Are You Confident of the Diagnosis?
Erythema nodosum (EN) is the most common form of panniculitis. EN is generally viewed as a cutaneous reactive process that can be seen as an isolated, idiopathic condition (sometimes referred to as primary EN), or more often as a skin sign of another disease (secondary EN). The list of associated diseases is lengthy and ever-expanding, and often a thorough history is the most crucial diagnostic test in determining the underlying etiology, and, thus, the optimal investigations and best therapeutic approach.
EN can be triggered by infections of all types, medications, hormones (including birth control pills and pregnancy), cancer, autoimmune diseases, chronic inflammatory diseases (sarcoidosis and inflammatory bowel disease chief among them), and more. The underlying disease often depends on the regional and seasonal setting in which the patient presents; in areas of endemic tuberculosis, for instance, TB may be a more likely trigger for EN than sarcoidosis; conversely, most cases are seen early in the calendar year, when streptococcal infections are most prevalent.
An exhaustive review of diseases associated with EN was compiled by Requena and Sanchez Yus in their 2008 review; readers are refered to Table I of that manuscript for a comprehensive list of disease associations.
|PhysicalRest, elevation, compression*||LocalIntralesional triamcinolone injections (generally 5mg/mL)||SystemicFirst LineNSAIDsSuggested: Indomethacin(50mg 3 times daily); naproxen (500mg twice daily)Alternatives: Ibuprofen (200 to 800mg every 6 to 8 hours); aspirin (325mg daily); celecoxib (100 to 200mg twice daily); ketoprofen (25 to 75mg every 6 to 8 hours); diclofenac (50 to 75mg twice daily)Potassium iodide:Tablet(400 to 900mg daily); Saturated solution (1000mg/mL; 1 to 3 drops daily, escalated weekly up to 5 drops 3 times daily)Second LineColchicine (0.6 to 1.2mg twice daily); prednisone (20 to 40mg daily); hydroxychloroquine (200mg twice daily); dapsone (25 to 100mg daily) AlternativesThalidomide; TNFα inhibitors; infliximab; etanercept; adalimumab; mycophenolate mofetil; cyclosporine; erythromycin; extracorporeal monocyte granulocytapheresis|
Characteristic findings on physical examination
Patients will often describe the sudden appearance of numerous red, tender nodules, typically located on the anterior, extensor lower extremities, and often seen bilaterally and symmetrically. Rarely the lesions will involve the thighs, trunk, or upper extremities. Individual lesions frequently resolve spontaneously over 2 to 6 weeks without ulceration or scarring, but persistent lesions are not uncommon, and recurrences are frequent, and intermittent recurrent disease occurs in a subset of patients (1/6 to 1/3). EN is often accompanied by systemic symptoms, which may vary depending on the underlying disease process but often include fever, fatigue, malaise, arthralgias or arthritis, cough, and gastrointestinal complaints.
Physical examination is generally similar across all patients with EN. Almost all patients present with symmetric, tender, slightly warm, raised and subcutaneous nodules and plaques that often start out with a frankly red color, which then evolve to duller earth tones with yellows and browns as the lesions become more chronic and flatten out, occasionally resembling a bruise (erythema contusiformis). Nearly all lesions are located between the ankles and knees, often on the anterior shins and anterolateral lower leg bilaterally (Figure 1, Figure 2). Fever and joint pain or swelling are not uncommon.
What you should be alert for in the history
A detailed history is integral in narrowing the scope of investigations and leading clinicians towards the most likely underlying trigger for EN. Secondary EN often requires treatment of the instigating underlying process to treat the skin lesions. Given the long list of potential triggers, physicians may be overwhelmed by the potential investigations to identify the cause. Infection, sarcoidosis, connective tissue disease, inflammatory bowel disease, medications, pregnancy, and malignancy are the most common associations. Some clues that EN may be presenting as a reaction to an underlying systemic disease may be the history of fever, cough, sore throat, diarrhea, arthritis; leukocytosis, increased inflammatory markers (ESR, CRP), or pulmonary pathology may also suggest secondary EN.
