Bronchiectasis in African children: Prevalence, etiology, and clinical spectrum at a pediatric tertiary hospital in Cape Town, South Africa

INTRODUCTION

Bronchiectasis is a respiratory disease characterized by irreversible airway damage.^[1] The cardinal features of bronchiectasis include stasis of infected airway secretions, regional or diffuse airway wall dilation, thickening, and destruction with loss of structural integrity. The associated prolonged neutrophilic inflammatory and secretory response within the airways cause a chronic or recurrent “wet” cough in young children and mucopurulent sputum expectoration in older children. This clinical phenotype, coupled with radiological findings of airway wall thickening and dilatation on chest computerized tomography (CT), is the criteria for making a diagnosis of bronchiectasis.^[2-4]

Bronchiectasis lies at the end of a continuum of suppurative airway inflammation that ranges from an inconsequential protracted bacterial bronchitis to chronic suppuration that leads to irreversible airway damage and results from different disease processes or insults.^[5,6] These include diseases that impair mucociliary clearance (e.g., primary ciliary dyskinesia [PCD] and cystic fibrosis [CF]), altered immune responses (e.g., primary or acquired immunodeficiency, chronic infections), structural abnormalities of the airways (e.g., bronchomalacia, trachea-esophageal fistula, and bronchomegaly), or disease associated with systemic inflammation. Similarly, events that result in chronic inflammation and airway damage, such as recurrent lower respiratory tract illness, foreign body aspiration, or recurrent aspiration, may cause bronchiectasis.^[7]

The global burden of bronchiectasis has increased over the past decades despite improved living conditions and access to antibiotics and medical facilities.^[8] Further, it has now been recognized that there is a global increase in the prevalence of bronchiectasis with associated morbidity and mortality, especially in children living with social deprivation.^[9-11] The global prevalence of bronchiectasis is reported to range from 0.7/1000 to 14.7/1000 in children <15 years of age.^[10] The true global pediatric prevalence is difficult to determine due to diagnostic criteria needing chest CT, which is not easily accessible in low- to middle-income settings. Hence, there remains a paucity of data from these settings and minimal data from Africa despite the high burden of pediatric chronic lung disease.^[9]

Management of bronchiectasis in our setting mirrors global guidelines^[12] which include airway clearance, sputum surveillance, treatment of acute exacerbations with antibiotics, and regular follow-up at our pediatric respiratory outpatient clinic.

Current evidence suggests that early diagnosis and treatment improves outcomes. However, delayed diagnosis remains common and mainly arises from low awareness of bronchiectasis among clinicians and limited access to diagnostic tools.^[13,14] Further, there is evidence that the majority of adult lung disease has its origins in childhood.^[15,16] It is hence important that we improve our knowledge of bronchiectasis in African children, where risk factors for disease are common and may differ from other settings. This would facilitate the development of effective prevention, diagnostic, and management strategies with the potential to improve the trajectory of the disease.

This study aimed to describe the burden of disease, etiology, and clinical spectrum of bronchiectasis, not including CF, in children attending a tertiary pediatric respiratory service at Red Cross War Memorial Children’s Hospital (RCWMCH) in Cape Town, South Africa.

MATERIALS AND METHODS

RESULTS

Disease burden

There were a total of 627 clinic attendances (including repeat visits) for the period January to December 2019. A total of 337 patients attended the clinic. A total of 58 patients met the inclusion criteria for the study, giving a clinic prevalence for bronchiectasis of 17.2%, as shown in Figure 1. The diagnosis of bronchiectasis was based on chest radiographs for 30 (53.4%) and on CT in 28 (46.6%) participants, as shown in as shown in Table 1.

Close

Figure 1:

The protocol that was used to determine the disease burden in the respiratory clinic. A total of 337 patients attended the clinic between January and December 2019 and 58 had bronchiectasis, amounting to 17.2% disease burden.

Export to PPT

Table 1: Demographic and anthropometric data of cohort (n=58).

