In 1958, Edward L. Kaplan and Paul Meier collaborated to publish a seminal paper on how to deal with incomplete observations. Subsequently, the Kaplan-Meier curves and estimates of survival data have become a familiar way of dealing with differing survival times (times-to-event), especially when not all the subjects continue in the study. “Survival” times need not relate to actual survival with death being the event; the “event” may be any event of interest. Kaplan-Meier analyses are also used in non-medical disciplines. The purpose of this paper is to explain how Kaplan-Meier curves are generated and analyzed. Throughout this article we will discuss Kaplan-Meier (K-M) estimates in the context of “survival” before the event of interest. Two small groups of hypothetical data are used as examples in order for the reader to clearly see how the process works. These examples also illustrate the crucially important point that comparative analysis depends upon the whole curve and not upon isolated points.

This study extended the ®ndings of Ketten et al. [Ann. Otol. Rhinol. Laryngol. Suppl. 175:1±16 (1998)] by estimating the three-dimensional (3D) cochlear lengths, electrode array intracochlear insertion depths, and characteristic frequency ranges for 13 more Nucleus-22 implant recipients based on in vivo computed tomography (CT) scans. Array insertion depths were correlated with NU-6 word scores (obtained one year after SPEAK strategy use) by these patients and the 13 who used the SPEAK strategy from the Ketten et al. study. For these 26 patients, the range of cochlear lengths was 29.1±37.4 mm. Array insertion depth range was 11.9±25.9 mm, and array insertion depth estimated from the surgeon's report was 1.14 mm longer than CT-based estimates. Given the assumption that the human hearing range is ®xed (20± 20,000 Hz) regardless of cochlear length, characteristic frequencies at the most apical electrode (estimated with Greenwood's equation [Greenwood DD (1990) A cochlcar frequency ± position function of several species ± 29 years later. J Acoust. Soc. Am. 33: 1344±1356] and a patient-speci®c constant a s ) ranged from 308 to 3674 Hz. Patients' NU-6 word scores were signi®cantly correlated with insertion depth as a percentage of total cochlear length (R = 0.452; r 2 = 0.204; p = 0.020), suggesting that part of the variability in word recognition across implant recipients can be accounted for by the position of the electrode array in the cochlea. However, NU-6 scores ranged from 4% to 81% correct for patients with array insertion depths between 4% and 68% of total cochlear length. Lower scores appeared related to low spiral ganglion cell survival (e.g., lues), aberrant current paths that produced facial nerve stimulation by apical electrodes (i.e., otosclerosis), central auditory processing dif®culty, below-average verbal abilities, and early Alzheimer's disease. Higher scores appeared related to patients' highaverage to above-average verbal abilities. Because most patients' scores increased with SPEAK use, it is hypothesized that they accommodated to the shift in frequency of incoming sound to a higher pitch percept with the implant than would normally be perceived acoustically.

Thirty-nine percent (29/75) of subjects with implants were dizzy after implantation. The majority of subjects experienced dizziness in a delayed episodic fashion. Dizziness was not related to implant activation. It seemed that delayed dizziness was not related to immediate surgical intervention but could result from chronic changes occurring in the inner ear; there was some suggestion this could take the form of endolymphatic hydrops.

Objectives: A new technique for determining the position of each electrode in the cochlea is described and applied to spiral computed tomography data from 15 patients implanted with Advanced Bionics HiFocus I, lj, or Helix arrays. Methods: ANALYZE imaging software was used to register 3-dimensional image volumes from patients' preoperative and postoperative scans and from a single body donor whose unimplanted ears were scanned clinically, with micro computed tomography and with orthogonal-plane fluorescence optical sectioning (OPFOS) microscopy. By use of this registration, we compared the atlas of OPFOS images of soft tissue within the body donor's cochlea with the bone and fluid/ tissue boundary available in patient scan data to choose the midmodiolar axis position and judge the electrode position in the scala tympani or scala vestibuli, including the distance to the medial and lateral scalar walls. The angular rotation 0 0 start point is a line joining the midmodiolar axis and the middle of the cochlear canal entry from the vestibule. Results: The group mean array insertion depth was 477 0 (range, 286 0 to 655 0). The word scores were negatively correlated (r =-0.59; P = .028) with the number of electrodes in the scala vestibuli. Conclusions: Although the individual variability in all measures was large, repeated patterns of suboptimal electrode placement were observed across subjects, underscoring the applicability of this technique.

To our knowledge, this is the first report of specific criteria for completing the SB system. It is also the first in-depth description of the location within the system in which the majority of variances occur.