Evaluation of all patients should include a detailed medication history and comprehensive review of systems. Any isolated complaint may warrant further investigation. Family history, foreign travel, exposures (pets, hobbies, sick contacts, work), and a detailed past medical history aimed at eliciting any potential underlying systemic disease is critical. The list of medications to which EN has been attributed is immense; the most common culprits are oral contraceptive pills, halogens (iodinated compounds, bromides), antibiotics (sulfonamides, penicillin/amoxicillin, and others), aspirin, and non-steroidal anti-inflammatory drugs (NSAIDs). A history of NSAIDs as potential triggers is important to elucidate, as those drugs are often utilized in the management of EN.
Expected results of diagnostic studies
Initial laboratory investigations should include a complete blood count (with differential), ESR and CRP, antistreptolysin O (ASO) titer or anti-DNAse titer, throat culture, urinalysis and urine pregnancy test, purified protein derivative (PPD) intradermal tuberculin test (the newer interferon gamma release assays, including the widely available Quantiferon (QFT) TB gold test, may be appropriate to substitute), and chest radiography.
If historical elements suggest alternate systemic disorders, targeted investigation is warranted (potentially including anti-nuclear antibodies or other rheumatologic serologic investigations, liver function tests and hepatitis serologies, hormone studies, stool culture for bacteria and ova and parasite analysis, respiratory virus panel screening for RSV, gastrointestinal track evaluation via radiography or direct visualization, evaluation of urine or serum for evidence of a fungal infection, eg, urine histoplasmosis antigen or serum histoplasmosis antibodies), skin biopsy to rule out mimickers or confounders, such as infectious panniculitides or Sweet’s syndrome, which may mimick EN.
Generally the white blood count is normal or slightly elevated (10,000 per mm3), and the ESR and CRP are often high initially but fade as the lesions evolve. The ASO and anti-DNAse titers correspond with a recent streptococcal infection. While ASO titers may be falsely elevated (in liver disease or in cases of certain infections), a positive ASO in the setting of EN or with a history of recent throat symptoms should be considered highly suspicious for a recent streptoccal infection.
PPD tests must be interpreted based on the accepted guidelines, which depends on the patient’s exposure history and immunocompetency; IFN-gamma release assays may be more specific for tuberculosis in populations with high rates of prior BCG vaccination. A chest X-ray is essential to evaluate for pulmonary infectious processes, but is used primarily to asses for hilar adenopathy; the presence of hilar adenopathy in the setting of EN may suggest Löfgren’s syndrome, particularly if there is fever and arthritis. This acute form of sarcoidosis, more common in patients of Northern European ancestry, generally portends a good prognosis and benign course.
Seasonal and regional disease variations are important to keep in mind. Streptococcal infections are the most common associated cause of EN overall; in countries with high rates of TB, EN secondary to TB may be seen with some regularity; conversely, in populations with a high incidence of sarcoidosis, that may be a more likely cause. While Beçhet’s is a rare disease, when evaluating a patient in Turkey, EN may be a presenting sign of Behçet’s.
A skin biopsy is not required to make the diagnosis of EN in patients with classic presentation, but may be helpful in confirming the diagnosis. Erythema nodosum is the hallmark septal pannicullitis. While pathology can vary depending on the acuity of the lesion sampled, in nearly all cases of EN the septa are thickened and fibrotic, with an inflammatory infiltrate, which may spill over into the periseptal adipocytes (Figure 3). The panniculitis may be accompanied by a less intense superficial and deep perivascular lymphocytic infiltrate as well.
Early lesions of EN may show a predominantly neutrophilic infiltrate, often with some hemorrhage and early septal thickening, while later lesions are more likely to reveal frank fibrosis, lymphocytes, histiocytes, and multinucleated giant cells. “Miescher’s radial granuloma” are small collections of histiocytes, which may be scattered throughout the septae or grouped in focal areas (Figure 4, Figure 5). EN typically spares the vessels, although rare cases with small vessel vascular inflammation and fibrin thrombi have been reported, and the presence of vasculitis should prompt consideration for alternate diagnoses.
EN has a somewhat specific clinical appearance and course, and often the challenge is determining the cause of EN, rather than distinguishing EN itself from other panniculitides. Evaluation of histologic specimens is helpful in determining whether clinical lesions suspect for EN show a predominantly septal or lobular panniculitis; the presence of a lobular panniculitis will often guide the clinician down an alternate diagnostic pathway.
