WJH 6 41

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Dermatol Ther (Heidelb) (2016) 6:649–658

DOI 10.1007/s13555-016-0144-3

ORIGINAL RESEARCH

Cardiovascular Disease Outcomes Associated


with Three Major Inflammatory Dermatologic
Diseases: A Propensity-Matched Case Control Study
Vincent D. Marshall . Farah Moustafa . Spencer D. Hawkins .
Rajesh Balkrishnan . Steven R. Feldman

Received: August 17, 2016 / Published online: September 22, 2016


Ó The Author(s) 2016. This article is published with open access at Springerlink.com

ABSTRACT dermatitis are independent risk factors for CVD


1 year following diagnosis.
Introduction: Inflammation is an established Methods: Using a large commercial claims
component of cardiovascular disease (CVD) and database of 21,801,147 lives, we employed a
an underlying factor of several dermatologic propensity-matched logistic regression to
conditions including rosacea, atopic dermatitis, evaluate the association between diagnoses of
and psoriasis. Identifying potential associations rosacea, psoriasis, or atopic dermatitis and a
between these dermatologic and cardiovascular 1-year risk of being diagnosed with
diseases can better inform holistic healthcare cardiovascular disease. Control patients were
approaches. The objective of this study was to matched based on health-care utilization, age
determine whether rosacea, psoriasis or atopic and overall health status as defined by a
modified Deyo–Charlson comorbidity index.
Results: The analysis included 2105 rosacea,
Enhanced content To view enhanced content for this
article go to http://www.medengine.com/Redeem/ 622 atopic dermatitis, 1536 psoriasis, and 4263
AFD6F06058FA9064. control patients. Compared to
Electronic supplementary material The online propensity-matched controls, the adjusted
version of this article (doi:10.1007/s13555-016-0144-3) odds of cardiovascular disease were not higher
contains supplementary material, which is available to
authorized users. in patients with rosacea (odds ratio: 0.894,

V. D. Marshall R. Balkrishnan
College of Pharmacy, University of Michigan, Department of Public Health Sciences, University of
Ann Arbor, MI, USA Virginia, Charlottesville, VA, USA

F. Moustafa S. R. Feldman
Department of Dermatology, Brown University, Department of Pathology, Wake Forest School of
Providence, RI, USA Medicine, Winston-Salem, NC, USA

S. D. Hawkins (&)  S. R. Feldman S. R. Feldman


Department of Dermatology, Center for Department of Public Health Sciences, Wake Forest
Dermatology Research, Wake Forest School of School of Medicine, Winston-Salem, NC, USA
Medicine, Winston-Salem, NC, USA
e-mail: spencerhawkins@gmail.com
650 Dermatol Ther (Heidelb) (2016) 6:649–658

