October 28, 2004
The incidence of dizziness and dysequilibrium following head and/ or neck injury lies between 40-60%, even following mild or moderate head injuries not requiring acute hospitalization. Accordingly, most practitioners should have a fairly extensive experience with this large outpatient group. Yet, in spite of their familiarity with this common problem, many clinicians routinely use a diagnostic approach which is not pathophysiologically based and a treatment approach which is often confined to attempts at providing symptomatic relief with vestibular suppressants. Indeed, the tendency to attribute most post-traumatic dizziness to the post-concussion syndrome without diagnostic consideration of specific vestibular pathologies has been the cause of a significant and largely unnecessary increase in morbidity in this population. What follows is a review of those specific pathologies which, taken together, cause the great proportion of post-traumatic symptoms and which should rightly form the basis for a diagnostic and management approach.
TEMPORAL BONE FRACTURE
Basal skull fractures may be of two kinds depending upon the relationship of the fracture line to the long axis of the petrous bone. These fractures can be well demonstrated on thin section temporal bone CT. Eighty percent are longitudinal and twenty percent are transverse. As longitudinal fractures run thru the axis of the middle ear, they often produce tympanic membrane tears with otorrhagia, and conductive or mixed hearing loss. They also can cause facial palsies. Given that the fracture line does not directly involve the inner ear, vestibular symptoms are somewhat less common in this type of fracture, but they can often occur due to concurrent labyrinthine concussion, perilymphatic fistula, or benign paroxysmal positional vertigo (see below). Transverse fractures extend thru the inner ear and produce damage to cochlear and/or vestibular labyrinthine neural structures directly with or without hemorrhage. They often produce hemotympanum and sensorineural hearing loss, and less commonly facial palsies. They also can be complicated by CSF leak and/or meningitis. Vestibular symptoms consist acutely of severe vertigo and ataxia (and possibly vegetative symptoms), which progressively improves due to CNS compensation, but in most cases persists to some degree well into the outpatient course. The nature of these persistent vestibular symptoms depends also upon the possible concurrent presence of a perilymphatic fistula, BPPV or the development of delayed hydrops (see below).
Successful outpatient management of the post-traumatic vertigo in this group of patients involves judicious restraint in the use of symptomatic treatment with vestibular suppressants (so as not to interfere with the development of CNS compensation), early resort to vestibular rehabilitation as the mainstay of treatment, and vigilance in detecting the concurrent existence or development of BPPV, perilymphatic fistula or delayed hydrops. Ongoing follow-up, including serial exams and timely utilization of vestibular laboratory diagnostic studies should be routine in patients not demonstrating progressive, if not rapid, recovery.
In spite of all such efforts, however, a small but significant percentage of patients with uncomplicated temporal bone fracture (estimates vary from 10-30% depending upon severity criteria) will remain with some degree of vestibular symptomatology indefinitely due to incomplete CNS compensation. This is the likely prognosis for those patients whose recovery during vestibular rehabilitation plateaus for two months in the absence of any other concurrent treatable vestibular post-traumatic pathology. In such cases, liberal use of vestibular suppressants long term remains the only option.
Blunt head injury can concuss the membranous labyrinth against the otic capsule (much as the brain is concussed against the skull in cerebral concussion). This results in acute hypofunction of some portion of the vestibular neural substrate within the affected labyrinth. As labyrinthine concussions and transverse temporal bone fractures both produce acute unilateral (or asymmetrical) vestibular hypofunction, their clinical manifestations and course are much the same. There is the acute onset of vertigo and ataxia whose severity is proportional to the degree of hypofunction and there is some variable degree of associated vegetative symptomatology. Symptoms are most severe at the time of the head injury and invariably improve but with a temporal profile which can vary from days to months, with a minority of patients having some degree of residual symptomatolgy which persists indefinitely. There are, however, some differences which can be seen between labyrinthine concussions and fractures. Concussions are somewhat more likely to produce reversible neural insults which can result in a very abbreviated course of recovery. They are less likely to produce associated perilymphatic fistulas and to lead to the development of delayed hydrops, but are more likely to be associated with coexistent BPPV (perhaps because fractures more often result in complete ablation of all vestibular function), and hearing loss is a less common accompaniment. Although radiological confirmation of involvement of vestibular structures is necessarily absent in uncomplicated concussions, several types of vestibular diagnostic studies are now available to confirm and quantify the severity of these lesions and even to measure the degree of CNS compensation. Outpatient management considerations are otherwise the same for labyrinthine concussions as described above for fractures.
