Wolf-Henning Boehncke, Department of Dermatology, Johann Wolfgang Goethe-University, Theodor-Stern-Kai 7, D-60590 Frankfurt am Main, Germany, Tel./Fax: (+49) 69 6301-5743/5117, e-mail: Boehncke@em.uni-frankfurt.de
Abstract: There is increasing awareness that psoriasis is more than ‘skin deep’. Several recent reviews focussed on biomarkers indicating the systemic dimension of psoriasis and the aspect of comorbidity psoriasis shares with other chronic inflammatory diseases, such as Crohn’s disease and rheumatoid arthritis. Of emerging significance is the relationship to cardiovascular disease, as this contributes substantially to the patients’ increased mortality. In this viewpoint, we examine currently available evidence favouring the concept of a causal link between psoriasis and cardiovascular disease: systemic inflammation may cause insulin resistance, which in turn triggers endothelial cell dysfunction, leading to atherosclerosis and finally myocardial infarction or stroke. While this ‘psoriatic march’ is not yet formally proven, it raises clinically and academically relevant questions, and gains support by recent observations of numerous investigators.
Psoriasis is a common, chronic inflammatory skin disease. There is increasing awareness that psoriasis is more than ‘skin deep’ and that it has important systemic manifestations that are shared with other chronic inflammatory diseases, such as rheumatoid arthritis (1). Of emerging significance is the relationship between cardiovascular disease and severe psoriasis, as this may explain the increased mortality of the latter (2). The term ‘severe psoriasis’ in this context refers to all patients, who – at any point in time – received either systemic therapy or phototherapy, or were treated on an in-patient basis (3). As much as one-third of all psoriasis patients meet these criteria (4).
A recent review outlined a scenario on how the march of psoriasis unfolds from gene to clinic (5): genetic factors drive disease-specific processes (step 1), possibly triggered by environmental factors, involving both innate and adaptive immune responses (step 2), and leading to disease expression (step 3); comorbidity would then ‘likely … result from chronic inflammation’ (step 4). This viewpoint focuses primarily on that final step, breaking it down to a cascade of events comparable with the manifestation of type 2 diabetes mellitus – given our initial findings on insulin resistance among psoriasis patients (6) – which we coined the ‘psoriatic march’ (7) (admittedly unaware of the legend of figure 8 in Ref. 5): according to this hypothesis, severe psoriasis is a chronic systemic inflammatory disorder. It increases the inflammatory burden of the patient and causes a state of insulin resistance, resulting in endothelial cell dysfunction and atherosclerosis. At the level of coronary, carotid or cerebral arteries, this cascade will result in myocardial infarction or stroke (Fig. 1).
Here, we examine step by step the currently available information for evidence to support causality of this cascade in psoriasis; we also highlight open questions that need to be addressed. But before we do so, we wish to point towards …
Some arguments for why reading this paper could be a waste of time
• A recent cohort study documented a comparable risk for ischaemic heart disease among psoriasis patients and matched controls, concluding that psoriasis may not be an independent risk factor in this regard (8). This study sparked heated debates regarding its validity (9,10).
• A case-control study found no difference in endothelial function between psoriasis patients and controls (11). However, this was a cross-sectional study, and 70% of the psoriasis patients received systemic therapy, which may improve endothelial function (see next).
• The increased cardiovascular mortality may simply be a result of the cumulation of ‘traditional’ risk factors, as these are often more common among psoriasis patients when compared with controls, this may at least in part be due to shared genetics. On the other side of this argument is that the largest most recent Genome Wide Association Scan did not reveal any susceptibility loci common with obesity or cardiovascular disease. Only 1 locus of 15 may be close to a minor locus for type I diabetes mellitus. Overall, however the genetic susceptibility loci of psoriasis and cardiovascular disease appear to be separate (12). Discussing data from studies controlling their impact, we will provide evidence that this aspect cannot fully explain the increased mortality.
• Focussing on this unidimensional cascade may result in gross oversimplification. We acknowledge this point and admit that – among other potentially important aspects – the role of genetic susceptibility (5) is not discussed in depth here. But this aspect, as well as others, could well be incorporated while elaborating on the backbone of our hypothesis (13).
