Identifying risk factors for cutaneous disease among solid organ transplant recipients: A retrospective review

Abstract

Background

As solid organ transplant recipient (SOTR) life expectancy lengthens, the risk of developing other chronic diseases also increases.

Objective

To determine the cutaneous pathologies for which SOTRs are at an increased risk.

Methods

We performed a retrospective review of SOTRs seen by dermatology from January 1, 2012 and June 1, 2022. Data were analyzed using multivariate logistic regression. Benjamini Hochberg-adjusted P values were examined for multiplicity.

Results

Five hundred and thirty SOTRs were identified. Patients had cutaneous malignancy (38.3%), precancerous lesions (32.5%), inflammatory (35.5%), and infectious diseases (33.1%). Odds of precancerous lesions were higher with increased age at transplant (odds ratio [OR], 1.04; adjusted P =.006), and lower with female sex (OR, 0.505; adjusted P =.006) and African American race (OR, 0.027; adjusted P =.006). Odds of inflammatory lesions were lower with increased age at transplant (OR, 0.979; adjusted P =.023). Odds of infectious diseases were higher with prednisone use (OR, 2.615; adjusted P value =.023).

Limitations

This study is retrospective and was not able to capture patients seen by dermatology outside of our institution.

Conclusions

SOTRs at risk of cutaneous lesions should be referred to dermatology because these conditions may place a significant burden on the quality of life.

Capsule Summary.

• •Immunosuppression after receiving transplant is associated with a risk of cutaneous pathologies, particularly neoplastic transformation.
• •Transplant recipients with increased age at transplant have increased odds for precancerous lesions. Female sex and African American races are protective.
• •Patients prescribed prednisone after receiving transplant have increased odds of cutaneous infection.

Introduction

The incidence of solid organ transplant (SOTs) continued to increase annually. The United States performed 41,354 transplants in 2021; a 5.9% increase compared with 2020.¹ Advances in immunosuppressive therapies have significantly improved transplant success and survival rates.² However, chronic immunosuppression decreases immune mediated surveillance,^3,4 which is associated with an increased incidence of cutaneous neoplastic lesions.5, 6, 7

The reduction in immune cell survival and functionality secondary to immunosuppressive therapy is also associated with an increased prevalence of precancerous, infectious, and inflammatory conditions.^3,8 Within 4 years after transplantation, infectious and inflammatory conditions comprise >50% of cutaneous complications.^9,10 It is common for the reactivation of latent viruses, such as human herpes virus 6, human papillomavirus, and cytomegalovirus, to contribute.^8,11, 12, 13, 14 Although some conditions, such as psoriasis, are less common because of immunosuppressive drug regimens, other inflammatory lesions, such as acneiform eruptions and dermatitis, occur with increased frequency.^6,9,15 Both inflammatory and infectious conditions have the potential to affect quality of life and subsequent adherence to chronic immunosuppressive therapy.^16,17

The existing literature primarily examines the incidence of cutaneous neoplastic lesions in White SOT recipients (SOTRs). However, the characterization of risk factors for the development of precancerous, inflammatory, and infectious pathologies is lacking; therefore, we aimed to investigate the effect of immunosuppressive therapy on the development of a broad range of skin pathologies among SOTRs.

Patients and methods

After institutional review board approval was obtained by IRB-I at the Medical University of South Carolina, we conducted a retrospective chart review of all SOTRs at our institution who had been evaluated by dermatology between January 1, 2012, and June 1, 2022.

We collected data regarding patient demographics, comorbid conditions, family history, transplant course, and dermatologic course. Cutaneous pathologies were categorized as precancerous, inflammatory, or infectious. Precancerous lesions included actinic keratoses (AKs). Inflammatory conditions included acne, atopic dermatitis (AD), contact dermatitis, folliculitis, hidradenitis suppurativa (HS), nummular dermatitis, psoriasis, rosacea, and seborrhoeic dermatitis. Infectious conditions included abscess, cellulitis, deep fungal infection, dermatophytosis, herpes simplex virus (HSV), impetigo, intertrigo, molluscum contagiosum, onychomycosis, shingles, and verruca, including plana, plantar, and vulgaris.

