Skull Base, Petrous Apex, Infection 

Updated: Jun 22, 2020
Author: Andrea H Yeung, MD; Chief Editor: Arlen D Meyers, MD, MBA 

Overview

Practice Essentials

The petrous portion of the temporal bone lies in a complicated anatomic position and has critical relationships to important neural and vascular structures. Consequently, infections arising within or spreading to the petrous apex can result in severe clinical sequelae. When such infections have been radiographically documented, culturing middle ear effusions is important to help discern the microbiologic etiology and institute appropriate antibiotic therapy.

Before the advent of antibiotics in the 1940s, infections of the petrous apex, termed petrous apicitis, commonly progressed to meningitis, brain abscess, cavernous sinus thrombosis, and death. Since the introduction of antibiotics, the prevalence of such serious complications has been drastically reduced. By the late 1950s, experienced clinicians already noted this decreasing prevalence; DeWeese cautioned physicians to remain vigilant of petrous apicitis, "Lest we forget that this condition still occurs."[1]

DeWeese's admonition is still valid. Patients still occasionally present with petrous apicitis, and the clinician needs to be wary of the condition's presenting features to prevent possible life-threatening complications.

An image depicting skull base anatomy can be seen below.

Internal anatomy of the skull base, lateral view, Internal anatomy of the skull base, lateral view, and base of the skull.

Workup in petrous apicitis

In cases of radiographically documented petrous apicitis, culturing middle ear effusions is important to help discern the microbiologic etiology and institute appropriate antibiotic therapy.

Additional important lab studies include a complete blood count (CBC), electrolyte tests (eg, sodium, potassium, chloride), and a blood glucose determination.

Computed tomography (CT) scanning has replaced plain-film tomography as the standard diagnostic study for assessing skull base processes. Among the temporal bone features diagnostic of petrous apicitis on CT scan are opacification of the mastoid air cell system, including the petrous apex; enhancement of the cavernous sinus; and bony erosion within the petrous apex. With contrast, cavernous sinus enhancement may also occur.

Magnetic resonance imaging (MRI) provides additional important information unavailable with CT scanning. Single-photon emission computed tomography (SPECT) scanning may be useful when CT scanning or MRI is nondiagnostic.

Management of petrous apicitis

In the absence of a severe life-threatening complication or chronic ear disease, the first line of therapy for petrous apicitis includes intravenous antibiotics for the treatment of the underlying infection.

Surgical intervention is warranted for patients whose symptoms do not respond to appropriate antibiotics (eg, continued fevers, otorrhea, headache) or who develop complications from the infection, including cranial nerve (CN) deficits, abscess formation, or venous sinus thrombosis.

In most patients, the petrous apex can be drained via a transmastoid approach to the temporal bone. This approach involves a complete mastoidectomy.

In patients with infections within the posterior aspect of the petrous apex, identifying fistulous tracts along the sinodural angle, the subarcuate air cell tract, and air cells inferior to the posterior semicircular canal is often possible. Enlarging these structures provides adequate drainage of the petrous apex.

Infections in the anterior portion of the petrous apex may be open to a hypotympanic-subcochlear approach or an approach via the air cell tract below the posterior semicircular canal and superior to the jugular bulb.

A complete petrous apicectomy is reserved for patients who require additional exposure to the anterior petrous apex.

History of the Procedure

Gradenigo syndrome, ie, petrous apicitis in combination with the clinical triad of headache, abducens nerve (cranial nerve [CN] VI) palsy, and otorrhea, is named after Giuseppe Gradenigo. In 1904, Gradenigo related his experience with this disorder in a manuscript entitled "Uber circumscripte Leptomeningitis mit spinalensymptomen und über Paralyse des N. Abducens otitischen Ursprungs," in the Archiv für Ohrenheilkunde.[2] In the paper, Gradenigo summed up the disease process through its 3 principal symptoms, which "... dearly and mutually cohere: acute inflammation of the middle ear, continuing pain in the temporal and parietal area, and paralysis the N Abducens, which is indicated usually by the patient himself."

