Non-invasive magnetomyographic recordings of the uterus to predict labor in term pregnancies


      To predict labor in term pregnancies by using transabdominal magnetomyographic (MMG) recordings of uterine contractions.

      Study design

      72 MMG examinations were conducted serially on ten low-risk patients beginning at 38 weeks until delivery by using a device with 151 magnetic sensors spaced 3 cm apart. The entire maternal abdomen rests against the smooth surface of the array. The clinical staff were blinded to MMG results and there were no elected cesarean deliveries. Two tracking parameters were used, one to quantify the intensity of the magnetic field related to electrophysiological activity (Peak Power [PP]) and the other to determine the spread of activity over the uterus (Percentage of Sensors Active [PSA]).


      Spectral analysis of MMG signals revealed that peak power on 72 recordings ranged from 1 to 10 pT with a frequency of range 0.15-0.2 Hz. PSA was <33% on the first day of recordings, implying that only 1/3 of the recording area was magnetically active over the abdomen. The number of serial recordings per patient ranged from 1 to a maximum of 19. Three patients with less than three serial recordings were excluded from longitudinal analysis; however, all three had PP>5 pT, implying that they were already in preparatory phase of labor. Six of seven patients with increased PP and /or PSA delivered within 4 days of their last examination where a 50% increase in PP or PSA was considered significant. One patient who underwent 19 days of recordings up until 3 days before delivery showed no increase in either PP or PSA and was induced for delivery post-date.


      Magnetomyographic evaluation provides a new non-invasive method for prediction of labor. Further studies are in progress to determine the temporal relationship of changes in MMG with onset of labor.