Given these specific inclusion and exclusion criteria and the specific purpose of this study, only 29 cases qualified for inclusion from an original screen of approximately 4000 cases. An initial search of our database resulted in a larger number of cases that would theoretically qualify; however, further investigation revealed the need to exclude a great number of cases. The reasons for exclusion of these cases were as follows: scatter on the CBCT images due to fixed prosthodontics, unclear CBCT images, poor charting that did not allow for proper data gathering, no follow-up radiographs, not enough of the sinus being visible in the image, diagnosed periapical pathosis in the examined areas, implants not being placed in the area of the maxillary sinus, or no grafting completed in the maxillary sinus. Although these factors greatly reduced our sample size, this, in turn, created a stronger data set for analyses.

Subject privacy protection

The study required access to University of Michigan Protected Health Information (PHI). PHI was necessary in order to track and coordinate the CBCT data and dental and medical history for each subject. Corresponding subject charts and electronic records were reviewed for retrieval of relevant implant placement and restorative history, medical history, and demographic information, including gender and age and smoking history. Also, any pertinent dental treatment was received and response to treatment were reviewed and recorded. No other personal information was retrieved. The use of PHI involved no greater than minimal risk because each subject was assigned a coded number that was used for all data analyses, tables, and reports.

Measurement methodology

Using CBCT images, the mucosal thickness/height was measured at the point of maximum height using sagittal views, which were perpendicular to the underlying sinus floor at edentulous sites. Using these sagittal views, measurements were taken at four points: ¼, ½, and ¾ of the widest distance of the maxillary sinus from anterior to posterior.