The longest data sets available for estimating thermospheric temperature trends are those from ground-based ionosondes, which often begin during the International Geophysical Year of 1957, close to a solar activity maximum. It is important to investigate inconsistencies in trend estimates from these data sets so that trends can be clearly determined. Here we use selected ionosonde stations to show that one of the most significant factors affecting the trend estimates is the removal of the solar cycle. The stations show trend behavior that is close to the behavior of a theoretical model of damped harmonic oscillation. The ringing features are consistent with the presence of solar cycle residuals from the analysis with an amplitude of 2.5 km. Some stations do not show trend behavior that is close to either the average behavior of the stations studied here or the theoretical model of oscillation. Four European stations (Poitiers, Lannion, Juliusruh, and Slough), three of which are closely located in western Europe, were analyzed with the expectation that their trend should be similar. Only Poitiers and Juliusruh showed an evolution that was close to the average behavior of other stations, while the other two were significantly different. The primary cause of this appears to be changes in the M(3000)F2 parameter and demonstrates the importance of incorporating consistency checks between neighboring ionosondes into global thermospheric trend estimates.