Quantitative Ultrasound Assessment of the Cervix

By Barbara L. McFarlin PhD CNM RDMS.

Preterm birth is a significant public health problem in the United States. The incidence of preterm birth has risen 16% since 1990 (Hamilton, Martin, & Sutton, 2004) and 29% over the last two decades (Martin et al., 2003). Prevention programs and pharmacologic agents have focused on controlling the active phase of extensive uterine activity, rather than cervical ripening. Presently there is not an objective instrument available to determine cervical ripening, which precedes preterm labor and birth.

For many women, the changes that lead to cervical ripening occur without any noticeable contractions or physical signs, such as vaginal discharge. The cervix ripens from the internal portion ( Fig.1, pink arrow) to the external portion (yellow arrow). The green arrow in Figure 1 indicates the fetal head next to the cervix. Digital examination of the cervix is not sensitive to these changes until the external portion of the cervix is thin and dilated.

Figure 1. Human transvaginal
cervical length ultrasound.

Conventional B-mode (brightness-mode) ultrasound images created with a clinical ultrasound system are created (processed) from radio frequency (RF) echo signals.. The RF echoes are created by reflections from interfaces between acoustically different regions and by incoherent scattering from tissue microstructures (Shung & Thieme, 1993). The RF echoes contain frequency-dependent information about the smaller-scale tissue microstructure (less than the wavelength of sound). Conventional B-mode ultrasound processing removes the frequency-dependent information. B-mode ultrasound images are good at displaying information about larger scale structures (those that are larger than the wavelength). In order to resolve and quantify the smaller-scale tissue microstructure, the frequency-dependent information must be utilized. Thus, the processing of frequency-dependent, or RF, information yields three unique quantified tissue-based parameters: scatterer diameter, scatterer acoustic concentration, and scatterer strength factor. Figure 2 displays an example of quantitative ultrasound output for a day 21 pregnant rat cervix.

Figure 2. Quantitative ultrasound images of a day 21 rat cervix. The image on the left measures
scatterer diameter and the image on the right scatterer strength factor. Each pixel represents a
mean values for that region of interest.

A new ultrasound technology, quantitative ultrasound, examined the microstructure of tissues, was used to determine cervical ripening in one group of non-pregnant rats and five groups of pregnant rats. There were 13 rats in each group (non-pregnant, day 15, 17, 19, 20 and 21 of pregnancy). The results indicated that the ultrasound variables of acoustic concentration of the scatterers, scatterer diameter, and scatterer strength factor were significantly different as pregnancy progressed and were correlated to hydroxyproline in the cervix. The hydroxyproline concentration in the cervix decreased as pregnancy progressed and the hydroxyproline content increased. Quantitative ultrasound was significantly associated with the changes in hydroxyproline concentration and content. Water content of the cervix samples was not significantly different between the groups. Discriminant analysis was used to build a model to predict gestational age from the ultrasound variables.

The results of this study confirmed the following hypotheses of the conceptual framework of this study: (a) the hydroxyproline content ( g hydroxyproline/mg cervix) of the cervix increased as the pregnancy advanced in the rat; (b) the hydroxyproline concentration (% hydroxyproline) decreased as pregnancy advanced; (c) ultrasound acoustic concentration of the scatterers decreased as the pregnancy advanced and (d) hydroxyproline had significant correlations with rat groups and with ultrasound variables of scatterer diameter, acoustic concentration, and scatterer strength factor.

This project was funded by:

American Institute of Ultrasound in Medicine
March of Dimes
National Institutes of Health Individual National Research Service Award F31 NR08191-01A1