Inglês (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Enviar este artigo por email
Citado por SciELO 
Similares em
SciELO 
Permalink
Introduction
Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide, accounting for 6% of all deaths, according to the World Health Organization’s Global Health Estimates. This condition is increasingly recognized as a critical public health issue1.
COPD is a complex, multi-component disease characterized by persistent respiratory symptoms and airflow limitation that is not fully reversible1. However, there is increasing evidence that COPD encompasses more than just airflow obstruction, with systemic inflammation observed in patients that extends beyond the effects of cigarette smoke alone. It’s still unclear whether this systemic inflammation is the result of spillover from inflammation in the peripheral lung, leading to the release of inflammatory mediators into the circulation (meaning the central process is still in the lung) or whether it’s related to a comorbid condition that creates a pro-inflammatory state that affects the lung (meaning the main process here is the systemic inflammation itself). It is now known that this systemic inflammation leads to extrapulmonary manifestations and can also cause or exacerbate comorbid conditions2.
COPD has a significant impact on health status at all stages, even in milder diseases, and its extrapulmonary effects and comorbidities contribute to its burden2,3. More than 70% of patients have an additional chronic conditions and a further 40% have two or more, with prevalence increasing with age3-5.
Comorbidities have a major impact on healthcare costs, symptom burden, and quality of life and may contribute to reduced life expectancy. Some of these (such as cardiovascular disease, malnutrition, anemia, osteoporosis, and depression) are well recognised2-4. However, less known conditions associated with COPD also exist. Milkowska-Dymanowska et al., in their article published in 2015, discuss these under-recognized comorbidities of COPD and include “skin changes” as one of them5.
The skin is the largest organ in the human body and is a multifunctional unit that acts as a protective physical barrier, absorbing UV radiation, preventing the penetration of microorganisms and chemicals, and controlling the passage of water and electrolytes; it also has thermoregulatory, immunological, sensory and autonomic functions3.
Smokers are known to have accelerated skin aging with characteristic facial features such as prominent wrinkles, gaunt face, grey appearance of the skin, and a puffy complexion5,6. It can be postulated that the pathophysiology of both COPD and accelerated skin aging lies in the systemic inflammation caused by tobacco exposure. Interestingly, one study showed that facial wrinkling was not only associated with COPD risk, but its severity was also congruent with the extent of emphysema on CT scans6.
Majewski et al. compared the biophysical properties of the skin in non-smokers, healthy smokers, and COPD patients and found that there were significant differences in skin temperature, melanin index, sebum content, and hydration levels between COPD patients and healthy non-smokers. These differences correlated with the BODE index (a marker of COPD severity), showing that skin condition in COPD patients worsens as the disease progresses3.
Evidence on skin and/or nail changes in COPD are lacking and the available evidence is outdated. Data on the link between clubbing and COPD are also inconsistent, with some authors saying it’s a sign of COPD7 and others saying it should prompt a search for another diagnosis, such as lung cancer1.
An association between psoriasis or rosacea and COPD has also been described8,9 but further research is needed to confirm this.
The aim of this study is to investigate skin and nail changes in COPD and associated dermatological conditions, and whether these associations are independent of the presence of respiratory insufficiency and/or neoplastic disease.
Methods
Cross-sectional cohort study with prospective data collection of all patients admitted to the Internal Medicine ward of a tertiary hospital in Lisbon from January 2023 to June 2023, defining cases as patients with a diagnosis of cigarette smoke-induced COPD (ICD-10 criteria J40 to J44) and controls as patients without a diagnosis of COPD. Controls were selected by consecutive sampling until the number of controls matched the number of cases.
The following inclusion criteria were defined to establish a diagnosis of cigarette smoke-induced COPD:
− History of tobacco use (mandatory)
− Symptoms consistent with chronic bronchitis, such as chronic cough and/or chronic sputum production and/or dyspnea odds ratio (OR)
− Imaging evidence of emphysema in the absence of another most likely cause OR
− Spirometry showing a non-reversible obstructive pattern OR
− A well-established diagnosis of COPD, documented in medical records, with appropriate medication and follow-up.
