Purpose: This study aimed to examine the characteristics and correlation of knee muscle strength and body composition (fat and muscle mass) among elderly women aged 60–70 years with knee osteoarthritis. The present study hypothesized that the muscle mass and the peak torques of the knee joints were considerably low in the knee osteoarthritis (KOA) group. Methods: A total of 47 elderly women aged 60–70 years were recruited from Yangpu District in Shanghai and assigned to the knee osteoarthritis (n = 25, KOA) or healthy control group (n = 22, CON). The knee extension/flexion isokinetic strength measurements were conducted on an isokinetic dynamometer at angular velocities of 90°/s. Dual-energy X-ray absorptiometry was used to measure the body composition (fat and muscle mass in the whole body and lower limbs). The independent sample t-test was employed to determine the effects of knee osteoarthritis on each variable, and the Pearson correlation analysis was used to investigate the correlation between the body composition and knee muscle strength. Results: Compared with the CON, the KOA exhibited the following: (1) Lower absolute peak knee extension torque (66.02 ± 10.57 vs. 56.61 ± 14.69 Nm), relative peak knee extension (1.11 ± 0.19 vs. 0.89 ± 0.26 Nm/kg), and flexion torque (0.62 ± 0.15 vs. 0.54 ± 0.16 Nm/kg, p < 0.05); (2) greater relative peak torque ratio of the knee extension and flexion (0.55 ± 0.08 vs. 0.62 ± 0.15, p < 0.05); and (3) lower total body muscle mass percentage (63.24% ± 4.50% vs. 59.36% ± 3.94%), particularly in the lower limbs (19.96% ± 1.51% vs. 18.47% ± 1.49%, p < 0.05). Furthermore, the total body fat mass percentage was negatively associated with the relative peak knee extension and flexion torque regardless of the group (p < 0.05). The total body muscle mass percentage was positively associated with the relative peak knee extension torque in the two groups and the relative peak knee flexion torque in the CON (p < 0.05). Conclusion: For elderly women with knee osteoarthritis, knee muscle strength decreases significantly, especially for the extensor strength. Moreover, compared with fat mass, the index of muscle mass is more sensitive in detecting the decrease in knee joint torque. Therefore, rather than weight loss alone, the quadriceps muscle and the rear-thigh muscles, which maintain the stability of knee joints during rehabilitation training, should be strengthened emphatically to improve muscle mass.