The primary objective of this study was to identify cutaneous pathologies, other than skin cancer, for which SOTRs are at an increased risk. Secondary objectives were to determine risk factors associated with the development of such conditions, which included standard posttransplant medications and photosensitizing agents used to treat comorbid conditions. There were no missing data after data collection.

The study cohort was characterized using descriptive statistics, cutaneous pathologies were evaluated using one-sample z-testing, and risk factors were analyzed using univariate logistic regression. Multivariate logistic regression was conducted using a stepwise model to remove risk factors that were not significant at the P value of <.05 level in the reduced model. Benjamini Hochberg-adjusted P values were examined to account for multiplicity.¹⁸ All data were analyzed using SPSS Statistics (IBM).

Results

We identified 530 SOTRs seen by dermatology for inclusion in this study. Of all SOTRs included, most were White (63.58%) and men (70.19%), with a mean age of 49.44 years. Cutaneous pathology was present in 423 (79.8%) patients. Patients commonly had cutaneous malignancy (38.3%), precancerous lesions (32.5%), inflammatory (35.5%), and infectious conditions (33.1%). Multiple skin conditions were present in 224 (42.3%) of SOTRs. Among inflammatory conditions, folliculitis (10%), contact or irritant dermatitis (9.8%), seborrheic dermatitis (9.4%), AD (6%), and acne (6.2%) were the most common. Of the infectious conditions, verrucae (14%), dermatophytosis (12.1%), and deep fungal infections or abscess (5.1%) were the most common. All data were significant (P <.001) as compared with the general outpatient population.

SOTRs with cancerous lesions were White (95.1%), men (72.9%), kidney transplant recipients (64.0%) with a mean age of 54.33 (±11.52) years at transplant. SOTRs with precancerous lesions were White (96.5%), men (75.0%), and kidney transplant recipients (61.6%) with a mean age of 55.21 (±11.78) years at transplant. SOTRs with inflammatory disease were White (58.5%), men (61.2%), and kidney transplant recipients (70.7%), with a mean age of 45.81 (±17.93) years at transplant. SOTRs with infectious disease were White (60.0%), men (65.7%), kidney transplant recipients (77.1%), with a mean age of 47.17 (±17.02) years at transplant (Table I).

Table I.

Patient demographics for solid organ transplant recipients seen by dermatology posttransplant

	All SOTRs	SOTRs with cancerous lesions	SOTRs with precancerous lesions	SOTRs with inflammatory disease	SOTRs with infectious disease
Age (mean, SD)	49.44 (±16)	54.33 (±11.52)	55.21 (±11.78)	45.81 (±17.93)	47.17 (±17.02)
Sex (n, %)
Male	372 (70.2%)	148 (72.9%)	129 (75.0%)	115 (61.2%)	115 (65.7%)
Female	203 (29.8%)	55 (27.1%)	43 (25.0%)	73 (38.8%)	60 (34.3%)
Race (n, %)
Black or African American	177 (33.4%)	7 (3.4%)	4 (2.3%)	73 (38.8%)	61 (34.9%)
American Indian or Alaska Native	2 (0.4%)	1 (0.5%)	1 (0.6%)	0 (0%)	2 (1.1%)
Asian	7 (1.3%)	0 (0%)	0 (0%)	3 (1.6%)	3 (1.7%)
Caucasian or White	337 (63.6%)	193 (95.1%)	166 (96.5%)	110 (58.5%)	105 (60.0%)
Hispanic or Latino	5 (0.9%)	1 (0.5%)	1 (0.6%)	1 (0.5%)	3 (1.7%)
Pacific Islander	1 (0.2%)	0 (0%)	0 (0%)	0 (0%)	1 (0.6%)
Other	1 (0.2%)	1 (0.5%)	0 (0%)	1 (0.5%)	0 (0%)
Transplant type (n, %)
Kidney	374 (70.6%)	130 (64.0%)	106 (61.6%)	133 (70.7%)	135 (77.1%)
Liver	80 (15.1%)	37 (18.2%)	37 (21.5%)	26 (13.8%)	17 (9.7%)
Heart	85 (16.0%)	45 (22.2%)	38 (22.1%)	29 (15.4%)	27 (15.4%)
Lung	16 (3.0%)	10 (4.9%)	7 (4.1%)	6 (3.2%)	7 (4.0%)
Pancreas	49 (9.2%)	21 (10.3%)	16 (9.3%)	15 (8.0%)	13 (7.4%)

