Antimicrobial use among paediatric inpatients in a Nigerian tertiary hospital: A three-year point prevalence survey

Abstract

Aim

Antimicrobials are among the most widely prescribed therapeutic agents among paediatric population. Irrational use of these agents increases the risk of antimicrobial resistance. We described the prevalence and pattern of antimicrobial use among paediatric inpatients from 2017 to 2019.

Subject and Method

The study was a repeated point prevalence survey over a 3-year period using the global point prevalence standardized tools among all paediatric inpatients. The prevalence of antimicrobial use, the prescription indicators and patterns of antimicrobial use were estimated.

Result

Among 191 paediatric inpatients assessed, the 3-year period prevalence antimicrobial use was 85.9% (164/191) with prevalence of 80.6% in 2017, 94.6% in 2018, and 83.6% in 2019. Antimicrobial agents used ranged from one agent (20.1%) to five different agents (5.5%). Parenteral route (66.6%) was the preferred route of administration. The reason for the use of antimicrobial agents (92.6%) and the stop/review date (99.5%) were mostly well documented. Only 4.5% of the antimicrobial use were targeted. There were no antimicrobial guidelines or policy guiding the use of antimicrobial agents, except the national guideline on the treatment of malaria and tuberculosis. Ceftriaxone, a third generation cephalosporin was the most commonly used agent across the period under review. Community-acquired infection was the commonest indication for antimicrobial use.

Conclusion

Our findings of high antimicrobial prevalence has raised the attention for the need to develop hospital-based antimicrobial guideline and antimicrobial stewardship program to protect the vulnerable children, their contacts and the environment from the impact of antimicrobial resistance.

Background

Antimicrobial agents are prescribed for the purpose of treatment or as preventive measures and among the most widely prescribed therapeutic agents. An ideal antimicrobial use is targeted at the particular causative organism of a disease, and this is best determined through a microbiological test. However, in situations of limited diagnostic capabilities, empirical prescription of antimicrobials becomes an alternative and is best practiced within a given guideline (De Waele and Dhaese 2019).

Antibiotics can be a lifesaving treatment for children with bacterial infections and are the most commonly prescribed therapy among all medications given to children (Rogawski et al., 2017). However, the wrong use of these agents is a source of concern as a public health threat. It could become a risk factor for the development of resistant bacteria, reduce treatment effectiveness, increase risk of adverse events and lead to extra costs for both hospitals and patients (Chitashvili et al., 2017). Irrational use of antibiotics is widespread across healthcare settings and this includes use of antibiotics without clinical indications, treatment of a health condition with an antibiotic not recommended for that condition and incorrect dosing or route of administration (Chitashvili et al., 2017).

Antimicrobial use in children differs from that in adults due to a number of reasons such as a lack of data on pharmacokinetics, pharmacodynamics, efficacy, safety of drugs in children as well as the varying physiological spectrum among the different age groups in children (preterm neonates, full-term neonates, infants and toddlers, older children and adolescents). More so, children are more vulnerable to the majority of the diseases due to lower immunity to these microorganisms and the adverse effect of inappropriate use of antimicrobials than adults.

One of the best ways to monitor antimicrobial prescription in hospitals is through electronic monitoring; however, this technology is still not available in many hospitals in low- and middle- income countries. Prevalence survey is a useful alternative tool for this purpose and allows identifying areas for quality improvement (Zarb et al., 2011). Studies from various regions of the world showed varying patterns of antimicrobial prescription in paediatric populations in terms of prevalence of use, type of antimicrobials, indications and mode of administration (Labi et al., 2018; Versporten et al., 2015). Differences in practice of antimicrobial use in children around the world reflect variations in local medication policies, in barriers to access to care and preferences of healthcare providers and caregivers (Rogawski et al., 2017). Inappropriate antimicrobial use for childhood health conditions is more pronounced in low- and middle- income countries (Chitashvili et al., 2017).

In Nigeria, there is limited data on antimicrobial use among paediatric in-patients, and without such information, it is difficult to develop strategies for the infection prevention and containment of resistant pathogens in this population. We therefore described the prevalence of antimicrobial use among paediatric inpatients in a tertiary hospital and the patterns of antimicrobial prescription from 2017 to 2019.

Methods

Study area

The survey was conducted in the paediatric department of a tertiary hospital in Southeast Nigeria. The hospital is 720 bedded (Umeokonkwo et al., 2019). It has a fully functional paediatric department with different subspecialties including adolescent, cardiology, children emergency, endocrinology, gastroenterology, haemato-oncology, infectious disease, neonatology, neuorology and respiratory/community paediatric units. There is also a paediatric surgical unit. It runs speciality clinics and children outpatient clinics from Mondays to Fridays. The paediatric inpatient bed capacity for paediatric unit was 130. The department also operates the Institute of Child Health that provide immunization services and community health services to children in schools and communities. In addition to clinical services, the department provides platform for training of postgraduate doctors in different specialities in paediatrics and conducts clinical, biomedical and operations research.