Patient characteristic
Sex (number, %)
Female	32 (55.2)
Male	26 (44.8)
Age (months) at enrolment (Mean, SD)	92.2 (41.0)
Males	95.5 (47.9)
Females	89.3 (34.7)
Age (months) at diagnosis (Mean, SD)	34.2 (26.3)
Males	34.7 (31.6)
Females	33.9 (22.8)
HIV status (number, %)
HIV unexposed and uninfected	36 (62.0)
Positive	16 (28.0)
HEU	6 (10.0)
Gestation age at birth in weeks (mean SD)	37.5 (1.3)
Term (≥37 weeks)	53 (91.4)
Preterm (Below 37 weeks)	8 (13.8)
Received oxygen therapy at birth (number, %)	5 (9)
WHZ*, Mean (SD)	0.0 (1.6)
HAZ*, Mean (SD)	1.2 (1.6)
Number of chest infections ever, mean (SD)	2 (1.2)
Number of admissions ever, mean (SD)	2 (0.9)
Radiology basis for bronchiectasis diagnosis (number, %)
Chest X-ray	30 (51.7)
HRCT	28 (48.3)

*Calculated from WHO percentile charts, HIV: Human immunodeficiency virus, HEU: HIV Exposed Uninfected, WHZ: Weight for height z-score, HAZ: Height for age z score, HRCT: High-resolution computerized tomography

DISCUSSION

Bronchiectasis is an important cause of chronic lung disease in children. We found 17.2% of all patients attending our tertiary respiratory clinic with bronchiectasis. The most common cause of bronchiectasis in our study population was post-infectious; all children had a history of recurrent LRTIs. The most common cause of LRTI was TB and adenovirus, which differed by HIV status. This highlights the importance of strengthening diagnostic pathways for children at risk of bronchiectasis and pneumonia preventive strategies for all children. Chronic respiratory symptoms were common, with mean lung function of the children severely impaired over the study period. This study adds to the knowledge of bronchiectasis in African children.

Bronchiectasis prevalence was 17% in our tertiary respiratory clinic in the setting of a high HIV endemicity, with nearly 30% of all children in the clinic with HIV infection. A study in Zimbabwe of 84 adolescents living with HIV with abnormal HRCT, with a median duration of ART of 4.7 years, had a higher bronchiectasis prevalence of 33%.^[20] Studies in high-income countries, in contrast, found a much lower prevalence, ranging from 0.2–2.3/100,000 in European countries and 1.5/100,000 in New Zealand people of European descent. In disadvantaged and indigenous communities in New Zealand, however, the prevalence was higher: 15/100,000 and 17.8–18.3/100,000 population in Pacific islanders.^[9] There are no comparable population-level statistics in low-middle income settings with high HIV prevalence; however, we estimate that the high burden in our clinic population (172/1000 clinic attendees) reflects both the selected population of children with CLD and the setting of high-risk factors for bronchiectasis, notably TB, HIV, and high infective exposure. Community prevalence data from South Africa are needed.

Although the most common cause of bronchiectasis in our participants was attributed to post-infectious causes, nearly a quarter of our children had immune deficiencies, mainly arising from perinatally acquired HIV. These children all had a history of recurrent or severe LRTI. TB is endemic in South Africa, with 322 cases (95% CI 230–428) per 100,000 population^[21] and not surprising that it was the most common cause identified in children living with HIV (68.8%) and 19.3% in HIV uninfected children, highlighting the need for TB prevention. Adenovirus has been found to cause long-term sequelae, including bronchiectasis, following respiratory infections as a single infection or coinfection with other organisms.^[22,23] This is similar to our findings, where adenovirus was isolated as a single organism as well as a coinfection. The most common cause of bronchiectasis among indigenous poor communities in Australia was reported to be post-infectious/idiopathic (94.5%),^[24] similar to our results although our cohort was hospital-based. Other common causes, such as GORD, PID, and congenital or acquired airway anatomical abnormalities, are antecedents to recurrent respiratory infections, which may further result in bronchiectasis. PCD and PID have been recognized to be common among societies with a high frequency of consanguinity, as described in Turkey,^[25] while the incidence of CF, and PCD are unknown in many low- or middle-income countries (LMIC) due to limited diagnostic capacity.^[1,24] Determination of the etiology of bronchiectasis is important and key to the successful management of the disease.

Our children were younger at diagnosis (mean age 34 months) compared to children in New Zealand (median 5.4 years).^[10] Unlike in our cohort, where the cause was mainly infections, diagnosis of bronchiectasis in infancy has been associated with congenital lung malformation and other genetic diseases like PCD.^[6,26] The diagnosis of bronchiectasis in our cohort was made early, and this demonstrates that children who are at risk of bronchiectasis may present early and should be screened for bronchiectasis.

