Objective: To compare the cyclic fatigue resistance of nickel-titanium files made by 3 new heat treatment in simulated S-shaped root canals at different temperatures. Methods: Gold heat-treated nickel-titanium files TruNatomy (25 mm, tip size 26#/0.04) and ProTaper Gold (25 mm, tip size 25#/0.08) were selected as the experimental group, M wire technique nickel-titanium file ProTaper Next (25 mm, tip size 25#/0.06) was selected as the control group. It was speculated that the Gold technique used in TruNatomy nickel-titanium file was R phase separation technique, which included a complete intermediate R-phase, increasing its flexibility. ProTaper Gold was a CM wire nickel-titanium file and the increased phase transformation temperature by heat treatment introduced martensite at room temperature, while it underwent gold heat treatment on the surface, generating an intermediate R phase during phase transformation, providing hyperelastic. ProTaper Next used M wire technique, M wire included austenite at room temperature, where heat mechanical processing introduced hardened martensite, which was incapable of participating phase transformation. Because of the lower elastic modulus of hardened martensite than austenite, the flexibility of the file was increased. Twenty instruments of each nickel-titanium file were submitted to the cyclic fatigue test by using a simulated canal with double curvatures at room tem-perature (24 ℃) and 65 ℃, 10 instruments of each nickel-titanium file were selected at each temperature (n=10). At the same temperature, the number of cyclic fatigue (NCF) and fragment length were analyzed by using One-Way analysis of variance at a significance level of P < 0.05. NCF and fragment length of the same nickel-titanium file at room temperature and 65 ℃ were compared by paired sample t test and the significance level was α=0.05. Fractured surfaces were analyzed by using scanning electron microscope. Results: In double-curved canals, all the failure of the files due to cyclic fatigue was first seen in the apical curvature before the coronal curvature. At room temperature, in the apical curvature, NCF of TruNatomy was 344.4±96.6, ProTaper Gold was 175.0±56.1, ProTaper Next was 133.3±39.7, NCF of Tru Natomy was the highest (P < 0.05). In the coronal curvature, NCF of TruNatomy was 618.3± 75.3, ProTaper Gold was 327.5±111.8, ProTaper Next was 376.6±67.9, NCF of TruNatomy was also the highest (P < 0.05). There was no significant difference among the apical and coronal fragment length of the 3 nickel-titanium files (P>0.05). At 65 ℃, in the apical curvature, NCF of TruNatomy was 289.6±65.8, ProTaper Gold was 187.5±75.4, ProTaper Next was 103.0±38.5, NCF of TruNatomy was the highest (P < 0.05). In the coronal curvature, NCF of TruNatomy was 454.2±45.4, ProTaper Gold was 268.3±31.4, ProTaper Next was 283.8±31.7, NCF of TruNatomy was also the highest (P < 0.05). The apical fragment length of ProTaper Next was the highest (P < 0.05), and there was no significant difference among coronal fragment length of the 3 nickel-titanium files (P>0.05). Compared with room temperature, at 65 ℃, in the coronal curvature, NCF of TruNatomy decreased significantly (P < 0.05). The fractured surfaces of the three nickel-titanium files demonstrated typical cyclic fatigue. Conclusion: Gold heat-treated nickel-titanium file had better cyclic fatigue resistance than M wire nickel-titanium file in S-shaped root canals.