As there is a high prevalence of comorbidity already among children with psoriasis (14), there may not be anything like a march at all. Others, however, did find a higher propensity among psoriasis patients to develop comorbidity subsequent to their psoriasis diagnosis (15). Again, we wish to point out that our concept primarily aims to explain cardiovascular comorbidity in psoriasis.
Severe psoriasis is an independent risk factor for cardiovascular disease
The question of whether or not the ‘psoriatic march’ describes a cascade of events linked by causality may be answered on the basis of epidemiological data: The presence of so-called comorbidity, that is, a higher than expected prevalence of an entity in combination with another disorder, among psoriasis patients has been known for decades (16). Comorbidity of psoriasis comprises, among others, obesity, lymphomas, depression and cardiovascular disease (5,17–20). The relationship between the latter and psoriasis is of particular importance, as it explains in part the increased mortality of patients with severe psoriasis (2). This association, however, does not per se establish a causal relationship between those disorders, but the plot thickens. In Sweden, an increased cardiovascular mortality was only observed among psoriasis inpatients (probably a more severe group), but not in outpatients (21). In line with this finding, a recent population-based study identified severe psoriasis as an independent risk factor for myocardial infarction (3), and a case-control study showed substantially elevated levels of coronary artery calcification as an indicator for coronary artery disease among psoriasis inpatients when compared with controls matched for all major known cardiovascular risk factors (22). In a cohort study with nested case-control analysis, the adjusted odds ratio of developing myocardial infarction for patients with psoriasis aged <60 years was found to be 1.66 (23), and a large observational study observed an odds ratio for ischaemic heart disease of 1.78 among psoriasis patients (24). Still, one has to acknowledge that patients with severe psoriasis appear to have an excess of ‘traditional’ risk factors for the development of cardiovascular disease (25).
Taken together, there is increasing epidemiologic evidence to suggest that severe psoriasis is a relevant and independent cardiovascular risk factor, and it is time to consider possible pathways to better understand this link.
Severe psoriasis is a chronic systemic inflammatory disorder
Several of the epidemiological studies mentioned confirm that psoriasis per se goes more than skin deep. For example, the cardiovascular risk of psoriasis patients depends on the severity of their disease, as only inpatients (presumably suffering from rather severe psoriasis), but not outpatients (likely to have only mild psoriasis) exhibit increased mortality in a Swedish study (21). Similarly, a large population-based study found an increased risk for myocardial infarction in patients with severe, but not with mild psoriasis (3). Finally, a cross-sectional study measuring biomarkers for inflammation in general and adipokines – that is, cytokines produced by adipocytes – in particular, found the serum levels of psoriasis patients to correlate with their respective Psoriasis Area and Severity Index (PASI), but not the body mass index (BMI) (6).
Several other biomarkers indicating a state of systemic inflammation have been observed to be elevated in the blood of psoriasis patients, including C-reactive protein (CRP) (26), vascular endothelial growth factor (VEGF) (27) and indicators of platelet activation such as P-selectin (28,29). Interestingly, VEGF never drops to levels as low as in non-psoriatic controls, but remains somewhat elevated even in the absence of clinical signs of psoriasis (27), thus arguing in favour of a chronic smouldering, rather than really chronic-recurrent inflammation.
Noteworthy, not only classical markers for inflammation have been noted to be elevated, but also adipokines, most of them functioning as insulin antagonists, for example, resistin (6) and leptin (30). Collectively, the adipokine milieu in the blood of psoriasis patients is strikingly similar of that in prediabetic individuals, the latter exhibiting signs of insulin resistance.
