Heatwaves are defined as a period that temperatures exceed a threshold above normal in a given time and region. Most heatwave definitions, such as CTX90, have originated based on stationary assumptions. In this study, we suggest a modification for the CTX90 method and developed a new dynamic heatwave threshold (NS-Q₉₀), based on Generalized Additive Models for Location, Scale and Shape parameters (GMALSS) framework, in order to redefining and detecting heatwaves under nonstationarity conditions of the near-surface temperature. Some major heatwave attributes including frequency, heat-spell, duration, magnitude and hotness were estimated by both the original CTX90 (fixed thresholds) and modified CTX90 (dynamic thresholds) methods and were compared for summertime heatwaves in the Middle-East region. This study revealed that the measurement of heatwave attributes significantly depends on the threshold type and the modified CTX90 method with dynamic thresholds has a better ability to identify heatwave events which provides more realistic information for measuring heatwave attributes than the original CTX90 method. The outputs from the dynamic heatwave thresholds approach revealed that, in the Middle-East, the summertime heatwave attributes had no significant change, except for the hotness attribute that represented a significant increase over the last four decades. Based on the modified CTX90 method it is found that some countries in Middle-East have encountered the hottest heatwaves in the last decades with a hotness more than 45 °C on average. This study for the first time highlighted the significance of applying dynamic thresholds methodology under the existence of new nonstationary normalcy of temperature in the modern-era to detect heatwaves.