All patients’ electronic medical records were reviewed and the following data were collected: sex, age, and medical history (tobacco use, dermatological diseases, malignancy, home oxygen therapy [HOT], home non-invasive ventilation [NIV]).
Patients were defined as having chronic respiratory insufficiency (CRI) if HOT was prescribed OR NIV was prescribed, excluding prescription due to diagnosis of obstructive sleep apnea OR they had an arterial blood gas sample consistent with CRI after stabilization.
After obtaining informed consent, all patients underwent a thorough dermatological examination by a dermatology resident to detect skin manifestations such as xerosis, malar exanthema, Thinker’s sign, tobacco stains, nail clubbing, Beau’s lines, Muehrcke’s lines, koilonychia, longitudinal melanonychia, onycholysis, onychomycosis, onychorrexis and paronychia. These were recorded as present or absent.
Descriptive statistical analysis was performed for numerical variables using measures of central tendency and dispersion, while frequencies and percentages were used for qualitative variables. Chi-square and Fisher’s exact tests were used to compare variables between cases and controls, as appropriate. For statistically significant variables identified in the first analysis, their OR were calculated and compared in a stratified analysis according to the presence of CRI and the presence of a neoplastic diagnosis, using the Mantel-Haenszel test where appropriate. Statistical significance was defined as a two-tailed p < 0.05. Statistical Package for the Social Sciences version 20.0 was used.
The study protocol was approved by the CHULN Ethics Committee, aligning with the Declaration of Helsinki.
Results
Demographic data
A total of 187 patients were included, of whom 94 were diagnosed with COPD (Table 1), 60 (63.8%) men and 34 (36.2%) women, with a mean age of 76.3 ± 12.2 years. The remaining 93 patients without a diagnosis of COPD, consisted of 59 men (63.4%) and 34 women (36.6%), with a mean age of 80.9 ± 10.3 years.
Clinical data
The most common skin lesions were xerosis, onychomycosis, and clubbing (Table 2). Xerosis, nicotine stains, and clubbing were statistically significantly associated with COPD. Onycholysis on the other hand was associated with the absence of a COPD diagnosis.
Table 2 Skin and nail signs in patients with and without COPD
| Skin signs | Without COPD (n = 93) (%) | With COPD (n = 94) (%) | p |
|---|---|---|---|
| Xerosis (n = 69) | 27 (29) | 42 (44.7) | 0.027† |
| Malar exanthema (n = 15) | 4 (4.3) | 11 (11.7) | 0.06† |
| Onychomycosis* (n = 47) | 24 (25.8) | 23 (24.5) | 0.833† |
| Onycholysis (n = 6) | 6 (6.5) | 0 (0) | 0.014‡ |
| Clubbing (n = 35) | 9 (9.7) | 26 (27.7) | 0.002† |
| Nicotine stains (n = 25) | 0 (0) | 25 (26.6) | 0.000† |
| Onychorrhexis (n = 33) | 15 (16.1) | 18 (19.1) | 0.702† |
| Koilonychia (n = 6) | 2 (2.2) | 4 (4.3) | 0.682‡ |
*Of toe nails.
†Qui-squared test.
‡Fisher’s exact test. COPD: chronic obstructive pulmonary disease.
In terms of diseases (Table 3), the most common were CRI and neoplasms, with both HOT and NIV reaching statistical significance and being associated with COPD.