Similar to previous studies, chronic wet cough was the most common symptom.^[1] Other researchers have reported 28– 100% chronic wet cough in children from LMICs.^[1] Cough is the earliest and most important symptom to recognize and investigate in children with suspected bronchiectasis.^[9] Not recognizing this causes delays in diagnosis and treatment and has a negative impact on prognosis, increasing spending on national health budgets and family expenditures. Increasing productive cough is a sign of exacerbation and requires escalation of management interventions. Therefore, strengthening prevention, early diagnosis, and care has the potential to mitigate these untoward effects.

Wheeze was present in a quarter (27.6%) of our cohort, which was similar to findings in Tunisia (20%); the United Kingdom reported a lower prevalence of wheezing (10%), and Alaskan children had a high prevalence of wheezing (41%) which was attributed to RSV hospitalization.^[10] In our cohort, RSV was isolated as a coinfection in two HIV-negative children. The finding of wheezing is variable among different studies and needs further evaluation to determine the true prevalence and the underlying risks.

Features of severe bronchiectasis in children may include clinical features such as exercise intolerance, frequent exacerbations, hemoptysis, frequent attendances at clinics, failure to thrive, reduced lung function, use of home oxygen, and presence of finger clubbing.^[1,24,27] These features were not predominant in our cohort. Early diagnosis in our cohort and access to appropriate care may be responsible for this.

Diffuse or multilobar disease on the baseline chest radiograph was predominant compared to focal disease, which is sometimes used as a marker of disease severity. Focal disease may be associated with foreign body aspiration, though our cohort did not have that history. Despite this preponderance of diffuse disease on the baseline chest radiograph, our cohort did not have predominant severe disease. Chest radiograph changes in our patients were persistent, and thus, chest CT was not offered. In addition, it is not always feasible to recommend chest CT for all patients to diagnose bronchiectasis in LMICs as is recommended in high-income countries. This may have limited our ability to diagnose all patients with bronchiectasis. We, however, highlight the need for earlier diagnosis to prevent this extent of disease, and chest CT plays an important role in the detection of early disease and can facilitate targeted investigations and treatment.

Sputum surveillance results had few children with H. influenzae or H. parainfluenza, which were the predominant organisms in LMIC settings in children living with HIV.^[28] H. influenzae accounted for 47% of children having exacerbations in Australia.^[24] TB was isolated on surveillance sputum done in 2019 and can easily cause opportunistic infections in patients with bronchiectasis. Our cohort was on prophylactic azithromycin, which may be responsible for low bacterial load in sputum.^[9,29] BrookeHollidge et al. reported that low pathogen load in sputum was associated with the mildest form of the disease.^[30]

Although the mean z-FEV₁ and z-FVC showed severe dysfunction of >−3SD, our cohort showed lung function improvement between baseline and last test, probably due to physiological growth and improved health care as all children were being regularly followed up in a tertiary hospital in Cape Town. This ensured the children could get regular airway clearance, sputum surveillance, and prompt treatment of exacerbations.

This is one of the first studies to describe the etiology and clinical spectrum of bronchiectasis in South African children attending a general respiratory clinic. The strength included access to comprehensive clinical information and investigations. Limitations of the study include being a retrospective study; we relied on data from clinical records. Although these were available for all children and had been collected by the same team of experienced pulmonologists, there may be reporting bias. Further, there could have been a selection bias of children with severe disease attending a specialist respiratory clinic in a tertiary facility. Further, the research was conducted in a tertiary center and, so unlikely to reflect the general population. Further data are needed to describe disease burden, etiology, and clinical spectrum at the community level.

CONCLUSION

Bronchiectasis is common in our tertiary respiratory clinic setting and an important cause of lung morbidity in children. Post-infectious causes are the most predominant due to the high burden of early respiratory childhood infections in this setting. Post-TB bronchiectasis in children, including those living with HIV, is an important cause in TB-endemic settings and is the leading cause of bronchiectasis in children living with HIV. Prevention, early diagnosis, and management of TB in children can prevent the development of bronchiectasis. Clinical features of bronchiectasis vary depending on the extent of the disease. Chronic wet cough remains the hallmark clinical feature of bronchiectasis and any such cough in children should be evaluated further to facilitate early diagnosis and early intervention.