Purpose Total knee arthroplasty (TKA) is frequently performed for obese patients. TKA in this population shows a high rate of complication, particularly tibial component loosening. The aim of this study is to compare the survival rate of tibial components in obese population using TKA with stem versus without stem. Methods From a prospective database of 4216 TKA, obese patients [body mass index (BMI) > 30 kg m(2)] with primary TKA using a tibial short stem extension (30 mm) at a minimum follow-up of 2 years were retrospectively reviewed and compared to a matched control group (1:3 ratio) with a standard tibial stem. Inclusion criteria were BMI > 30 kg m(2), first knee surgery and 24 months minimum of follow-up. The primary outcome was revision for tibial aseptic loosening. Secondary outcomes were all-cause revisions and Knee Society Scores (KSS). Results The final study population consisted of 35 TKA with tibial extension stem versus 105 TKA with standard stem. The mean age was 69.2 and 69.5 years, respectively, with a mean follow-up of 52 months. Both groups were comparable before surgery. After 2 years of follow-up, we observed seven tibial loosening in the group without stem (6.6%) versus no tibial loosening in the stemmed group (p < 0.001). The difference in KSS knee score (83 versus 86; p = 0.06) and the KSS function score (73 versus 77; p = 0.84) were not statistically significant at the final follow-up. Conclusion Using stemmed TKA for obese patients significantly decreased tibial loosening rate at minimum 2 years of follow-up.