p = 0.2713), atopic dermatitis (OR 1.032, prevalence of high cholesterol (OR 1.29,
p = 0.8489), or psoriasis (OR 1.087, p\0.001), or a lifetime prevalence of
p = 0.4210). In univariate analysis, the hypertension (OR 1.31, p\0.001) [7]. Using a
unadjusted odds of cardiovascular disease was Taiwanese nationwide database, a retrospective
higher in patients with psoriasis (OR 1.223, population-based study identified 20,323
p = 0.0347). patients with AD and 20,323
Conclusions: Limitations of this study include comorbidity-matched subjects (age C20 years).
the short follow-up period and inclusion of only After adjusting for age, sex, comorbidities, and
commercially insured patients limit the medications used, the hazard ratio for
generalizability of these findings. In this large experiencing ischemic stroke over a 4-year
study of patients with rosacea, atopic follow-up period was 1.33 (p = 0.001),
dermatitis, and psoriasis, we did not detect an suggesting AD may be an independent risk
increased 1-year risk of cardiovascular disease factor of ischemic stroke [8].
after adjusting for confounders. For rosacea, a case control study done by
Duman et al. assessed the compared CVD risk
Keywords: Atopic dermatitis; Cardiac health; factors for 60 rosacea patients and 50 age- and
Eczema; Heart failure; Ischemic stroke; gender-matched controls. Rosacea patients were
Myocardial infarction; Psoriasis; Rosacea more likely than controls to have: high total
cholesterol (199 mg/dL vs. 163, p\0.001), LDL
INTRODUCTION (121 vs 101, p = 0.002), CRP levels (0.43 vs.
0.24 mg/L, p = 0.007), and a family history of
Most dermatologic conditions are readily premature CVD (p = 0.002) [9]. This study only
apparent to the physician taking care of the assessed risk factors and did not look at
patient and can be indicators of inflammatory cardiovascular endpoints. Also, the control
processes (i.e. atherosclerosis) occurring subjects were only age and gender matched, so
internally. Some authors discuss an association invariably there was residual confounding.
between psoriasis and cardiovascular disease Nevertheless, patients with rosacea may be
(CVD) [1–6]. Less clear associations exist more likely to have CVD risk factors but this
between other dermatologic inflammatory has not been shown to be an independent risk
conditions [atopic dermatitis (AD) and rosacea] factor for CVD.
and CVD. Similar to psoriasis, these conditions In this study, we evaluate whether three
are (1) common and (2) may reflect underlying dermatologic diseases that have inflammatory
systemic inflammation. Following this same components contribute an independent risk for
concept, these diseases are worth assessing for development of CVD outcomes at 1-year
their independent contribution to CVD risk. follow-up using a large commercial claims
Adults diagnosed with or self-identifying to database.
have AD are more likely to have risk factors
associated with CVD including: being a current METHODS
smoker (adjusted odds ratio (OR) 1.28,
p\0.001), being a current heavy drinker (OR The data were analyzed as a case–control study
1.58, p\0.001), having a lifetime prevalence of with matching on covariates using propensity
diabetes (OR 1.37, p\0.001), having a lifetime scores to quantify the association between the
Dermatol Ther (Heidelb) (2016) 6:649–658 651

dermatologic diseases and CVD outcomes. The corticosteroids ± anti-itching or anti-fungal/


matching done with propensity scores used a bacterial agents, and anti-itch treatments (such
Mahalanobis 1:1 algorithm using all the main as calamine lotion) for AD); and oral antibiotics,
effects from the regression except the metronidazole, isotretinoin, retinoids, azelic
dermatologic disease indicator variable [9]. acid, benzoyl peroxide, calcineurin inhibitors,
sulfacetamide products, permethrin and
Setting and Participants oxymetazoline. This method allowed for
creating a homogeneous set of recently
All variables come from the MarketScanTM diagnosed people with medication
Commercial Claims and Encounters database prescriptions for their illness.
for 2005–2007 [10]. No new studies with human The control group was comprised of people
or animal subjects were performed by any of the selected with a simple random sample from the
authors. The people included in the analysis database. None of the people in the control
were in the range of 30–64 years old and had at group had any of the three skin diseases. In
least 1 year of follow-up data and 6 months of order to match to the 1-year follow-up period
baseline data. We chose the lower age limit for the dermatologic disease group, the control
because CVD is less common in individuals group was required to have complete data for
younger than this age. We chose the upper age the whole of 2006.
limit to minimize loss to follow-up that could The cases of cardiovascular disease were
occur from 65 year olds switching from defined as having at least 2 diagnosis codes
commercial insurance to Medicare. (ICD 9-CM) or at least one diagnosis code plus a
The dermatologic diseases group was procedure code (CPT) for the disease.
composed of three arms: rosacea (ICD-9
695.3), psoriasis (ICD-9 696.1), or AD (ICD-9 Variables and Quantitative variables
691.XX, except for 691.0X). None of the arms
have people with any of the other two skin We examined several covariates for explanation
conditions. They were all defined with of variance in the outcome and for their
diagnosis codes, having at least 2 entries each. intrinsic interest. Potential confounders
The dermatologic disease group was selected included age and gender, number of inpatient
further using a washout period. We defined our days, number of days for which people had
washout period as at least a 180-day period after outpatient visits, and total health-care
a person’s first diagnosis on record since the utilization costs. Since diabetes is a large risk
beginning of 2005 during which they received factor for CVD, we included it as a predictor of
no treatment for their skin condition. CVD, defined by having at least 2 diagnosis
Treatments identified during this washout codes between 2005 and 2007. Overall
period included biologics, systemic treatments comorbid disease severity was examined with
(e.g., methotrexate and cyclosporine), vitamin the Deyo–Charlson Comorbidity Index with the
D analogs, calcineurin inhibitors, topical components related to the outcome removed.
corticosteroids, anthralin, coal tar, retinoids, Those components were CVD, peripheral
and ammonium lactate for psoriasis; vascular disease (PVD), and old myocardial
systemic treatments, antihistamines (oral and infarction (MI). We also examined the number
topical), calcineurin inhibitors, topical of comorbid conditions, which could be as high
652 Dermatol Ther (Heidelb) (2016) 6:649–658