Table 3 Other diseases found in inpatients with or without COPD
| Disease | Without COPD (n = 93) (%) | With COPD (n = 94) (%) | p |
|---|---|---|---|
| Psoriasis (n = 3) | 0 (0) | 3 (3.2) | 0.246 (Fisher) |
| Rosacea (n = 6) | 4 (4.3) | 2 (2.1) | 0.444 (Fisher) |
| Skin cancer (n = 1) | 1 (1.1) | 0 (0) | 0.497 (Fisher) |
| Eczema (n = 3) | 0 (0) | 3 (3.2) | 0.246 (Fisher) |
| Lung cancer (n = 12) | 5 (5.4) | 7 (7.4) | 0.577 (QQ) |
| Other cancers (n = 46) | 23 (24.7) | 23 (24.5) | 0.966 (QQ) |
| NIV (n = 15) | 3 (3.2) | 12 (12.8) | 0.016 (QQ) |
| HOT (n = 34) | 8 (8.6) | 26 (27.7) | 0.001 (QQ) |
| CRI (n = 43) | 11 (25.6) | 32 (74.4) | 0.000 (QQ) |
COPD: chronic obstructive pulmonary disease; NIV: non-invasive ventilation; HOT: home oxygen therapy; CRI: chronic respiratory insufficiency.
Stratified analysis
In the stratified analysis, the Mantel-Haenszel test showed that the association between clubbing and COPD is independent of the presence of respiratory insufficiency or neoplastic disease. On the other hand, the association between xerosis and COPD is dependent on the presence of a neoplastic disease. This test was not carried out on the variable “nicotine stains” because this nail sign is known to be associated with chronic tobacco smoking.
Discussion
In our study, out of the 94 patients with COPD, 63.8% were men, which is consistent with the current evidence of a 3:2 male-to-female ratio in the diagnosis of COPD10. The mean age of the COPD patients was 76.3 ± 12.2 years, whereas the mean age of the non-COPD patients was 80.9 ± 10.3 years, which may indicate that COPD patients have a slight tendency to require hospitalization at a younger age or may reflect a bias in the selection of control patients.
The present study found that xerosis (44.7%), clubbing (27.7%), and nicotine nail stains (26.6%) were common manifestations in patients with COPD and were statistically significantly associated with this disease. Xerosis has been previously described as being more prevalent in people with chronic lung disease, particularly emphysema11,12 Xerosis can be symptomatic and quite bothersome, manifesting primarily as pruritus. It is important for clinicians to be aware of this so that they can diagnose and manage it more effectively, advising skin moisturization, emollients, short showers with lukewarm water, and avoidance of skin irritants13.
As mentioned in the Introduction, clubbing has been described by some as being associated with COPD and by others as an alert for another diagnosis. This study found evidence to support clubbing as a common manifestation of cigarette smoke-induced COPD, independent of the presence of neoplastic disease.
We didn’t consider the association found between nicotine nail stains and cigarette smoke-induced COPD because these nail sign is known to be associated with chronic tobacco smoking14.
Onychorrexis was not found to be associated with COPD, although it has been described as such in other studies11.
Regarding skin and nail signs, there were some with significantly low numbers, such as onycholysis (3%) and koilonychia (3%), and some that were not observed in any patient such as the Thinker’s sign15, Beau’s lines, Muehrcke’s lines, longitudinal melanonychia and paronychia.
In concern to onycholysis and koilonychia, these low findings probably reflect an overall low prevalence of these nail changes in the population. There is currently no information on the prevalence of onycholysis or koilonychia in different age groups and sexes. This also means that the association found between onycholysis and lack of COPD diagnosis is probably spurious and should not be considered given the biological implausibility and the small sample size.
As for the other skin signs mentioned that were not found (Thinker’s sign, Beau’s lines, Muehrcke’s lines, longitudinal melanonychia, and paronychia), this may also reflect an overall low prevalence in the population. There is little to no evidence of the true prevalence of these conditions.
t was difficult to identify previous diagnoses of dermatological conditions (psoriasis, rosacea, eczema, skin cancer) from medical records, as most doctors don’t consider dermatological history to be worth mentioning on their records, especially if the patient is old. Patients were asked about a previous dermatological diagnosis, but due to their age, the information was not completely reliable. This may be a possible explanation for the low number of patients found with a previous diagnosis of chronic dermatological disease or skin cancer. However, our numbers are in line with the reported prevalence of psoriasis (0.5-1%)16, although a recently published article suggests that the prevalence of psoriasis in Portugal is much higher, about 4%17. Our numbers are also in line with the reported prevalence of rosacea (5%)18. This small sample size does not allow us to carry out credible statistical studies.