Study population and sampling technique

The study included all paediatric inpatients admitted in the hospital on the days of the survey irrespective of the reason for the admission. Those admitted after 8.00 am on the day of the survey were excluded. It was a total population survey. All eligible inpatients were included in the survey.

Study tool and duration of survey

This was a repeated point prevalence survey over a 3-year period 2017–2019 using the global point prevalence standardized tools. The tools – patient form and ward form – were used to capture standardized sets of patients’ demographic information (age, sex, weight), indication for each antimicrobial encounter, quality indicators for antimicrobial use and surveillance on some key antimicrobial resistant organisms.

Statistical analysis

Data captured on the forms were validated online for completeness and logical consistencies using the global point prevalence online data management platform. The antimicrobial point prevalence was estimated for the period and yearly. The antimicrobial prescription indicators were estimated for the period.

Results

A total of 191 paediatric inpatients were assessed during the period with 377 antimicrobial encounters. The 3-year period prevalence of antimicrobial use among paediatric inpatients was 85.9% (164/191). The prevalence rose from 80.6% in 2017, peaked at 94.6% in 2018 with a slight decline to 83.6% in 2019. The trend differed across the wards but, remained at 100% across the period in the newborn intensive care unit (Figure 1).

Figure 1.

Prevalence of antimicrobial use in the different paediatric wards.

The number of antimicrobial agents used ranged from one agent (20.1%, 33/164) to five different agents (5.5%, 9/164). Majority of the patients (84.0%) were on two antimicrobial agents. Parenteral route (66.6%) was the preferred route of administration of antimicrobial agents among the paediatric inpatients. The reason for the use of antimicrobial agents (92.6%) and the stop/review date (99.5%) were mostly well documented in the patient case notes. Only 4.5% of the antimicrobial use were targeted (i.e. guided by laboratory diagnosis of infection and sensitivity testing), the rest was based on empirical use. There were no departmental level or hospital level antimicrobial guidelines or policy guiding the use of antimicrobial agents. The only guideline in use was the national guideline on the treatment of malaria and tuberculosis (Table 1). Ceftriaxone, a third generation cephalosporin was the most common antimicrobial agent used among the paediatric inpatients across the period under review. This was followed by metronidazole in 2017 and 2019, while sultamicillin was the second most commonly prescribed antimicrobial in 2018 (Table 2). Community-acquired infection (64.7%) was the commonest indication for antimicrobial use among the group, and this was followed by medical prophylaxis (16.7%) and surgical prophylaxis (12.5%). Hospital-acquired infection accounted for only 5.6% of the antimicrobial use among the patients.

Table 1.

Antimicrobial prescription indicators among paediatric in-patients.

Prescription indicator	Frequency	Percent
Reason in note
Yes	349	92.6
No	28	7.4
Stop/review date
Yes	375	99.5
No	2	0.5
Guideline compliance
Yes	16	4.3
No	14	3.7
NA	347	92.0
Route
Oral	126	33.4
Parenteral	251	66.6
Treatment
Empirical	360	95.5
Targeted	17	4.5

NA = Not available.

Table 2.

Common antimicrobial agents used among paediatric inpatients.

2017		2018		2019
Agent	Frequency (%)	Agent	Frequency (%)	Agent	Frequency (%)
Ceftriaxone	24 (26.7)	Ceftriaxone	26 (19.1)	Ceftriaxone	33 (21.9)
Metronidazole	14 (15.6)	Sultamicillin	20 (14.7)	Metronidazole	18 (11.9)
Ampicillin and enzyme inhibitor	9 (10.0)	Ceftazidime	14 (10.3)	Ceftazidime	17 (11.3)
Ceftazidime	6 (6.7)	Artemether	11 (8.1)	Sultamicillin	15 (9.9)
Ciprofloxacin	5 (5.6)	Metronidazole	10 (7.4)	Amoxicillin and enzyme inhibitor	9 (6.0)

Discussion

The antimicrobial point prevalence among paediatric inpatients was high throughout the period of the review. This could be due to the setting and type of the cases that commonly present in the tropics which are mainly infectious diseases which are commonly treated with antibiotics. However, the level is comparable to those reported in studies in Ethiopia (Kebede et al., 2017), South Africa (Koopmans et al., 2018) and Georgia (Korinteli et al., 2019) but higher than observed in other studies (Bharathiraja et al., 2005; Choudhury and Bezbaruah 2013; Majhi et al., 2017; Pradeepkumar et al., 2017). In Italy, lower antibiotic usage of 52% was reported in paediatric populations under the age of 15 years (Resi et al., 2003). The absence of antimicrobial stewardship program in the hospital could also be a factor that led to unrestricted use of antimicrobial agents. As noted in the result, there was no departmental or hospital antibiotic policy nor guideline specifying how and when antibiotics were indicated, leaving the use of the antibiotics to the discretion of the clinician. The unrestricted use could also be responsible for the high prevalence and might also lead to higher risk of antimicrobial resistance.

