Hand, foot, and mouth disease (HFMD) with central nerve system complications may rapidly progress to fulminated cardiorespiratory failure, with higher mortality and worse prognosis. It has been reported that circadian rhythms of heart rate (HR) and respiratory rate are useful in predicting prognosis of severe cardiovascular and neurological diseases. The present study aims to investigate the characteristics of the circadian rhythms of HR, respiratory rate, and temperature in HFMD patients with neurological complications. Hospitalized HFMD patients including 33 common cases (common group), 61 severe cases (severe group), and 9 critical cases (critical group) were contrasted retrospectively. Their HR, respiratory rate, and temperatures were measured every 4 hours during the first 48-hour in the hospital. Data were analyzed with the least-squares fit of a 24-hour cosine function by the single cosinor and population-mean cosinor method. Results of population-mean cosinor analysis demonstrated that the circadian rhythm of HR, respiratory rate, and temperature was present in the common and severe group, but absent in the critical group. The midline-estimating statistic of rhythm (MESOR) (P = 0.016) and acrophase (P < 0.01) of temperature and respiratory rate were significantly different among 3 groups. But no statistical difference of amplitude in temperature and respiratory rate was observed among the 3 groups (P = 0.14). MESOR value of HR (P < 0.001) was significantly different in 3 groups. However, amplitude and acrophase revealed no statistical difference in circadian characteristics of HR among 3 groups. Compared with the common group, the MESOR of temperature and respiratory rate was significantly higher, and acrophase of temperature and respiratory rate was 2 hours ahead in the severe group, critical HFMD patients lost their population-circadian rhythm of temperature, HR, and respiratory rate. The high values of temperature and respiratory rate for the common group were concentrated between 3 and 9 PM, whereas those for the severe group were more dispersive. And the high values for the critical group were equally distributed in 24 hours of the day. Circadian rhythm of patients' temperature in the common group was the same as the normal rhythm of human body temperature. Circadian rhythm of patients' temperature, HR and respiratory rate in 3 groups were significantly different.