Recurrent perianal abscesses and fistulas heighten suspicion for Crohn disease (CD). Perianal disease often occurs in association with gastrointestinal symptoms; however, perianal disease alone can be the initial manifestation of CD. In patients who present only with recurrent perianal abscesses and fistulas, in the absence of a conclusive diagnosis of CD, further investigation is indicated to determine the etiology of the perianal disease. The following cases describe 3 patients with recurrent perianal abscesses, fistulas, or ulcerations who were subsequently diagnosed as having autoimmune neutropenia (AIN), and, once treated for AIN, had complete resolution of their perianal disease.
A 5-month-old African American boy presented to the surgery clinic with a nonhealing perianal abscess. The abscess had persisted despite treatment with amoxicillin for 5 days and amoxicillin-clavulanic acid for 2 weeks. He was thought to have a fistula-in-ano and underwent an anal fistulectomy as well as drainage of the site. Three weeks after the fistulectomy, he presented with a recurrent fistula, a new fistula, as well as 2 new abscesses, with these new findings remote from his initial fistula site. Upper endoscopy and colonoscopy, and biopsies obtained during these procedures were normal. Initial evaluation for immune deficiency demonstrated persistent neutropenia, with absolute neutrophil counts (ANCs) as low as 300 cells per microliter. Bone marrow biopsy was normal. Neutrophil antibody identification detected neutrophil-reactive IgG antibodies, confirming the diagnosis of AIN. At 8 months of age, the patient was started on granulocyte colony-stimulating factor (GCSF) 5 μg/kg 3 times per week. He had healing of his perianal disease and was continued on GCSF for 1 year. The patient is now 3 years 2 months old and has been off GCSF for 18 months. His ANCs have ranged from 400 to 3300 cells per microliter and he has had no recurrence of his perianal disease, nor any other significant infections.
A 4-month-old white boy presented to the surgery clinic with a perianal abscess that had been present for 3 months despite 2 courses of amoxicillin-clavulanic acid. His examination was significant for 2 perianal abscesses and 1 fistula. Because of these findings, an immunologic workup was initiated. Although his ANC and neutrophil oxidative burst were normal, he was positive for IgM anti-neutrophil antibodies. He underwent surgical drainage of his abscesses and an anal fistulectomy. He was started on GCSF 5 μg/kg twice per week and had complete healing of his perianal disease. His GCSF was discontinued after 2 months, with all ANCs continuing to remain greater than 1000 cells per microliter and no recurrence of his perianal disease. Neutrophil antibody screen 6 months after discontinuing GCSF was negative.
A 22-month-old African American girl with a history of pulmonary atresia status postheart transplant at 4 weeks of age, was admitted to the hospital secondary to a 3-month history of perianal ulcerations and fever. Her perianal ulcerations were consistent with the diagnosis of ecthyma gangrenosum, and cultures of both the perianal ulcerations and peripheral blood were positive for Pseudomonas aeruginosa, for which she was treated with intravenous ciprofloxacin and piperacillin-tazobactam. Before this admission, ANCs had been as low as 200 cells per microliter. Neutrophil antibody testing was significant for IgG anti-neutrophil antibodies; in addition, she also had anti-platelet IgM antibodies. The patient was given 1 dose of intravenous immunuoglobulin (IVIG) and started on GCSF 3 μg/kg every other day and had healing of her perianal disease. Because of her refractory autoimmune cytopenias, in addition to the GCSF, she received 4 doses of rituximab. Subsequent ANCs ranged from 2700 to 5300 cells per microliter and her platelet count normalized. She is now 2 years 9 months old and remains on GCSF at this time because of continued recurrent infections and persistent positive neutrophil antibody screens. She also continues on mycophenolate mofetil, tacrolimus, and prednisolone for immunosuppression for her heart transplant.