Systemic inflammation induces insulin resistance
Insulin resistance, that is, reduced uptake of glucose by metabolically active cells upon exposure to insulin, is reflected at the clinical level by the so-called Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) (31). It can be calculated, based on a single blood test, according to the formula: fasting insulin (mIU/l) × fasting glucose (mmol/l)/22.5. An index >2.5 reflects the clinical state of insulin resistance. A more sensitive method to look for insulin resistance is the so-called oral glucose tolerance test (OGTT) (32): patients consume 75 g of glucose; blood glucose, insulin and C-peptide as a cleavage product of insulin were measured at t = 0, 30, 60 and 120 min. Using these methods, two cross-sectional studies showed that psoriasis patients exhibit insulin resistance at clinical levels (6,33).
In the context of this viewpoint, the notion of insulin being a vasoactive hormone is of particular relevance: intravenously administered insulin enhances blood flow and vasodilation in a nitric oxide (NO)-dependent manner (34). This pathway involves activation of phosphoinositide-3-kinase (PI3K) and leads to phosphorylation of endothelial nitric oxide synthase (eNOS); it is therefore related to the pathway mediating insulin’s metabolic effects. On the other hand, insulin may also activate the pro-atherogenic mitogen-activated protein kinase (MAPK) pathway in endothelial cells (35) (Fig. 2). Insulin receptor substrate 1 (IRS-1) is one of the key proteins downstream of the insulin receptor for signalling to metabolic effects, for example, glucose uptake in fat cells and NO production in endothelial cells (Fig. 2).
Insulin resistance is driven by pro-inflammatory cytokines and adipokines (see before). Their effects are by no means restricted to metabolically relevant cells. For example, leptin is known to have immunomodulatory capacities and to upregulate adhesion molecules on endothelial cells (36).
Endothelial cell dysfunction refers to an imbalance in release of vasodilating factors such as NO and prostacyclin, and vasoconstricting factors such as endothelin-1 and angiotensin-II (37,38). Noteworthy, endothelin-1 levels are significantly elevated in psoriasis and endothelin appears to be produced by keratinocytes (39,40). When this balance is changed, it predisposes the endothelium towards an atherogenic milieu. This will result in rolling of leucocytes, smooth muscle growth, impaired coagulation, vascular inflammation, atherosclerosis and thrombosis (41). In atherosclerotic coronary arteries vasodilation is impaired and a paradoxical constriction may occur, suggestive of endothelial cell dysfunction (42). Several lines of evidence provide links between insulin resistance and endothelial cell dysfunction, for example, work on the IRS-1: IRS-1 is one of the key proteins downstream of the insulin receptor for signalling to metabolic effects, for example, glucose uptake in fat cells and NO production in endothelial cells (Fig. 2). Low cellular IRS-1 identifies individuals who are markedly insulin resistant, have high proinsulin and insulin levels and exhibit evidence of early atherosclerosis measured as increased intima media thickness in the carotid artery bulb (43). Moreover, low IRS-1 expression may not only be a marker for insulin resistance, but also for arterial stiffness (44). These data imply that shared stressors, such as hyperglycaemia, cause oxidative stress and downregulation of IRS-1 in fat cells and endothelial cells, leading to insulin resistance and endothelial cell dysfunction.
With regard to psoriasis, several groups found evidence for endothelial cell dysfunction, using ultrasound methods; in particular, flow-mediated dilation was impaired (45–48). In one of these studies, the HOMA-IR was assessed as well and found to be significantly higher when compared with non-psoriatic controls (47), again stressing the link between insulin resistance and endothelial cell dysfunction.
As stated before, a change in the balance of vasodilating and vasoconstricting factors predisposes the endothelium towards an atherogenic milieu. Atherosclerosis is now being regarded as an inflammatory disease (49). Early events comprise adhesion-mediated leucocyte extravasation, which is facilitated by activated platelets and followed by infiltrating macrophages, releasing cytokines and enzymes such as matrix metalloproteinases, thus degrading the connective tissue matrix. This step is followed by the formation of a more advanced fibrous lesion with accumulating lipid-rich necrotic debris and smooth muscle cells. This fibrous-capped plaque then develops into an advanced and very complex lesion. Continuing inflammation may alter the fibrous cap to create an unstable plaque, rupture of which would cause thromboembolic complications, such as myocardial infarction or stroke.
