Methods
A total of 102 cases of prevention of mother-to-child transmission of HIV in infants, who were treated in the Guangzhou Eighth People′s Hospital from July 2013 to July 2015,were randomly divided into observation group(52 cases) and control group(50 cases).The two groups began to be given formula feeding within 4 hours after birth, and were also given nevirapine suspension(1.5 mL/time, once a day),for antiviral treatment to prevent mother-to-child transmission of HIV. The amount of formula milk would be adjusted individually according to the neonatal feeding tolerance. If the blood sugar was low or energy intake through mouth was insufficient, it was necessary to give intravenous nutrition treatment. The observation group was added the treatment of pressing acupoints during the study period. If newborn babies in observation group had stable vital signs, the treatment of pressing acupoints would begin 24 hours after birth. Acupoints selecting: Zusanli, Zhongwan, Tianshu, Pishu and Weishu. Press time of each acupoint started from 1 minute each time; when newborn babies were adaptive, it would be gradually extended to 2 minutes each time, 2 times a day. Two groups were not given gastrointestinal promoting medicine during the study period. The observation indexes included the incidence of feeding intolerance, the speed to increase feeding quantity, the time for whole gastrointestinal feeding through mouth, and the time of returning to birth weight.

Results
Compared with the control group,the speed to increase feeding quantity was faster, the time for whole gastrointestinal feeding through mouth was shorter, the incidence of feeding intolerance was lower, and the time of returning to birth weight was shorter in the observation group, and these differences were statistically significant(P<0.05). The incidence rate of vomiting, abdominal distension, gastric retention and difficulty in increase of milk volume was lower in the observation group than in the control group, and the difference was statistically significant(P<0.05).