Recurrent perianal abscesses and fistulas raise suspicion for CD. Up to 62% of pediatric patients with CD will have perianal disease at some point during their disease course (1,2). Most often, the perianal disease will be present in addition to gastrointestinal symptoms at the time of diagnosis of CD, or will develop after the diagnosis of CD (3). However, in 20% to 36% of adults, perianal disease alone can be the initial presentation of CD (3,4). In patients who present only with recurrent perianal abscesses and fistulas, in the absence of endoscopic and histologic evidence of CD, further investigation is indicated to determine the etiology of the perianal disease. As described in this case series, children with recurrent perianal abscesses, fistulas, or ulcerations should be evaluated for AIN.
Primary AIN is an isolated autoantibody phenomenon of unknown etiology. The incidence in children is at least 1 in 100,000 (5). In two thirds of cases, AIN is diagnosed between 5 and 15 months of age, with 8 months being the average age of diagnosis. There is a slight female predominance, and the majority of patients are white. Eighty percent of the time, these patients have minor infections, most commonly upper respiratory tract infections, ear infections, skin infections, or fevers of unknown origin, although serious infections, such as pneumonia, meningitis, and sepsis, occur up to 12% of the time (6). An association with Pseudomonas perianal infections has been reported in females (7).
The ANC is often less than 500 cells per microliter, with this number varying considerably from day to day (6,7). It should be appreciated that in some patients with AIN, the ANC is normal, yet these patients are still at risk for serious infections. This is because neutrophil membrane glycoproteins are involved in both neutrophil adhesion and phagocytosis, and antibodies to these neutrophils can alter their number and function, which can subsequently result in defects of adhesion, aggregation, chemotaxis, phagocytosis, and metabolic activation (8).
Anti-neutrophil antibodies are present in 98% of patients with AIN (9). However, these tests have a low sensitivity, because one quarter of patients require repeated testing to detect the antibodies. Bux et al (6) and others (10) demonstrated that the most commonly represented neutrophil antigen-specific autoantibodies were to FCgRIIIb (CD16), CD11b (CD18), CD11a (CD18). In 85% of cases, antibodies are of the IgG class only, another 12% will have both IgG and IgM antibodies, and 3% will have IgM antibodies only. Bone marrow biopsy is not a prerequisite for the diagnosis of AIN; in cases in which the bone marrow is evaluated, biopsy findings typically show a normocellular or hypercellular marrow with a normal or decreased number of metamyelocytes, bands, and mature neutrophils (6,7).
Most patients with AIN do not have significant infections and do not require specific treatment of the AIN. Symptomatic treatment with antibiotics is sufficient in more than 90% of infectious events (6). For those patients with severe infections, or for patients requiring a surgical procedure, GCSF has been used successfully. Other options include high-dose IVIG and corticosteroids, although because of the potential side effects and inconsistent results of corticosteroids and the short-lasting effect of IVIG, GCSF is the preferred treatment (11,12).
Almost all children with primary AIN have eventual disappearance of the autoantibodies and will outgrow the disease. Bux et al (6) followed 240 children with AIN and found that 80% of the children recovered within 7 to 24 months, with a range of 1 to 38 months. Another study by Lalezari et al (9) reported the median duration of disease was 20 months, and 95% of children were in complete remission by 4 years of age.
It is important to discern the differences in the clinical presentation of children with recurrent perianal disease and AIN from those with routine perianal abscesses or fistulae-in-ano to know when an immunologic workup is warranted. Routine perianal abscesses are commonly caused by staphlococci and Gram-negative enteric organisms. They are treated with incision and drainage followed by sitz baths, with antibiotics used if there is a surrounding cellulitis. About 30% to 50% of perianal abscesses progress to fistula-in-ano. The main therapeutic option for a fistula is a fistulectomy, which allows secondary healing to occur quickly. Complete healing of the site is complete by 7 to 10 days and fistula recurrences are rare (between 1% and 2%).
Evaluation for AIN should be considered when perianal abscesses are unusual in some respect. Atypical scenarios include an unusual clinical course (as in the first patient who had fistula recurrence and presence of new abscesses after documented healing), multiple abscesses (as in the second patient), and unusual organisms (as in the third patient who had Pseudomonas, an organism not typically found in these abscesses).