Throughout the early 1900s, in the absence of antibiotics, various surgical procedures were developed to treat petrous apicitis. Procedures usually included a mastoidectomy with or without a labyrinthectomy. In 1930, Almour and Kopetsky describe following fistulous tracts into the petrous apex via a mastoidectomy with or without a labyrinthectomy.[3] Ramadier introduced the radical petrous apicectomy in 1933.[4]

In 1937, Lempert described the mastoidotympanoapicectomy, which is now the classic operation for exenteration of the anterior petrous apex.[5, 6] This radical operation required glenoid fossa exposure and dissection along the carotid canal within the skull base. Because of the morbidity associated with the operation, many other surgeons rejected it during this period. In 1973, Hendershot and Wood advocated the extradural middle fossa approach.[7]

In the 1930s, Profant and Lindsay described the drainage pathways of the petrous apex through several air cell tracts along with the routes whereby infection may come to involve the petrous apex.[8, 9]

Between the 1940s and the 1960s, the incidence of petrous apicitis was dramatically reduced, which is evidenced by the paucity of clinical manuscripts on the topic. This reduction was primarily due to the introduction of antibiotics. Most reports since this time are based on single clinical cases or very small clinical series.

Problem

Petrous apicitis is a rare infection of the petrous apex of the temporal bone that occurs as an extension of a middle ear or mastoid infection.

Epidemiology

Frequency

Petrous apicitis was commonly encountered before the introduction of antibiotics. It now appears quite rarely. Most reports on petrous apicitis in the literature are single clinical cases, and determining its frequency is difficult.

In a retrospective review of petrous apicitis cases occurring over a 40-year period, Gadre and Chole found that six out of 44 patients with the condition (13.6%) had Gradenigo syndrome.[10]

Etiology

Petrous apicitis is believed to result when organisms, typically pseudomonads, become trapped within the complex air cell system of the petrous apex. Blockage of this air cell system may result from acute or chronic inflammation or mechanical blockage from an obstructing lesion.

Direct extension of infection from the mastoid and middle ear through pneumatized air cell tracts into the petrous apex is thought to be the etiology of petrous apicitis, which can occur as a rare complication of acute or chronic otitis media.[11, 12] An undetected and poorly drained infected air cell of the petrous apex must trail through small air cell tracts into the middle ear and mastoid. These cell tracts consist of the infralabyrinthine air cell tract, the retrofacial tract, and the peritubal air cells superior to the eustachian tube.

Pathophysiology

Because of the extensive pneumatization and presence of rich bone marrow within the petrous apex, it is susceptible to infection or inflammation, typically in combination with mastoiditis. The proximity of the venous sinuses to the petrous apex is the reason for the historically high incidence of venous sinus thrombosis associated with petrous apicitis. The inflammation may extend into the Dorello canal, which transmits CN VI and the gasserian ganglion (CN V). Inflammation of the canal produces the triad of symptoms recognized by Gradenigo: lateral rectus (CN VI) palsy, retroorbital pain, and otorrhea.

Pseudomonads are most often responsible for petrous apicitis. In rare cases, tuberculosis has been identified as the cause, primarily in individuals younger than 20 years. Rare cases of fungal petrous apicitis have also been described including aspergillosis and blastomycosis infections.

Presentation

Petrous apicitis may follow an acute or a chronic course.[13]  The acute form typically develops rapidly and is caused by sudden obstruction of a normally pneumatized petrous apex air cell system. This obstruction can be caused by mechanical blockage from a lesion within the mastoid or by acute mastoid inflammation.

Chronic apicitis has a more indolent course and typically follows months to years of otorrhea. Patients may not appear acutely ill compared with those patients with acute petrositis. Chronic apicitis may result from chronic mastoid inflammation, or it may occur after a mastoid operation has led to blockage of the air cell system.

The triad of retroorbital pain, lateral rectus (CN VI palsy), and otorrhea is pathognomic for petrous apicitis. However, the presence of this triad is uncommon, since antibiotics typically halt the disease process before it involves dural structures. The presence of both otorrhea and deep pain should lead the examining physician to suspect petrous apicitis.

The anatomic relationship at the petrous tip may explain some of the symptoms of petrous apicitis. If the bony cortex of the anterior petrous apex is involved by the extension of infection, the infection may cause an epidural abscess in the region or damage nearby cranial nerves. On the superior aspect of the petrous tip lies the trigeminal or gasserian ganglion. Damage or irritation to the ganglion may explain the deep facial pain in some patients with apicitis. The petroclinoid ligament extends from the tip of the petrous apex to the clinoid is the petroclinoid ligament. The abducens nerve travels below the petroclinoid ligament in a small canal called the Dorello canal. The inflammation may extend into the Dorello canal, which transmits CN VI and the gasserian ganglion (CN V). Inflammation of the canal produces the triad of symptoms recognized by Gradenigo: lateral rectus (CN VI) palsy, retroorbital pain, and otorrhea.

