UNIFORM REPORTING PROTOCOL
Jack M Kartush, MD

In reviewing the results of ossicular chain reconstruction reported over the last forty years, it becomes clear that a uniform protocol for reporting results is needed. To enhance the accuracy of future reporting and allow meaningful study comparisons, a consensus should be obtained that 1) delineates essential data and 2) stratifies cases within various prognostic categories. These two goals can be accomplished by the application of a uniform Tympanoplasty Reporting Protocol and a Middle Ear Risk Index.

The Tympanoplasty Reporting Protocol (TRP) defines 1) pertinent pre- and intra-operative risk factors, 2) primary and adjunctive surgical procedures and 3) outcome reporting (FIGURE 18). In part, the TRP is an integration of two commonly used classifications (Austin and Bellucci) which, in and of themselves are insufficient measures of the severity of middle ear disease. Pertinent queries in establishing the TRP include:

1. What are the preoperative prognostic factors (perforation, drainage, cholesteatoma, ossicular status)?
2. What concomitant procedures are performed (tympanic membrane grafting, mastoidectomy)?
3. Is the procedure primary, staged or an unplanned revision?
4. What is the audiometric criteria for "success"?
5. How should the Air-Bone Gap (ABG) be measured (pre vs. post operative bone conduction)?
6. Which prosthesis is used for which condition?
7. What are the results beyond hearing (graft integrity, infection, recurrent cholesteatoma?
8. What is the duration of follow-up?

Most authors agree that preoperative abnormalities indicative of the severity of underlying Eustachian tube dysfunction and infection have a significant influence on prognosis. Black12 noted that "good risk" patients had a 67% chance of obtaining hearing within 10db versus only a 25% chance if they were considered a poor risk. Although the study by Brackmann et al4 did not demonstrate a significant correlation of preoperative risk factors to postoperative hearing, this was most likely due to their routine staging of ossiculoplasty wherein all unfavorable cases were deferred until conditions improved.

Often quoted is Wullstein’s 1956 article that proposed a five part system intended to help in determining the prognosis of hearing improvement based on the residual ossicular remnants and how they might be reconstructed. The limitation of this system was recognized by Belluci who proposed a dual classification which added an appraisal of middle ear pathology. Four groups were defined ranging from "good" to "very poor" prognosis: Group I. Dry ear, Group II. Occasional drainage, Group III. Persistent drainage with mastoiditis and Group IV. Persistent drainage with nasopharyngeal malformation (cleft palate or choanal atresia). Even using both of these classifications, however, many important parameters remain undefined. Austin developed a system which describes the residual ossicular remnants. Patients with a malleus are considered to have a good prognosis with (M+S+) or without (M+S-) a stapes superstructure. Those without a malleus (M-S+) or (M-S-) are considered a poor risk. For the reporting protocol, an additional category has been added to represent an intact ossicular chain (M+I+S+). This allows use of the same protocol for tympanoplasty surgery without ossiculoplasty.

In 1965, the American Academy of Oph thalmology and Otolaryngology presented a proposed standard classification for surgery of chronic ear infection. It highlighted the need for better reporting of gross pathology and postoperative results but it has not been well accepted as a means to compare different studies because some areas are too detailed (eg five types of perforations are listed) and because some areas are ill-defined (eg "describe the type and extent of pneumatization").

Otologists differ on what hearing level should be considered a measure of "success". Some consider closure to within 10 dB too strict a definition while others use 15 or 20 dB. Acknowledging the less favorable prognosis of patients missing both incus and stapes, Brackmann et al4 suggest two different criterion: 15 dB for PORP’s and 25 dB for TORP’s. A reasonable compromise might be to require stratification of results at 0 -10 dB "excellent", 10 - 20db "good", 20 - 30dB "fair" and more than 30 dB "poor". Authors could then choose to define success as representing a combination of the first two or three categories.

Clinicians also differ on whether an assessment of hearing results should compare the postoperative air conduction level to either the pre or postoperative bone conduction level. Using the preop bone line may be misleading particularly if the preoperative audiogram was not performed immediately before surgery because of the possibility of interval changes. Conversely, using the postoperative bone line may obscure an interval loss in the sensorineural hearing level. In balance, because substantial changes in the bone line are infrequent following ossicular chain reconstruction, we prefer to use the postoperative bone line. Patients with significant sensorineural hearing losses should be stratified and discussed separately.

The Tympanoplasty Reporting Protocol can be used to generate a numeric indicator of the severity of middle ear disease (Table IV). An index of Middle Ear Risk (MER) can be used to stratify patient groups and allow more meaningful comparisons between different studies. Examples of MER assessments are illustrated in Table V. A method for uniform data presentation is shown in Table VI.

Suggested major risk categories can be derived from the MER index as follows:

Normal = 0

Mild pathology = 1-3

Moderate pathology = 4-6

Severe pathology = 7-12

Note that allocation of these risk groups are weighted with fewer categories predicting a favorable prognosis. A computerized database application for these analyses is available.