記住我
Abstract
Background: Joint replacement is an important surgery for replacing a damaged joint with prosthesis. Implants used for joint replacement are made of metal, plastic, and ceramic. Skin reactions, such as dermatitis, can occur due to a hypersensitivity to these external substances. Aims: The aim of this study was to find the relationship between joint replacement and dermatitis. Methods: A nationwide population-based retrospective cohort study was performed using the National Health Insurance Service Database of the Republic of Korea. A total of 40,218 patients who underwent joint replacement were enrolled as the operation group and 40,218 controls were also enrolled. A cox proportional hazard regression model, and Fine and Gray regression model were used to compare the risk of dermatitis between the two groups. Results: Dermatitis occurred in 9.2% of the operation group and 9.1% of the control group, and no statistical difference was observed between the two groups. According to the Cox proportional hazard regression model, and Fine and Gray regression model, the risk of dermatitis did not increase in the operation group compared to that in the control group. However, the risk of dermatitis increased 1.20-fold in the operation group compared to that in the control group aged <60 years according to the Fine and Gray regression model (95% confidence index (CI) = 1.05–1.37, P = 0.0008). Conversely, no difference in dermatitis risk was observed between the two groups aged ≥60 years. Conclusions: We found that the risk of dermatitis increased after joint replacement in those aged <60 years.
Keywords: Dermatitis, joint replacement, retrospective study
Introduction 
Joint replacement is an operation that replaces joints damaged by various factors, such as arthritis. Plastic, metal, and ceramic are used in artificial joints, and new materials such as polymeric nanocomposites and biomimetic composites are being studied recently.[1]
Allergic contact dermatitis is known as a type of T lymphocyte-mediated hypersensitivity reaction. Sensitization and elicitation periods are known to be the major phases of allergic contact dermatitis.[2] First exposure to an allergen causes sensitization, and then a re-exposure triggers the elicitation phase, causing various skin reactions such as erythema, itching, and blisters. Immunologic or hypersensitivity reaction to the prostheses may occur after joint replacement.[3] Allergic dermatitis caused by these prostheses was first reported in 1966.[4] Hypersensitivity reaction to the prostheses can cause a skin reaction in various forms such as eczematous dermatitis, urticaria, vasculitis, and erythema. In addition, extracutaneous complications such as joint pain, joint loosening, and implant failure may occur.[5]
Allergic complications that occur after joint replacement have been reported previously.[6],[7],[8] In a population-based study conducted by Wu et al.,[9] the risk of eczema increases in patients who have undergone joint replacement. However, a larger-scale cohort study has not been conducted yet, and no population-based study has been conducted in the Republic of Korea. This is the first large-scale cohort study in the Republic of Korea. This study aimed to investigate the relationship between joint replacement and dermatitis by considering various covariates and matching the control group with the operation group.
Subjects and Methods Data sources
This was a nationwide population-based retrospective cohort study using the National Health Insurance Service (NHIS) database of the Republic of Korea. The NHIS program obligatorily covered almost the entire (approximately 97%) Korean population (51 million).[10] The NHIS database contains all types of insurance claims, demographic data, and diagnosis according to the International Classification of Disease, tenth revision (ICD-10).[11] This study was approved by the institutional review board (IRB) of Hallym University Sacred Heart Hospital (IRB no. 2020-02-013). Informed consent from the subjects was waived because of the de-identification of the NHIS database that follows the guidelines of the Personal Data Protection Act.
Study design
This was a retrospective cohort study of patients who underwent joint replacement surgery (including hip, shoulder, knee, interphalangeal, elbow, wrist, ankle, and jaw joint replacement) from 1 January 2009, to 31 December 2018. The study was conducted on patients who were >20 years of age and had never been diagnosed with atopic dermatitis (ICD-10 diagnostic code L20) during the study period. The years 2009–2013 was set as the washout period. During this period, patients who underwent joint replacement or had been diagnosed with allergic contact dermatitis or unspecified dermatitis (ICD-10 diagnostic code L23, L30.9) were excluded. Moreover, patients diagnosed with dermatitis prior to joint replacement during the cohort study period were excluded. Thereafter, patients with missing variables or abnormalities in the follow-up period were excluded.
The control group included patients who had not undergone joint replacement in the same period. The control group was enrolled after 1:1 propensity score matching with the operation group according to the index date, age, sex, income, underlying disease (hypertension, diabetes, dyslipidemia, liver disease, cancer, chronic kidney disease, systemic lupus erythematosus (SLE), rheumatoid arthritis, allergic rhinitis, asthma), and Charlson comorbidity index (CCI) (0–2, ≥3).
The index date was set as the date of joint replacement in the operation group and the same date as the control group. The follow-up period was until the patient died or was diagnosed with dermatitis or the end of the study (31 December 2018). Diagnosis of dermatitis was defined as diagnosis of allergic contact dermatitis or unspecified dermatitis (ICD-10 L23, L30.9) at least three times after joint replacement [Figure 1].
