Volume 1, Issue 3 , Pages 207-212, September 2006
Treatment of a clivus abscess in a child using image-guidance
Article Outline
Summary
Osteomyelitis and abscess of the clivus are rare conditions thought to arise from contiguous spread of infection from the paranasal sinuses [D.L. Hoistad, A.J. Duvall III. Sinusitis with contiguous abscess involvement of the clivus and petrous apices. Case report, Ann. Otol. Rhinol. Laryngol. 108 (1999) 463–466] or the temporal bone and petrous apex [S.A. Aziz, P.B. Fayad, F. Fulbright, M.L. Giroux, S.G. Waxman. Clivus and cervical spinal osteomyelitis with epidural abscess presenting with multiple cranial neuropathies, Clin. Neurol. Neurosurg. 97 (1995) 239–244], but have also been found to occur with no known source [R. Rusconi, S. Bergamaschi, A. Cazzabillan, V. Carnelli. Clivus osteomyelitis secondary to Enterococcus faecium infection in a 6-year-old girl, Int. J. Ped. Otorhinolaryng. 69 (2005) 1265–1268]. This report represents the second case of a clivus abscess in a child and demonstrates an innovative approach to surgical localization and treatment of abscesses in this area and also potentially in the nearby retropharyngeal region.
Keywords: Clivus, Abscess, Retropharyngeal space, Osteomyelitis, Image-guidance, Navigation
1. Report of a case
A 3-year-old female with no significant past medical history presented to the emergency department with a 4-day history of fever as high as 100.6
°F. Initially, she was playful and without complaint but progressed to have exquisite neck pain with any neck movement. There was no inciting event such as pharyngeal trauma or sinonasal complaints. Immunizations were up to date. The child lived with both parents and attended daycare. On examination, her temperature was 100.2°F and she was ill-appearing. Her oropharyngeal exam was significant for white exudates on erythematous tonsils. Her neck exam revealed bilateral cervical lymphadenopathy with no dominant mass. Cervical range of motion was quite limited to flexion, extension, and turning to the left. The remainder of her exam was normal.
Laboratory findings included a white blood cell count of 14,900 (6000–17,000) with 4% banded neutrophils and 67% segmented neutrophils. C-reactive protein was 8.3 (0.0–1.0) and erythrocyte sedimentation rate was 45 (0–30). The blood chemistry profile was normal. A blood culture was obtained. A lumbar puncture was performed and laboratory results were normal. A rapid Strep screen by throat swab was negative and culture was subsequently negative. A throat swab for virus and an Epstein Barr Virus (EBV) serology was also sent.
The patient was admitted to the pediatrics service for further evaluation. With no improvement the following day, because of concerns for possible neck abscess, the patient was started empirically on intravenous clindamycin (10mg/kg every 8h). An urgent computed tomography scan (CT) with intravenous contrast was obtained that demonstrated a 1.7cm×1cm×1.2cm low-attenuating mass posterior to the adenoids and extending into the clivus (Fig. 1). There was suspicion for slight epidural extension. At this point, the otolaryngology service was promptly consulted. In addition, clindamycin was discontinued and intravenous meropenem (40mg/kg every 8h) was started to provide broad polymicrobial coverage and optimal central nervous system penetration.
Fig. 1.
(a) Axial and (b) sagittal images of a preoperative computed tomography scan with contrast of the neck demonstrating a 1.7
cm×1cm×1.2cm low-attenuating, rim-enhancing mass posterior to the adenoids and extending into the clivus (arrow). Also demonstrated is epidural extension into the spinal canal (block arrow).
The position of the abscess presented a difficult dilemma because of its deep location and the potential for injury to vital structures, namely the brainstem and spinal cord, with blind dissection. On the day of surgery, a computed tomography scan with contrast with a protocol for image-guidance with the BrainLAB Vector Vision Compact Unit (BrainLAB, Inc., Westchester, Illinois) was obtained. In the operating room, the patient was registered to the BrainLAB system and nasal endoscopy performed. The patient was found to have prominent adenoids that were then removed using Thompson-St. Claire forceps. For localization of the abscess, a reference star was attached to an 18gauge spinal needle on the end of a 10cm3 syringe (Fig. 2). The needle was then carefully passed through the nose to the nasopharynx and using a 0° telescope and the image-guidance system, the level of the abscess was localized and the needle inserted into the abscess and about 1cm3 of pus was aspirated (Fig. 3). With the abscess localized, a sickle knife and suction tip were then used to enter the abscess and the abscess was opened widely with an elevator. Upon opening the abscess, the bony defect in the anterior clivus could be visualized using the endoscope. Several biopsies were taken from the abscess wall.
Fig. 2.
An 18 gauge spinal needle attached to a syringe with affixed BrainLAB reference star.
Fig. 3.
Screen from BrainLAB Vector Vision Compact Unit during localization of the clivus abscess using a reference star attached to syringe with an 18 gauge needle. Note that the end of the needle, as represented by the cross-hairs, is within the abscess cavity.