SOTR, Solid organ transplant recipient.

Regarding the development of precancerous lesions in the full model, SOTRs with increased age at the time of transplant were 1.041 times higher odds per each year older (P ≤.001; 95% CI, 1.026-1.056). Odds of developing precancerous lesions were 2.008 times higher in liver transplant recipients (P =.005; 95% CI, 1.238-3.256) and 1.876 times higher in heart transplant recipients (P =.009; 95% CI, 1.169-3.012). SOTRs with any smoking history had 1.844 times the odds (P =.001; 95% CI, 1.275-2.667). Odds were 3.555 times higher among SOTRs with pretransplant history of skin cancer (P ≤.001; 95% CI, 2.039-6.201). Use of cyclosporine (CsA) resulted in 1.983 times higher odds than SOTRs not prescribed CsA (P =.009; 95% CI, 1.187-3.313). Odds were lower among SOTRs taking tacrolimus (48.1%, P =.005; 95% CI, 0.327-0.822) and prednisone (49.3%, P =.002, 95% CI, 0.334-0.772).

Women were 58.7% less likely to develop precancerous lesions than men (P ≤.001; 95% CI, 0.276-0.617). Similarly, kidney transplant recipients were 46.1% less likely (P =.002; 95% CI, 0.365-0.796). African American patients were 97.6% less likely than White patients (P ≤.001; 95% CI, 0.009-0.006) to develop precancerous lesions. SOTRs with type 2 diabetes mellitus (T2DM) were 36.5% less likely to develop precancerous lesions (P =.024; 95% CI, 0.428-0.941).

In the reduced model, age at transplant remained a risk factor for precancerous lesion development (odds ratio [OR], 1.040; adjusted P value =.006; 95% CI, 1.024-1.056). Female sex (OR, 0.505; adjusted P value =.006; 95% CI, 0.318-0.802) and African American race (OR, 0.027; adjusted P value =.006; 95% CI, 0.010-0.074) were protective against precancers. The receiver operator characteristic (ROC) area under the curve (AUC) was 0.823 (Table II).

Table II.

Univariate logistic regression risk factors associated with the development of precancerous lesions among solid organ transplant recipients

	Exp (B)	95% CI for exp (B) Lower	95% CI for exp (B) Upper	Rank	P value	Benjamini Hochberg-adjusted P value
Age at transplant	1.041	1.026	1.056	1	<.001∗	.006∗
Sex
Male	Reference group
Female	0.413	0.276	0.617	2	<.001∗	.006∗
Race
Black or African American	0.025	0.009	0.066	3	<.001∗	.006∗
American Indian or Alaska Native	1.036	.064	16.702	21	.980	1.000
Asian	.000	.000		22	.999	1.000
Caucasian or White	Reference group
Hispanic or Latino	0.259	0.029	2.342	13	.229	.420
Pacific Islander	.000	.000		23	1.000	1.000
Transplant type
Kidney	0.539	0.365	0.796	6	.002∗	.007
Liver	2.008	1.238	3.256	8	.005∗	.014
Heart	1.876	1.169	3.012	10	.009∗	.020
Lung	1.645	0.602	4.494	15	.332	.494
Pancreas	1.010	0.540	1.891	20	.975	1.000
Ever smokers	1.844	1.275	2.667	4	<.001∗	.006
Pretransplant skin cancer	3.555	2.039	6.201	5	<.001∗	.006
Posttransplant medications
Tacrolimus	0.519	0.327	0.822	9	.005∗	.014
Cyclosporine	0.009	1.187	3.313	11	.009∗	.020
Azathioprine	0.777	.203	2.965	18	.711	.920
Prednisone	0.507	0.334	0.772	7	.002∗	.007
Sirolimus	1.042	0.413	2.632	19	.930	1.000
Comorbid conditions
Hypertension	0.804	0.532	1.215	14	.301	.494
Hyperlipidemia	1.147	0.789	1.667	17	.473	.650
Type 2 diabetes mellitus	0.635	0.428	0.941	12	.024∗	.048
Coronary artery Disease/cardiomyopathy	1.219	0.814	1.825	16	.337	.494

