Lower Airway Infection in Preschool Children with Cystic Fibrosis: An International Comparison.

To the Editor:

We read with interest the paper by Breuer and colleagues published in the Journal in September 2019, which demonstrated a declining prevalence of Pseudomonas aeruginosa, Haemophilus influenzae, and Staphylococcus aureus lower airway infection in children with cystic fibrosis (CF) over an 18-year period (1). Like the authors, we have been conducting BAL surveillance in preschool children with CF at three of the six specialist CF centers in Ireland, as part of SHIELD CF (The Study of Host Immunity and Early Lung Disease in Cystic Fibrosis). Here, we present data on 335 BAL samples from 110 children with CF that were collected from 2010 to 2018. We also find a reassuring reduction in the prevalence of P. aeruginosa infection in BAL from Irish children over time.

There are, however, two clear differences between our data and those published by Breuer and colleagues that warrant discussion. Although the prevalence of infection with P. aeruginosa we observe does not differ significantly from that reported by Breuer and colleagues, S. aureus and H. influenzae prevalence is significantly higher and Aspergillus prevalence is significantly lower in our cohort (Table 1). A closer look at the data (Table 2) reveals striking differences in the prevalence of S. aureus and H. influenzae in the first 2 years that decreases with age, and an increasing difference in the rates of infection with Aspergillus, which only becomes significant in the older cohort.

Table 1.

Prevalence of Pathogenic Organisms in the Lower Airways of Preschool Children with Cystic Fibrosis in Irish Cystic Fibrosis Centers, 2010–2018

	2010–2012 (n = 87)	2013–2015 (n = 149)	2016–2018 (n = 99)	2012–2018 (n = 285)	Breuer et al. (1) 2012–2018 (n = 866)	P Value
P. aeruginosa	3 (3.4)	8 (5.4)	1 (1)	10 (3.5)	51 (5.9)	0.11
P. aeruginosa PCR	4/65 (6.2)	16/101 (15.8)	2/80 (2.5)	19 (8.9)	NA	NA
S. aureus	14 (16.1)	35 (23.5)	24 (24.2)	66 (23.2)	78 (9.1)	<0.001
H. influenzae	30 (34.5)	40 (26.8)	25 (25.3)	74 (26)	72 (8.3)	<0.001
Aspergillus sp.	5 (5.7)	4 (2.7)	4 (4)	9 (3.2)	97 (11.2)	<0.001

Definition of abbreviations: H. influenzae = Haemophilus influenzae; NA = not available; P. aeruginosa = Pseudomonas aeruginosa; S. aureus = Staphylococcus aureus.

Comparison is made with data from Breuer and colleagues (1). Data are shown as n (%).

Table 2.

Prevalence of Pathogenic Organisms in Irish and Australian Cohorts (2012–2018) according to Age at the Time of Sampling

	0–2 Years			3–4 Years			5–6 Years
	Ireland (n = 122)	Australia (n = 304)	P value	Ireland (n = 117)	Australia (n = 236)	P value	Ireland (n = 96)	Australia (n = 326)	P value
P. aeruginosa	4 (3.3)	13 (4.3)	0.63	3 (2.6)	14 (5.9)	0.16	5 (5.2)	24 (7.4)	0.06
S. aureus	20 (16.4)	13 (4.3)	<0.001	24 (20.5)	25 (10.6)	0.01	29 (30.2)	40 (12.3)	0.08
H. influenzae	32 (26.2)	4 (1.3)	<0.001	38 (32.5)	23 (9.7)	<0.001	25 (26.0)	45 (13.8)	0.6
Aspergillus sp.	1 (0.8)	11 (3.6)	0.14	5 (4.3)	26 (11.0)	0.03	7 (7.3)	60 (18.4)	<0.001

Definition of abbreviations: H. influenzae = Haemophilus influenzae; P. aeruginosa = Pseudomonas aeruginosa; S. aureus = Staphylococcus aureus.

Comparison is made with data from Breuer and colleagues (1). Data are shown as n (%).

These data are somewhat surprising at first sight given the seemingly homogeneous approach to treatment of young children with CF and, indeed, the processing of airway specimens in the clinical laboratories of specialist centers in the developed world. Like our Australian colleagues, we approach airway clearance and treatment of intercurrent infection aggressively in this young cohort. We also routinely treat our patients <2 years of age with antistaphylococcal prophylaxis—in our case, flucloxacillin as opposed to co-amoxiclav. We routinely use co-amoxiclav for treatment of low-grade exacerbations throughout childhood and also use azithromycin prophylaxis in selected children in this age group. We use 28 days of inhaled tobramycin for eradication of P. aeruginosa and only use oral or intravenous antipseudomonal therapy if this fails on more than one occasion.

The Australian data show much lower rates of infection with S. aureus and H. influenzae in the first 2 years of life. The fact that this difference decreases with age suggests that it may decline as a result of antibiotic prophylaxis and that co-amoxiclav is more effective than flucloxacillin in this regard. The difference in prevalence of Aspergillus species between the two datasets is the opposite of what is seen with S. aureus and H. influenzae, where an increasing difference is evident over time. This may be related to cumulative antibiotic exposure, and although direct comparisons of total antibiotic exposures are not possible, the description of antibiotic use in the paper by Breuer and colleagues would suggest that more antibiotics are used overall in Australian children than in Irish children.

Another potential factor in the differing prevalence of airway pathogens is the local environment. Airway P. aeruginosa infection was previously shown to be linked to environmental conditions (2). It is possible that different environmental conditions in Ireland and Australia have an effect on the prevalence of Aspergillus and that treatment of suspected airway P. aeruginosa infection might contribute to this (3), particularly as this seems to be more aggressive in the Australian cohort.

Independently of the etiology of the significant differences in lower airway infection between our countries, these data raise the question of which is the greater evil, a higher prevalence of bacteria or a higher prevalence of Aspergillus? This can only be understood with further clinical research and underlines the importance of structured longitudinal clinical research programs and international comparisons and collaboration in CF.