Modules to be added to the backbone of the ‘psoriatic march’
As stated before, the concept outlined so far is a gross oversimplification, intended to spark further activities to investigate in detail the interdependences of psoriasis and associated comorbidity.
An aspect that needs to be incorporated into this hypothesis is obesity. Several epidemiologic studies identify obesity as a risk factor for psoriasis (50,51). A feasible pathogenetic link may be the true nature of obesity as state of smouldering systemic inflammation causing immune dysregulation. Abdominal fat represents an active endocrine organ, visceral adipocytes are the source of pro-inflammatory mediators, and there is an intimate relationship and cross-talk between the immune and the metabolic response system (52,53).
This inflammation may well represent a trigger factor for the manifestation of the psoriatic phenotype in genetically susceptible individuals. Xenogeneic transplantation studies in mice have shown that grafts from non-lesional skin of psoriasis patients spontaneously develop all histological hallmarks of psoriatic plaques under the influence of tumor necrosis factor- α (TNF-α) (54). Although, in this setting, TNF-α is most likely secreted by dendritic cells in the skin, it is also known to be overexpressed in adipose tissue of obese patients (55). As cytokines act in networks, no single mediator is likely to determine such complex processes. Still, this TNF-α connection historically provided the first link between obesity, insulin resistance and inflammation (56).
As stated earlier, genetics are another important field, as there is increasing evidence for common genetic risk factors for psoriasis and atherosclerosis, with a role for TH17-cells in general and interleukin (IL)-12 in particular in both diseases. Noteworthy, polymorphisms in the IL-12B gene signal susceptibility for psoriasis (57).
Finally, systemic inflammation may well backfire on psoriatic skin lesions, as insulin resistance substantially perturbates epidermal homeostasis (C. Bürger et al., manuscript submitted).
Clinical consequences: a concept for comprehensive monitoring of patients with severe psoriasis
Having summarized the currently available evidence in favour of the concept of the ‘psoriatic march’, it is still unclear if all relations discussed so far are really causal in nature (see next). However, the epidemiological data alone point out that patients with severe psoriasis often have comorbidity, most importantly premature cardiovascular disease. The management of patients with severe psoriasis therefore requires more than the assessment and treatment of skin symptoms alone.
Recently, the National Psoriasis Foundation published screening recommendations for psoriasis patients (58). With regard to cardiovascular comorbidity, these recommendations refer to those of the American Heart Association, but do not take into account the fact that psoriasis as such is an independent risk factor for cardiovascular diseases. The latter has been considered in another publication that defines – depending on the number of additional cardiovascular risk factors – different target ranges for those parameters assessed in psoriasis patients (59). Both recommendations are summarized in Table 1, which may serve as a template to monitor psoriasis patients in daily dermatological practice.
Table 1. A checklist to monitor psoriasis patients for cardiovascular risks factors (58,59)
Recommendation by the American Heart Association (58)
• Evaluate at least every 2 years • Target <120/80 mm Hg
• <140/90 mm Hg in all patients with psoriasis and ≤2 major risk factors for CAD • <130/80 mm Hg in patients with previous CVD, diabetes mellitus, chronic renal disease or ≥3 major risk factors
Body mass index
• Evaluate at least every 2 years • Target <25 kg/m2
• Evaluate at least every 2 years • Target ○ <102 cm (males) ○ <88 cm (females)
Evaluate at least every 2 years
Fasting blood lipids
• Evaluate at least every 5 years or every 2 years if risk factors1 are present • Total cholesterol ≤ 200 mg/dl • LDL ○ Optimal: <100 mg/dl ○ Near optimal: 100–129 mg/dl ○ Borderline: 130–159 mg/dl ○ High: 160–189 mg/dl ○ Very high: ≥190 mg/dl
• Evaluate at least every 5 years or every 2 years if risk factors1 are present • Target: <100 mg/dl
As a foot note, we wish to point out that – although formally correct – some of the suggested screening and/or monitoring procedures are not in line with guidelines of the respective societies. Actually, some parameters may be assessed in an easier way than suggested in Table 1. This is true for example in the case of monitoring blood glucose. The German Diabetes Society suggests that occasional measurements of blood glucose rather than fasting glucose may well serve as screening tool for diabetes. Noteworthy, HbA1C has been identified as a biomarker for cardiovascular risk in patients without diabetes (60).