^∗Indicates statistically significant result.

Regarding the development of inflammatory pathologies, protective factors included age at the time of transplant (2.2% lower odds, P ≤.001; 95% CI, 0.968-0.989) and presence of comorbid T2DM (36.7% lower odds, P =.020; 95% CI, 0.432-0.929) or hyperlipidemia (HLD) (33.2% lower odds, P =.035; 95% CI, 0.459-0.973). In the reduced model, only age remained a protective factor (OR, 0.978; adjusted P value =.023; 95% CI, 0.968-0.990) with an ROC AUC of 0.590 (Table III).

Table III.

Univariate logistic regression risk factors associated with the development of inflammatory lesions among solid organ transplant recipients

	Exp (B)	95% CI for exp (B) lower	95% CI for exp (B) upper	Rank	P value	Benjamini Hochberg-adjusted P value
Age at transplant	0.978	0.968	0.989	1	<.001∗	.023∗
Sex
Male	Reference group
Female	1.035	0.718	1.492	19	.853	.992
Race
Black or African American	1.435	0.986	2.090	4	.059	.399
American Indian or Alaska Native	.000	.000		22	.999	1.000
Asian	1.534	0.337	6.971	13	.580	.992
Caucasian or White	Reference group
Hispanic or Latino	0.511	.056	4.628	12	.551	.992
Pacific Islander	.000	.000		23	1.000	1.000
Transplant type
Kidney	1.014	0.686	1.498	21	.947	1.000
Liver	0.856	0.516	1.420	11	.547	.992
Heart	0.931	0.572	1.518	17	.776	.992
Lung	1.095	0.391	3.060	20	.863	.992
Pancreas	0.785	0.416	1.483	10	.456	.992
Ever smokers	0.814	0.570	1.163	8	.258	.742
Pretransplant skin cancer	0.647	0.354	1.183	5	.157	.679
Posttransplant medications
Tacrolimus	1.214	0.750	1.966	9	.430	.992
Cyclosporine	0.897	0.525	1.532	14	.691	.992
Azathioprine	0.398	0.085	1.861	7	.242	.742
Prednisone	1.353	0.872	2.100	6	.177	.679
Sirolimus	1.125	0.458	2.765	18	.798	.992
Comorbid conditions
Hypertension	0.924	0.614	1.389	15	.703	.992
Hyperlipidemia	0.668	0.459	0.973	3	.035∗	.268
Type 2 Diabetes Mellitus	0.633	0.432	0.929	2	.020∗	.230
Coronary Artery Disease/Cardiomyopathy	0.930	0.622	1.392	16	.724	.992

^∗Indicates statistically significant result.

Odds of developing infection were 1.638 times higher in kidney transplant recipients (P =.020; 95% CI, 1.080-2.485). SOTRs taking prednisone had 2.615 times the odds (P ≤.001; 95% CI, 1.579-4.329). There was a trend toward significantly increased odds among SOTRs taking sirolimus (2.314, P =.061; 95% CI, 0.963-5.558).