Statistical analysis
The baseline characteristics were presented as mean and standard deviation for continuous variables and as number and percentage (%) for categorical variables. Student's t-test and Chi-squared test were used to analyse the difference between the operation and control groups. The incidence rate for eczema was determined by dividing the number of events with the total follow-up duration (person–years). The incidence rate graph was estimated using Kaplan–Meier curves and evaluated using the log-rank test and Gray's test for competing risk with death.[12] A cox proportional hazard regression analysis was used to determine the hazard ratio (HR) and 95% confidence interval (CI) for the incidence of dermatitis between the operation and control groups. The proportional hazard assumption was assessed by the Schoenfeld residuals test using the logarithm of cumulative hazard function based on Kaplan–Meier estimates. There was no interference with the assumption of proportional hazard risk over time. The multivariable-adjusted Cox regression model was used to estimate the adjusted HR after adjusting for various confounding factors including age, sex, income, underlying disease, and CCI. In addition, the Fine and Gray regression model for competing risk was used to consider competing risk with death in estimating the HR and 95% CI for the incidence of eczema.[12] A 1:1 propensity score matching was performed between the operation and control groups, to reduce the confounding by covariates. After matching, the standardized mean difference was used to evaluate the difference in covariates between the two groups.
The SAS software version 9.4 (SAS institute, Cary, NC) was used for data analysis, and P value <0.05 were considered statistically significant.
Results Flow of the study population
According to the NHIS database from 2009 to 2018, there were a total of 306,870 patients who received joint replacement, had not been diagnosed with atopic dermatitis, and were older than 19 years. After exclusion, a total of 65,313 patients remained. After 1:1 propensity score matching (index date, age, sex, income, underlying disease, CCI), 40,218 patients in the operation group and 40,218 patients in the control group were enrolled [Figure 2]. The standardized mean difference in the variables between the two groups was measured to be <0.01.
Baseline characteristics of the study population
The mean age of the operation and control groups was 69.0 ± 11.0 years and 69.1 ± 10.7 years, respectively. The proportion of men in the operation group was 27.4% and that of women was 69.3%, which was the same in the control group. No statistically significant differences in age, sex, income, underlying disease, and CCI were observed between the operation and control groups [Table 1]. The incidence of dermatitis was 9.2% in the operation group and 9.1% in the control group, and no statistically significant difference was observed between the two groups. The number of deaths in the operation group was higher than that in the control group (12.5% and 9.3%, respectively; P < 0.0001).
Association between joint replacement and dermatitis
During the study period, the dermatitis incidence rate (per 1,000 person–years) was 36.09 and 37.34 in the control and operation groups, respectively [Table 2]. [Figure 3] shows the result of comparison of the cumulative incidence for dermatitis between the control and operation groups using the Kaplan–Meier curve. No significant difference in the log-rank test and the Gray test between the two groups was observed (P = 0.17 for the log-rank test, P = 0.81 for Gray test). According to Cox proportional hazard regression analysis, the HR for dermatitis in the operation group was 1.03-fold higher compared to that in the control group, but it was not statistically significant (95% CI = 0.99–1.08, P = 0.17). Moreover, in the multivariate Cox regression model, no difference in dermatitis risk was observed between the two groups (95% CI = 0.98–1.08, P = 0.22). According to the Fine and Gray regression model considering the competing risk with death, crude HR was 1.01 (95% CI = 0.96–1.05, P = 0.77) and adjusted HR was 1.01 (95% CI = 0.96–1.05, P = 0.78). No significant differences were observed [Table 3].
Figure 3: Kaplan–Meier curves for the incidence of dermatitis according to operation
Table 3: The Fine and Gray regression model considering competing risk with death for incidence of dermatitisIn addition, subgroup analysis was performed by dividing the participants into two groups: those aged ≥60 years and those aged <60 years. In the ≥60-year age group, no significant difference in dermatitis risk was observed between the control and operation groups in Cox regression model, and Fine and Gray regression model. However, in the <60-year age group, the dermatitis risk was 1.20-fold higher in the operation group than that in the control group (Fine and Gray regression model, 95% CI = 1.05–1.37, P = 0.0008). We performed additional analyses (student's t-test) to compare the period until the dermatitis occurs. After joint replacement, dermatitis occurred earlier in the younger age (those aged ≥60: 1.48 ± 1.34 years, those aged <60 years: 1.27 ± 1.18 years, P = 0.0003). There was no difference in dermatitis risk between the control and operation groups according to gender.