as 14 (as opposed to the 17 conditions usually diagnosis codes or a diagnosis code plus a
examined with Deyo–Charlson). In addition to cardiovascular-related procedure code
the number of days spent in the hospital, we (Supplementary Table 1). We used a 365-day
examined the total number of hospitalizations follow-up period for the exposure and control
for the year but did not include this in the in order to standardize the results. Thus,
regression in order to reduce multi-collinearity. variables such as cost and healthcare
We also looked at the medication costs directly utilization will also be generalizable to other
related to the skin conditions as defined in the studies.
washout period, for a total of 365 days The propensity scores were generated for the
post-treatment. In addition to summing the three dermatologic disease arms versus the
costs per person, we looked at the number of control group, and certain people were
prescription drug refills related to these costs for selected from the randomly created control
that year. We further examined the total group using Mahalanobis matching [11]. All
medical costs divided into yearly inpatient, variables, except for the exposure variable, were
outpatient, and medication costs. included in the propensity score calculation. In
order to look for a balanced propensity score set,
Study Size we compared the covariates individually with
effect sizes and with a global test designed by
The sample size was determined by the Hansen and Bowers [11, 12]. Continuous data
exclusion criteria and the eventual propensity were reported as mean (standard deviation) and
score selection. The control group originally categorical data were reported as n (%). Analyses
was comprised of 20,000 people and many were were performed in SAS 9.4 (SAS Institute, Cary,
not included because they did not match well to NC, USA) and R [12].
the dermatologic disease group. Given the large
sample size, we did not perform a power Compliance with Ethics Guidelines
analysis prior to the study to estimate the
effect size we could detect. This article does not contain any new studies
with human or animal subjects performed by
Statistical Methods any of the authors.

The study was designed to examine the RESULTS


independent risk of CVD among the
dermatologic conditions compared with a Participants
general population of people in a commercial
insurance database. Cardiovascular disease The sample sizes for the disease and control
outcomes included ischemic heart disease groups were arrived at separately. For the
(ICD-9: 410–414), transient cerebral ischemia dermatologic disease group, we began with
(ICD-9: 435), heart failure (ICD-9: 428), people having either outpatient or inpatient
occlusion and stenosis of pre-cerebral arteries visits with ages inclusive between 30 and
(ICD-9: 433), and occlusion of cerebral arteries 64 years (n = 21,801,147). After excluding
(ICD-9: 434). Cases were required to have two patients that lacked diagnosis criteria,
Dermatol Ther (Heidelb) (2016) 6:649–658 653

Fig. 1 Flow diagram to identify exposure groups

sufficient wash-out period, at least 1 year of Descriptive Data and Outcome Data
follow-up, and at least one outpatient record,
there were 4263 remaining patients (Fig. 1). For Our sample was comprised of people between
the control group, we began with a simple the ages of 30–64 years old with an average age
random sample of 20,000 people from the 2006 of 49.1 years [standard deviation (SD) 9.0].
database. After excluding those of age less than Women comprised 67% of our total sample,
30 years or greater than 64 years, with less than and 13% of the population had either type I or
12 months of continuous enrollment, or with a type II diabetes. The sample had an average of
diagnosis of the dermatologic diseases 15.7 days of outpatient visits (SD 15.1), 0.3 days
investigated, there were 14,162 patients. of inpatient visits (SD 1.5), and 0.08
Following the propensity score matching with hospitalizations (SD 0.34) per year. The
the combined dermatologic disease group, the average number of comorbidities was 0.28 (SD
control sample size was 4263 patients. Our 0.57) and the average score of the modified
sample consisted of n = 8526 people, matched Charlson Comorbidity index was 0.34 (SD 0.8).
1:1 between the dermatologic diseases and For health-related expenses, there were an
control groups. average US$8508 of total costs (SD 16,069),
654 Dermatol Ther (Heidelb) (2016) 6:649–658