Symptoms of petrositis are usually subtle. Typically, a patient who has had previous mastoid surgery complains of persistent infection and deep facial pain. Chole and Donald found that the most common presenting symptoms in petrous apicitis, as found in 22 patients from 1976-1995, were as follows:

  • Deep pain and headache – 13 patients (59%)

  • Otalgia – 16 patients (72%)

  • Otorrhea – 13 patients (59%)

  • Fever – 5 patients (22%)

  • Coma – 2 patients (9%)

  • Cranial nerve paralysis

    • Nerve V – 15 patients (68.2%)

    • Nerve VI – 4 patients (18.2%)

    • Nerve VII – 6 patients (27.3%)

    • Nerve VIII – 9 patients (40.9%)

    • Nerve IX – 1 patient (4.5%)

    • Nerve X – 1 patient (4.5%)

Another study, the aforementioned report by Gadre and Chole, found that out of 44 patients with petrous apicitis, 24 (54.5%) had severe retroorbital pain, 27 (61.4%) had otitis media (including 16 [36.4%] with purulent otorrhea) at the time of presentation, 37 (84.1%) had facial pain along with otitis media, eight (18.2%) had fever, and seven (15.9%) had abducens nerve palsy.[10]

Indications

Indications for surgical drainage of the petrous apex in patients with petrous apicitis include failure to respond to medical (antimicrobial) therapy, the development of CN deficits as a result of extension of the infection, development of a petrous apex, epidural or parenchymal brain abscess, and the development of another life-threatening complication as a result of the infection.

Relevant Anatomy

Wedged between the occipital bone and the greater wing of the sphenoid, the petrous portion of the temporal bone is shaped like a 3-sided pyramid. The pyramid's base is the medial wall of the middle ear. Two of the sides of the pyramid constitute the anterior floor of the middle cranial fossa and the anterolateral wall of the posterior fossa. The jugular bulb and inferior petrosal sinus line the inferior aspect of the petrous bone.

Anterior and superior to the petrous apex lie CN V and the gasserian ganglion. CN VI traversing through the Dorello canal also lies in this region. At its apex lie the carotid artery and cavernous sinus.

Air cells within the petrous portion of the temporal bone communicate with the eustachian tube and nasopharynx via the middle ear, providing a route of pressure equalization and drainage. In 1931, Profant described 2 separate air cell tracts within the temporal bone, an epitympanic tract leading from the antrum and a hypotympanic tract traveling beneath the cochlea.[8]  A study by Lee et al suggested that pneumatization of the petrous apex may not be influenced by major temporal bone structures but may instead be primarily impacted by the anterior saccule of the saccus medius.[14]

The petrous apex is the most surgically inaccessible portion of the temporal bone. A coronal plane through the internal auditory canal may arbitrarily bisect the apex. This plane divides the apex into anterior and posterior portions. The anterior apex, which is pneumatized in 9% of patients, is a peritubal area anterior and medial to the cochlea. The carotid artery traverses the anterior petrous apex. The posterior petrous apex, which is pneumatized in 30% of patients, is a perilabyrinthine area just medial to the semicircular canals.

Contraindications

If the patient is medically unstable and unable to tolerate general anesthesia, high-dose IV antibiotics may be attempted to eradicate the infection in place of surgery.

 

Workup

Laboratory Studies

In cases of radiographically documented petrous apicitis, culturing middle ear effusions is important to help discern the microbiologic etiology and institute appropriate antibiotic therapy. This may be accomplished via paracentesis through an intact drum or culture of any purulent otorrhea.

Additional important lab studies include a complete blood count (CBC), electrolyte tests (eg, sodium, potassium, chloride), and a blood glucose determination.

Imaging Studies

See the list below:

  • Computed tomography (CT) scanning has replaced plain-film tomography as the standard diagnostic study for assessing skull base processes.

