Abstract: AIM To explore variation of P wave dispersion(Pd) and P wave amplitude (Pa) in high altitude heart disease(HAHD)patients and clinical value of the occurrence of atrial arrhytthmia. METHODS Totally 87 patients with high altitude heart disease in our hospital were chosen.According to whether the occurrence of atrial arrhyttmia or not,35 patients with atrial arrhytthmia were regarded as atrial arrhytthmia groupand 52 patients without atrial arrhytthmia were regarded as no-atrial arrhytthmia group. Maximum P wave duration(Pmax) and minimum P wave duration(Pmin) in three groups were measured, then Pd were calculated by the formula(Pd=Pmax-Pmin).Synchronous measurement of Pa were obtained.The heart rate variability(HRV) in the frequency domain was analyzed. Pd,Pa and HRV frequency domain index were respectively compared in three group. Multivariable logistic regression analysis was performed to predict the risk of the occurrence of atrial arrhythmia. RESULTS When compared withno-atrial arrhythmia group, Pd in atrial arrhythmia group were significantly higher (all P<0.05), nevertheless there were no statistical difference in Pa between the aforementioned two groups(P>0.05). Compared with no-atrial arrhythmia group, low frequency(LF) and LF/HF ratio in atrial arrhythmia group significantly increased, while high frequency(HF) obviously decreased (all P<0.05).Multivariable logistic regression analysis revealed that Pd was an independent predictor for atrial arrhythmia. Frequency of atrial arrhythmia. was respectively positive correlated with Pd, LF and LF/HF, nevertheless which was negative correlation with HF. CONCLUSION Increased Pd in HAHD patients is closely related with the occurrence of atrial arrhythmia. sympathetic apparent activation plays a leading role in inducing it, however Pa fails to predict the occurrence of atrial arrhytthmia.