BENIGN PAROXYSMAL POSITION VERTIGO (BPPV)
Because BPPV is the most common cause of vertigo in general and of post-traumatic vertigo in particular, clinicians need to be well versed in its varied clinical presentations and their treatment options.
Canalolithiasis and Cupulolithiasis are the underlying mechanisms of BPPV In either case, otoconia are displaced from an injured uticular membrane into one of the semicircular canals. Even relatively mild blunt head trauma can provide enough shearing force to the utricular membrane to displace critical numbers of otoconia. There is also much anecdotal evidence that the acceleration- deceleration forces produced in whiplash injuries even without head injury may also suffice to produce such displacements. Because the posterior canal is situated directly inferior to the utricule when the head is upright, the otoconia settle into that canal in 90+ % of cases, but the lateral (5-8%) or the anterior canal (1-3%) can be sites of deposition as well.2
In Canalolithiasis - the most common cause of BPPV- the calcium carbonate crystals that compose the individual otoconia tend to form clumps that are free to move within the lumen of the involved canal. Changes in head position - especially in the axis of the canal - cause a displacement of the otoconial plug which in turn causes a displacement of the cupula. This produces a brief paroxysm of vertigo, nystagmus and ataxia. These paroxysms of positional vertigo continue to occur until a critical portion of the otoconial mass moves back out of the canal. This egress can eventually occur as a result of the random head movements characteristic of unrestrained daily activities or much more rapidly with more purposeful head positionings performed as part of the treatment regimen (see below). However, a damaged utricular membrane may contain numbers of less than adequately adherent otoconia which will provide a source of future displacements causing a potential for recurrences. Indeed, recurrences occur in 30% of cases of BPPV within the first year and 50% within five years.
BPPV as a cause of post-traumatic vertigo can be a straightforward diagnosis in many but not all cases. It often presents with the classical history of isolated paroxysms of positional vertigo and the equally classical paroxysmal positional nystagmus can be documented with Dix-Hallpike positional testing. However, there are many instances, in which the presentation and course can be confusing. One or more types of post-traumatic vestibular pathology can coexist. For example, if BPPV and labyrinthine concussion are both present (a fairly common event), the paroxysms of positional vertigo may be less evident in the context of the ongoing symptoms of motioninduced dizziness commonly seen as a result of the resolving effects of the concussion alone. If the BPPV is not recognized, then delayed or limited recovery will be unnecessarily attributed to a refractory case of concussion. Further complicating management, there may be a hiatus between the time of injury and the onset of BPPV. This is postulated to be due to the time required for partially displaced crystals towork loose, enter the affected canal and coalesce into a critical mass. This delay in onset of symptoms can certainly be weeks and may be months. Thus, ongoing follow-up and re-examination/retesting is especially necessary in patients who experience unexplained setbacks or refractory courses.
Recognition of BPPV as the sole or contributing cause of post- traumatic vertigo opens up the opportunity for treatment which is most often easily effected and rapidly successful. Canalith Repositioning Maneuvers (Epley/modified Epley Maneuvers) have success rates reported to be between 80-95% following one to three repetitions (with occasional cases requiring several more). Abrupt cessation of all symptoms is common, but other possible outcomes result in residual symptoms which give the appearance of failure but which can often be easily treated if recognized. In this regard, one must note that many patients will experience residual lightheadedness, motion sensitivity and dysequilibrium following an essentially successful maneuver which will clear following a short course of habituation/balance retraining exercises. These patients can be distinguished by the absence of positional nystagmus on retesting. In some cases, both ears are involved and must be treated sequentially. Additionally, care must be taken to distinguish lateral canalolithiasis or anterior canalolithiasis from the much more common posterior canal variant.3 Further, in what may be upwards of 5-10% of cases, Epley maneuvers can actually displace the otoconial plug from the posterior into the horizontal or the anterior canal/' These variants are a potential cause for confusion unless one recognizes the very different type of positional nystagmus produced by each of the three types of canalolithiasis. Because specific maneuvers are required for these respective variants, failure to determine that one is present in a given case will inevitably lead to treatment failure.
