Résumé : Background/Objectives: The prevalence of diabetes in very old people is rising sharply worldwide, due not only to obesity, nutritional and sedentary lifestyles, but also to aging per se. Diabetes is associated with a higher incidence of sarcopenia, malnutrition and physical disabilities. However, many age-specific issues in the clinical management of very old diabetic patients remain unstudied. Methods: This is a case–control prospective study including 162 very old hospitalized diabetic patients and 301 controls. We explored the impact of diabetes on the prevalence of sarcopenia according to the EWGSOP2 criteria, using Jamar handgrip to assess muscle strength, BIA-derived fat-free mass index to assess muscle mass, and the timed up and go test to assess physical performance. We also explored factors associated with sarcopenia in both groups in multiple logistic analysis. Results: Mean age was 84.8 ± 6.0 years. We found a prevalence of sarcopenia of 8.0% and 16.7% in the diabetic and the control groups, respectively (p = 0.010). BMI was independently associated with sarcopenia in both groups, explaining 25% of the model in the diabetic group and 33% of the model in the control group. Conclusions: Sarcopenia was less prevalent in diabetic hospitalized older patients than in other patients, indicating that old frail patients are not the same patients as those that are in epidemiological studies on sarcopenia in diabetes. These results should be confirmed in further studies.