The Association between Atopic Disorders and Keloids: A Case-control Study.

BACKGROUND: Keloids and atopic disorders share common inducing and maintaining inflammatory pathways that are characterized by T-helper cell 2 cytokines. AIMS AND
OBJECTIVES: The objective of this study was to test for associations between keloids and atopic eczema, asthma and hay fever.
MATERIALS AND METHODS: This was a case-control study with 131 patients diagnosed with keloids at our dermatology outpatient clinic between 2000 and 2012. Controls were 258 partners of keloid or sarcoidosis patients. Patient who reported life time prevalences of atopic eczema, asthma and hay fever were assessed using a questionnaire based on The European Community Respiratory Health Survey (ECRHS) and The International Study of Asthma and Allergies in Children (ISAAC).
RESULTS: The prevalence of asthma was lower in keloid patients (19/131 vs. 20/258, P = 0.035), as was being diagnosed with asthma by a physician (18/131 vs. 19/258, P = 0.039) and using inhalators for asthma (13/131 vs. 7/258, P = 0.02). After adjusting for age and non-European descent the odds ratio for having a keloid was (adjusted OR = 4.44; 95% CI 1.59-12.40) in asthmatics using inhalators. There were no clear and consistent associations found for keloids with atopic eczema or with hay fever.
CONCLUSION: In conclusion, our study shows that keloids may be strongly associated with atopic asthma. Atopic eczema and hay fever do not seem to be correlated with keloid. Further studies are warranted to assess the validity of atopic asthma as a risk factor for the development of keloid scars.

What was known?

Keloids are common but no modifiable risk factors are known.

Introduction

Keloids are fibroproliferative dermal tumors resulting from skin trauma and subsequent abnormal scar formation.[1] Besides being a cosmetic problem, keloids are often pruritic and painful. Therefore, patients with keloids may suffer from physical impairments and a decreased health-related quality of life.[2] The life-time prevalence of keloids has been reported to be around 10% in Africans, whereas keloids are rarely seen in individuals of European genetic ancestry.[3] Besides the ancestry and young age no other risk factors for the development of keloids are known. To date, there are no effective treatments and there are no satisfactory means of prevention. On the cellular level, keloids are characterized by an overproduction of extracellular matrix and a high activity of fibroblasts during the inflammatory and remodelling phases of scar formation.[45] Recent experimental studies have demonstrated that keloid development is driven by three dysfunctional pathways: Altered apoptosis, altered transcription of growth factors such as transforming growth factor beta (TGF-β) and vascular endothelial growth factor (VEGF) along with deregulation of the immune system. Evidence for the involvement of the immune system in keloid formation is found in the increased numbers of mast cells in the edges of keloids and the high CD4:CD8 tissue T-lymphocyte ratio.[6] The key players in this inflammatory pathway are the T-helper 2 (Th-2) cytokines interleukin IL-13 and IL-4, which induce collagen production by the fibroblasts.[7]

Upregulation of the Th-2 cytokines is also strongly linked to atopy. Atopic disorders include atopic eczema, hay fever and asthma. Atopy results from a defect in the epidermal barrier function and a hyperactive, Th-2 skewed, immune response with increased serum IgE levels.[8] The main cytokines driving the atopic inflammation are also IL-13 and IL-4. Thus, cytokine milieus in keloids and atopic disorders appear to be similar, and therefore, patients with atopic disorders may have an increased risk of developing keloids. The objective of this case-control study was to investigate whether atopy is a risk factor for the development of keloids.

Materials and Methods

Study design and study population

This case-control study was conducted at our Department of Dermatology. Cases were patients diagnosed with keloids between 2000 and 2012 at our dermatology outpatient clinic. The diagnosis of keloid was made by a dermatologist on the basis of clinical findings of a raised tumorous scar extending beyond the wound borders. The partners of the keloid patients were asked to serve as controls. In addition, available data from a similar postal survey were also included. From this data-set, partners of patients diagnosed with sarcoidosis at our hospital [Figure 1] were included as controls. This additional control group of partners of sarcoidosis patients was considered eligible because the data collection was similar and demographic characteristics ethnic background, residence, age and sex were comparable to the keloid group. Controls were excluded if they had had keloid. Both groups were approached by a postal questionnaire. The non-responders were contacted by telephone and sent a second mailing. The Local Medical Ethical Committee waived the board review because the study was single-subject and questionnaire based.