When evaluating EN, the differential diagnosis often includes erythema induratum, also known as nodular vasculitis, which generally occurs on the posterior lower legs and ulcerates far more frequently than EN, with histologic evidence of vasculitis another distinguishing feature. Lupus profundus or lupus panniculitis, a deep presentation of cutaneous lupus, can present with nodules; often these lesions will have an overlying “dell” to the skin and may show features of more common types of cutaneous lupus, such as epidermal change.
Subcutaneous panniculitic T-cell lymphoma can present with nodules; these nodules are usually not as widespread and numerous, and are less apt to self-resolve. Atypical mycobacterial infection and certain sporotrichoid infectious processes may mimic the red lower leg nodules of EN; a detailed history and the lack of bilateral involvement may be clues. Other panniculitides can present with red-brown nodules on fatty areas such as the lower legs (alpha1-antitrypsin deficiency, pancreatic panniculitis, etc), but histologic evaluation often distinguishes these processes from the characteristic septal panniculitis of EN.
Erythema nodosum migrans, subacute nodular migratory panniculitis of Vilanova and Pinol, and chronic erythema nodosum were once conisdered separate entities, but are now felt to be variant presentations of classic EN; patients with these inflammatory patterns may develop unilateral leg lesions, which can expand in an enlarging circular fashion, or individual nodules may run together into larger, broad raised and subcutaneous plaques. These variants are unusual and represent the exception to the more common, classic presentation of EN.
Who is at Risk for Developing this Disease?
EN occurs in all ages, races, and ethnicities, affecting people in urban areas and rural areas alike. The overall incidence is estimated at 1 to 5 cases per 100,000 persons per year, with some variation based on country. There is a strong seasonal shift, with many cases occurring during the winter, the majority presenting in the first half of the calendar year. In adults, EN is more common in females, with a male:female ratio of 1:5 or 6; in children, the ratio is nearly normal. The peak incidence is in the 20s to 30s, though EN can occur at any age. Pregnant women and women on oral contraceptive pills are at relatively increased risk.
What is the Cause of the Disease?
Although numbers vary, generally in half of the patients with EN, an underlying etiology can be determined (infections, medications, systemic diseases, as noted elsewhere). The actual cause of the septal inflammation itself, though, remains uncertain. Many feel that the panniculitis is a form of a type IV delayed hypersensitivity reaction in response to a triggering antigen. The wide range of associated diseases and potential antigens suggests EN is a nonspecific inflammatory reaction pattern, potentially in response to multiple stimuli.
There is some evidence that circulating immune complexes may deposit within the subcutaneous fat and activate complement. Recent studies have shown that there are increased reactive oxygen species in patients with EN, with levels of reactive oxygen intermediates correlating with disease activity. A genetic polymorphism in the TNF-alpha gene promoter was found in patients with sarcoidosis and EN, suggesting a potential role for TNF-alpha; other proinflammatory cytokines have been linked to the development of EN as well. The predeliction for the anterior shins in nearly all patients suggests that physical factors may play a role in disease development.
Overall, EN likely represents a cutaneous reaction pattern of septal inflammation and panniculitis due to an array of triggering factors and stimuli that set off an inflammatory cascade leading to early neutrophil activation and progressive fibrosis.
Systemic Implications and Complications
As noted, EN is a cutaneous reaction pattern that may be seen as an isolated, idiopathic occurrence or in association with a tremendous variety of underlying systemic diseases and triggers. While the precise numbers vary somewhat from study to study, overall approximately one half of all cases of EN are associated with a systemic process beyond the panniculus; investigations may identify triggers including infection, sarcoidosis, connective tissue disease, inflammatory bowel disease, pregnancy, or malignancy, or an identifiable medication associated with the onset of skin lesions.
Requena and Sanchez Yus compiled a comprehensive list in their 2008 review on the subject. The most common and important systemic disease associations are discussed here.
Overall, infections are the most often identified trigger of EN, particularly streptococcal pharyngitis. Other notable infectious triggers of EN include gastroenteritis due to bacterial pathogens (Yersinia, Salmonella, and Campylobacter), Mycoplasma, Chlamydia pneumonia and C trachomatis, and deep fungal infections (histoplasmosis, coccidioidomycosis). Viral triggers include the hepatidites (B more often than C, and the virus and the vaccine may induce EN) and Epstein-Barr virus.
