socio-demographic and clinical characteristics of the study participants

Out of the potentially eligible 750 HIV-infected children, nasopharyngeal samples were collected from a total of 183 HIV-infected children at the ALERT hospital. Among the study participants, 50.8 %(93/183) were girls; the mean age of the study participants was 10.78±2.68 years. The median [interquartile range (IQR)] CD4 +T cell count of the study participants was 73(524-924) cells/mm³ and 95.6 %(175/183) of the children have received full dose of PCV10 (Table 1).

Table 1.

Socio-demographic, environmental and clinical characteristics and their association with S. aureus colonization among HIV-infected children aged <15 years at ALERT hospital, Addis Ababa, Ethiopia

Variable	Category	All children (n=183)	Colonized with S. aureus (n=50)	Bivariate analysis
		No. (%)	No. (%)	Cor (95 % CI)	P§
Gender	Male	90 (49.2)	25 (50)	1.05 (0.55–2.01)	0.892
	Female	93 (50.8)	25 (50)	Ref
Age category (in years)	0–4	3 (1.6)	2 (4)	5.18 (0.45–58.75)	1.86
	5–9	51 (27.8)	12 (24)	0.8 (0.37,1.69)	.550
	10–14	129 (70.4)	36 (72)	Ref
Number of household members	<=3	37 (20.2)	11 (22)	0.65 (0.24–1.76)	.396
	4–6	113 (61.7)	26 (52)	0.41 (0.17–0.97)	.065
	>=7	33 (18)	13 (26)	Ref
Number of rooms in the house	1	69 (37.7)	16 (32)	0.71 (0.36–1.41)	0.330
	>=2	114 (62.3)	34 (68)	Ref
Siblings aged <5 years	Yes	68 (37.2)	13 (26)	0.5 (0.24–1.03)	0.58
	No	115 (62.8)	37 (74)	Ref
Mother's level of education	Illiterate	68 (37.2)	22 (44)	2.1 (0.18–24.59)	0.555
	First level	72 (39.3)	18 (36)	1.25(0.12–13)	0.852
	Secondary	39 (21.3)	9 (18)	1 (0.09–10.23)	1.000
	College and above	4 (2.2)	1 (2)	Ref
Father's level of education	Illiterate	60 (32.8)	19 (38)	1.1 (0.26–5.03)	0.860
	First level	58 (31.7)	13 (26)	0.82 (0.2–3.27)	0.773
	Secondary	53 (29)	14 (28)	0.58 (0.15–2.23)	0.426
	College and above	12 (6.6)	4 (8)	Ref
Monthly income (in Ethiopian Birr)	<1000	108 (59)	31 (62)	1.61 (0.43–6.10)	0.483
	1000–2500	43 (23.5)	11 (22)	1.38 (0.33–5.8)	0.664
	2500–3500	17(9.3)	5(10)	1.67(0.32–8.6)	0.541
	>3500	15 (8.2)	3 (6)	Ref
Exposure to cigarette smoking	Yes	7 (3.8)	1 (2)	0.37 (0.04–3.1)	0.355
	No	176 (96.2)	49 (98)	Ref
Reason for hospital visit	ART follow-up	85 (46.6)	26 (52)	1.1 (02–6.05)	0.911
	Asthma	1 (.5)	1 (2)	–	1.000
	Dermatological problems	18 (9.8)	7 (14)	1.59 (0.24–10.56)	0.63
	LRTI symptoms	72 (39.3)	14 (28)	0.6 (0.11–3.44)	0.570
	UTI	7 (3.8)	2 (4)	Ref
Antiretroviral treatment status	Follow-up	149 (81.4)	41 (82)	0.9 (0.39–2.09)	0.812
	Initiation	34 (18.6)	9 (18)	Ref
PCV10 vaccine status	Full dose†	175 (95.6)	45 (90)	3.58 (0.92–13.93)	0.065
	Incomplete‡	8 (4.4)	5 (10)	Ref
S. pneumoniae carriage	Yes	79 (43.2)	15 (30)	0.46 (0.23–0.93)	0.029
	No	104 (56.8)	35 (70)	Ref
URTI in the last 3 months	Yes	25 (13.7)	8 (16)	1.81 (0.61–5.4)	0.495
	No	158 (86.3)	42 (84)	Ref
LRTI in the last 3 months	Yes	6 (3.3)	2 (4)	1.34 (0.24–7.58)	0.738
	No	177 (96.7)	48 (96)	Ref
CD4+ T cell count (cells/mm3)	>350	164 (89.6)	46 (92)	0.73 (0.23–2.34)	0.600
	<350	19 (10.4)	4 (8)	Ref
Previous hospitalization in the last 3 months	Yes	6 (3.3)	47 (94)	2.77 (0.54–14.18)	0.223
	No	177 (96.7)	3 (6)	Ref

*LRTI symptoms represent (cough, difficulty in breathing and runny nose).

