Development of an ultrasonic method to detect cervical remodeling in vivo in full-term pregnant women

Abstract

The objective of this study was to determine whether estimates of ultrasonic attenuation could detect changes in the cervix associated with medically induced cervical remodeling. Thirty-six full-term pregnant women underwent two transvaginal ultrasonic examinations separated in time by 12 h to determine cervical attenuation, cervical length and changes thereof. Ultrasonic attenuation and cervical length data were acquired from a zone (Zonare Medical Systems, Mountain View, CA, USA) ultrasound system using a 5-9 MHz endovaginal probe. Cervical attenuation and cervical length significantly decreased in the 12 h between the pre-cervical ripening time point and 12 h later. The mean cervical attenuation was 1.1 ± 0.4 dB/cm-MHz before cervical ripening agents were used and 0.8 ± 0.4 dB/cm-MHz 12 h later (p < 0.0001). The mean cervical length also decreased from 3.1 ± 0.9 cm before the cervical ripening was administered to 2.0 ± 1.1 cm 12 h later (p < 0.0001). Cervical attenuation and cervical length detected changes in cervical remodeling 12 h after cervical ripening administration.

Keywords: Cervical length; Cervical remodeling; Cervical ripening; Ultrasonic attenuation.

Figure 1. Step by step process for estimating cervical attenuation coefficient

(a) IQ data obtained from machine, (b) IQ data converted to RF data, (c) RF data decomposed into regions of interest (ROIs) based on the pulse length and number of echoes. Spectral difference algorithm used; hence proximal and distal windows for each ROI are identified, (d) Spectrograms of proximal and distal windows are calculated, (e) Estimate the attenuation coefficient for the entire image, (f) IQ data converted to B-mode image, (g) User selects the cervical region from the B-mode image, (h) Attenuation coefficients estimated for the selected cervix region.

Figure 2. Scan of the cervix displaying tissue architecture via attenuation mapping

The colored pixels represent attenuation values the stroma of the cervix and the blue pixels in the cervical canal (mucus which is almost totally water) have an attenuation value close to 0, consistent with the attenuation of water.

Figure 3. Thirty-six subjects each scanned twice and approximately 12-hours apart, for which ultrasonic attenuation (Attn) and cervical length (CL) were determined

Attn1 and CL1 denote before a cervical ripening agent was administered, Attn2 and CL2 denote 12 hours after the administration of a cervical ripening agent, and Attn2 – Attn1 and CL2 - CL1 denote the changes between the two attenuation and cervical length outcomes. Fig. 3A and 3C are ordered by increasing Attn1 and CL1. Fig. 3B and 3D segmented by parity into nulliparous and multiparous, and then for each, into their respective ordering by increasing Attn1 and CL1.