Atopic eczema is the most common paediatric skin disease in developed countries with a considerable impact on children’s quality of life.1 It is a chronically relapsing dermatitis, and dry skin, not only in lesional areas but also elsewhere, is a cardinal feature. It is thought that xerosis is primarily caused by a loss of integrity in the stratum corneum, resulting in an increase in transepidermal water loss (TEWL).2 The key gene involved in skin barrier function is the filaggrin gene (FLG), which is found in the epidermal differentiation complex on chromosome 1q21.3 Filaggrin is an important component of the granular cell layer of the epidermis. It aggregates keratin filaments, leading to keratinocyte compaction and formation of the stratum corneum.2 Up to 50% of patients in the U.K. with moderate to severe eczema are heterozygous for one of the FLG loss-of-function mutations, and these findings have been replicated in a number of white European populations.4 However, it is currently unclear whether skin barrier impairment and an increase in TEWL precedes phenotypic eczema in FLG mutation carriers, or whether it is primarily an epiphenomenon of disease activity. It is also not yet established at what age FLG mutations manifest clinically, for instance through an increase in TEWL and dry skin, as well as eczematous skin inflammation.5
Background Filaggrin loss-of-function (FLG) mutations are associated with eczema and skin barrier impairment, but it is unclear whether skin barrier impairment precedes phenotypic eczema in FLG mutation carriers.
Objectives To study the association between FLG mutations, skin barrier impairment and clinical eczema at 3 months of age.
Methods A total of 88 infants were examined for eczema. Disease severity was determined by the SCORAD eczema severity score. Transepidermal water loss (TEWL) was measured on unaffected forearm skin. Venous blood samples were screened for the four most common FLG mutations found in the U.K. white population (R501X, 2282del4, R2447X and S3247X). Median SCORAD and TEWL measurements in children with and without eczema and FLG mutations were compared.
Results Thirty-three per cent (29/88) of children had clinical eczema. Median SCORAD was 10·6 (range 3·5–31·0). TEWL (g m−2 h−1) was higher in children with eczema compared with unaffected infants (median TEWL 14·24 vs. 11·24, P < 0·001). Higher TEWL was associated with more severe disease (r = 0·59, P < 0·001, median TEWL, SCORAD < 15, 13·1 vs. 29·6, SCORAD ≥ 15, P = 0·029). Clinically dry skin was associated with higher TEWL, even in the absence of eczema (median TEWL 17·55 vs. 11·08, P = 0·008). Seventeen per cent (15/88) of children carried at least one FLG mutation. FLG mutation carriers were significantly more likely to have clinically dry skin, even in the absence of eczema [odds ratio (OR) 8·50, 95% confidence interval (CI) 1·09–66·58, P = 0·042]. FLG mutation carriers were also more likely to have eczema by 3 months of age (OR 4·26, 95% CI 1·34–13·57, P = 0·014). FLG mutations were significantly associated with higher median TEWL (all children, FLG‘yes’ 21·59 vs. FLG‘no’ 11·24, P < 0·001), even without clinical eczema (FLG‘yes’ 15·99 vs. FLG‘no’ 10·82, P = 0·01).
Conclusions By the age of 3 months, FLG mutations are associated with an eczema phenotype, dry skin and TEWL. The observation that TEWL is elevated in unaffected FLG mutation carriers suggests that skin barrier impairment precedes clinical eczema.
Materials and methods
A total of 88 three-month-old infants were recruited from the general population between October 2009 and January 2010 as part of an interventional birth cohort study, which involves randomization to either the introduction of six potentially allergenic foods plus breastfeeding or exclusive breastfeeding until 6 months (http://www.eatstudy.co.uk). Following written consent and prior to the start of the study intervention, children were examined for clinical eczema using the U.K. diagnostic criteria-based photographic protocol of the International Study of Asthma and Allergies in Childhood (ISAAC) Phase Two.6 Disease severity was determined by the Scoring Atopic Dermatitis (SCORAD) index.7 TEWL was measured with the Biox Aquaflux® AF200 (Biox Systems Ltd., London, U.K.) closed condenser chamber device on the unaffected skin of the volar aspect of the forearm.8 Participants’ parents were advised not to use any skin-care products on the infants’ arms for the preceding 24 h. Measurements were performed in our environmentally controlled clinical research facility (ambient temperature 20 ± 2 °C, relative room humidity 32–50%), after at least 20 min of acclimatization. Measurements were not taken if the child was visibly distressed or crying. In all children we calculated the mean of three separate TEWL measurements. Venous blood samples were screened for the four most common FLG mutations found in the U.K. white population (R501X, 2282del4, R2447X and S3247X), using the TaqMan allelic discrimination assay (ABI 7900 HT; Applied Biosystems, Foster City, CA, U.S.A.). We used the Mann–Whitney U-test and Spearman correlation coefficients to compare median SCORAD and TEWL measurements in children with and without eczema and with and without FLG mutations, using SPSS version 17 (SPSS, Chicago, IL, U.S.A.). Where possible, we also calculated odds ratio (OR) estimates and corresponding 95% confidence intervals (CI).