Finally, a more comprehensive approach to psoriasis treatment is needed. Even today, there is evidence for a better clinical outcome in patients treated accordingly. In a controlled clinical trial, obese patients with moderate-to-severe psoriasis had a better response to low-dose cyclosporine if a calorie-reduced diet was included in their treatment regimen (61).
Academic consequences: questions for today’s (and tomorrow’s) research agenda
The question of whether or not psoriasis and cardiovascular comorbidity are directly and causally related is academically challenging and deserves an answer in its own right. Indirect evidence comes from two retrospective analyses, one showing that long-term continuous methotrexate treatment reduces cardiovascular morbidity among psoriasis patients (62), the other documenting reduction of CRP as a biomarker for cardiovascular risk in a registrational study with the TNF-α blocking biologic etanercept (63). However, prospective studies are needed to directly address this question. Conclusive evidence will only come from well-designed, large-scale, long-term multi-centre studies, focussing on hard clinical end points (e.g. myocardial infarction or death) rather than biomarkers; the additional baggage of other cardiovascular risk factors such as excessive smoking, hypertension, obesity or alcohol excess will have to be properly controlled. These studies will be difficult to perform as treatment intervention may obscure the results. Registries which collect data on traditional risk factors may be suitable to achieve this goal. These would also answer the question of whether we can stop the ‘psoriatic march’. To this end, we and others were able to demonstrate that successful continuous systemic anti-psoriatic therapy ameliorates biomarkers for cardiovascular risk, including cytokines, adipokines and endothelial cell dysfunction (64–67).
This viewpoint summarizes evidence that psoriasis is causally related to cardiovascular comorbidity, and suggests a pathogenetic concept to explain this link. Based on this concept, it is tempting to speculate that certain anti-psoriatic drugs may be more potent to stop the psoriatic march than others, depending on their respective mode of action. A particularly attractive target seems to be TNF-α: it is a well-known adipokine and known to induce endothelial cell dysfunction. Noteworthy, blockade of TNF-α normalizes vascular elasticity in vivo (68), and patients with rheumatoid arthritis treated with TNF-α inhibitors exhibit reduced cardiovascular morbidity and mortality (69,70). As there is some evidence that other cytokines targeted by biologics approved for the treatment of psoriasis – such as IL-12 and/or IL-23 targeted by ustekinumab – may also play a role in atherosclerosis (71), it will be interesting to compare the impact of such drugs not only on psoriatic skin lesions, but also on developing cardiovascular comorbidity.
The aforementioned clinical studies need to be backed up by in vitro experiments working out the molecular details of how the psoriatic inflammation induces endothelial cell dysfunction and how anti-psoriatic therapies may normalize it. Thus, experiments on altered insulin receptor signalling in endothelial cells under the influence of pro-inflammatory cytokines in general and TNF-α, IL-12, IL-17 and IL-23 in particular are needed. Of interest is not only the impact on kinase phosphorylation patterns, but also on functional consequences such as expression of adhesion molecules and – carrying the projects back into the clinic – vasoreactivity.
In summary, the concept of psoriasis being causally related to cardiovascular comorbidity, or – in more general terms – the idea of seemingly organ-specific inflammation driving atherosclerosis is supported by an increasing number of studies. More efforts both at the level of clinical and basic research are needed to turn this concept into the foundation upon which we can base a comprehensive approach to the management of psoriasis. This will remain a hot topic for ambitious scientists in cutaneous biology and related fields for years to come.
Conflicts of interest
WHB has received honoraria as a speaker or advisor for the following companies: Abbott, Biogen Idec, Essex, Janssen Cilag and Pfizer. BK has acted as a consultant for Abbott, Serono, Schering-Plough, Pfizer (formerly Wyeth) and Janssen-Cilag. He is in receipt of unrestricted research grants from Abbott and Janssen-Cilag.