An aetiological analysis of erythema nodosum in a community hospital
K N Sarveswari
No.1, Kamalam Apertments,15'V' Block, 4th Main Road. Anna Nagar, Chennai (Madras)-600 040
|How to cite this article:
Sarveswari K N, Swamikannu M, Ramakrishnan S. An aetiological analysis of erythema nodosum in a community hospital. Indian J Dermatol Venereol Leprol 2001;67:287-289
AbstractErythema nodosum (EN) is a reaction pattern in the skin characterized by septal panniculitis. Infectious diseases are most common cause of EN. This study was aimed to find out the most common causes for EN as seen in a community hospital over a period of two years.
Fifteen patients with EN were subjected to detailed clinical and laboratory investigations to establish the etiology. Immuno - fluorescence and PCR for Mycobacterium tuberculosis were performed in relevant patients.
Seven cases were diagnosed to be secondary to tuberculosis that was confirmed by a complete resolution of the lesions after a full course of anti TB chemotherapy. In one patient, the EN was drug induced, two were due to polyarteritis nodosa, one was due to SLE, one was due to streptococcal infection, and three were idiopathic in aetiology. Though the number of patients is small, it still establishes the fact that tuberculosis continues to be the most common cause of EN.
Erythema nodosum (EN) is a cutaneous reaction pattern, characterized clinically by erythematous tender nodules mainly on lower limbs and histologically by septal panniculitis. The disease runs an acute course,characterized during regression, by bruise-like skin color changes, subsiding in 3-6 weeks without scarring or atrophy. Various conditions have been associated with erythema nodosum (EN), most of them being infectious diseases.
Subjects and Methods
This is a 6 - month prospective study of patients with erythema nodosum attending the outpatient service facility between May′ 98 - May 2000 at Sundaram Medical Foundation, a community hospital, in Chennai (Madras), south India. The purpose of this study was
- To establish the frequency of various disorders associated with EN and to discuss their possible roles in causing EN. The data was then compared with those obtained from literature.
- To attempt to develop a cost effective protocol -that would aid in detection of associated aetiological factors, with minimal investigations.
- Patients with typical EN lesions, characterized by the presence of crops of tender erythematous nodules and papules without ulcers on legs were included. A detailed history of symptoms of arthralgia, pharyngitis, drug intake and diarrhoea was taken. As a first step, complete hemogram, anti-streptolysin 0 titer, Mantoux intradermal test and skiagram of chest was done in all patients [Table - 1].
Subsequently, to be cost effective,
- In those with positive Mantoux (more than 10mm induration) and a raised erythrocyte sedimentation rate (ESR more than 40 mm/first hr), the biopsy specimen was sent for polymerase chain reaction test (PCR) for Mycobacterium tuberculosis (MTB).
- In those with negative Mantoux test, but severe arthralgia, antinuclear antibody test was done and the biopsy of the specimen was also sent for immunofluoresecence.
Since PCR was negative in all specimens that we had sent, the diagnosis of tuberculosis was based on the following criteria:
(a) A raised ESR,(b) Positive Mantoux,(c) Therapeutic response to anti tuberculosis therapy (ATT) within 8 weeks.
Patients were treated according to the disease/infections detected. Those in whom no cause was found, empirical treatment with choroquine (or) anti - inflammatory drugs was given. All patients were reviewed every month for a period of six months.
A total of 15 cases of EN were seen over a two-year period. Six were males, eight were females and one male child. The mean age was 31 years, and range was from 9 years to 53 years. All patients presented with erythematous nodules and plaques over legs. One patient had nodules on the forearm. Arthralgia occurred in 6 patients, fever in 2 patients and pharyngitis in one. Based on clinical and laboratory evaluation, an aetiological factor could be implicated in 11 patients [Table - 2].
Seven patients were diagnosed to have EN secondary to tuberculosis. Complete resolution of symptoms with ATT, was seen in 5 patients with no recurrence of EN. One patient stopped treatment within 2 months since he was free from symptoms, and there was no recurrence of EN. One patient was lost to follow up.
