Nasopharyngeal abscess and facial paralysis as complications of petrous apicitis: A case report.

Abstract

An 83-year-old man with a lifelong history of intermittent otorrhea and hearing loss was referred for management of a facial paralysis of 3 weeks' duration. Computed tomography (CT) of the head detected a neoplasm of the nasopharynxalong with chronic otomastoiditis. A followup CT suggested the development of a nasopharyngeal abscess, which was confirmed by needle aspiration. A later coronal-projection CT showed definite bone destruction in the anterior petrous apex, confirming suspicions that a petrous apicitis was responsible for the facial paralysis and abscess. This article describes the management of this patient and reviews the historical, medical, and surgical aspects of petrous apicitis.

Introduction

Petrous apicitis, once a relatively common complication of otomastoiditis, is much less prevalent today, as are other complications of infectious ear disease. Antibiotics have probably played a major role in this reduced incidence, and they have improved our ability to cure this complication when it does occur. It also seems fair to state that antibiotics are a mixed blessing in that the cardinal symptoms of petrous apicitis--deep retrobulbar pain and abducens paralysis with otomastoiditis (Gradenigo's syndrome)--are rarely present these days, which often contributes to a delayed diagnosis.

Reports in the modem literature caution us of the vagaries of these symptoms and recount instances of delayed diagnoses following partial antibiotic and surgical management.[1,2] The occurrence of facial paralysis as a sequela of petrous apicitis has been reported. [2,3] Nasopharyngeal abscesses, also relatively uncommon today, were once caused primarily by tuberculosis in adults and nasopharyngitis in children. [4] A less common although not unusual cause of abscess formation was indeed petrous apicitis. [5] Today, the diagnosis can be more difficult to make solely by assessing the response to partial antibiotic treatment, but high-resolution computed tomography (CT) scanners have made the radiologic diagnosis easier.

Case report

An 83-year-old man was referred by his internist for management of Bell's palsy. The patient reported a lifelong history of intermittent right otorrhea and hearing loss. Three months earlier, the patient had noticed the onset of a purulent otorrhea, which had lasted 6 weeks. This drainage had been treated by the internist with antibiotic eardrops. Three weeks prior to referral, the man had developed a sudden peripheral facial paralysis and postauricular pain on the right. He denied any antecedent viral illness, trauma, or vertigo. At the time of referral, the patient had been on prednisone at 30mg/day for 2 weeks.

Physical examination revealed that the man was alert and had a complete peripheral facial paralysis on the right. His lacrimation was normal, and the maximum stimulation test showed only a slight decrease in the right facial nerve function. Examination of the right ear revealed a central perforation of the inferior tympanic membrane and scant clear otorrhea. There was no cholesteatoma noted, but the ear's appearance was similar to that of chronic adhesive otitis media. The left ear was normal, and the other cranial nerves were intact. A mirror examination of the nasopharynx was normal, as was digital palpation.

An audiologic evaluation revealed a moderate to severe mixed hearing loss on the right and a moderate sensorineural hearing loss on the left. The patient reported that he had been using a hearing aid in the right ear for the previous 2 years. His stapedial reflexes were absent bilaterally. CT of the head with intravenous contrast revealed air fluid levels in the right mastoid and a 4-cm densely enhancing mass in the right nasopharynx, but no protrusion into the nasopharyngeal airway (figure 1). Central hypodense areas consistent with necrosis and obliteration of the fascial planes were evident, but no bony destruction was noted in the nasopharynx or petrous apex. The radiologic diagnosis was carcinoma of the nasopharynx and chronic otomastoiditis.

The patient was admitted to the hospital for an examination and biopsy of the nasopharynx under general anesthesia. His admission laboratory tests showed a white blood count (WBC) of 16,000/[mm.sup.3] and a markedly elevated zeta sedimentation ratio (ZSR) of 69%. The results of a visual and digital examination of the nasopharynx by a head and neck oncologic surgeon were negative. An open biopsy of the nasopharynx was performed; microscopic examination of the biopsied tissue revealed no pathology, and tissue cultures were reported as "no growth."

A complete mastoidectomy with facial nerve exploration was performed. The facial nerve was decompressed from the stylomastoid foramen up to the geniculate ganglion, and it appeared to be normal. The mastoid air cells were intact and lined with moderately thickened mucosa, but no pus was noted. The middle ear contained dense adhesions, and the tympanic membrane was stuck to the medial wall of the middle ear. The eustachian tube area was not explored because of the lack of inflammatory or infectious findings in the mastoid, middle ear, and attic. There was no indication of osteitic bone in the mastoid, attic, middle ear, or zygomatic process. No attempt was made to enter the petrous apex.

Postoperatively, there was no improvement in the patient's facial function. All cultures of the middle ear and mastoid tissue were negative. The patient had been on intravenous moxalactam at 2 g every 8 hours prophylactically, but this was stopped after 3 days because of the lack of growth on culture. Blood cultures were also negative. The patient had a transient low-grade fever, but 2 days postoperatively he was again afebrile. Three days postoperatively, his WBC was down to 9,600/[mm.sup.3], and his ZSR was only moderately elevated at 64%. It was decided to discharge the patient and perform a repeat CT to evaluate the nasopharynx in 1 week, when the surgical edema could be expected to have subsided.