†Full dose PCV10 indicates all three doses.

‡Incomplete PCV10 vaccination includes ≤ 2 doses.

§P value of binary logistic regression analysis, Values in bold: P < 0.05.

CAP, community-acquired pneumonia; CI, 95% confidence interval; COR, Crude odds ratio; LRTI, lower respiratory tract infection; PCV, pneumococcal conjugate vaccine; URTI, upper respiratory tract infection; UTI, urinary tract infection.

Prevalence of S. aureus and S. pneumoniae colonization

The overall prevalence of S. aureus nasopharyngeal colonization among the HIV-infected children in this study was 27.3 %(50/183) whereas. S. pneumoniae was isolated from 43.2 %(79/183) of the children (Table 1).

Prevalence of MRSA colonization and detection of mecA

mecA was detected in 10 %(5/50) of the S. aureus isolates and MRSA colonization rate was therefore 2.7 %(5/183).

Risk factors associated with S. aureus colonization

According to multivariable logistic regression analysis, there was a statistically significant association between S. aureus and S. pneumoniae nasopharyngeal colonization [aOR, 0.49; CI, (0.24–0.99); P=0.046] (Table 2).

Table 2.

Risk factors associated with S. aureus colonization among HIV-infected children aged <15 years at ALERT hospital, Addis Ababa, Ethiopia

	Bivariate		Multivariable
Variable	cOR (Cl)	P*	aOR (CI)	P*
PCV10 vaccine status	3.58 (0.92–13.93)	0.065	2.81 (0.7–11.34)	0.148
S. pneumoniae carriage	0.46 (0.23–0.93)	0.029	0.49 (0.24–0.99)	0.046
Number of household members (4–6)	0.46 (0.2–1.05)	0.065	0.49 (0.21, 1.15)	0.104

^∗ P value of binary logistic regression analysis. Values in bold: P<0.05.

aOR, adjusted odds ratio; CI, 95% confidence interval; COR, Crude odds ratio; PCV, pneumococcal conjugate vaccine.

Antibiotic susceptibility pattern of S. aureus isolates

Antimicrobial susceptibility patterns were determined for all the 50  S. aureus isolates. The highest rate of resistance was seen against tetracycline (32 %) followed by erythromycin (24 %) and trimethoprim-sulfamethoxazole (24 %) (Fig. 1).

Fig. 1.

Antimicrobial susceptibility pattern of S. aureus isolates among HIV-infected children aged <15 years at ALERT hospital, Addis Ababa, Ethiopia.

Among all MRSA isolates the highest percentage of drug resistance was seen against tetracycline (35.2 %), erythromycin (29.4 %), clindamycin (29.4 %), and gentamicin (11.7 %), sulfamethoxazole-trimethoprim (5.9 %), chloramphenicol (5.9 %) and ciprofloxacin (5.9 %). The overall multidrug-resistance (MDR) rate in this study was 12 %.

Discussion

Nasopharyngeal colonization and infection with S. aureus and MRSA is higher in HIV-infected children than those without HIV [22, 23]. Determining the S. aureus carriage rate and antibiotic resistance profiles is important in order to identify risk factors associated with S. aureus infection.

In the current study, we found that 5–7 years after introduction of PCV10 in Ethiopia, prevalence of S. aureus nasopharyngeal carriage was 27.3 %(50/183). Our result is in agreement with a similar study done in Northern Ethiopia 29 %(88/300) [19]. Similar prevalence has also been reported from South Africa, prior to the introduction of PCVs 25.6 %(91/355) [24]. Our finding was however lower than a report from results from a study in Brazil (45.16 %) before the introduction of PCVs [23]. This might be because of geographical differences, sample size, the use of antiretroviral drugs and the introduction of PCVs. Our results were however higher than S. aureus nasopharyngeal carriage prevalence among HIV-uninfected under-5 children reported from Uganda, 19.4 %(144/742) [25] and Ghana, 23.2 %(95/210) [26]. This might be due to the difference in the age group as most (70 %) of the children in the current study are aged 10–14 years.

