The concept of perilymphatic fistula (PLF) as a cause of hearing loss and imbalance has evolved slowly over the last 30 to 40 years. In 1962 Farrior (1) first described a postsurgical PLF as a complication of polyethylene tube stapedectomy, stating, “The polyethylene strut placed over the mobilized footplate may gradually work its way into the inner ear so that the perilabyrinthine fluid escapes submucosally up the polyethylene tube. The perilabyrinthine fluid may escape all the way to the incus, producing submucosal blebs and resultant separation of the incus and polyethylene strut.” Subsequently, Steffen et al. (2) described “slipped strut syndrome” following stapedectomy. They found 24 of 52 cases (46%) explored for drop in hearing to have PLF, “usually in the nature of perilymph ascending through the lumen of the tube.” In 1967 Harrison et al. (3) brought attention to the vestibular component of PLF in their report of 46 cases of PLF confirmed by surgical exploration for revision of stapes surgery. According to Harrison et al. (3), “The symptoms in the proven cases are so similar to those of labyrinthine hydrops that it makes the differential diagnosis very difficult, if not impossible.” They observed vertigo in 35%, tinnitus in 28%, fluctuating progressive hearing loss in 87%, imbalance in 39%, and fullness or aural pressure in 35% of patients with surgically confirmed fistulas. Their surgical findings included:

a punched out funnel-shaped or round hole in a pliable membrane, to a crescent-shaped opening conforming to the semicircular shape of the beveled cut end of the polyethylene prosthesis. A slitlike opening was often observed after the stainless steel wire with Gelfoam or fat graft techniques. The perilymph escaped freely in a few cases with continued rapid refilling of the oval window niche, while in others the perilymph could be seen to pulsate and slowly seep through passages not clearly defined (3).

Their results state that “prognosis for hearing improvement after fistula repair is guarded at best. Only 11 patients (24%) achieved a practical level of 30 dB or better after fistula repair” (3). They conclude that “those patients with a definite sudden onset of symptoms of moderate severity explored early had the best prognosis. Those explored and repaired after symptoms had been present over a long period of time did not obtain good results” (3). In addition to postsurgical PLF, several groups noted congenital PLF in the region of the stapes (4, 5, 6, 7, 8). In all these cases the patients presented with recurrent meningitis or frank cerebrospinal fluid (CSF) otorrhea.

A major conceptual advance in understanding the role of hydrodynamics in damage to the inner ear was the theory of membrane breaks in sudden hearing loss proposed by Simmons (9). In this landmark paper, he reviewed animal work by Lawrence and McCabe (10), in which intense sound caused disruption of Reissner’s membrane, and his own demonstration in cats of intracranially applied endolymphatic duct pressure causing sudden loss in electrical activity of the ear without signs of dysequilibrium (11). He also reported 15 clinical cases meeting specific criteria:

1) Each could document exactly when the loss occurred. 2) Each had previously normal hearing in that ear and (with two exceptions) had a negative history for any auditory disorder other than childhood otitis. 3) None had vertigo associated with his sudden hearing loss or developed dizziness subsequently. 4) None had a history or physical examination positive for recent respiratory infection, cardiovascular disease, diabetes, or other serious systemic disease (11).

He noted that prognosis for recovery is strongly correlated with time of initial examination. Specifically, “If a patient was examined less than two days after hearing loss, likelihood of recovery (regardless of therapy) was about 50%. By three weeks, chance for recovery was about 5%. Almost no one seen later than two months recovered their hearing completely. No patient over 45 years of age recovered their hearing” (9).

Fee (12) gave the first report of PLF resulting from head trauma in patients with otherwise normal ears. Based on his three cases, Fee (12) stated that the “most common combination of presenting complaints resembles the syndrome of labyrinthine hydrops and tinnitus, fluctuating hearing loss, fullness, and vertigo. Less commonly, the patient will complain of sudden hearing loss, popping, gurgling, or a sensation of sudden linear movement.” He stated that audiometric findings in traumatic PLF are consistent with hydrops, showing low-tone SNHL progressing to flat loss. Discrimination scores fluctuate initially and then drop disproportionately low. Reduced discrimination may be the only audiometric finding in a patient with primarily vestibular symptoms. He observed that the “most consistent labyrinthine finding, regardless of the presenting complaint, is a direction fixed positional nystagmus usually directed toward the affected ear when it is lowermost. A canal paresis in response to caloric stimuli is usually present early in the course of fistula” (12). He recommends exploration of any ear with persisting positional vertigo and hearing loss following trauma, regardless of whether the hearing loss was conductive or sensorineural.