Mycobacterial triggers of EN have been reported, including Mycobacterium tuberculosis (there are reports of EN occurring following PPD placement and after BCG vaccination). It is important to distinguish EN from erythema induratum, which occurs as recurrent groups of tender violaceous nodules on the posterior lower legs, which often ulcerate and drain, leaving behind scarring and dyspigmentation. Rarely EN may occur in association with other infectious triggers, including rickettsial disease, syphilis, human immunodeficiency virus, Herpes simplex virus, and parasitic infestations (amebiasis, giardiasis).
Evaluation for infection requires a rational, cost-effective approach, dictated primarily by a thorough history, review of systems, and physical examination. All patients with EN warrant evaluation for streptococcal infection (throat culture, ASO, DNAse titers). Patients with a history of travel to areas endemic for deep fungal infections (such as to the Mississippi/Ohio River valley in the United States for histoplasmosis, or the Southwest US for coccidioidomycosis) may require serologic testing, although in immunocompetent hosts pulmonary infection with those agents may be self-limited.
While all patients warrant evaluation for
M tuberculosis infection, particular attention should be paid to patients who are immunocompromised or may demonstrate anergic response to the intradermal PPD test, as well as patients with a clear exposure history or who have been institutionalized for any reason. If there is a high clinical suspicion for tuberculosis, acid-fast bacilli sputum analysis may be considered.
Patients with a history of gastrointestinal complaints, diarrhea, bloody diarrhea, or known exposures may require stool cultures; identifying the specific organisms in question to the microbiology laboratory may increase culture yield in some cases. Further investigation for infectious agents should be determined by the clinician on a case-by-case bases depending on the constellation of presenting symptoms.
EN is typically seen in sarcoidosis in the setting of Löfgren’s syndrome, an acute form of the disease where patients present with fever, arthritis, bilateral hilar adenopathy, and EN. This variant is more common in patients of Northern European descent. EN can occur in the setting of other variants of sarcoidosis as well. All patients presenting with EN should have chest radiography performed; the presence of hilar adenopathy should prompt clinicians to consider sarcoidosis.
Importantly, M tuberculosis, deep fungal infections, and lymphoma may all induce EN and can show hilar adenopathy on chest imaging. The presence of arthritis and fever may be helpful, but in cases where the diagnosis is in question, clinicians are advised to maintain a broad differential diagnosis and consider further evaluations.
Inflammatory Bowel Disease
EN may occur in the setting of both Crohn’s disease and ulcerative colitis; traditionally EN is more commonly associated with ulcerative colitis. Flares of the intestinal inflammation may be associated with the development of EN, and control of the colitis may lead to resolution of EN and prevention of new lesions. Epidemiologically the peak incidence of EN coincides with one of the common ages when patients present with IBD; a more helpful clue is symptoms of gastrointestinal distress. Patients with frequent or loose bowel movements, or a history of bloody diarrhea, may warrant referral to a gastroenterologist for consideration of direct intestinal visualization (colonoscopy and/or endoscopy).
Other chronic inflammatory diseases
EN has been described as a nonspecific cutaneous inflammatory pattern seen with a broad range of systemic inflammatory conditions. Two of the more noteworthy diseases that may be more strongly associated with EN are Behçet’s disease and Sweet’s syndrome. Many patients with Behçet’s develop lesions that resemble EN clinically; notably, histologic evaluation of these lesions often reveals a vasculitis (similar to that seen in the primary disease lesions of Behçet’s disease). In areas of high prevalence, such as Turkey, or in patients with a history of ocular or genital lesions and a reported history of pathergy, consideration of Behçet’s is appropriate.
The relationship between Sweet’s syndrome and EN is similarly confusing; both entities can occur in association with a number of overlapping underlying systemic processes, and both Sweet’s and EN are neutrophil-mediated processes in the acute setting; both diseases respond to similar treatments. The coexistence of Sweet’s and EN has been reported in a number of series, and it may be in some cases that the two diseases are actually cutaneous reactive processes to a common underlying trigger in some patients.
The diagnostic criteria for Sweet’s syndrome have many overlapping features with the symptoms of patients with EN. The presence of “juicy” edematous plaques with dense, more superficial neutrophilic inflammation and dermal edema may suggest the presence of Sweet’s in patients with EN (the minor criteria for Sweet’s syndrome of preceding fever or infection, fever, arthralgia, leukocytosis, response to steroids, and elevated ESR may all be seen in patients with EN alone).