$1049 of inpatient costs (SD 6756), $3925 of dermatologic conditions: psoriasis, AD, and
outpatient costs (SD 8735), and $1987 of rosacea contribute to an independent risk of
medication costs (SD 3927). For patients cardiovascular at 1 year following diagnosis.
diagnosed with AD, psoriasis, or rosacea, an While psoriasis is associated with an
average of 3 dermatologic medications (SD 3) increased risk for cardiovascular outcomes, in
were filled, resulting in an average annual cost this study we did not detect independent
of $840 (SD 3261). All covariates, except the increased risk at 1-year follow-up after
number of inpatient and number of outpatient adjusting for confounders. Prior studies have
days, had statistically significant differences shown patients with psoriasis and CVD share
between the groups (Table 1). many risk factors and the heterogeneity of
The propensity score matching reduced the existing studies makes it difficult to determine
Hansen and Bowers global test for balance if psoriasis is truly an independent risk factor for
between treatment and control populations CVD [1]. Association of a disease with
from a Chi-square value of 1155 (p\0.0001) confounders does not establish any association
to 12.4 (p = 0.0528). with the corresponding disease itself. In our
study, patients with psoriasis had an increased
Logistic Regression Model risk of developing CVD 1 year after diagnosis
before accounting for comorbidities, but this
We considered first an effects modification difference vanished after accounting for
model, and found that the overall test for comorbidities. In our study, limited follow-up
the variable was insignificant (F = 3.5118, of 1 year may also play a role in not identifying
p = 0.3912), and we reported only the main psoriasis as an independent risk factor for CVD.
effects model. The model was highly predictive The risk between other inflammatory skin
of the outcome with an AUC value of 0.7935. conditions (AD and rosacea) has received only
None of the three dermatologic disease groups limited evaluation.
had statistically significant increased odds of We found that patients with AD did not have
experiencing a CVD event at one year compared statistically significant independent increased
to the control group, after controlling for all odds of being diagnosed with CVD compared to
other covariates (Table 2). In univariate our propensity-matched control population at
analysis, the unadjusted odds of cardiovascular 1 year following diagnosis. Silverberg et al.
disease was higher in patients with psoriasis. found that patients with atopic dermatitis are
more likely to have comorbid (high cholesterol,
DISCUSSION hypertension, and diabetes) and environmental
habits (increased smoking and drinking with
Systemic inflammation is a risk factor for decreased activity) that would pre-dispose AD
cardiovascular disease as inflammation patients to developing CVD [7]; however, AD
accelerates atherosclerosis [13–15]. does not appear to be a significant independent
Rheumatoid arthritis, psoriasis and other risk factor. Our findings do contrast with the
diseases associated with chronic systemic findings from a Taiwanese retrospective
inflammation have been associated with population-based study, which found patients
increased risk of CVD [16–23]. Our study with AD having an elevated risk for ischemic
assessed whether three inflammatory stroke (hazards ratio of 1.33) [8], while patients
Table 1 Demographic and health care utilization for each exposure group
All people All treatment Control Dermatitis Psoriasis Rosacea Significance
Sample size 8493 4230 4263 619 1521 2090
CVD 407 (5%) 173 (4%) 234 (5%) 27 (4%) 72 (5%) 74 (4%) 0.0075
Gender (F) 5658 (67%) 2792 (66%) 2866 (67%) 417 (67%) 787 (52%) 1588 (76%) \0.0001
Dermatol Ther (Heidelb) (2016) 6:649–658