Cupulolithiasis represents a much less common mechanism for BPPV, and is another potential cause for treatment failure if not recognized. In these cases, the otoconia adhere to the gelatinous cupula rather than remaining free within the lumen of the canal. Once again, the pattern of positional nystagmus is different. As all Epleybased maneuvers are designed to roll the loose plug of otoconia out of the canal, they are unhelpful in cases of cupulolithiasis. Liberatory maneuvers-maneuvers that utilize rapid, high amplitude head accelerations-are necessary to loosen the otoconia from the cupula.
Those responsible for the outpatient management of patients with post-traumatic BPPV must keep in mind the above considerations. Epley maneuvers can easily be performed in the office setting. Success rates of 80-90% should be routinely expected. These are cases in which vestibular suppressants have a minimal role (confined to providing tolerance for maneuvers and/or exercises). Persistence of symptoms demands further exploration of pathogenesis and/or utilization of more appropriate vestibular rehabilitation techniques.
In 1% of cases or less, surgical intervention is warranted. Posterior canal plugging is a relatively effective and safe procedure which "dams off" the cupula from exposure to the otoconial plugs in the canal.5 Singulectomy is a procedure which deafferents the cupula in the posterior canal alone, and is less commonly performed.
PERILYMPHATIC FISTULA (PLF)
Blunt head trauma with or without temporal bone fracture can cause rupture of the oval or round windows which separate the perilymphatic space from the middle ear. This can produce a variety of symptoms of vestibular and/or cochlear dysfunction. Unfortunately, most of these clinical presentations mimic closely those of other post-traumatic vestibular syndromes.
Patients with PLF can present with sudden unilateral sensorineural hearing loss (with or without tinnitus and aural blockage) and /or acute onset of persistent, gradually diminishing vertigo and ataxia which is indistinguishable from the clinical picture of temporal bone fracture or inner ear concussion. PLF has also been reported to produce positional vertigo which can be strikingly paroxysmal by history - although reports of positional nystagmus typical of BPPV are hard to substantiate as unrelated to coexistent canalolithiasis. PLF often produces paroxysms of spontaneous vertigo and/ or fluctuations in hearing loss, much as in endolymphatic hydrops (although the latter would more likely be of delayed onset when seen post-traumatically). PLF can produce sound- induced vestibular symptoms (Tullio phenomenon) as well as pressure- induced vestibular symptoms including vertigo, oscillopsia and ataxia. Patients often report that the latter symptoms occur during activities that effectively induce Valsalva maneuvers.
The sensitivity of labyrinthine structures to changes in pressure across the middle ear forms the basis for the one objective finding that is somewhat specific for PLF (although it, too can be seen in hydrops). This finding - the Hennebert sign - can be produced by a "pressure fistula test" in which pressure is introduced by pneumatic otoscope or tympanogram into the external auditory canal. A positive sign consists of ocular deviation or nystagmus observed directly or recorded as part of an ENG study. Subjective responses, including vertigo, oscillopsia or postural sway are considered suggestive but not definitive responses.6
More definitive diagnosis involves performing exploratory tympanotomy with direct observation for perilymph leakage thru the oval or round window. However, the inevitable presence of tissue fluid in the middle ear during the procedure makes even intraoperative confirmation of the diagnosis uncertain. As the definitive surgical treatment of placing a soft tissue graft over the fistula can easily be done once the tympanotomy has been performed, it is routinely carried out even in the absence of convincing intraoperative findings.
Most cases of suspected post-traumatic PLF are managed conservatively. Patients are put at bed rest for durations of a week or two, and instructed to avoid any activities that would produce Valsalva type maneuvers. Opinions vary as to whether, with the notable exception of unstable hearing, persistence of symptoms for further weeks or even months warrants exploration.7 In general, much controversy surrounds the subject of PLF. It is generally accepted as a viable post-traumatic entity (as opposed to occurring spontaneously), but even then there is wide disagreement in terms of diagnostic criteria and management.
