September 2004

The guest columnist for this month’s “What’s Your Diagnosis?” is Jared J. Lund, a fourth-year medical student (MS-4) at the Texas A&M University System Health Science Center College of Medicine. In addition to being an excellent student, Lund is involved in many extracurricular activities, including being one of a select few students who serve on the College of Medicine’s Admissions Committee. Jared’s hometown is Flower Mound, Texas, and he graduated from Brigham Young University prior to matriculation at A&M. He plans to become a pediatric dermatologist.

An 11-year-old African-American girl is air-evacuated to our institution from an outside hospital with a one-day history of ataxia, slurred speech and multiple episodes of emesis, with reported computed tomography (CT) findings showing a brain stem mass. Upon arrival, her mother also mentions a one-month history of poor oral intake with minor weight loss.

The patient’s past medical history is significant for cerebral palsy with static encephalopathy, left-sided hemiparesis and a severe seizure disorder associated with multiple hospitalizations since the age of 6 months. She takes multiple seizure medications including lamotrigine (Lamictal, GlaxoSmithKline), gabapentin and levetiracetam (Keppra, UCB). Her immunizations are up to date, and there has been no recent travel or animal exposure.

Examination reveals a well-appearing female with stable vital signs. Pertinent findings include facial and left-arm edema as well as bilateral pretibial edema. It is difficult to assess her gait because of her hemiparesis. Her skin shows both hypopigmented and hyperpigmented macular lesions in a mottled appearance as seen in Figures 1-4. These lesions have been present for quite some time, according to her mother.

Laboratory studies, including a complete blood count, serum electrolytes (sodium, potassium, chloride, bicarb and glucose), blood urea nitrogen and creatinine are all normal. However, her erythrocyte sedimentation rate (ESR) is elevated at 120. Initial urinalysis (UA) shows trace protein. A chest radiograph is normal. Additional imaging studies, including CT angiography, show no mass but do reveal some acute bleeding in the posterior fossa and a 2- to 3-mm saccular aneurysm of the posterior inferior cerebellar artery. Cerebral arteriography additionally revealed a small aneurysm along the right anterior inferior cerebellar artery as well as inflammatory changes along the vessel walls.

On day 2 of hospitalization, the patient is noted to be hypertensive, with a blood pressure of 190/100, and febrile, with a temperature of 102.6°F. A grade II/VI systolic murmur is heard on examination. A repeat UA is again unremarkable. Cultures including blood, cerebrospinal fluid and urine for fungal and bacterial pathogens are negative. Additional studies such as cytoplasmic antineutrophilic cytoplasmic antibodies (C-ANCA) and perinuclear antineutrophilic cytoplasmic antibodies (P-ANCA) are also negative. Echocardiogram is also normal. The result of a skin biopsy is pending.

What’s Your Diagnosis?

1. Wegener’s granulomatosis
2. Microscopic polyangiitis
3. Bacterial endocarditis
4. Polyarteritis nodosa
5. Henoch-Schönlein purpura

Answer

This is an example of a patient with polyarteritis nodosa (PAN). In 1866, Adolf Kussmaul and Rudolf Maier (both German physicians) described this disease during an autopsy that revealed multiple nodules along the medium-sized arteries, with histological examination showing focal inflammatory exudates. The famous Viennese pathologist Karl Rokitansky may have been the first to describe PAN in 1852 in a patient at autopsy but lacked histological confirmation to recognize it as an inflammatory process (Matteson E. “A History of Idiopathic Vasculitis.” Rochester, Minn.: Mayo Clinic Press; 1999). The American College of Rheumatology developed criteria for the diagnosis of PAN in 1990 based mainly on clinical findings as seen in the table below. In 1992, The Chapel Hill Consensus Conference classified systemic vasculitides based on histological characteristics and defined PAN as a systemic focal necrotizing vasculitis affecting the small- and medium-sized arteries without glomerulonephritis or involvement of arterioles, venules or capillaries (Jennette J, et al. Arthritis Rheum. 1994;37:187-192).

