Purpose: Knowledge of the early mortality patterns in edentulous implant patients is limited. This study aimed to report patient mortality within the first year after surgery and compare the cause of death with preexisting conditions reported before surgery.

Materials and methods: In this retrospective cohort study, data from the Swedish National Death Register on patients, consecutively treated in the edentulous arch between 1986 and 2013, were compared with information from the patient files regarding the preexisting health conditions of the deceased patients. One-year survival rates were calculated and compared with expected mortality in a Swedish reference population based on three age groups: young (< 45 years of age), middle-aged (45 to 64 years of age), and old patients (> 64 years of age). Proportions of mortality between study groups and reference populations were tested by means of a log-rank test, and agreement between diagnoses before surgery and cause of death was tested by means of kappa test.

Results: Altogether, 3,877 patients were included, of whom 60 patients died within 1 year after implant surgery (1.5%). The expected mortality in the Swedish reference population was 2.1% (P < .05). Mortality was higher for middle-aged (P = .02) but lower for old patients (P = .0001) compared with the Swedish reference populations. Eight of the deceased patients (13%) had no preexisting conditions, while 48 patients reported a health diagnosis before implant surgery. The most common of these were related to the circulatory system (ICD 10-I), which was the cause of death for 30 patients. A "none to slight agreement" between presurgical diagnoses and cause of death was observed in the population (kappa: 0.152).

Conclusion: Edentulous implant patients presented overall lower mortality than expected in the general population during the first year after surgery. However, middle-aged patients showed a higher proportion of deceased patients compared with control people of the same age. Cardiovascular diseases were the cause of death in 50% of the group, and the agreement between presurgical and cause of death diagnoses was poor.