In this study, we aimed to clarify whether the skin and nail changes that were statistically significantly associated with COPD were related with CRI, to clarify whether this was something specific to COPD or a possible common finding in chronic respiratory diseases in general. Due to the above-mentioned conflicting evidence regarding clubbing and COPD, this study also aimed to clarify whether those skin changes were associated with the presence of a diagnosis of lung cancer and/or other cancer diagnoses. Stratified analysis using the Mantel-Haenszel test showed that the association between clubbing and COPD was independent of the presence of respiratory insufficiency or neoplastic disease. On the other hand, the association between xerosis and COPD was dependent on the presence of neoplastic disease (Tables 4 and 5). The pathophysiology behind COPD and skin or nail changes could be related to the previously mentioned systemic inflammation, which is known to be present in COPD patients. Whether this systemic inflammation is due to spillover from inflammation in the peripheral lung (caused by tobacco smoke), a parallel abnormality, or related to a comorbid disease that also affects the lung, the truth is that the presence of increased circulating cytokines, chemokines, and acute phase proteins may eventually trigger these skin and nail changes2. Further research is needed to clarify this issue.
Table 4 Stratified analysis of xerosis and clubbing by CRI
| Skin findings | OR | OR (with CRI) | OR (without CRI) | Mantel-haenszel |
|---|---|---|---|---|
| Xerosis | 1.97 (1.09-3.62), p = 0.027 | 1.20 (0.29-4.94), p = 1.000 | 2.25 (1.13-4.51), p = 0.032 | 1.99 (1.07-3.71), p = 0.043 |
| Clubbing | 3.57 (1.57-8.12), p = 0.002 | 1.50 (0.27-8.45), p = 1.000 | 4.38 (1.70-11.32), p = 0.002 | 3.39 (1.48-7.81), p = 0.005 |
OR: odds ratio; CRI: chronic respiratory insufficiency.
Table 5 Stratified analysis of xerosis and clubbing by the coexistence of neoplasia
| Skin findings | OR | OR (with ND) | OR (without ND) | Mantel-haenszel |
|---|---|---|---|---|
| Xerosis | 1.97 (1.08-3.62), p = 0.027 | 5.08 (1.50-17.24), p = 0.007 | 1.39 (0.68-2.81), p = 0.471 | 1.95 (1.07-3.56), p = 0.029 |
| Clubbing | 3.57 (1.57-8.12), p = 0.002 | 1.50 (0.23-9.76), p = 1.000 | 4.51 (1.77-11.46), p = 0.002 | 3.65 (1.60-8.33), p = 0.002 |
OR: odds ratio; ND: neoplastic disease.
Limitations
The present study has a few limitations.
One significant issue is that patients were seen in the hospital in an acute or decompensated phase (whether from COPD or another disease), which means that some findings need to be considered with caution, and others could not be assessed in this setting (such as the presence of central cyanosis).
Another important aspect concerns the study’s small sample size, which is always a limitation, but even more so when we wanted to evaluate the relationship between certain dermatological diseases with a low prevalence in the population and COPD.
To explore this issue further, another study should be designed in a Primary Care setting, which would probably allow a larger sample size and easier access to previously known diseases, while assessing patients in a stable phase.
Conclusion
The present study was carried out in a tertiary hospital in Lisbon, Portugal. Bearing in mind previously mentioned limitations; it still allows us to conclude that clubbing is a manifestation of COPD, independent of the presence of CRI or neoplastic disease. This means that clubbing in a COPD patient should not prompt suspicion of another diagnosis, as previously reported.
Xerosis is also a common manifestation of COPD, although the diagnosis of neoplastic disease seems to be a confounder in this association.
These skin and nail signs should be at the forefront of the clinicians’ minds when dealing with these patients and should be treated promptly if necessary.
The relationship between certain chronic dermatological conditions and COPD warrants further investigation.