SOTRs who were older at the time of transplant were 1.3% less likely to develop infectious lesions (P =.002; 95% CI, 0.976-0.998). Similarly, liver transplant recipients were 50.1% less likely (P =.017; 95% CI, 0.282-0.881), and SOTRs with pretransplant skin cancer were 51.8% less likely (P =.031; 95% CI, 0.249-0.935) to develop cutaneous infections. In the reduced model, prednisone was the only contributing factor to the development of cutaneous infection (OR, 2.615; adjusted P value =.023; 95% CI, 1.579-4.329) with an ROC AUC of 0.574 (Table IV).

Table IV.

Univaraite logistic regression risk associated with the development of infectious lesions among solid organ transplant recipients

	Exp (B)	95% CI for exp (B) Lower	95% CI for exp (B) Upper	Rank	P value	Benjamini Hochberg-adjusted P value
Age at transplant	0.987	0.976	0.998	4	.022∗	.127
Sex
Male	Reference group
Female	0.773	0.530	1.128	7	.182	.549
Race
Black or African American	1.167	0.793	1.717	13	.433	.766
American Indian or Alaska Native	3ˆ9	.000		22	.999	1.000
Asian	1.664	0.366	7.568	15	.510	.782
Caucasian or White	Reference group
Hispanic or Latino	3.329	0.548	20.217	8	.191	.549
Pacific Islander	3ˆ9	.000		23	1.000	1.000
Transplant type
Kidney	1.638	1.080	2.485	3	.020∗	.127
Liver	0.499	0.282	0.881	2	.017∗	.127
Heart	0.934	0.568	1.536	19	.788	.954
Lung	1.602	0.586	4.375	11	.358	.730
Pancreas	0.711	0.367	1.378	10	.313	.720
Ever smokers	1.114	0.775	1.600	16	.561	.806
Pretransplant skin cancer	0.482	0.249	0.935	5	.031∗	.143
Posttransplant medications
Tacrolimus	0.840	0.524	1.347	14	.470	.772
Cyclosporine	1.360	0.807	2.290	9	.248	.634
Azathioprine	1.711	0.515	5.685	12	.381	.730
Prednisone	2.615	1.579	4.329	1	<.001∗	.023∗
Sirolimus	2.314	0.963	5.558	6	.061	.234
Comorbid conditions
Hypertension	0.996	0.656	1.512	21	.986	1.000
Hyperlipidemia	0.986	0.678	1.434	20	.943	1.000
Type 2 diabetes mellitus	1.088	0.746	1.586	17	.663	.897
Coronary artery disease/cardiomyopathy	1.080	0.720	1.620	18	.711	.908

^∗Indicates statistically significant result.

Discussion

In recent years, there have been significant advancements in immunosuppressive therapies prescribed to SOTRs.² As the number of transplants performed annually within the United States continues to rise, and SOTRs live longer given improved therapies, the number of SOTRs affected by complications of chronic immunosuppression will also rise. Although the link between chronic immunosuppression and cutaneous malignancy is well established, SOTRs are at increased risk for other cutaneous conditions.⁵ Most commonly, these conditions include cutaneous malignancy, precancerous lesions, as well as inflammatory and infectious diseases.^6,7 As demonstrated by our cohort, SOTRs tend to be older, White, men, and kidney transplanted is most frequently performed.¹

Risk factors for developing skin cancer after transplantation include White race, age >50 years, thoracic transplant (ie, heart or lung), male sex, pretransplant history of skin cancer, and smoking status.¹⁹ The use of certain immunosuppressive medications have also been associated with an increased risk of skin cancer, including calcineurin inhibitors, such as tacrolimus and CsA, as well as azathioprine.²⁰ Other therapies, such as mammalian target of rapamycin (mTOR) inhibitors, are believed to have a decreased risk of skin cancer relative to calcineurin inhibitors, although risk still exists.²¹

By type, cutaneous squamous cell carcinoma (cSCC) is the most common among SOTRs.²¹ AKs are thought to be precursor lesions to cSCC, with an estimated risk of progression between <0.1% and 20%.²² Not surprisingly, the risks for developing precancerous lesions were similar to those for developing skin cancer. Odds of precancerous lesions were increased among SOTRs who were White and men. Naturally, being Black or women was protective. Odds were also increased among SOTRs who had an increased age at transplant, received a heart transplant, had a history of smoking, had pretransplant skin cancer, or who were prescribed CsA as part of their posttransplant immunosuppression.