Discussion Joint replacement is a surgical procedure for joint problems using various materials such as plastic, metal, and ceramic.[1] It is known that an immunologic and hypersensitivity reaction to various foreign substances may occur in implants.[3] Allergic reactions to the implant may cause impairment of implant function, as well as dermatitis.[3],[7] In fact, about 10%–15% of the general population has hypersensitivity to metal.[13] However, about 20%–25% of patients who have undergone total joint arthroplasty have metal hypersensitivity.[7] Therefore, after joint replacement, dermatitis risk may increase due to metal hypersensitivity. Particularly, nickel, chromium, and cobalt ions in implants can cause dermatitis. Dermatitis can appear through various skin conditions, including generalized eruption, eczema, urticaria, and vasculitis.[14] In a population-based cohort study conducted in Taiwan, the rates of eczema were 38% higher in patients who received joint replacement than that in the control group.[9]
Our study has several differences with previous studies. We attempted to reduce heterogeneity between the operation and control groups by matching them according to age, sex, index date, income, underlying disease, and CCI. Moreover, a difference in the survival rate would be possible between the two groups. Therefore, the Fine and Gray regression model was used considering the competing risk with death to analyse the HR. In fact, the rate of death within the study period was 12.5% and 9.3% in the operation and control groups, respectively. In addition, this study excluded atopic dermatitis. Atopic dermatitis is known to be caused by various factors such as genetic changes, alteration of immune response, and skin barrier dysfunction.[15] The frequency of allergic contact dermatitis increases in patients with atopic dermatitis.[16] We believe that atopic dermatitis in patients may affect the study outcome. Therefore, more accurate results may be expected when patients with atopic dermatitis are excluded.
In this study, no significant difference was observed between the operation and control groups in terms of dermatitis incidence rate (operation group 9.2%, control group 9.1%). Even when the HR was evaluated, no significant difference was observed between the two groups. An additional subgroup analysis was performed by dividing the participants by age: ≥60 years and <60 years. In those aged ≥60 years, no difference in eczema risk was found between the operation and control groups. However, in those aged <60 years, the operation group had a 1.20-fold greater eczema risk than that in the control group. According to Wick et al.,[17] there is impairment in immune system function in the elderly. Lymphocyte depletion, decrease signal transduction of T cells, and impairment of cellular immune reaction are known to appear with aging.[17] Allergic contact dermatitis is a delayed hypersensitivity reaction that occurs after a period of sensitization and elicitation, which is mainly mediated by T lymphocytes.[2] Therefore, this hypersensitivity reaction is expected to be reduced in the elderly due to T lymphocyte dysfunction. In general, several patients who underwent joint replacement were elderly. Thus, sufficient hypersensitivity reaction to the surgical implant might not be induced in these elderly patients, which might be the reason why no difference was observed in the dermatitis risk between the control and operation groups. However, in younger patients (aged <60 years), the immune response to the implant was sufficiently induced, leading to dermatitis. In addition, the association between covariates and dermatitis was analysed. According to the Fine and Gray regression model as the age increased by 1 year the dermatitis risk decreased by 0.99-fold (95% CI 0.99, P < 0.001). Therefore, dermatitis might occur when there is sufficient immune reaction to the implant after joint replacement. We think that patch testing and investigating of metal allergy before joint replacement might predict this allergic reaction. If an allergic reaction after joint replacement is expected, using hypoallergenic materials might be considered to reduce the risk.
Our study has several limitations. First, it is a retrospective study using the NHIS database. The database included almost the entire Korean population (51 million), however it has some limitations. Data such as location, severity and photographs of dermatitis, materials for joint replacement, location of joint replaced, etc., are not provided. Dermatitis might be associated with materials or location of joint replacement. However in our study, we cannot examine this correlation. Second, considering various covariates that may affect the risk of dermatitis, there may be other covariates that have not been taken into account. Third, the occurrence of dermatitis after joint replacement was evaluated using the ICD-10 diagnosis code. To reduce the bias, dermatitis was considered when the dermatitis code was used more than three times. However, evaluating the occurrence of dermatitis through the diagnosis code is only considered as a limitation. The occurrence of dermatitis might have been underestimated in patients who actually had dermatitis but did not visit hospital. Although pathological findings and patch test results play an important role in diagnosing dermatitis, we do not know whether skin biopsy or patch test have been performed to diagnose dermatitis. Therefore, we believe that additional prospective studies and consideration of more covariates are needed.
In conclusion, the risk of dermatitis did not increase after joint replacement in the operation group compared to that in the control group. However, it was increased in patients younger than 60 years who underwent joint replacement. Therefore, physicians should consider the risk of dermatitis after joint replacement in younger patients.
Key messages
Dermatitis increased after joint replacement in younger patients, but not in older patients.
Financial support and sponsorship
This research was supported by Hallym University Research Fund 2019 (HURF-2019-27).
Conflicts of interest
There are no conflicts of interest.
References 
留言 (0)