Unknown etiology

PAN is a rare disease in children with an unknown etiology. A clear association with hepatitis B exists, but this is more common in adults. Prior infections with such agents as group A streptococcus, cytomegalovirus, parvovirus, Epstein-Barr virus and tuberculosis have been linked to PAN. Such associations suggest a possibility of a post-infectious autoimmune response in susceptible patients (“Nelson’s Textbook of Pediatrics.” 17th ed. 829-830).

The initial presentation of PAN may be variable because of multiple organ system involvement. Patients usually present with symptoms of fever, malaise, anorexia and myalgias. Abdominal pain, nausea and vomiting frequently indicate gastrointestinal tract involvement. Neurological symptoms such as transient monocular blindness, peripheral neuropathy and cerebral ischemia may also occur. Subcutaneous nodules, palpable purpura, painful skin ulceration and livedo reticularis (seen in the photos of the patient presented, and those of another in Figures 5 and 6) are some of the skin manifestations seen in PAN. As seen in these contrasting patients, having dark skin makes the rash more subtle.

Clinical symptoms result from minimal to severe vessel inflammation and sometimes aneurysm formation. This causes disruption of blood flow that may lead to organ ischemia, infarction and hemorrhage in any of the organ systems involved. Because this disease may have such an insidious course, such severe symptoms as seen in our patient may be the initial presentation.

Laboratory studies are nonspecific and usually demonstrate anemia, leukocytosis, thrombocytosis and an elevated ESR. Anti-neutrophilic cytoplasmic antibodies (ANCA) are a marker that is commonly positive in certain vasculitic syndromes and has been seen in childhood PAN. However, this is the exception rather than the rule, as most cases of PAN are ANCA-negative (Ozen S. Best Pract Res Clin Rheumatol. 2002;16:411-425). In ANCA testing, human neutrophils are incubated with the patient’s serum, and through the process of indirect immunofluorescence microscopy, a pattern of staining is seen. A diffuse, granular stain in the cytoplasm is interpreted as a positive C-ANCA, as autoantibodies react with the enzyme proteinase 3 (PR3). This is characteristically seen in Wegener’s granulomatosis, with specificity above 90% in active disease. A positive P-ANCA produces a perinuclear pattern of staining when autoantibodies bind to the enzyme myeloperoxidase. This is often seen in patients with microscopic polyangiitis and necrotizing glomerulonephritis; however, the sensitivity of this test for these diseases is quite low.

The suspected diagnosis of PAN is confirmed with biopsy of skin lesions (Figure 7) or other possible sites such as peripheral nerves (sural nerve), as seen in Figure 8. If the results are negative or biopsy sites inaccessible, angiography is another tool useful in determining the diagnosis.

Once diagnosed, corticosteroids remain the first line of treatment in patients without renal, gastrointestinal or neurological involvement. When these systems are involved, it is recommended that other immunosuppressive agents such as cyclophosphamide be used in addition to steroids. Treatment should always be in consultation with a pediatric rheumatologist. Prognosis is variable depending on the time of diagnosis and the severity of the disease.

Other choices

Wegener’s granulomatosis (WG) is a systemic vasculitis affecting the small- to medium-sized vessels characterized by necrotizing granulomas involving the upper and lower respiratory tracts with focal glomerulonephritis. Children may present with constitutional symptoms, cough, congestion, persistent rhinorrhea, hemoptysis or dyspnea. Because the current patient lacked respiratory involvement and had a negative C-ANCA and a clear chest x-ray, it is unlikely that she had WG.

Microscopic polyangiitis (MPA) differs from PAN due to extensive glomerular involvement and small vessel inflammation. Patients with MPA present with hematuria, proteinuria and hypertension. Pulmonary symptoms are present in some cases. Nongranulomatous inflammation on biopsy is helpful in distinguishing MPA from WG; however, the treatment is the same for the two diseases. Although our patient did develop hypertension, she did not display the characteristic findings of glomerulonephritis commonly seen in MPA.

Bacterial endocarditis may cause embolic findings and ruptured infectious aneurysms, but this is a late manifestation. Both of the major criteria set forth in the Duke Criteria (two positive blood cultures and evidence of endocarditis on echocardiogram) were negative. While children may have bacterial endocarditis without these findings, the skin manifestations also do not match the late vasculitic findings of endocarditis, which include Janeway lesions, Osler nodes and splinter hemorrhages.