Diabetes 1068 (13%) 524 (12%) 544 (13%) 68 (11%) 241 (16%) 215 (10%) \0.0001
Age 49.1 (9) 49 (8.8) 49.2 (9.1) 46.6 (8.9) 48.7 (9.1) 49.9 (8.3) \0.0001
Number outpatient days 15.7 (15.1) 16 (14.5) 15.3 (15.7) 16.7 (15.5) 16.3 (15) 15.5 (13.9) 0.0553
Number inpatient days 0.3 (1.5) 0.3 (1.5) 0.3 (1.5) 0.4 (1.7) 0.3 (1.6) 0.2 (1.4) 0.2124
Hospitalizations 0.08 (0.34) 0.09 (0.35) 0.08 (0.33) 0.12 (0.44) 0.08 (0.33) 0.08 (0.34) 0.0497
Inpatient costs $1049 (6756) $1323 (8610) $778 (4151) $1597 (7714) $1306 (9652) $1254 (8040) 0.0017
Outpatient costs $3925 (8735) $4317 (10,765) $3537 (6061) $4514 (7489) $4445 (8038) $4166 (13,074) 0.0004
Dermatitis medication costs $840 (3261) $840 (3261) (NA) $185 (895) $2012 (5197) $181 (280) \0.0001
Medication costs $1987 (3927) $2746 (4661) $1235 (2835) $1964 (3385) $4226 (6465) $1901 (2771) \0.0001
Total costs $8508 (16,069) $8386 (15,939) $8629 (16,198) $8075 (13,626) $9977 (15,227) $7321 (16,960) \0.0001
Dermatitis drug count 3 (3) 3 (3) (NA) 2 (2.4) 4 (3.7) 2.6 (2.3) \0.0001
Comorbidities 0.28 (0.57) 0.27 (0.56) 0.29 (0.58) 0.27 (0.58) 0.31 (0.61) 0.24 (0.52) 0.0002
Charlson comorbidity index 0.34 (0.8) 0.33 (0.82) 0.34 (0.78) 0.35 (0.92) 0.38 (0.87) 0.3 (0.76) 0.0309
Statistical significance determined using Chi-squared test or ANOVA test across the 4 treatment groups. Data reported as n (%) or mean (standard deviation). Costs
in US dollars
655
656 Dermatol Ther (Heidelb) (2016) 6:649–658

Table 2 Odds ratios of cardiovascular disease events


Crude odds ratio (95% CI) Adjusted odds ratio (95% CI) p valuea
Disease group
Rosacea 0.833 (0.691–1.004) 0.894 (0.732–1.091) 0.2713
Atopic dermatitis 0.893 (0.664–1.202) 1.032 (0.744–1.432) 0.8489
Psoriasis 1.223 (1.015–1.475) 1.087 (0.887–1.332) 0.4210
Control
Gender (male versus female) 2.330 (2.014–2.695) 2.172 (1.853–2.547) \0.0001
Age 1.104 (1.093–1.116) 1.092 (1.080–1.105) \0.0001
Diabetes 3.698 (3.133–4.365) 2.103 (1.644–2.464) \0.0001
Number of outpatient days 1.026 (1.022–1.030) 1.016 (1.011–1.022) \0.0001
Number of inpatient days 1.273 (1.224–1.325) 1.137 (1.085–1.192) \0.0001
Modified Charlson comorbidity index 1.593 (1.487–1.708) 1.074 (0.982–1.175) 0.1174
Total expenses 1.000 (1.000–1.000) 1.000 (1.000–1.000) 0.0006
CI confidence interval
a
For multivariate odds ratio

identified with severe AD had a more CV risk factors such as smoking and alcohol
pronounced risk of developing ischemic stroke abuse [25].
(hazards ratio of 1.7) over a follow-up period
of 1–4 years compared to their Limitations
comorbidity-matched counterparts. We
suspect that the longer follow-up period in The large sample size gave our study power to find
their study may explain the potential even small clinical effect size. There were several
discrepancy with our findings. prominent potential confounders for which the
In our study, patients with rosacea were not database does not collect data, including obesity
statistically more likely to have comorbidities or and smoking and drinking history. Consequently,
an independent risk for cardiovascular disease. we expect a degree of residual confounding that
This corroborates recent findings of Thyssen may bias our estimate upward from the null. The
et al. which also found patients with rosacea washout period can potentially select a
were not more likely to have an independent population with lower severity because it looks
risk of CVD or an increase in CVD risk factors for new treatment patients, but this may be
(high cholesterol, LDL, and CRP) and balanced by the fact that they are at least being
environmental habits (increased smoking and treated by prescription medications. The study
drinking) [24]. A Taiwanese study did find that had a short follow-up period of 1 year, but there
rosacea was significantly associated with CV was a cross-section of many patient ages. With
comorbidities including dyslipidemia, younger patients, 1 year may be insufficent to
hypertension, and coronary artery disease, detect an increased risk of CVD. We used a large
although it lacked adjustment for important commercial claims database that is applicable to a
Dermatol Ther (Heidelb) (2016) 6:649–658 657