Figure 1

Study population

Patient reported atopic features

Atopic status was established through a validated questionnaire comprising questions from the European Community Respiratory Health Survey (ECRHS) and International Study of Asthma and Allergies in Children (ISAAC) protocol.[910] Patients were considered to have atopic eczema, asthma or hay fever if they reported having one of these conditions. However, the diagnoses were further explored by inclusion of questions on physician-made diagnoses and physician prescribed medication. Thus, patients reporting eczema were questioned about the use of topical corticosteroids, asthma patients were asked about the use of inhalators, and patients with hay fever were asked about their use of anti-histamine drugs. Patients were also questioned about the key-symptoms of these conditions. Family history was positive for these disorders if one first-degree relative had been reported to have either eczema, asthma or hay fever.

Ancestry

Non-European ancestry and darker skin phenotypes are important determinants of the risk of keloids.[1] The ancestry was established by documenting the birth country of both parents and self-identification by the subject. Patients were classified to be either of African, Asian or European ancestry. If one of the parents was born in an African or Asian country, the subject was considered to be of African or of Asian descent, respectively. Considering that all ethnicities other than the European confer a higher risk of keloids, we dichotomized the ancestry to European or non-European in the final multivariate model.

Keloid properties

Patients were asked to report the number of keloids and anatomical location of their keloids. The modes of induction of the keloid were predefined: Surgery, acne, spontaneous, piercing, burns and other. Patients also graded on a visual analogue scale (VAS) on pain and pruritus symptoms caused by their keloids. Both symptoms highly correlate with physical impairment and loss of quality of life.[2] In order to further assess the symptom severity and its effect on quality of life, we used the validated Skindex-29 questionnaire to measure dermatology-specific health-related quality of life.[11] These measurements were collected in order to test for a correlation between the keloid severity and the presence of atopy.

Statistical analysis

For sample size calculations, we assumed a prevalence of 10% for patient-reported asthma and 14% for patient reported hay fever in our control group, as found previously.[12] This would make up for an estimated prevalence of about 20% of any atopic disorder in controls. We expected at least twice this percentage in keloid patients.[13] For a power of 80% and an alpha level of 0.05, a sample size of 126 cases and 252 controls was required. The Chi-squared test was used to test for statistical differences in proportions and the Mann Whitney-U test for non-parametric testing. Multivariate logistic regression models were used to calculate adjusted odds ratios (OR) and 95% confidence intervals (95% CI). Ancestry was considered a confounder in the possible association between atopy and keloids. Therefore we also performed analyses with stratification by ancestry. In the multivariate analysis, atopic eczema, asthma and hay fever were analyzed separately. However, additional analyses were done with a composite variable where subjects were considered atopic if they had reported either eczema, asthma or hay fever. No additive or multiplicative interactions were found between asthma and non-European descent. Values of P < 0.05 were considered significant. All data were analyzed using IBM SPSS version 19.0 (Chicago, Illinois, 2011).

Results

Patient characteristics

We identified 207 keloid patients through medical records diagnosed between 2000 and 2012. Of the 207 patients, 131 (response rate 63%) sent back the questionnaire. Eighty-one (81) partners of keloid patients also replied [Figure 1]. One hundred and seventy-seven (177) partners of sarcoidosis patients were added to the control group which totalled 258 controls. The 76 non-responder keloid patients had an average age of 42 years (SD 17) and 36% of them were male. The keloid patients were on average younger than the controls (41 vs. 48 years, P < 0.001) [Table 1]. The cases did not differ significantly from the controls in terms of sex distribution and a positive family history of atopy. However, the cases with keloid were significantly more frequently of non-European ancestry (81% vs. 51%, P < 0.001) [Table 1]. Keloids were most frequently found on the chest (44%) and on the back and the shoulders (16%). The most frequently self-reported cause of keloid was surgery (40%) followed by spontaneous that is unprovoked onset (17%).

