Purpose: The purpose of this study was to consider the therapeutic decision whether to use advanced surgery or short implants based on data concerning the use of these implants found in follow-up studies. Materials and Methods: The MEDLINE database was consulted for follow-up studies published between the years 1980 and 2004. For those studies that met the inclusion and exclusion criteria, data concerning the number of implants 7, 8.5, or 10 mm long placed and lost, the time at which the failure occurred, and related risk factors were gathered for 33 studies arranged in tables and subjected to analysis. The studies included 16,344 implant placements with 786 failures (4.8%). Implants were analyzed according to the time of failure (ie, before or after prosthesis seating) and risk factors implicated in the failures. Results: The total rate of failures was 4.8%. Implants 3.75 mm wide and 7 mm long failed at a rate of 9.7%, compared to 6.3% for 3.75 3 10-mm implants. It was found that 54.9% of failures occurred before the prosthesis connection. Finally, 66.7% of all failures were attributed to poor bone quality, 45.4% to the location (maxilla or mandible), 27.2% to occlusal overload, 24.2% to location within the jaw, and 15.1% to infections (an implant could be associated with multiple risk factors). Discussion: The analysis revealed that among the risk factors, poor bone quality in association with short implants seemed to be relevant to failure. The use of implants 4 mm in diameter appeared to minimize failure in these situations. The 3.75 3 7-mm implant presented the highest failure rate (9.7%) of 1,894 implants analyzed (excluding implant designs with higher failure rates but few total implants). Conclusion: Short implants should be considered as an alternative to advanced bone augmentation surgeries, since surgeries can involve higher morbidity, require extended clinical periods, and involve higher costs to the patient. Parole chiave: bone quality, risk factors, short implants