From Congenital Dermal Sinus Tract to Extensive Epidural Abscess of the Spine

This is the first report of a large epidural abscess ranging from the cervical to the lumbosacral spine in a child with a very short history of clinical symptoms and a previous history of imperforate anus. A 2-year-old boy presented with a history of fever and fatigue for 5 days. The last two days, he was unwilling to walk or lie on his back. His parents reported on diarrhea with diaper dermatitis followed by the onset of fever. In his past history, the boy underwent a posterior sagittal anorectoplasty for an imperforate anus with a rectourethral fistula two years prior to the current presentation, and an inconspicuous sacral dermal sinus tract (DST) was diagnosed, but he had not yet undergone surgery.

On initial physical examination, the child was hemodynamically stable, without meningeal irritation. The boy had a painful swelling of the sacral bone. Laboratory tests revealed an elevated C-reactive protein (CrP) of 176 mg/l, leukocytosis of 21 000/μL, and an erythrocyte sedimentation rate (ESR) of 120 mm after 60 min. Blood chemistry and an urinanalysis were within normal limits.

An empiric antibiotic treatment with intravenous cefuroxime was initiated for fever of unknown cause. The following day, nuchal rigidity occurred and immediately a lumbar puncture was performed that revealed purulent liquid. Further evaluation included magnetic resonance imaging that demonstrated an extensive epidural abscess ([Fig. 1a–c]).

Fig. 1 Preoperative images: Sagittal T1-weighted image after i. v administration of gadolinium (a) and sagittal T2-weighted images of the total spine demonstrating the septal enhancement of an epidural abscess with compression of the spinal cord extending from the foramen magnum to the sacrum (b, c). At the sacral end there is an abscess formation (asterix) showing a clear connection to the intraspinal epidural abscess (a, b). Postoperative images: T1-weighted sagittal images after i. v administration of gadolinium 10 weeks after surgical decompression and antibiotic treatment demonstrating the normal spinal cord without compression or pathological enhancement (d, e).

At that time, the patient was referred to our hospital and underwent emergency surgery with resection of the DST and lavage of the epidural abscess formation. The antibiotic therapy was changed to meropenem and vancomycin and was continued, when bacterial cultures revealed a co-infection with P. mirabilis, E. faecalis and B. fragilis. After 4 weeks, intravenous antibiotic treatment was changed to an oral application of amoxicillin plus clavulanic acid, and discontinued after 2 further weeks when ESR and CrP levels normalized. Further imaging studies 10 weeks after the initial diagnosis showed no evidence of a residual abscess formation ([Fig. 1d–e]).

A congenital dermal sinus tract (DST) of the spine is a rare form of spinal dysraphism with an incidence of about 1 in 2 500 births. The tract is epithelium-lined and usually corresponds to the intradural space (58–60%), followed by the extradural space in 10–20%. In 6–7%, it ends dorsal to the spine. DST may present with either neurologic symptoms or with infectious complications (Ackermann LL and Menzes AH, Pediatrics 2003; 112: 641–647). Since most DST are connected to the intradural space, subarachnoidal and intradural as well as intramedullary abscesses have been reported frequently (Morandi X et al., Childs Nerv Syst 1999; 15: 202–206). Infectious complications include meningitis as well as subarachnoidal and intramedullary abscess formation in the majority of cases, since the DST usually correspond with the intradural space. Staphylococcus aureus, Streptococcus species, Actinomyces species, Proteus mirabilis and Eschericha coli are the most common infectious pathogens, and empiric antibiotic therapy should be selected with respect to these pathogens. Most authors have recommended prolonged intravenous application between 4 and 9 weeks. Thus, we recommend early surgery, especially in patients with large abscess formation.

To avoid infectious complications, an early resection of the DST is associated with an improved outcome and is strongly recommended in literature (Ramnarayan R et al., Childs Nerv Syst 2006; 22: 1220–1224). In the presence of an infection, an exclusive conservative approach is also not justified, but the timing of surgery is still under debate. Some authors recommend delayed surgery following an intravenous antibiotic therapy until the infection is under control. In the present case, immediate surgery led to a stable recovery, no dissemination of the infection was encountered and intravenous antibiotics could be discontinued after 4 weeks.

Frequently, a DST is associated with a dermoid and/or a tethered-cord. Our patient had a history of an imperforate anus. This malformation is associated with occult spinal dysraphisms, but to our knowledge an association with a DST is so far undescribed (Tsakayannis DE et al., J Pediatr Surg 1995; 30: 1010–1012). If DST is suspected, spinal sonography can easily be used to detect a DST with a high sensitivity. Nevertheless, a contrast-enhanced MRI scan remains the investigation of choice in preoperative work-up (Barkovich AJ et al., AJNR Am J Neuroradiol 1991; 12 (1): 123–129). However, in the case of infection, interpretation may be difficult.

Main statements:

We report a patient with a previous history of an imperforate anus and a DST complicated by a large epidural abscess formation. Early diagnosis and resection of a DST before complications occur is warranted. Physicians should especially be aware of associated spinal dysraphisms in patients with anorectal malformations and should suspect serious infectious complications when fever occurs. A register for spinal dysraphisms should be implemented to standardize diagnostic and therapeutic procedures and to monitor outcome.