Table 1

Subject characteristics

Atopic asthma

Both self-reported and physician-diagnosed asthma were more prevalent among keloid patients than in controls (15% vs. 8%, P = 0.035 and 14% vs. 7%, P = 0.039) respectively. This observation was confirmed by the increased use of inhalators in the keloid patients (10% vs. 3%, P = 0.002). This difference was observed in the European subjects but not in the non-Europeans.

In the univariate analysis only asthma was associated with keloids (crude OR 2.03;95% CI 1.04–3.96) [Table 2] but this association did not remain statistically significant in the model adjusted for age and ancestry (adjusted OR 1.91;95% CI 0.92–3.93). Being treated for asthma with inhalators was significantly associated with keloids in both the crude (crude OR 4.00; 95% CI 1.56–10.30) and the adjusted model (adjusted OR 4.44; 95% CI 1.59–12.40). Having any atopic feature also increased the risk of developing a keloid (crude OR 1.64; 95% CI 1.07–2.50). This positive association remained after adjusting for age and ancestry (adjusted OR 1.34; 95% CI 0.85–2.12) but was no longer statistically significant. In the stratified analysis of only European subjects, the effect of treated asthma remained significant after adjusting for age; patients on inhalants had a seven-fold increased odds of having a keloid (adjusted OR 7.08; 95% CI 2.00–25.25). The percentage of subjects reporting any atopic feature was higher in keloid patients (54% vs. 42%, P = 0.023) and having both hay fever and asthma was more frequently found in keloid patients (3% vs 8%, P = 0.034) [Table 3].

Table 2

Univariate and multivariate logistic regression analysis of keloid risk factors

Table 3

Atopic features in keloid patients and controls stratified according to ancestry

Atopic eczema

Of the atopic eczema related outcomes only flexural itching rash was significantly more frequently reported by the keloid patients compared to the controls (24% vs. 12%, P = 0.004) [Table 3]. Keloid patients reported having had eczema more frequently but this difference was only borderline significant (31% vs. 24%, P = 0.125). The prevalence of the physician made diagnosis of atopic eczema was similar in both groups (26% vs 23%, P = 0.568). No difference was found in the use of topical corticosteroids. Atopic eczema was not associated with keloids in the adjusted model (adjusted OR 1.27; 95% CI 0.76–2.13) [Table 2].

Hay fever and keloids

The prevalence of self-reported hay fever was similar in keloid patients and controls (33% vs. 29%, P = 0.463). The prevalence of physician diagnosed hay fever was higher in keloid patients (26% vs. 20%, P = 0.173) but this difference was borderline significant. Keloid patients reported having a congested runny nose without having a cold more frequently than the controls (51% vs. 39%, P = 0.019). Nevertheless, no association between hay fever and keloids was found in the multi multivariate model (adjusted OR 1.04 95% CI 0.63–1.17) [Table 2].

Atopic versus non-atopic keloid patients

The atopic keloid patients did not differ from the non-atopic keloid patients in terms of family history, ancestry, age, and age at onset of keloids [Table 4]. The visual analogue scale scores for itch (6.4 vs. 5.4 P = 0.615) and pain (1.2 vs. 0.9, P = 0.838) as well as the Skindex-29 scores (45 vs 45, P = 0.962) were similar across the groups.

Table 4

Comparison of atopic with non-atopic keloid patients

Discussion

Keloids and atopic disorders share common inducing and maintaining inflammatory pathways that are characterized by T-helper cell 2 (Th-2) cytokines. In this case control study, we assessed the association between atopic disorders and keloids. Of the three atopic disorders, asthma was significantly, consistently and strongly associated with keloids. However, there were no consistent associations found for keloids with atopic eczema or with hay fever. There seems to be no difference in clinical presentation and disease impact between atopic and non-atopic keloid patients. In addition, this study has confirmed the traditional risk factors young age and non-European ancestry.