EN has been reported in association with autoimmune diseases ranging from antiphospholipid antibody syndrome to systemic lupus erythematosus to polychondritis to rheumatoid arthritis to systemic vasculitides and more. A focused evaluation based on patient’s presenting complaints, or the absence of a clear systemic trigger for EN after appropriate initial evaluation, may prompt further workup aimed at elucidating these diseases.
EN has been reported as a skin sign of malignancy for a number of tumors but is most often seen in patients with lymphoma or leukemia. The presence of EN, or flares of EN, may signal disease progression or relapse. Patients with a variety of solid organ tumors have developed EN. Age-appropriate screening is always indicated; more thorough, targeted, and specific investigations depend on the clinical scenario. The diagnosis of lymphoma may be challenging, as “B” symptoms may initially be mistakenly ascribed to the presence of EN, and if there is hilar adenopathy that, too, may be attributed to other sources.
Patients with night sweats, weight loss, or lymphadenopathy on examination, or those with unexplained abnormalities on complete blood count analysis, should undergo further evaluation. Patients whose EN remains chronic or refractory to therapy may require further workup to identify the presence of an underlying disease. Overall large-scale epidemiologic studies are lacking and precise figures are unknown, but EN is rarely associated with malignancy.
All women of child-bearing age who present with EN should undergo pregnancy testing (urine pregnancy test). The EN is thought to occur as a result of either estrogen levels or the altered ratio of estrogen and progesterone. The occurrence of EN in association with female hormones, particularly birth control pills, is well established. The connection with estrogen may explain the markedly increased female-to-male incidence ratio of EN. Besides pregnancy and oral contraceptive pills, hormone replacement therapy has been identified as a potential trigger of EN.
A thorough medication history is important in all patients; it is not uncommon for young women to neglect to mention birth control pills as part of their medication history, and targeted questioning about any hormonal therapy is essential in the evaluation of patients with EN.
Beyond estrogen-containing compounds, there is an extensive list of other medications of multiple drug classes that can trigger EN. Antibiotics, particularly penicillins (amoxicillin in particular) and antibacterial sulfonamides, are frequent culprits. Halogenated compounds (bromides, iodides) are less commonly prescribed medications but have been reported to trigger EN. Notably, potassium iodide is frequently employed in the management of EN (see below). It is important to note that many other drugs used in the treatment of EN have also been identified as potential triggers of EN, including NSAIDs, dapsone and thalidomide.
The list of drugs inducing EN is ever-expanding, with recent reports of proton pump inhibitors and leukotriene inhibitors as culprits; even echinacea supplements have been reported to cause EN. When evaluating EN it is critical to review every prescription drug, herbal supplement, over-the-counter medication, and recreational drug that patients are taking, and to consider any and all as potential suspects. In cases where physicians initiate treatment with NSAIDs and the lesions of EN flare, it is recommended that the treating clinician reassess the history and initial presentation for any clues as to whether the onset of EN initially may have been related to medication use.
If there is any concern that medications being used to manage the EN may in fact be worsening the disease, clinicians should maintain a high index of suspicion and low threshold to adjust therapy accordingly.
Table 1. Treatment options for EN
Optimal Therapeutic Approach for this Disease
The management of EN is two-fold: first, identify whether there is an underlying trigger, and remove or treat that condition as necessary. In cases of medication-induced EN, removing the inciting agent may lead to spontaneous remission. Infection-associated EN will often respond to therapy targeted at the underlying infectious process. In many systemic conditions associated with EN, controlling the underlying disease will lead to improvement and resolution of EN. It is worth re-emphasizing that many cases of EN will improve and often resolve on their own; while recurrences are not uncommon, the lesions themselves are often self-limited and resolve without sequelae of atrophy or scarring.
Treatment options for EN are summarized in the Table. When deciding on a specific therapy to target EN itself, it is essential to weigh patient-specific risks and benefits for each option. Notably many of the suggested treatments are not safe to use in pregnancy, and the importance of ruling that out prior to initiating therapy is re-emphasized. Additionally, many of the therapeutic options have been described as potential causative agents of EN; in cases where a drug is suspect, neither that drug nor members of that drug class are suggested in the management of drug-related EN. In cases where malignancy or infection is suspected, immunosuppressive medications should be avoided.