Stroud and Calcaterra (13) were the first to propose spontaneous oval window PLF. A summary of clinical features in their four cases suggests significant similarity to Meniere’s disease. However, age of onset was younger than typically seen in Meniere’s; three fourths of cases began with an episode of increased intracranial or venous pressure such as bending over, laughing, or singing; and all cases showed predominantly high frequency hearing loss with disproportionate decrease in discrimination, in contrast to the low-tone loss often seen in early Meniere’s. All cases showed progressive canal paresis on caloric testing, consistently negative fistula test with pneumatic otoscope, and no positional nystagmus above and beyond that already present spontaneously in some patients. They postulated that, unlike Simmon’s presumption of inner ear membrane rupture, patients with increased intracranial pressure will transmit it equally to all membranous compartments of the inner ear, resulting in a pressure gradient across the stapediovestibular ligament or round window membrane leading to rupture and perilymph leakage. According to Stroud and Calcaterra (13), “If a congenital weakness of the mucoendosteal articulation of stapes footplate and vestibular fenestra were combined with a rapid equilibration system between the cerebrospinal fluid space and the perilymphatic space, it would be easy to explain how transient rises in intracranial pressure could occasionally produce perilymph fistulas.” They go on to say,

Although the incidence of spontaneous perilymph fistula is completely unknown, we suspect that it occurs only rarely. If this were commonly to mimic Meniere’s disease, it most certainly would have been discovered during the many thousands of destructive labyrinthine operations. It is possible that these leaks occur intermittently, and the fistulas heal over between attacks of symptoms (13).

Goodhill (14) presented three cases of oval and/or round window perilymphatic fistulas resulting from exertion. All presented with sudden hearing loss. Some had immediate or delayed vertigo, and some had no vestibular symptoms. Goodhill (14) advanced the now famous theory of implosive and explosive membrane rupture routes. The explosive route transmits increased intracranial pressure to the perilymphatic space via a patent cochlear aqueduct or internal auditory canal and cribrose area, resulting in rupture of the stapediovestibular ligament or round window membrane and perilymph volume shift into the middle ear. The implosive route transmits increased air pressure (e.g., Valsalva maneuver) from the nasopharynx via the eustachian tube and middle ear to the round and oval windows, with resultant membrane rupture driving air into the inner ear and displacing perilymph into the middle ear. Goodhill (14) concludes, “Where there is a distinct exertion history, and where spontaneous improvement is not observed at bed rest, exploratory tympanotomy and careful search for an oval window or round window rupture may be indicated.” In this theory, as in explanations of PLF after trauma or stapedectomy, there is an assumption of perilymph volume shift from the inner ear to the middle ear (with or without pneumolabyrinth) to account for hearing loss or imbalance.

A variety of basic science studies have been undertaken that bear directly on the validity of Goodhill’s theory of explosive rupture of the round window membrane or stapediovestibular ligament. Anatomic studies of human cochlear aqueduct patency have observed variable patency with a general trend to progressive soft tissue occlusion of the aqueduct lumen with increasing age (15, 16, 17, 18, 19). Jackler and Hwang (20) emphasized that previous reports identifying a dilated cochlear aqueduct as a possible explanation for PLF have only assessed configuration of the medial end of the aqueduct. Their more thorough evaluation of the radiographic anatomy looked at the diameter of the different segments in computed tomography (CT) scans of 100 ears. The medial end is highly variable in dimension. The portion that traverses the wall of the otic capsule was so small, however, that it was only radiographically identifiable in 56% of cases and never exceeded 2 mm in diameter. Others have studied the pressure relationships between intracranial and perilymphatic spaces (21, 22, 23, 24, 25, 26, 27). In cases of wide open disruption of stapediovestibular ligament or round window membrane, it is conceivable that there is a volume shift of perilymph from inner ear to middle ear associated with increased intracranial pressure. However, the pressure transfer function across the inner ear should be markedly mitigated by the narrowness of the vestibular and cochlear aqueducts in most patients.