    • Temporal bone features diagnostic of petrous apicitis on CT scan include opacification of the mastoid air cell system, including the petrous apex; enhancement of the cavernous sinus; and bony erosion within the petrous apex. With contrast, cavernous sinus enhancement may also occur.

    • Acute petrositis may appear as an expanding lesion with irregular margins. In contrast, chronic petrous apicitis may demonstrate hypopneumatization and sclerosis.

  • Magnetic resonance imaging (MRI) provides additional important information unavailable with CT scanning.

    • In acute apicitis, a high-resolution MRI with gadolinium through the temporal bone demonstrates a low-intensity (ie, hypointense) signal on T1-weighted images, shows a high-intensity (ie, hyperintense) signal on T2-weighted images, and has ring enhancement with gadolinium dye.

    • In chronic apicitis, findings resemble chronic mastoiditis, showing a hyperintense signal on T2-weighted images. Some enhancement may occur with gadolinium, although not to the same degree as in acute apicitis.

    • These findings are key to diagnosis and help distinguish petrous apicitis from other lesions of the petrous apex, such as cholesterol granuloma (hyperintense on T1- and T2-weighted images, no gadolinium enhancement), cholesteatoma (hypointense on T1-weighted images, hyperintense on T2-weighted images, no gadolinium enhancement), and neoplasia (hypointense on T1-weighted images, hyperintense on T2-weighted images, and gadolinium enhancement).

  • Single-photon emission computed tomography (SPECT) scanning and nuclear imaging studies

    • SPECT scanning may be useful when CT scanning or MRI is nondiagnostic. SPECT scanning also can assist in identifying subtle petrous apex inflammation.

    • SPECT-scan findings in petrous apicitis include focal uptake of signal within the affected petrous apex.

 

Treatment

Medical Therapy

In the absence of a severe life-threatening complication or chronic ear disease, the first line of therapy includes intravenous antibiotics for the treatment of the underlying infection. The aforementioned study by Gadre and Chole found that over a 40-year period, antibiotic treatment remained the primary therapy for petrous apicitis, with a reduction in surgical treatment.[10]

Antibiotics should be directed at the offending organism, which is typically one of the pseudomonads. In rare cases, tuberculosis has been identified as the cause, usually in individuals younger than 20 years.

If the patient is medically unstable and unable to tolerate general anesthesia, high-dose IV antibiotics may be attempted to eradicate the infection in place of surgery. Underlying sepsis as a result of the infection may be life threatening, and the patient should be stabilized as quickly as possible for anticipated surgical drainage of the infection.

Surgical Therapy

For patients whose symptoms do not respond to appropriate antibiotics (eg, continued fevers, otorrhea, headache) or who develop complications from the infection, including CN deficits, abscess formation, or venous sinus thrombosis, surgical intervention is warranted.

Because of the complicated anatomic relationships involved with the petrous apex, various surgical approaches to petrous apicitis have been developed. No single standard approach to the petrous apex exists, and any of these approaches may be required depending on the clinical presentation.

In most patients, the petrous apex can be drained via a transmastoid approach to the temporal bone. This approach involves a complete mastoidectomy. In patients with infections within the posterior aspect of the petrous apex, identifying fistulous tracts along the sinodural angle, the subarcuate air cell tract, and air cells inferior to the posterior semicircular canal is often possible. Enlarging these structures provides adequate drainage of the petrous apex.

Infections in the anterior portion of the petrous apex may be open to a hypotympanic-subcochlear approach or an approach via the air cell tract below the posterior semicircular canal and superior to the jugular bulb. Typically, the margins of the exposure include the cochlea superiorly, the carotid artery anteriorly, and the internal jugular vein and bulb posteriorly. These approaches may require removal of the posterior canal wall for adequate exposure. A high or anteriorly placed jugular bulb may render this approach unfeasible.

In either of these cases, some surgeons advocate placing a drain, such as silastic draped from the infected site into the mastoid or hypotympanum, to maintain the patency of the drainage pathway and prevent recurrence of the infection.

A complete petrous apicectomy is reserved for patients who require additional exposure to the anterior petrous apex not afforded by the above routes. This classic procedure was originally described by Ramadier and was popularized by Lempert. The exposure involves removal of the anterior canal wall and condyle of the mandible; exposure of the epitympanum; avulsion of the tensor tympani; opening of the tensor semicanal; and then dissection in the triangle between the carotid artery (posterior), the cochlea (superior), and the middle fossa dura (anterior). This approach can be modified by preserving the anterior canal wall and condyle.