Our finding indicates that the prevalence of MRSA nasopharyngeal colonization in HIV-infected children was 2.7 %(5/183). Our finding is similar to nasal MRSA colonization reported among children and adults in Northern Ethiopia 2.4 %(6/249) [27]. Our findings were however lower than the pharyngeal colonization reported in a similar group of HIV-infected children in Northern Ethiopia 9.7 %(29/300) [19]. In the study by Mulu et al. and colleagues, methicillin resistance was determined using cefoxitin discs whereas in the current study, we used mecA amplification to determine methicillin resistance. Although Cefoxitin resistance is known as a reliable surrogate marker for mecA-mediated methicillin resistance [28], there are studies that have questioned its accuracy [29]. Additional reasons for discrepancies in prevalence of MRSA could be differences in study locations and periods, differences in the implementation of infection prevention among the hospitals and rational antibiotic use practices in the two settings. In addition, in both studies, it is important to note that the reported prevalences are of the study cohorts and do not necessarily represent the overall situation in the country.

In this study, nasopharyngeal carriage of S. pneumoniae was 43.2 %(79/183). The results are higher than pharyngeal carriage reported among similar group of children in Northern Ethiopia 31(10.3 %) [27]. The study by Mulu and colleagues from Northern Ethiopia reports pharyngeal carriage and samples might not have been taken from the nasopharynx. Our results were however lower than those reported from the same hospital as the current study at a later time point (2018–2019), 52 %(26/50) [30]. The differences in carriage prevalence might be due to differences in the median ages of the study populations, which was the 53.5 months' range (24–89) in the study by Lemma and colleagues while in the current study the median age is the 132 months' range (36–168). Our results were also similar to those reported among HIV-uninfected children of a similar age range in Southern Ethiopia [31].

In multivariable analyses, there was an inverse association between S. aureus and S . pneumoniae nasopharyngeal colonization in this group of HIV-infected children 5 to 7 years after introduction of PCV10 in the country. Accordingly HIV-infected children who are colonized with S. pneumoniae are less likely to be colonized with S. aureus . A negative association between carriage of S. aureus and PCV7 vaccine type S. pneumoniae was first reported close to two decades ago by two studies from Israel [32] and the Netherlands [33] before the introduction of pneumococcal conjugate vaccines. Since then, other studies from different parts of the world have reported similar results [34, 35] mainly among healthy or HIV-uninfected children. However, studies among HIV-infected children from South Africa indicated lack of association between the nasopharyngeal colonization of S. aureus and S. pneumoniae [14, 24]. The possible reasons for the lack of competition between the two bacteria in HIV-uninfected children included reduced mucosal immunity and therefore decreased immunological pressure and increased exposure to respiratory pathogens [24]. Other authors however suggested that a secondary host-related mechanism associated mainly with CD4⁺ T cells might play an important role in the pathway of interaction between S. pneumoniae and S. aureus [36]. The studies from South Africa did not study the correlation between CD4⁺ T cell counts and bacterial colonization or interaction. In the current study, the median (IQR) CD4 +T cell count of the study participants was 732 (524-924) cells/mm³.

According to the Ethiopian national guideline for HIV prevention, care and treatment, for all children living with HIV, ART is given as early as possible regardless of their WHO clinical stages and CD4⁺ T cell counts/percentage [37]. Studies indicate that early HIV diagnosis and early ART reduce infant mortality and HIV progression [38]. In addition, in HIV-infected infants with early ART treatment CD4⁺ T cell counts stabilize at the highest levels possible [39]. We therefore hypothesize the inverse interaction between S. aureus and S. pneumoniae seen in the HIV-infected children in this study, similar to previous reports on the inverse interaction between S. aureus and S. pneumoniae in HIV-uninfected children, is due to the impact of ART treatment.

People living with HIV are at increased risk of acquisition and infection with drug-resistant bacteria [6]. In this study, among both MSSA and MRSA isolates, the highest level of resistance was seen against tetracycline, which was consistent with a similar study among HIV-infected children in Northern Ethiopia [40] although a higher percentage of resistance to tetracycline among MRSA isolates (72 %) compared to our study (35.2 %) were reported. The low rates of resistance to most of the antibiotics tested observed in this study might suggest that most of the MSSA and MRSA were community-acquired strains [41].

Our study had some limitations. Firstly, prevalence of S. aureus colonization might have been underestimated since we sampled only the nasopharynx. Since our results originate from only one hospital, the results might not be fully representative of children in the whole of Addis Ababa or Ethiopia. Because serotyping was not performed for the S. pneumoniae isolates, we could not determine the relationship between vaccine type and non-vaccine type S. pneumoniae colonization and S. aureus colonization. In addition, our focus was only on colonization and the study was not designed to identify disease caused by these pathogens.

Conclusions

The results of this study suggest that 5 to 7 years after introduction of PCV10 in Ethiopia and more than 12 years after introduction of ART in Ethiopia, the nasopharyngeal colonization rate of S. aureus in HIV-infected children at ALERT hospital, Addis Ababa, Ethiopia was 27.3 and 10 % of the S. aureus isolates were MRSA. In addition, there was an inverse relationship between the colonization of S. aureus and S. pneumoniae .