Henoch-Schönlein purpura (HSP) is the most common systemic vasculitis of childhood. This hypersensitivity vasculitis is often preceded by upper respiratory tract infections. Our patient lacked the characteristic rash of HSP, which begins as blanching pink maculopapules that evolve into petechiae or purpura. Palpable purpura is commonly seen on dependent and pressure-bearing regions such as the legs and buttocks (Figures 9 and 10). Patients with HSP may present with abdominal pain caused by edema and damage to the gastrointestinal tract or intussusception. Microscopic hematuria and proteinuria may also occur secondary to renal involvement.

Other diagnoses in patients with febrile exanthems include erythema multiform major (Stevens-Johnson syndrome, Figures 11 and 12) or toxic epidermal necrolysis, especially in those on anticonvulsants, as in this case, or antibiotics, especially sulfas. Juvenile rheumatoid arthritis may also be in the differential. However, the rash (Figures 13 and 14) is typically a pink, evanescent rash that exacerbates with fever. In any case, the diagnosis of these patients is always challenging at best.

The rest of the story

The biopsy of the skin lesions and sural nerve showed a necrotizing vasculitis involving the small and medium arteries with fibrinoid necrosis, consistent with PAN. Our patient was transferred to a pediatric rheumatologist for further evaluation and treatment. While she was there, further arteriograms revealed aneurysms along her superior mesenteric artery with involvement of the left renal vasculature. She was treated with IV dexamethasone and prednisolone and started on cyclophosphamide given in monthly doses over seven months. Since completing this treatment, she has been maintained on etanercept (Enbrel, Immunex), a tumor necrosis factor (TNF) receptor blocker, which has provided continued successful remission of her PAN. Repeat arteriograms showed resolution of her multiple aneurysms. However, one month following her final dose of cyclophosphamide, she developed a pain in the left midaxillary area followed by the appearance of a vesicular rash on the left side of the chest spreading to her back. She was seen by her rheumatologist and diagnosed with herpes zoster, most likely resulting from her immunosuppressive therapy. She was treated with acyclovir and the rash resolved soon after. Treating with acyclovir will not only shorten the duration of the illness (if started early enough) but also may decrease the risk of developing post-herpetic neuralgia.

Columnist comments

For those of you who save these columns, I reviewed PAN in the March 1999 issue, using the case from which the pictures in Figures 5 and 6 came.

I would like to thank Jared Lund for his work on this column. Jared is a great example of the fine, young talent in medical schools today across our country, both allopathic and osteopathic. The volume of material they have to learn makes the challenge us “old-timers” faced look like a walk in the park. The same is true of residents today. This is what I think of every time I hear someone complain about how hard “we” had it and how easy residents today have it, just because there are limits on the number of hours they can work per week. Just take a look at the mere size of the textbooks we used in medical school and residency compared to the same books today. The PDR and Nelson’s are nearly twice the size now as the first ones I had in the early 1970s. They not only have to learn more information, they need to be able to practice without ANY mistakes, or else the litigation monster descends upon them (and us as attendings) like an avalanche. So, the next time you see a weary resident or a sleep-deprived student, be nice to him or her, as you may one day look up from your hospital or emergency room bed to see that same guy or girl looking down upon you as a patient.

We have received a number of letters from you in response to the recent two-part series of cases from Afghanistan by Dr. Giangiulio. This was a good look at what goes on medically day after day in these combat zones. As long as there are soldiers deployed in these areas of the world, pediatricians and other military physicians will continue to quietly, and without fanfare, perform extraordinary feats of caring and compassion. And as I alluded to previously, this is diplomacy in action. These patients and their relatives are usually not aware of, or likely interested in, what is going on in the halls of governments, but are keenly aware of who is helping them through their most difficult time. And this is how you chip away at the foundation of terrorism. Keep those calls and letters coming, and write a soldier when you can. — James H. Brien, DO

For more information:

• James H. Brien, DO, Pediatric Infectious Disease, Scott and White’s Children’s Health Center and Associate Professor of Pediatrics, Texas A&M University, College of Medicine, Temple, Texas. E-mail: jhbrien@aol.com. Jared J. Lund, is a fourth-year medical student, Texas A&M University, College of Medicine, Temple, Texas.