Interestingly, liver transplant recipients also had increased odds for precancers. This is in contrast to the risk of skin cancer, of which liver transplant recipients are believed to have the lowest risk relative to other transplant types.²³ Additionally, in contrast to skin cancer risk factors, the use of tacrolimus and prednisone actually decreased the odds of precancerous lesion development. This may be because of the fact that most patients were prescribed these medications among the precancerous lesion cohort, so a risk for precancerous development was not statistically detectable.

SOTRs with T2DM also had lower odds of developing precancerous lesions. Metformin is a first-line therapy among patients with diabetes. It has been reported that patients with diabetes taking metformin have a decreased risk of developing both cSCC and basal cell carcinoma.²⁴ Our results suggest that metformin may also provide protection against precancerous lesions, although further studies are warranted.

Inflammatory dermatologic diseases, particularly acne, AD, HS, psoriasis, and rosacea, are known to have a profound impact on psychological well-being and overall quality of life.25, 26, 27, 28, 29 In our cohort, there was a trend toward significance in Black patients having higher odds of developing inflammatory disease. Thus, we suggest that transplant providers monitor Black patients for dermatologic changes and refer to dermatology should such symptoms arise.

Protective factors from the development of inflammatory disease included increased age at transplant, T2DM, and HLD. It is known that immune system functionality decreases with age. This coupled with the prevention of the inflammatory response because of immunosuppressive therapies may lead to increased age as a protective factor.^6,15 Metformin has also demonstrated utility in the treatment of acne, HS, psoriasis, and other inflammatory dermatologic conditions.³⁰ The nature of HLD as a protective factor for inflammatory disease is unclear. Perhaps, similar to T2DM, there is a component of HLD therapy that lends itself to protection from inflammatory disease, or there is another confounding factor among SOTRs with HLD. Further studies in larger cohorts are needed to validate this finding.

Bacterial, fungal, and viral opportunistic infections may occur at any point after transplantation,¹⁴ and are reported most frequently one-month posttransplant.³¹ The most common infectious agents during this period include Staphylococcus, Streptococcus, human papillomavirus, and herpes viruses.^10,14,32 Furthermore, coexistent viral infection (ie. cytomegalovirus and human herpes virus 6) and degree of immunosuppression are risk factors for opportunistic fungal infections.⁸ Common fungal agents include Candida, Aspergillus, and Cryptococcus species^33,34 Posttransplant cutaneous infection contributed to increased morbidity, and in some cases, mortality, as infections among SOTRs are more likely to recur, are associated with more severe symptoms, tend to be recalcitrant to treatment, and have a higher risk for disseminated disease.³⁵

Factors associated with increased odds for cutaneous infection posttransplant included kidney transplant and the use of prednisone. There was also a trend toward significance with the use of sirolimus. Similar to inflammatory conditions, increased age at the time of transplant decreased the odds of developing infectious pathology. Liver transplant and pretransplant history of skin cancer were also found to be protective factors.

Limitations of our study include the retrospective nature of this review, as well as the small sample size for patients with skin of color, other than Black SOTRs. We were also unable to capture patients who were seen by the dermatology outside our institution. Future studies that investigate the duration of immunosuppressive therapy at the time of cutaneous pathology development are also warranted.

Our study demonstrates that SOTRs are at risk for cutaneous pathology beyond skin cancer. Most commonly, these include precancerous lesions, inflammatory, and infectious diseases. SOTRs should be educated on their risk for such conditions, and referrals to dermatology should be placed. Patients with risk factors for the development of these conditions may warrant earlier referral to dermatology to focus on intervention that may improve QoL and reduce overall posttransplant morbidity and mortality among SOTRs.