As noted by Chole, dissection between the cochlea and carotid must be avoided when attempting to preserve hearing because of the limited distance between these 2 structures. Infections within the anterior-most portion of the petrous apex, such as an epidural abscess, are probably best drained via a middle fossa approach.

Ogahlai has recently described a circumferential petrosectomy, a technique that removes the maximum amount of infected temporal bone while preserving the integrity of the peripheral auditory pathway and facial nerve. The circumferential petrosectomy removes most of the temporal bone around the external, middle, and inner ear. A combined retrolabyrinthine-apical petrosectomy is performed in conjunction with the fallopian bridge technique using a transmastoid and middle cranial fossa approach. A split temporalis muscle flap is used to bring vascularized tissue to the mastoid, jugular foramen, and petrous apex.

Improved proficiency in sinus endoscopy has led to increased acceptance of a transsphenoidal approach for lesions in this location. The transsphenoid approach to the petrous apex, a surgical procedure described for the first time by Montgomery in 1977.[15] However, the presence of venous sinuses between the petrous apex and sphenoid, such as the cavernous sinus, can make this approach challenging. It can be considered for lesions located in the medial section of the petrous apex abutting and/or prolapsing into the posterior wall of the sphenoid sinus.

Postoperative Details

Even with surgical drainage, prolonged postoperative antibiotics are usually recommended, typically for 2-3 weeks.

Patients with chronic petrous apicitis may have accompanying osteomyelitis; if so, 3-6 weeks of IV antibiotics may be required.

Follow-up

If a canal wall–down procedure was required for surgical drainage of the apicitis, long-term care of a mastoid bowl cavity is required.

Follow-up visits should include evaluation to ensure the infection is completely eradicated because recurrence may occur if the surgical drainage pathways to the petrous apex become reobstructed.

Obtain a follow-up CT scan or MRI to ensure that the petrous apex is adequately drained. If the patient is asymptomatic, adequate drainage should occur within several months of discharge from the hospital. If the patient is symptomatic, follow-up imaging should occur sooner.

Outcome and Prognosis

Although death was a common outcome of these infections in the preantibiotic era, the use of antibiotics and improved surgical techniques have dramatically improved survival rates in patients with petrous apicitis. Because of the small number of patients who present today, no large series have examined outcomes of patients with petrous apicitis. Furthermore, the widespread availability of antibiotics and improved surgical techniques in use today make comparisons with older series unhelpful.

In a series of 8 patients by Chole and Donald, 7 patients had resolution of their infection, while 1 died of infection.[16] Of these patients, 1 required a reexploration for continued drainage and another had persistent deficits of CN IX-XI. The remainder experienced resolution of their otorrhea and CN deficits.

Another report of 2 patients with Gradenigo syndrome treated with antibiotics and mastoid drainage demonstrated complete recovery of CN VII with treatment in both patients.

Resolution of CN palsies typically occurs over 3-4 weeks when adequately treated. Surgery in combination with antibiotics typically results in a more rapid resolution of CN deficits than antibiotics alone.

In a study of two cases of petrous apicitis that were refractory to medical therapy and involved progressive cranial neuropathy, Patel et al reported that preoperative cranial nerve deficits improved 24-48 hours after an anterior petrosal (Kawase) approach to petrous apex resection and debridement.[17]

Overall, hearing results in patients undergoing surgery for all petrous apex lesions are quite good. In a large series examining hearing results in surgery for primary petrous apex lesions, hearing was preserved in approximately one half of patients, improved in approximately one third, and worsened in 4% (1 patient of 25). Although this group presented with various lesions within the petrous apex, including cholesterol granulomas, cholesteatomas, mucoceles, and eosinophilic granulomas, it is reasonable to believe that hearing results would be comparable, if not better, in patients with apicitis.

Future and Controversies

Although surgery has historically been the mainstay of therapy, some authors are advocating IV antibiotics as a first-line therapy for petrous apicitis. These authors point out the successful treatment of certain brain abscesses with medical therapy alone, as well as some patients with petrositis whose condition resolved over a prolonged period with IV antibiotics alone. However, the authors agree that a poor clinical response to antibiotics is an indication for surgical drainage.