Purpose: This study of short-span fixed dental prostheses (Ss-FDPs) versus long-span FDPs (Ls-FDPs) evaluated the long-term efficacy and determined the frequencies and causes of failures. Materials and Methods: A total of 236 Ss-FDPs and 86 Ls-FDPs made in an undergraduate university clinic for 149 and 70 patients, respectively, were evaluated over a 20-year period. Kaplan-Meier analysis with a 95% confidence interval was used to estimate the survival probability. Failures of the FDPs were divided into irreversible (loss of FDPs or finish line involvement) or reversible (FDPs and abutments intact after conservative treatment) complications and into biologic and technical/patient-related failures. Results: The overall survival estimations for Ss-FDPs (70.8%; 95% CI: 63%-79%) and Ls-FDPs (52.8%; 36%-70%) at year 20, were statistically significantly different (P = .030). There was no statistically significant difference (P = .126) for the survival estimations for Ss-FDPs (60.4%; 48%-73%) versus Ls-FDPs (59.0%; 44%-74%) at year 19 in the root-canal treated (RCT) group. For the Ss-FDPs group there was a statistically significant difference (P = .009) between the vital (82.4%; 73%-92%) and RCT (60.4%; 49%-73%) groups at year 20. The reason for failure in the Ss-FDP group was of biologic origin in 55.6% to 66.7% of cases, but for the Ls-FDP group the failures were of technical origin in 56.0% to 84.0% of cases. Conclusion: The survival of Ss-FDPs and Ls-FDPs over a 20-year period was favorable. The overall survival estimation for Ss-FDPs was statistically significantly better than for Ls-FDPs at year 20. The use of an RCT abutment becomes more significant in fixed prosthetic restorations with 4 or more units. Occurrence of a previously reversible complication is a predictive factor for an irreversible complication later on. A reversible complication within the first 2 years for an Ss-FDP will lead to an irreversible complication.