POST-TRAUMATIC MENIERE'S SYNDROME (HYDROPS)
A syndrome clinically indistinguishable from idiopathic Meniere's Syndrome can begin post-traumatically.8 ' There can be a hiatus of months and perhaps even years between the injury and the onset of symptoms (delayed hydrops). The clinical picture of fluctuating/ progressive unilateral hearing loss, tinnitus, aural fullness and episodes of vertigo, ataxia and vegetative symptoms occurring spontaneously and lasting minutes to hours is classical. This clinical profile can often be difficult to distinguish from that of perilymphatic fistula-especially in the acute and subacute stage following the trauma when PLF is more likely. Management considerations are identical to the idiopathic form of Meniere's Syndrome.
Head trauma can produce focal lesions that involve the temporal or parietal association cortex which receives vestibular projections. These lesions can occasionally form seizure foci which can lead to simple or partial complex sensory seizures whose manifestation is episodic vertigo. The episodes usually last no more than seconds or minutes. They can be associated with mild nausea but not vomiting. Nystagmus can be an accompaniment due to stimulus of contiguous cerebral oculomotor control areas.10 Tinnitus can at times be associated. Contralateral paresthesias and/or olfactory and gustatory symptoms are occasionally reported. Unconsciousness will follow only if the seizure becomes generalized.
Epilepsy represents a relatively uncommon cause of post- traumatic vertigo. Seizure workups are not routinely fruitful except in selected cases in which other symptoms suggestive of sensory seizures are present or, certainly, in which loss of consciousness occurs-something that only rarely occurs in aural vertigo due to intense vegetative symptoms leading to vagal syncope.
Vestibular symptoms can possibly be produced from damage to either cervical vascular or cervical neural structures. Trauma to a vertebral artery can result in thrombosis or dissection with resultant brainstem/cerebellar ischemia." The case for vertigo mediated by C1-3 posterior roots following whiplash injury to deep cervical soft tissue is less clear.12 There is ample evidence that these roots mediate cervical proprioceptive inputs that actually synapse within the vestibular nuclei and which would provide an anatomical substrate capable of mediating vestibular-like symptoms. Further, there are many experimental models in lower species in which deep cervical lesions produce objective manifestations indistinguishable from primary vestibular lesions (nystagmus and ataxia). However, these objective findings are not reproducible in primates. Human experiments document subjective sensations of dizziness and perceptions of lateropulsion, but no measurable objective parameters that could form the basis for a diagnostic test for cervical vertigo have yet to be demonstrated. In many cases, there are cervical trigger points that produce subjective vertigo with deep palpation. In all cases in which the diagnosis is suspected, treatment is limited to cervical soft tissue physical therapy modalities
Headache is one of the most common post-traumatic symptoms, and upwards of 25% of cases meet the criteria for migraine. Vertigo associ\ated with both idiopathic and post-traumatic migraine has been widely reported, but criteria have not yet been established for the diagnosis of vestibular migraine.13 Headache may not be temporally associated with the vertigo in upwards of a third of these reported cases. Episodic vertigo, positional vertigo and non- vertiginous dizziness have all been described; and some authors accept durations of chronic unremitting symptoms lasting days or weeks as consistent with the diagnosis. Many studies report convincing results of therapeutic trials of medication and/or other modalities used in the treatment of other types of migraine.14
POST-CONCUSSION SYNDROME (PCS)
Substantial numbers of patients who have sustained mild or moderate head injuries present with a similar cluster of symptoms. These most often include headache, visual symptoms, and cognitive changes. Dizziness, nausea and, at times, tinnitus and hyperacusis are also symptoms that are quite commonly included in this syndrome. However, a diagnosis of PCS for a patient presenting with any of the latter symptoms, presumes that the pathophysiology is confined to traumatic brain injury and not to any of the specific vestibular and/ or auditory system pathologies described above. The pathology of traumatic brain injury includes focal lesions, diffuse axonal injury (DAI), and scattered petechial hemorrhages. These latter injuries, however, are rarely found in patients in which the duration of loss of consciousness is less than 30 minutes, and attributing symptoms to PCS in milder head injuries is probably inappropriate. Further, irregardless of the severity of injury, dizziness and vertigo as well as auditoty symptoms are due to other specifically identifiable pathologies in the overwhelming majority of patients. That being the case, PCS should be a diagnosis of exclusion. Once the diagnosis of PCS has been made, it still remains unclear how long symptoms can appropriately be attributed to it. Many authors suggest consideration should be given to post-traumatic stress disorder when symptoms extend beyond six months to a year.