This first postoperative CT showed an increased soft tissue swelling in the right nasopharynx and a large fluid-filled central cavity, which was interpreted as an abscess (figure 2). This suspicion was confirmed when needle aspiration yielded 2.5 ml of yellow pus. On culture, the pus eventually grew Pseudomonas aeruginosa, which is sensitive to a combination of tobramycin and ticarcillin.

The patient was readmitted and started on IV tobramycin at 65 mg every 8 hours and IV ticarcillin at 3 g every 4 hours. The patient began to develop a clouded mentation, but there was no evidence of an intracranial inflammatory process, and his spinal fluid was normal. His readmission WBC had fallen to 8,000/[mm.sup.3], and his ZSR had risen to 76%.

One week later, a second postoperative CT was obtained (figure 3). This scan revealed a reaccumulation of pus in the abscess, but again no bony destruction. The patient was taken back to the operating room and administered general anesthesia. The abscess cavity was surgically drained into the nasopharynx. Biopsies were read as "chronic inflammation without evidence of neoplasm."

Following the second surgery, the patient experienced a slight improvement in his facial function and a gradual improvement in his mental status. A followup CT 1 week later showed an increased soft tissue mass effect that was compatible with surgical edema, but no evidence of a persistent abscess cavity (figure 4). A gallium-67 ([Ga.sup.67]) scan was read as a "marked uptake in the right nasopharynx" (figure 5).

While the patient continued to receive IV tobramycin and ticarcillin, the radiologist rescanned his head, this time using a coronal projection rather than the standard axial projection (figure 6). With the coronal view, the radiologist was able to determine that there was indeed a definite bony destruction in the anterior petrous apex, along with a widening of the petrosphenoid suture line. This was in contrast to the earlier axial scans, which showed no bone destruction.

At this point, the medical team had a complete clinical picture: chronic otomastoiditis had developed into petrous apicitis, which caused facial paralysis. In retrospect, the nasopharyngeal abscess had undoubtedly been the egress point for a "silent" petrous apicitis. Further surgery was not deemed to be prudent (see Discussion), and antibiotics were continued. At this point, the patient's WBC was 7,300/[mm.sup.3] and his ZSR was 79%.

One week later (4 weeks into the second admission), a repeat [Ga.sup.67] scan (figure 7) showed that there was a slight improvement compared with the [Ga.sup.67] scan that had been performed 2 weeks earlier. This was the first encouraging sign that there was some resolution of this persistent infection. Another CT performed at this point showed no change from the one performed 2 weeks earlier. A few days later, the patient's ZSR fell to 68%, reflecting his improving clinical status. The ototoxic potential of the tobramycin was monitored by caloric tests and audiograms every 2 weeks, and mild toxicity was observed. Within 1 week, the patient developed acute congestive heart failure, which was probably related to the sodium content of the ticarcillin. The heart failure was easily managed with fluid restriction and diuretic therapy, and the patient was switched to piperacillin to reduce the sodium load.

At just over 6 weeks into treatment, the patient's ZSR continued to gradually decrease, but his creatinine level began to slowly rise. His tobramycin dosage was adjusted appropriately, but his creatinine level by this point had risen to 2.2 mg/dl, and repeat caloric tests showed a moderate reduction in his vestibular response. When a new [Ga.sup.67] scan showed only a slight uptake (figure 8), it was decided to stop the tobramycin, but the piperacillin was continued for 10 more days. During his final 10 days of hospitalization, the patient developed problems with aspiration, which caused his ZSR to rise again, but clinically he was asymptomatic.

The patient was discharged on no antibiotics after 7.5 weeks of hospitalization. A final followup Ga [67] scan 2 weeks following discharge was normal (figure 9). The patient's facial function gradually improved, and his ZSR had fallen to within the normal range 2 months after discharge.

Discussion

Petrous apicitis appears to be rare today. When it does occur, the initial clinical picture is often clouded by antibiotics. Radiologic examination is the primary means of supporting a clinical suspicion. Nevertheless, as we see in this case, radiologic changes in the petrous apex might not e readily apparent until several weeks into the course of the illness, even with the most sophisticated CT scanner. Because the coronal-projection CT provides a more instructive view, our institution now uses it routinely for temporal bone studies. To give credit where it is due, without CT we would not have been able to diagnose this patient's nasopharyngeal abscess until much later because it was not clinically apparent, even when the patient was under anesthesia.

The initial diagnosis of nasopharyngeal carcinoma was justified because the CT appearance of aggressive inflammatory disease is identical to that of an invasive malignant process--that is, there is an obliteration of the paranasopharyngeal fascial planes and an enhancing lesion. [6] Once the infectious process in our patient progressed to abscess formation, the diagnosis was certain. But before the abscess had formed, the clinical picture could have been that of a malignant nasopharyngeal process that was causing eustachian tube obstruction and resulting in otomastoiditis, rather than the reverse.