All patients should incorporate bed rest and leg elevation (above the level of the heart) as part of the treatment approach. Compressive therapy is advised whenever possible (gradient support stockings, ACE-compression wraps, or Coban self-adherent wraps), to a 15 to 20mm Hg pressure (unless contraindicated due to severe arterial insufficiency or circulatory compromise).
NSAIDs are effective at treating the skin lesions and often the systemic symptoms of EN in many patients. Indomethacin and naproxen have the most evidence and experience supporting their use. These are generally well tolerated, although patients with renal disease, gastritis, or esophagitis may develop drug-related complications. Starting with indomethacin 50mg 3 times daily along with compression, elevation, and rest is an effective initial treatment and will lead to good results in the majority of patients with idiopathic EN. NSAIDs should not be used late in pregnancy.
Potassium iodide is an effective, generally well-tolerated, non-immunosuppressive treatment that can achieve rapid disease control. It is generally administered as a saturated solution (SSKI); patients should start with 1 to 2 drops (often dissolved in orange juice to increase patient tolerability) taken 3 times daily. The dose can be increased by 1 drop per dose weekly up to 5 drops 3 times daily. Prior to initiation of therapy, patients should be queried for a history of thyroid disease. Patients who are on SSKI for 1 month or more should have monthly thyroid-stimulating hormone levels checked. If hypothyroidism is detected, treatment should be stopped.
Patients on certain medications (amiodarone, sulfonamides, lithium) are at increased risk for developing thyroid disease. As SSKI contains potassium, patients with known renal disease or who are taking medications that could raise their potassium should be counseled for signs and symptoms of hyperkalemia and monitored appropriately for this rare but dangerous toxicity. In patients with extensive disease or who are not responsive to NSAIDs, or for whom NSAIDs are contraindicated, SSKI (or the tablet form of potassium iodide) is a reasonable option.
In patients who fail to respond to rest, elevation, compression, and/or NSAIDs or SSKI, alternative therapies may be considered. Individual recalcitrant or refractory lesions may respond quickly to intralesional steroid injections. While 5mg/mL is a reasonable starting dose, certain larger lesions may require higher doses; high concentrations of intralesional steroids may lead to atrophy and hypopigmentation. Colchicine is particularly effective in patients with Behçet’s-associated EN. Colchicine frequently causes gastrointestinal complaints, which are typically dose-related, but therapy is generally well tolerated with low risk for severe systemic adverse events (bone marrow suppression). Colchicine is contraindicated in pregnancy.
Prednisone is a very effective drug at controlling EN, and in severe, widespread, or very symptomatic patients it may be appropriate to consider initial treatment with prednisone to achieve more rapid disease control. Infection and malignancy should be appropriately evaluated and ruled out prior to initiating systemic immunosuppressive therapy.
Patients should generally not be maintained on chronic prednisone for EN alone. Certain cases of chronic EN respond well to hydroxychloroquine therapy; this drug may be particularly appropriate in rare cases of lupus-associated EN. Dapsone is very effective at treating certain cases of neutrophilic-mediated inflammation and neutrophilic vasculitides, and some patients with EN resistant to other treatment approaches will respond to dapsone therapy.
Some patients with severe, refractory, recalcitrant disease may require alternative immunosuppressive and immunomodulatory agents. TNF alpha likely plays a role in sarcoidosis-associated EN and may be an important cytokine in EN in general. Thalidomide and the newer biologic TNF alpha-inhibitors have been used with success in treating limited numbers of patients with EN. Mycophenolate mofetil, cyclosporine, erythromycin, and extracorporeal monocyte granulocytapheresis have been reported to induce remission of EN in limited numbers of patients.
Patients with EN should undergo an initial thorough history, physical examination, and detailed review of systems, with close attention paid to any medications or exposures. Any identifiable underlying potential disease triggers should be investigated and, if possible, treated or removed. A basic initial laboratory workup and chest radiography is indicated in all patients (complete blood count, ESR and CRP, ASO or anti-DNAse titer, throat culture, urinalysis, urine prenancy test, and PPD (or interferon gamma release assay). Additional targetted investigations are warranted in select patients on a case-by-case basis based on their presenting signs, symptoms, and history.
A skin biopsy may be helpful in supporting the diagnosis or ruling out other entities.