large proportion of the population. Because this educational grant from Galderma Laboratories,
claims database did not include Medicaid L.P. S. R. Feldman is a consultant and speaker
patients, the magnitude of effect may not be for Galderma, Stiefel/GlaxoSmithKline, Abbott
generalizable for indigent populations. Insurance Labs, Warner Chilcott, Janssen, Amgen,
claims data do not provide information on Photomedex, Genentech, BiogenIdec, and
systemic inflammatory burden beyond the Bristol Myers Squibb. S. R. Feldman has
diagnosis. Consequently, we have no way of received grants from Galderma, Astellas,
measuring the systemic inflammation burden of Abbott Labs, Warner Chilcott, Janssen, Amgen,
skin disease. Photomedex, Genentech, BiogenIdec, Coria/
Valeant, Pharmaderm, Ortho Pharmaceuticals,
Aventis Pharmaceuticals, Roche Dermatology,
CONCLUSIONS
3M, Bristol Myers Squibb, Stiefel/
Among patients diagnosed with one of three GlaxoSmithKline, Novartis, Medicis, Leo,
common inflammatory dermatologic HanAll Pharmaceuticals, Celgene, Basilea, and
conditions— psoriasis, AD, and rosacea—we Anacor and has received stock options from
did not detect an independent increased risk Photomedex. S. R. Feldman is the founder and
of CVD 1 year following diagnosis when holds stock in Causa Research. V. D. Marshall, F.
compared with a random sample of Moustafa, S. Hawkins and R. Balkrishnan have
propensity-matched patients with no evidence nothing to disclose.
of these three diseases. Before accounting for
Compliance with Ethics Guidelines. This
covariates, patients with psoriasis appeared to
article does not contain any new studies with
be at an increased risk for CVD 1 year following
human or animal subjects performed by any of
diagnosis, but this was not significant after
the authors.
adjusting for confounders. Physicians should
be aware of the higher rates of covariates in
Open Access. This article is distributed
patients with psoriasis which can predispose to
under the terms of the Creative Commons
an increased risk of CVD.
Attribution-NonCommercial 4.0 International
License (http://creativecommons.org/licenses/
by-nc/4.0/), which permits any noncommercial
ACKNOWLEDGMENTS
use, distribution, and reproduction in any
No funding or sponsorship was received for this medium, provided you give appropriate credit
study or publication of this article. All named to the original author(s) and the source, provide
authors meet the International Committee of a link to the Creative Commons license, and
Medical Journal Editors (ICMJE) criteria for indicate if changes were made.
authorship for this manuscript, take
responsibility for the integrity of the work as a
whole, and have given final approval for the REFERENCES
version to be published.
1. Miller IM, Ellervik C, Yazdanyar S, Jemec GB.
Meta-analysis of psoriasis, cardiovascular disease,
Disclosures. The Center for Dermatology and associated risk factors. J Am Acad Dermatol.
Research is supported by an unrestricted 2013;69(6):1014–24.
658 Dermatol Ther (Heidelb) (2016) 6:649–658

