In shortterm analysis of Heart Rate Variability （HRV）, it is hard to get the consistent results. In order to explain this phenomenon, the nonstationary interference on analysis parameters of shortterm HRV has been studied. In virtue of correlation analysis, the results showed that although all the four parameters （HFnorm, LFnorm, SE, BE） considered fluctuate with the lapse of time, only some of them （HFnorm and SE） correlate significantly with nonstationary. Therefore, on the one hand, the consistence of shortterm analysis of HRV is impacted by nonstationary interference, and the equal environmental impacts are prerequisite in collecting HRV data from different subjects to make them comparable. On the other hand, the low frequency fluctuations of HRV contain not only the nonstationary interference, but also the intrinsic cardiac dynamic characteristics. Thus, shortterm analysis methods, which discard absolutely the low frequency components, can not reveal the allsided information of HRV. This is the inherent shortage of shortterm HRV analysis methods, which contributes to the results inconsistence.