Initial treatment with rest, elevation, and compression, often supplemented with NSAIDs (indomethacin 50mg 3 times daily) for symptomatic (and EN-specific) relief should be initiated. While lesions normally recede spontaneously, usually over weeks, often patients on therapy will improve within 7 to 14 days. Patients should be monitored closely for associated systemic diseases and appropriate investigations are warranted. It is generally recommended that all patients are up-to-date on age-appropriate cancer screening.
If patients experience a disease flare or relapse, it is worthwhile to re-evaluate the entire case, attempting to identify any triggers, extracutaneous symptoms or potential disease associations, or other signs of systemic disease.
Patients should be counseled about the broad range of possibilities that can be associated with EN. It is imperative to ensure the patient informs the clinician of any new or changing symptoms. Patients should be counseled about the frequently relapsing nature of EN and should be told to expect recurrent lesions.
Unusual Clinical Scenarios to Consider in Patient Management
Erythema nodosum migrans, subacute nodular migratory panniculitis, and chronic erythema nodosum are now generally accepted as variants of EN. These forms of disease may present with atypical clinical features, including large lesions or lesions that run together. There have been atypical variants of EN described in children with isolated involvement of the palms or soles, which may be unilateral; biopsy of these lesions shows classic features of EN and may be helpful in making the diagnosis.
What is the Evidence?
Requena, L, Sanchez Yus, E. “Erythema nodosum”. Dermatol Clin. vol. 26. 2008. pp. 425-38. (Comprehensive review article covering the etiologies of EN in depth, as well as describing the pathogenesis, clinical and histologic features, laboratory findings, and basic treatment outline.)
Schwartz, RA, Nervi, SJ. “Erythema nodosum: a sign of systemic disease”. Am Fam Physician. vol. 75. 2007. pp. 695-700. (Good overview of the diseasethat covers the most common features of EN, with useful tables summarizing key features.)
Gilchrist, H, Patterson, JW. “Erythema nodosum and erythema induratum (nodular vasculitis): diagnosis and management”. Derm Ther. vol. 23. 2010. pp. 320-327. (Excellent comparison of EN and erythema induratum, with a detailed discussion of EN treatment options.)
Papagrigoraki, A, Gisondi, P, Rosina, P, Cannone, M, Girolomoni, G. “Erythema nodosum: etiologic factors and relapses in a retrospective cohort study”. Eur J Dermatol. vol. 20. 2010. pp. 773-7. (Retrospective analysis of 124 patients with EN with focus on etiology.)
Mert, A, Ozaras, R, Tabak, F, Pekmezci, S, Demirkesen, C, Ozturk, R. “Erythema nodosum: an experience of 10 years”. Scand J Infect Dis. vol. 36. 2004. pp. 424-7. (Retrospective analysis of 50 patients with EN with a focus on identifying factors that predicted primary/idiopathic EN vs secondary/disease-associated EN.)
Mert, A, Kumbasar, H, Ozaras, R, Erten, S, Tasli, L, Tabak, F. “Erythema nodosum: an evaluation of 100 cases”. Clin Exp Rheum. vol. 25. 2007. pp. 563-570. (The follow-up to the above study, looking at 100 patients in an attempt to analyze features suggesting secondary EN. Excellent discussion reviewing systemic disease associations with EN.)
Mana, J, Marcoval, J. “Erythema nodosum”. Clin Dermatol. vol. 25. 2007. pp. 288-94. (Discussion of EN with a focus on Löfgren syndrome.)
García-Porrúa, C, González-Gay, MA, Vázquez-Caruncho, M, López-Lazaro, L, Lueiro, M, Fernández, M. “Erythema nodosum. Etiologic and predictive factors in a defined population”. Arthritis Rheum. vol. 43. 2000. pp. 584-92. (Study of 106 patients with EN in Spain looking at etiologies of the disease.)
Cribier, B, Caille, A, Heid, E, Grosshans, E. “Erythema nodosum and associated diseases.A study of 129 cases”. Int J Dermatol. vol. 37. 1998. pp. 667-72. (A large study looking at 129 cases of EN and exploring disease associations.)
Allen, RA, Spielvogel, RL, Lebwohl, MG, Heymann, WR, Berth-Jones, J, Coulson, I. “Treatment of skin disease: comprehensive therapeutic strategies”. Erythema nodosum. 2010. pp. 223-5. (Textbook with therapeutic options divided nicely into first-, second-, and third-line treatments, with annotated references for each therapy.)
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.