Thirty-three per cent (29/88) of the children had clinical eczema, with a median SCORAD of 10·6 (range 3·5–31·0). Twenty-three of the 29 children (79%) had a SCORAD < 15 and six (21%) had a SCORAD ≥ 15. Median TEWL (g m−2 h−1) in infants with neither eczema nor xerosis was 11·42 [interquartile range (IQR) 9·81, 13·88]. TEWL in children with eczema was significantly higher compared with unaffected children [14·24 (IQR 11·25, 21·78) vs. 11·24 (IQR 9·80, 14·29), P < 0·001]. Higher TEWL was associated with more severe disease [r = 0·59, P < 0·001, median TEWL, SCORAD < 15, 13·1 (IQR 11·1, 17·2) vs. 29·6 (IQR 18·2, 43·0), SCORAD ≥ 15, P = 0·029] (Fig. 1). Clinically dry skin was associated with higher TEWL, even in the absence of eczema [median TEWL 17·55 (IQR 13·95, 24·99) vs. 11·08 (IQR 9·73, 13·26), P = 0·008]. Seventeen per cent (15/88) of the children carried at least one FLG mutation (Table 1). As described in other U.K. white populations, R501X and 2282del4 were the commonest FLG mutations seen (four of 15 and eight of 15, respectively). Fourteen out of the 15 FLG mutation carriers were heterozygotes, one child was homozygote for 2282del4, and nine out of the 15 mutation carriers had clinical eczema at 3 months. FLG mutation carriers were significantly more likely to have clinically dry skin, even in the absence of eczema (OR 8·50, 95% CI 1·09–66·58, P = 0·042). FLG mutation carriers were also more likely to have eczema by 3 months of age (OR 4·26, 95% CI 1·34–13·57, P = 0·014), and this tended to be more severe, although the difference between FLG mutation and wild-type carriers was small [median SCORAD FLG‘yes’ 12·5 (IQR 10·7, 18·1) vs. FLG‘no’ 7·5 (IQR 7·0, 14·0), P = 0·016]. In addition, FLG mutations were significantly associated with higher median TEWL [children with eczema, FLG‘yes’ 29·06 (IQR 18·85, 35·46) vs. FLG‘no’ 12·21 (IQR 10·84, 16·32), P = 0·003; children without eczema, FLG‘yes’ 15·99 (IQR 11·57, 30·22) vs. FLG‘no’ 10·82 (IQR 9·62, 13·26), P = 0·01] (Fig. 2).
|FLG mutation||FLG null heterozygotes||Eczema at 3 months||FLG null homozygotes||Eczema at 3 months||Total|
To the best of our knowledge, this is the first report on the association between FLG mutations, skin barrier impairment (TEWL) and eczema in infants. While Nikolovski et al.9 have previously suggested that the skin barrier of infants is less mature than that of older children and adults, our findings suggest that unaffected 3-month-old infants without the FLG mutation lose water at rates similar to those found in older children, with little intersubject variability.9,10 The differences between our study and that of Nikolovski et al. may be explained by the fact that their population was pooled from different cohorts and FLG status was not assessed.
As expected, both FLG mutations and skin barrier impairment (TEWL) were significantly associated with dry skin, eczema phenotype and severity. However, others have not found such associations among older children and adults and, where present, these tended to be weaker.11–13 The differences between our and other studies are likely to be due to differences in study design, such as the fact that we recruited young infants from a population base, which resulted in our children having only mild to moderate eczema, rather than comparing a highly selective group of much older patients with eczema with unaffected controls. In addition, not all studies tested for the same number of FLG mutations. In this context, the use of different TEWL machines may also be important, as for instance open chamber devices are more prone to environmental influences, such as changes in ambient temperature and room humidity, compared with the closed condenser chamber device we used.8
FLG mutation carriers had a more than four times higher risk of developing eczema by 3 months of age compared with children without FLG mutations, and this has also been suggested previously by a number of cross-sectional studies in older children, which enquired about age of eczema onset.14,15
Most importantly, TEWL was not only elevated in children with eczema but also in unaffected FLG mutation carriers. This suggests that skin barrier impairment precedes clinical eczema, rather than being purely a secondary phenomenon associated with eczematous skin inflammation. Further follow-up of these infants is needed and is currently underway to confirm this hypothesis. Should skin barrier impairment precede clinical eczema, then this would provide an important therapeutic target for eczema prevention and potentially even respiratory allergies, as there also seems to be a link between FLG mutations, allergic sensitization and asthma.16,17
What’s already known about this topic?
- • FLG mutation carriage and eczema phenotype as well as severity are significantly associated, based on cross-sectional evidence from studies among older children and adults.
- • It is currently unclear whether skin barrier impairment and an increase in transepidermal water loss (TEWL) precede phenotypic eczema in FLG mutation carriers, or whether it is primarily an epiphenomenon of disease activity.
- • It is also not yet established at what age FLG mutations manifest clinically, for instance through an increase in TEWL and dry skin, as well as eczematous skin inflammation.
What does this study add?
- • FLG mutation carriage was positively associated with early-onset eczema at 3 months of age and eczema severity (SCORAD).
- • TEWL was even increased in FLG mutation carriers without clinical eczema, compared with wild-type infants.
- • This suggests that skin barrier impairment precedes clinical eczema, rather than being purely a secondary phenomenon associated with eczematous skin inflammation.
We would like to thank the parents and children of the EAT Study for taking part. We are also grateful to Prof. Michael Cork (The Academic Unit of Biomedical Genetics-Dermatology, University of Sheffield, U.K.) for scientific advice in the run-up to this study.