2. Mehta NN, Li K, Szapary P, Krueger J, Brodmerkel C. 14. Libby P. Inflammation in atherosclerosis. Nature.
Modulation of cardiometabolic pathways in skin 2002;420(6917):868–74.
and serum from patients with psoriasis. J Transl
Med. 2013;11:194. 15. Kaplan RC, Frishman WH. Systemic inflammation
as a cardiovascular disease risk factor and as a
3. Gisondi P, Girolomoni G. Cardiometabolic potential target for drug therapy. Heart Dis.
comorbidities and the approach to patients with 2001;3(5):326–32.
psoriasis. Actas Dermosifiliogr. 2009;100(Suppl
2):14–21. 16. Dhawan SS, Quyyumi AA. Rheumatoid arthritis and
cardiovascular disease. Curr Atheroscler Rep.
4. Friedewald VE Jr, Cather JC, Gordon KB, Kavanaugh 2008;10(2):128–33.
A, Ridker PM, Roberts WC. The editor’s roundtable:
psoriasis, inflammation, and coronary artery 17. Wolfe F, Mitchell DM, Sibley JT, Fries JF, Bloch DA,
disease. Am J Cardiol. 2008;101(8):1119–26. Williams CA, et al. The mortality of rheumatoid
arthritis. Arthritis Rheum. 1994;37(4):481–94.
5. Wakkee M, Thio HB, Prens EP, Sijbrands EJ,
Neumann HA. Unfavorable cardiovascular risk 18. Solomon DH, Goodson NJ, Katz JN, Weinblatt ME,
profiles in untreated and treated psoriasis patients. Avorn J, Setoguchi S, et al. Patterns of
Atherosclerosis. 2007;190(1):1–9. cardiovascular risk in rheumatoid arthritis. Ann
Rheum Dis. 2006;65(12):1608–12.
6. Alexandroff AB, Pauriah M, Camp RD, Lang CC,
Struthers AD, Armstrong DJ. More than skin deep: 19. Fischer LM, Schlienger RG, Matter C, Jick H, Meier
atherosclerosis as a systemic manifestation of CR. Effect of rheumatoid arthritis or systemic lupus
psoriasis. Br J Dermatol. 2009;161(1):1–7. erythematosus on the risk of first-time acute
myocardial infarction. Am J Cardiol.
7. Silverberg JI, Greenland P. Eczema and 2004;93(2):198–200.
cardiovascular risk factors in 2 US adult
population studies. J Allergy Clin Immunol. 20. Neimann AL, Shin DB, Wang X, Margolis DJ, Troxel
2015;135(3):721–8. AB, Gelfand JM. Prevalence of cardiovascular risk
factors in patients with psoriasis. J Am Acad
8. Su VY, Chen TJ, Yeh CM, Chou KT, Hung MH, Chu Dermatol. 2006;55(5):829–35.
SY, et al. Atopic dermatitis and risk of ischemic
stroke: a nationwide population-based study. Ann 21. Sin DD, Man SF. Why are patients with chronic
Med. 2014;46(2):84–9. obstructive pulmonary disease at increased risk of
cardiovascular diseases? The potential role of
9. Duman N, Ersoy ES, Atakan N. Rosacea and systemic inflammation in chronic obstructive
cardiovascular risk factors: a case control study. pulmonary disease. Circulation.
J Eur Acad Dermatol Venereol. 2014;28(9):1165–9. 2003;107(11):1514–9.

10. Danielson E. White paper: Health research data for 22. Mosca S, Gargiulo P, Balato N, Di CL, Parente A,
the real world: The MarketScanÒ databases. Truven Paolillo S, et al. Ischemic cardiovascular
Health Analytics. Accessed July 7, 2014. involvement in psoriasis: a systematic review. Int J
Cardiol. 2015;15(178):191–9.
11. Feng W, Jun Y, Xu R. A method/macro based on
propensity score and Mahalanobis distance to 23. Evensen K, Slevolden E, Skagen K, Ronning OM,
reduce bias in treatment comparison in Brunborg C, Krogstad AL, et al. Increased subclinical
observational study. 2005. http://www.lexjansen. atherosclerosis in patients with chronic plaque
com/pharmasug/2006/publichealthresearch/pr05. psoriasis. Atherosclerosis. 2014;237(2):499–503.
pdf. Accessed Apr 01, 2015.
24. Egeberg A, Hansen P, Gislason G, Thyssen J.
12. R Core Team. R: a language and environment for Assessment of the risk of cardiovascular disease in
statistical computing. Vienna: R Foundation for patients with rosacea. J Am Acad Dermatol.
Statistical Computing. http://www.R-project.org/. 2016;75(2):336–9.
Accessed Apr 01, 2015.
25. Hua TC, Chung PI, Chen YJ, et al. Cardiovascular
13. Willerson JT, Ridker PM. Inflammation as a comorbidities in patients with rosacea: a
cardiovascular risk factor. Circulation. nationwide case-control study from Taiwan. J Am
2004;109(21 Suppl 1):II2–10. Acad Dermatol. 2015;73:249–54.

You might also like