Possible mechanism to explain the association between keloids and atopic asthma is the shared inflammatory (Th-2) pathway but also the activity of fibrocytes in the injured tissues which play a role in tissue remodelling.[1415] The Th-2 cytokine pathway may interact similarly with fibrocytes in asthma and keloids. The interleukins 4 and 13, which are archetypical Th-2 cytokines, are of paramount importance in atopic disorders, mostly so in atopic asthma.[1617] They are also known to play a pivotal role in fibrosis.[18] They augment collagen formation in lung fibroblasts in asthma and in conjunctival fibroblasts.[1920] In human skin, mechanical damage polarizes skin dendritic cells towards a Th-2 cytokine secreting profiles where IL-13 and IL-4 are known to induce chemotaxis and collagen production in fibroblasts.[212223] It is also possible to induce alpha-smooth muscle activity in skin fibroblasts through sensitization with allergens. This is mediated by TGF-β and IL-13 mainly secreted by eosinophils.[24]

Of all atopic disorder, asthma embodies the most inflammatory and the only one where remodelling of tissue plays a role. This may also explain the lack of an association of keloids with atopic eczema and hay fever. The common denominator of atopic asthma and keloids is that both disorders start out with an inflammatory phase followed by untameable and inadequate tissue remodelling. If our findings represent a true association, asthma could be considered a very important risk factor for keloids and could have clinical relevance as it helps stratify the risk better.

Two observational studies have suggested the existence of an association between keloids and atopy. In a small (N = 38) questionnaire-based case control study from 1987, the authors found a prevalence of 85% of atopic and allergic symptoms in keloid patients versus 49% in healthy controls.[13] Another study by the same group observed that 6 out of 18 (33%) keloid patients had serum IgE levels above 150 ng/mL.[25] This study was uncontrolled but the authors expected 5% of the patients to have increased IgE levels based on previous measurements. Both studies were conducted at the time when the concept of atopy was not yet fully developed.[1325] The authors hypothesized that increased mast cell activity in allergic individuals may predispose them to keloids.

Although our keloid patients did not report having eczema or hay fever more often, they did report symptoms compatible with these diagnoses (itchy flexural rash and congested or runny nose in the absence of common cold). This may reflect underdiagnosing of these conditions in keloid patients of whom most are immigrants. Patients with asthma may suffer less from this phenomenon because of the more severe symptomatology which may prompt them to seek medical attention. The retrospective study design is the main limitation of our study. In addition, the questionnaire-based reporting may have led to information bias with subsequent underdiagnosing of eczema and hay fever but this bias is of non-differential type. Finding a suitable control group is a challenge because keloids are mostly seen in non-Europeans. To our surprise, the vast majority of their partners were actually of European ancestry causing a mismatch in ethnicity between cases and controls. We suspected ancestry to be the greatest confounder in the association between atopy and keloids. The prevalence of asthma has been shown to be higher in immigrants from Suriname when compared to the Dutch in a study based on administrative data (5.9% vs. 4.2%, P < 0.001 for all ages).[26] The prevalence of dermatologist diagnosed atopic eczema is also higher in Caribbean immigrant children in the UK than in their white counterparts (16.3% vs. 8.7%, OR 2.1 95% CI 1.1–3.9).[27] As a way of a sensitivity, analysis we explored the distribution of atopic features across subjects of European and non-European descent in the total study population. No difference was observed in the distribution of reported asthma and eczema related features in our groups. Also, the ancestry stratified analysis showed that the association between keloid and asthma is strongest in European subjects completely excluding mixing of effects by more prevalent asthma in non-European keloid patients.

This is to our knowledge the first study to assess the association between atopic features and keloids with validated questionnaires in dermatologist-diagnosed keloid patients and comparing it to a control group. We found that keloids are strongly associated (4–5-fold increased odds) with physician treated atopic asthma. Further exploring of this association necessitates confirmation in a larger study population preferably in an ethnically homogenous group or with matched controls. The ascertainment of atopy should be accomplished with more robust methods such as serum IgE-measurements or skin prick testing.

What is new?

Atopic asthma may be an easily identifiable risk factor for keloids.