Postoperatively, the patient's symptoms improved and fevers resolved. A peripherally inserted central venous catheter was placed for prolonged intravenous antibiotic therapy. Culture of the adenoid tissue grew α-hemolytic Streptococci, susceptible to cephalosporins, clindamycin, and meropenem, but resistant to penicillin. Culture swabbed from the abscess contents grew Peptostreptococcus species, susceptible to penicillin, amoxicillin, and meropenem. Blood and cerebral spinal fluid cultures were sterile. EBV serology was consistent with acute infection. Pathologic evaluation of tissue obtained in surgery showed only inflammation and follicular hyperplasia. A magnetic resonance imaging scan (MRI) of the brain with and without gadolinium contrast obtained 4 days after surgery showed continued enhancement of the retropharyngeal soft tissues and clivus with resolution of the soft tissue component of the abscess cavity (Fig. 4). She was discharged 10 days after surgical therapy to complete an additional 3 weeks of intravenous meropenem therapy at home. Prior to discharge, her C-reactive protein level had normalized. A follow-up MRI scan of the brain with and without gadolinium contrast obtained 2 weeks after the first postoperative scan showed continued but decreased enhancement of the clivus and nearby epidural space with complete resolution of any rim-enhancing fluid collection (Fig. 5). The patient received 4 weeks of intravenous meropenem followed by an additional 2 weeks of oral amoxicillin/clavuanate. When last seen at follow-up, the patient continues to do well without any sequelae.
Fig. 4.
(a) Axial spoiled gradient with contrast, (b) sagittal T1 fat-suppressed with contrast, and (c) coronal T1 with contrast images of a magnetic resonance image of the brain with gadolinium contrast performed on postoperative day 4 demonstrating continued enhancement of the retropharyngeal soft tissues and clivus (arrow) with resolution of the soft tissue component of the abscess cavity.
Fig. 5.
(a) Axial spoiled gradient with contrast, (b) sagittal T1 fat-suppressed with contrast, and (c) coronal T1 spin echo with contrast images of a magnetic resonance image of the brain with gadolinium contrast performed on postoperative day 19 demonstrating interval resolution of the rim-enhancing fluid collection (arrow) with persisting nasopharyngeal soft tissue and clivus enhancement. The previously noted dural enhancement and thickening along the clivus is also reduced (block arrow).
2. Comment
This report represents the second case of clivus abscess reported in a child. Whether this abscess began as a retropharyngeal abscess and extended into the clivus or started as a primary skull base osteomyelitis with extension into the surrounding soft tissues is not clear. The latter scenario is suspected due to the relatively short period of symptoms involving the soft tissue of the neck. The previous report in a child likewise had an unclear etiology [3]. In a series of peritonsillar, retropharyngeal, and parapharyngeal abscesses published by Brook, 78 anaerobes (5.6 per specimen) and 26 aerobes (1.9 per specimen) were isolated from 14 children with retropharyngeal abscess [4]. Peptostreptococcus species were the most common single isolate (18 isolates). With Peptostreptococcus species being isolated from our patient, it is possible that her abscess had an oropharyngeal source. More common bacteria involved in skull base osteomyelitis include Pseudomonas aeruginosa when the source is the temporal bone [1], [2] or Staphylococcus aureus when the sphenoid sinus [1] or a spinal epidural abscess [1], [2] are the source. Coagulase-negative Staphylococcus has also been reported [1].
Appropriate selection of empirical antibiotic therapy for clivus osteomyelitis and abscess should take into account the source of primary infections (e.g., sinusitis or retropharyngeal abscess) and possible complications, (e.g., epidural abscess). Meropemen was chosen because of its broad spectrum of activity against a variety of pathogens including gram-positive aerobic cocci (such as Staphylococcus aureus and Streptococci spp.) and anaerobes.
Localizing this abscess for drainage in the operating room was a challenge. Located deep to the adenoid bed, with potentially thinned and softened skull base bone behind the abscess cavity, and with the clivus-C1 space nearby, it was imperative that the abscess be localized and opened carefully. To aid in this effort, the image-guidance system was utilized. Frequently used at our institution for both endoscopic sinus and neurosurgical applications, the system allows the surgeon to correlate the location of the end of various instruments to their corresponding location on a multi-planar computed tomography scan. For our purposes, the advantage of the image-guidance system in use at our institution is the ability to attach a reference star to any instrument and register that instrument to the system. In this case, a reference star was attached to a syringe with a spinal needle and was used to locate and aspirate the abscess before then incising and draining the remainder of the purulent material. When attaching a reference star to a semi-rigid instrument such as a spinal needle, care must be taken to avoid bending the needle during insertion as this can result in inaccurate guidance. Whenever possible, a rigid instrument would be preferable. Early clinical recognition and surgical drainage in conjunction with long-term antibiotic therapy has contributed to this patient's recovery from a potentially debilitating or even lethal condition.
3. Conclusions
Although osteomyelitis and abscess of the clivus are very rare in children, clinicians should become familiar with these potentially serious conditions. Contiguous spread from adjacent infections such as from the paranasal sinuses or temporal bone is thought to occur in some instances. Other primary sources of infection may include retropharyngeal abscess or spinal epidural abscess and rarely via hematogenous spread. Due to its location deep to the nasopharynx and its close proximity to the brainstem and cervical spinal cord, the surgical approach to an abscess of the clivus can be difficult and risky. This case demonstrates the utility of an image-guidance system in localizing and treating an abscess of the clivus safely and effectively.
References
- . Sinusitis with contiguous abscess involvement of the clivus and petrous apices. Case report. Ann. Otol. Rhinol. Laryngol. 1999;108:463–466
- . Clivus and cervical spinal osteomyelitis with epidural abscess presenting with multiple cranial neuropathies. Clin. Neurol. Neurosurg. 1995;97:239–244
- . Clivus osteomyelitis secondary to Enterococcus faecium infection in a 6-year-old girl. Int. J. Ped. Otorhinolaryng. 2005;69:1265–1268
- . Microbiology and management of peritonsillar, retropharyngeal, and parapharyngeal abscesses. J. Oral. Maxillofac. Surg. 2004;62:1545–1550
PII: S1871-4048(06)00060-8
doi:10.1016/j.pedex.2006.03.007
© 2006 Elsevier Ireland Ltd. All rights reserved.
Volume 1, Issue 3 , Pages 207-212, September 2006
