Purpose: This study investigated the survival of complete crowns in relation to periodontal variables on a long-term basis. Materials and Methods: A total of 1,037 complete crowns made in an undergraduate clinic for 456 patients were evaluated over an 18-year period. The study population was a mixture of periodontally affected and non-periodontally affected patients, which is comparable to the population group in a private practice. Patients were offered a supportive maintenance program. Periodontal variables were measured, including Plaque Index (PI), bleeding on probing (BOP), probing pocket depth (PPD), and attachment level. The Community Periodontal Index for Treatment Needs (CPITN) was calculated per sextant in a full-mouth assessment. Results: The estimated survival rate was 78% at year 18. For the surviving restorations, the improved PI over time was statistically significant (P = .001). Odds ratios were 1.00 for both PI and BOP. For the CPITN, odds ratios were 3.00 to 3.83. Caries was the most frequent reason for failure, followed by periodontal disease. Conclusion: PI and BOP were not directly related to the frequency of failures. Patients with a high CPITN at baseline had a 3.8 times greater likelihood of losing a complete crown (and abutment tooth) than patients with a low CPITN. The higher failure rate was related not only to periodontal disease, but also to a wide range of biologic and technical problems. In relation to complete crown survival, caution is needed in patients with a high CPITN at baseline. Prosthetic work should be preceded by periodontal examination and prophylactic and periodontal treatment if needed.