The majority of the indications for use of antimicrobial agent were due to community-acquired infections. This is expected in the tropics since infectious diseases are still the most common cause of morbidity and mortality. However, it is worrisome that there is high use of antibiotics for medical prophylaxis, including the neonatal intensive care units especially since majority of the therapeutic antimicrobial use were not targeted (based on laboratory diagnosis and sensitivity testing) nor were they guided by a guideline. This calls for the need to develop a holistic antimicrobial guideline based on antibiotic sensitivity pattern in the setting. Similar high proportion of antimicrobial use due to community-acquired infection had earlier been reported (Versporten et al., 2015). The high empirical use of antibiotics has been reported in previous studies in Nigeria (Efunshile et al., 2019; Fowotade et al., 2020; Ogunleye et al., 2019) and elsewhere (Versporten et al., 2018) despite availability of laboratory support. The hospital has a functional laboratory support for diagnosis of infection despite low utilization observed in the study. A policy on antibiotic use will ensure that appropriate steps are taken to properly diagnose suspected infections and institute guided treatment based on the laboratory investigation results. Porta et al. reported that in London hospitals, antibiotics were most commonly prescribed in paediatrics for surgical prophylaxis (21%) and sepsis or suspected sepsis (66%) in neonates (Porta et al., 2012).

The greater proportion of the paediatric inpatients was receiving two or more antimicrobial agents with a maximum of five antimicrobial agents in one patient. A maximum of four agents used in a patient has been reported earlier (Choudhury and Bezbaruah 2013). In this study, the most common agent prescribed among the patients was third generation cephalosporins. Similar findings have been reported (Bharathiraja et al., 2005; Choudhury and Bezbaruah 2013; Majhi et al., 2017; Versporten et al., 2015). In some studies in Europe, the most commonly prescribed antibiotics in paediatric population were beta-lactam antibiotics including cephalosporins (Porta et al., 2012; Quaak et al., 2018; Resi et al., 2003). The preference for third generation cephalosporin as the most common agent for treatment of community-acquired infections is worrisome especially as the use is neither based on sensitivity pattern in the setting nor on any established antimicrobial protocol guiding its use. The common justification is often that the patient might have taken other common classes of antibiotics which are easily accessed over-the-counter in an unrestricted manner. This, however, could have been validated by ensuring that microbiological diagnosis is done, and sensitivity of the organism determined to guide use of the relevant antimicrobial. This is worrisome because a high use of such broad spectrum antibiotics has been associated with rapid development of antimicrobial resistance (Khan et al., 2018). The preference for the broad spectrum antibiotics could also be linked to the empirical use of antibiotics. Broad spectrum antibiotics are often used because the clinician is neither guided by the specific bacteriological diagnosis nor by the sensitivity pattern of the organism, so the clinician prefers broad spectrum agent in order to ensure wider range of organisms are taken care of. This could also explain the polypharmacy where the clinician prescribes multiple antibiotics to cover a broad range of organisms suspected to be involved in the suspected infection. Establishing an antimicrobial stewardship program together with having a functional antibiotic policy for the hospital which ensures efforts are made to properly diagnose all suspected infections and follow a well-documented guideline in the use of antibiotics will ensure that the antibiotics are used rationally.

Apart from the empirical use of antimicrobial and preference for the use of parenteral route over oral route of administration, the antimicrobial practice observed among clinicians was good. There was good documentation of reason for antimicrobial use and indication of date for stop or review of the antimicrobial. These practices have been found to be suboptimal in some previous studies (Labi et al., 2018; Versporten et al., 2018). These are commendable and if combined with use of laboratory support in making diagnosis and the antibiotic stewardship program will ensure improved antibiotic use.

The major limitation of this study was that the outcome of care was not documented due to the study design. However, the study called attention to the need to encourage development of antimicrobial guideline to protect the vulnerable children from the impact of antimicrobial resistance.

ORCID iDs

Chukwuma D Umeokonkwo https://orcid.org/0000-0002-0050-2874

Cosmas K Onah https://orcid.org/0000-0002-3315-7851

Azuka S Adeke https://orcid.org/0000-0002-6737-0378

Adanna A Umeokonkwo https://orcid.org/0000-0003-2518-7425

Oyinlola O Oduyebo https://orcid.org/0000-0002-2894-4367