Connective tissue disorders seemed to be the aetiological cause in 3 patients. Of these, a 11 - year -old boy with typical EN nodule, arthralgia and gangrene of little finger showed histopathological features of EN and vessel wall thickening suggestive of polyarteritis nodosa (PAN). Another 53 - year - old lady with erythematous plaque on the right shin, had raised serum creatinine level, mild hypertension, and showed histopathological features of polyarteritis nodosa. The boy responded to oral steroids and the lady to dapsone. A 24 - year -old lady presented with erythematous nodules on leg and forearm antibody with severe arthralgia and fever. Her Mantoux was negative but anti nuclear antibody (ANA) was positive.Her biopsy and immunofluorescence (IF) were consistent with systemic lupus erythematosis (SLE). She responded to steroid therapy but had recurrence of EN in 2 months time.
A lady who had preceding pharyngitis and raised ASO titre responded to antibiotics without any recurrence, showing a possible streptococcal aetiology. Only one patient gave a history of oral sulphonamides for upper respiratory infection (URI).
His ASO titre was not raised and the nodules subsided, without recurrence,on stopping the drug. For the rest of the 3 cases, the cause could not be identified. Empirical treatment with anti-inflammatory drugs in two and chloroquine in one cleared the lesion. As there was no recurrence,no further investigations were carried out.
This study, though small, was done in a community hospital (catering to middle income group of people) and point out certain interesting facts. The age and sex incidence and the clinical characteristics of EN seen in our series, were in accordance with literature findings.
The cause of EN was suggestive in 11 patients, (nearly 80%). The rest 20% responded to empirical treatment without recurrence.
Tuberculosis is no longer a cause of EN in USA and Europe. But in our series, tuberculosis was the major culprit. Nearly 7 out of 11 showed tuberculous aetiology. Though streptococcal infections are the common cause of EN in western literature, we had only one case with preceding pharyngitis and raised ASO titre, which responded to antibiotics.
Many infectious diseases associated with EN have been described, but it is virtually impossible to investigate all possible causes in patients presenting with EN, without specific symptoms. It should be noted that in cases with diarrhoea, culture for Yesinia enterolitica, salmonella and shigella has to be carried out., Since none of our cases had diarrhoea, this was not done.
Similarly, newer causes of EN described in the past decade inculde Human immunodeficiency virus (HIV) and Hepatitis C virus infections (HCV). It was not done in our cases as there was no clinical evidence of these.
Sarcoidosis, the second most common cause of EN in western literature, was not seen in our patients.
Other systemic diseases presenting with EN include collagen vascular disorders and malignancies. We had only one lady with SLE and 2 cases with polyarteritis nodosa. No malignancies were seen as the aetiological factor in our patients.
Drugs like halides, sulphonamides, oral contraceptives and antibiotics have been reported to cause EN. Only one of our patients had a suggestion of drug-induced EN.
Prof. Patrik Yesudian, Dr. T N Srinivasan, Mrs. Vijayalashmi Pandit.
Ryan TI. Cutaneous vasculitis. In; Champion RH, Burton JL, Burns DA, Breathnach SM. Textbook of Dermatology, Oxford: Blackwell Scientific Publications 1998;2155-2225.[Google Scholar]
Doxiadis SA. Aetiology of erythema nodosum. Br Med J 1949 ii : 844-845.[Google Scholar]
Vesery CMR, Wilkinson DS. Erythema nodosum. Br J Dermatol 1969;81:139-155.[Google Scholar]
Hannuksela M. Erythema nodosum. Arch Dermatol 1986; 4:8895.[Google Scholar]
Yus ES, Requena L. Erythema nodosum and nodular vasculitis. In: Arndt KA, Leboit PE, Robinson JK, Wintroub BU, eds. Cutaneous Medicine and Surgery. 1st edn, Philadelphia, USA: WB Saunders Co., 1996: 591- 598.[Google Scholar]
Domingo P, Ris J, Martinez E, et al. Erythema nodosum and hepatitis C. Lancet 1990;336:1377.[Google Scholar]
Shelly WB, Dorinda Shelly E. Erythema nodosum In: Advanced Dermatological Diagnosis, Philadelphia, USA. WB Saunders Co, 1996; 512-515.[Google Scholar]
Baden HP, Holcomb FD. Erythema nodosum from oral contraceptives. Arch Dermatol 1968;98: 634.[Google Scholar]
Bridges Al, Graziano FM, Calhoun W, et al. Hyperpigmentation, neutrophilic alveolitis and erythema nodosum resulting from minocycline. J Am Acad Dermatol 1990;22: 959 -962.[Google Scholar]