Facial paralysis in the clinical setting of a chronically draining ear should automatically lead us away from a diagnosis of Bell's palsy. A cholesteatoma is often the cause of facial paralysis in chronic otitis media. In fact, this is what we had suspected in this case, and it is the reason that we performed a complete facial nerve exploration. Inflammatory changes alone have been responsible for the development of facial paralysis, but in our case there was a paucity of active inflammatory findings in the mastoid and inner ear.

We also had to keep in mind one other etiology for a facial paralysis in a chronically infected ear: primary squamous cell carcinoma. [7,8] Although we had a strong suspicion of petrous apicitis, we had no radiologic evidence. It was not until 3 weeks into our patient's second hospitalization that we obtained the coronal CT view of the skull and for the first time had radiologic proof of an apicitis. At that point, we were faced with a difficult decision. Should we surgically approach the petrous apex to aid in the resolution of this chronic infection? We had three avenues of surgical approach available: superior to the eustachian tube, [9] under the cochlea, [10] and through the middle fossa. [2] Regardless of the approach, ourpatient was 83 years old, and thus we ran the risk of causing a carotid blowout, dural tear, and/or cochleovestibular injury that might have crippled or killed him. Because our patient had been tolerating his antibiotics well without adverse effects, we tried to sterilize the bony inf ection with long-term antibiotics. [11] Within 4 weeks, our decision was justified by his improving clinical condition, improving Ga [67] scan, and improving ZSR.

Once the decision was made to treat with long-term antibiotics, we needed some criteria for stopping therapy--an "exit strategy," if you will. Judging from the literature on malignant external otitis media, [11,2] we believed that we would need at least 6 weeks of antibiotic therapy. When we reached that point, however, our patient's ZSR was no longer a valid indicator because it had risen in response to his aspiration pneumonitis. Therefore, we relied mainly on the Ga [67] scan, which at 6 weeks was almost normal. [13] We had hoped to administer 8 weeks of antibiotic therapy, but we had to stop the tobramycin at just over 6 weeks because the patient began experiencing clinical nephrotoxic and vestibulotoxic reactions.

Obviously, the literature predating 1970 is not completely relevant to today's handling of petrous apicitis because there were no potent antibiotics against Pseudomonas prior to then. Without a doubt, surgical debridement remains a cornerstone in the treatment of osteomyelitis of the petrous bone, but other factors such as age, response to antibiotics, and potential complications of surgery all must be considered in the process we call clinical judgment.

From the Capital Ear Group, Washington, D.C.

Reprint requests: Dennis C. Fitzgerald, MD, Capital Ear Group, 106 Irving St., Suite 4600, Washington, DC 20010. Phone: [phone]; fax [phone].

References

(1.) Chole RA, Donald PJ. Petrous apicitis. Clinical considerations. Ann Otol Rhinol Laryngol 1983;92:544-51.

(2.) Glasscock ME III. Chronic petrositis. Diagnosis and treatment. Ann Otol Rhinol Laryngol 1972;81:677-85.

(3.) Linthicum FH, Graham MD. Facial paralysis due to petrous abscess: Case report. J Laryngol Otol 1972;86:1163-7.

(4.) Paparella MM, Shumrick DA. Otolaryngology. 2nd ed. Philadelphia: W.B. Saunders, 1980.

(5.) Shambaugh GE, Glasscock ME III. Surgery of the Ear. 3rd ed. Philadelphia: W.B. Saunders, 1980.

(6.) Hoover LA, Hanafee WN. Differential diagnosis of nasopharyngeal tumors by computed tomography scanning. Arch Otolaryngol 1983; 109:43-7.

(7.) Conley J, Schuller DE. Malignancies of the ear. Laryngoscope 1976;86:l 147-63.

(8.) Wagenfeld DJ, Keane T, van Nostrand AW, Bryce DP. Primary carcinoma involving the temporal bone: Analysis of twenty-five cases. Laryngoscope 1980;90:912-9.

(9.) Almour R. Surgical therapy for the release of suppurations of the petrous pyramid. Laryngoscope 193l;41:405-6.

(10.) Mayer O. Die Pyramiden Zelleneiterungen. Z Hals Nasen Ohrenheilkd 1937;42:1-86.

(11.) Neal GD, Gates GA. Invasive Pseudomonas osteitis of the temporal bone. Am J Otol 1983;4:332-7.

(12.) Doroghazi RM, Nadol JB Jr., Hyslop NE Jr., et al. Invasive external otitis. Report of 21 cases and review of the literature. Am J Med 1981;71:603-14.

(13.) Noyek AM, Kirsh JC, Greyson ND, et al. The clinical significance of radionuclide bone and gallium scanning in osteomyelitis of the head and neck. Laryngoscope 1984;94(Suppl 34):l-21.