All 3 children in this case series were treated with GCSF and had improvement in their perianal disease. Two were successfully weaned off GCSF within 2 to 12 months, with resolution of the AIN by the ages of 6 to 20 months, with no recurrence of perianal disease. The other child continues on GCSF for treatment of the AIN. (See Tables 1 and 2 and Figure 1A–C for detailed immunologic evaluation and a summary of clinical characteristics and clinical course.) Immunology tests were performed at the BloodCenter of Wisconsin.
Recurrent perianal abscesses, fistulas, or ulcerations are not commonly seen in AIN, with only a few other cases documented in the literature. Lejkowski et al (13) reported 2 cases of persistent perianal abscesses in early infancy as a presentation of AIN. It is unclear why this presentation is unusual, because the association of perianal disease with neutropenia is well known in oncology patients. It is possible that these patients with AIN and recurrent perianal disease have underlying genetic and/or immune risk factors.
It is important to recognize the association of perianal disease and AIN. Although CD is high on the differential diagnosis in children with recurrent perianal abscesses and fistulas, in the absence of endoscopic and histologic evidence of CD, evaluation for AIN should be initiated. Investigation for AIN should also be considered when perianal abscesses are atypical in some respect; this would include children with multiple abscesses, unusual organisms, or an unusual clinical course. Neutropenia is the hallmark laboratory finding in AIN; however, some patients with AIN can have a normal ANC and still be at risk for serious infections. Fortunately, the majority of patients with AIN will have minor infections and will have spontaneous disappearance of anti-neutrophil antibodies within the first few months to years after diagnosis. In the subset of patients with recurrent perianal disease and AIN, treatment with GCSF can result in resolution of perianal disease and improved quality of life.
1. Markowitz J, Daum F, Aiges H, et al
. Perianal disease in children and adolescents with Crohn's disease. Gastroenterology 1984; 86:829–833.
2. Palder SB, Shandling B, Bilik R, et al
. Perianal complications of pediatric Crohn's disease. J Pediatr Surg 1991; 26:513–515.
3. Sangwan YP, Schoetz DJ, Murray JJ, et al
. Perianal Crohn's disease: results of local surgical treatment. Dis Colon Rectum 1996; 39:529–535.
4. Williams DR, Coller JA, Corman ML, et al
. Anal complications in Crohn's disease. Dis Colon Rectum 1981; 24:22–24.
5. Lyall EGH, Lucas GF, Eden OB, et al
. Autoimmune neutropenia of infancy. J Clin Pathol 1992; 45:431–434.
6. Bux J, Behrens G, Jaeger G. Diagnosis and clinical course of autoimmune neutropenia in infancy: analysis of 240 cases. Blood 1998; 91:181–186.
7. Jonsson OG, Buchanan GR. Chronic neutropenia during childhood: a 13-year experience in a single institution. AJDC 1991; 145:232–235.
8. Shastri KA, Logue GL. Autoimmune neutropenia. Blood 1993; 81:1984–1995.
9. Lalezari P, Khorshidi M, Petrosova M. Autoimmune neutropenia of infancy. J Pediatr 1986; 109:764–769.
10. Bruin MCA, von dem Borne AEG, Tamminga RYJ, et al
. Neutrophil antibody specificity in different types of childhood autoimmune neutropenia. Blood 1999; 94:1797–1802.
11. Capsoni F, Sarzi-Puttini P, Zanella A. Primary and secondary autoimmune neutropenia. Arthritis Res Ther 2005; 7:208–214.
12. Smith MA, Smith JG. The use of granulocyte colony-stimulating factor for treatment of autoimmune neutropenia. Curr Opin Hematol 2001; 8:165–169.
13. Lejkowski M, Maheshwari A, Calhoun DA, et al
. Persistent perianal abscess in early infancy as a presentation of autoimmune neutropenia. J Perinatol 2003; 23:428–430.