The vestibular-like symptoms that commonly follow head injury are most often due to one of a number of specific pathologies affecting vestibular, CNS or cervical structures. These pathologies can be readily identified in the majority of cases by appropriate testing in the examining room, and in vestibular function or neurodiagnosic laboratories. Wheteas vestibular suppressants have a roll, they often delay recovery and have been supplanted as the mainstay of treatment in most cases by vestibular rehabilitation techniques. Given the large number of patients requiring outpatient management following head injury, it is appropriate that most generalists and many subspecialists should be aware of the differential diagnosis, workup and treatment options for post-traumatic vertigo.
...THE TENDENCY TO ATTRIBUTE MOST POST-TRAUMATIC DIZZINESS TO THE POST-CONCUSSION SYNDROME WITHOUT DIAGNOSTIC CONSIDERATION OF SPECIFIC VESTIBULAR PATHOLOGIES HAS BEEN THE CAUSE OF A SIGNIFICANT AND LARGELY UNNECESSARY INCREASE IN MORBIDITY IN THIS POPULATION.
1. Wennmo C, Svensson C. Temporal bone fractures. Vestibular and other related ear sequele. Acta Otolaryngol Suppl 989:468:379-83.
2. Honrubia V, Baloh RW, Harris MR, Jacobson KM. Paroxysmal positional vertigo syndrome. Am J Otol 1999:20:465-70.
3. Baloh RW, Jacobson K, Honrubia V. Horizontal semicircular canal variant of benign positional vertigo. Neural 1993:43:2542-9.
4. Herdman SJ, Tusa RJ. Complications of the canalith repositioning procedure. Arch Otolaryngol Head, Neck Surg 1996:122:281-6
5. Walsh RM, Bath AP, Cullen JR, Rutka JA. Long-term results of posterior semicircular canal occlusion for intractable benign paroxysmal positional vertigo. Clin Otolaryngol 1999:24:316-23.
6. Kohut RI. Perilymph fistulas. Clinical criteria. Arch Otolaryngol Head Neck Surg 1992; 118:687-92.
7. Glasscock ME 3rd, Hart MJ, Rosdeutscher JD, Bhansali SA. Traumatic perilymphatic fistula: How long can symptoms persist? Am J Otol 1992:13:333-8.
8. Shea JJ Jr, GcX, Orchik DJ.Traumatic endolymphatic hydrops. Am J Otol 1995;16:235-40.
9. DiBiase P, Arriaga MA. Post-traumatic hydrops. Otolaryngol Clin North Am 1997;30:1117-22.
10. Furman JM, Crumrine PK, Reinmuth OM. Epileptic nystagmus. Ann Neural 1990;27:686-8.
11. Showalter W, Esekogwu V, Newton KI, Henderson SO. Vertebral artery dissection. Acad Emerg Med 1997;4: 991-5.
12. Brandt T, BronsteinAM. Cervical vertigo. J Neural Neurosurg Psychiatry 2001:71:8-12.
13. Dieterich M, Brandt T. Episodic vertigo related to migraine (90 cases): Vestibular migraine? J Neural 1999:246:883-92.
14. Johnson GD. Medical management of migraine-related dizziness and vertigo. Laryngoscope 1998;108(1 pt 2):1-28.
JULES M. FRIEDMAN, MD
Jules M. Friedman, MD, is Assistant Professor of Neurology, Boston University School Of Medicine.
Jules M. Friedman, MD
Center For Vestibular And Balance Disorders
130 Waterman Street
Providence, RI 02906
Copyright Rhode Island Medical Society Oct 2004