Effect of gel-instillation sonography on Doppler ultrasound findings in endometrial polyps

Authors


Abstract

Objective

Saline infusion sonohysterography has been reported to suppress the color signal within the endometrium at color or power Doppler evaluation. The aim of this study was to evaluate if gel-instillation sonography (GIS) affects the power Doppler signal in patients with endometrial polyps.

Methods

Ultrasound volumes of the uterus, obtained by three-dimensional ultrasound imaging of 25 women with histologically confirmed endometrial polyps, were assessed offline by six gynecologists with a special interest in gynecological ultrasound. Each woman contributed four volumes: one gray-scale volume and one power Doppler volume before GIS, and one gray-scale volume and one power Doppler volume at GIS. Power Doppler features before and after gel infusion were compared.

Results

At unenhanced ultrasound a pedicle artery was seen in 27–46% of cases, whereas, after gel infusion the examiners reported a pedicle artery in 30–46% of cases (Exact McNemar's test P-values ranged from 0.50 to 1.00). The level of agreement between unenhanced ultrasound and GIS ranged from 59 to 91% (Cohen's kappa values ranged from 0.17 to 0.79). There was no tendency for a pedicle artery to be identified less often at GIS than before gel instillation.

Conclusion

Gel infusion does not affect the power Doppler signal in patients with endometrial polyps. Copyright © 2011 ISUOG. Published by John Wiley & Sons, Ltd.

INTRODUCTION

Endometrial polyps are common and are easily diagnosed at hysteroscopy. After menopause the prevalence of endometrial polyps is estimated to be about 12–17%, with similar percentages in women presenting with abnormal bleeding and in asymptomatic women1–3. Endometrial polyps are often suspected at transvaginal ultrasound examination: typically the interface between the endometrium and the polyp is highly echogenic, and often small cysts are visible inside the polyp4. At color Doppler imaging the presence of a distinct vascular pedicle inside the endometrium is highly suggestive for a polyp5, 6. For detection of endometrial polyps the diagnostic accuracy of fluid instillation sonography is similar to that of hysteroscopy7.

Most endometrial polyps are benign: only 1% of polyps contain a malignancy1, 8. The influence of age, menopausal status, the presence of hypertension8–11 and polyp size are inconsistently reported as risk factors1, 8–13. Because neoangiogenesis is a typical feature of cancer, evaluation of the vascularity of polyp-like lesions in the uterine cavity by color Doppler imaging might be relevant when assessing the risk of malignancy14–16. However, saline contrast sonohysterography has been found to suppress the color signal within the endometrium at color Doppler examination17.

The aim of this study was to evaluate if gel-instillation sonography (GIS) influences the power Doppler signal in patients with endometrial polyps.

METHODS

A total of 111 consecutive women presenting at the ‘Bleeding Clinic’ of the University Hospital of Leuven underwent two-dimensional (2D) and three-dimensional (3D) transvaginal gray-scale and power Doppler ultrasound examination of the uterus and endometrium with and without enhancement with gel instillation. GIS was performed using a 2-mm-diameter neonatal suction catheter through which about 2–5 mL of gel warmed to 37 °C (Endosgel® or Instillagel®; Farco-Pharma GmbH, Köln, Germany) was instilled, as previously described18. The ultrasound examinations were carried out using a GE Voluson E8 ultrasound system (GE Healthcare Ultrasound, Milwaukee, WI, USA) equipped with a 6–12-MHz 3D endovaginal probe. From each woman four ultrasound volumes of the uterus were acquired and stored: one gray-scale volume and one power Doppler volume before gel instillation, and one gray-scale volume and one power Doppler volume after gel instillation. All volumes were acquired starting with a sagittal section of the uterus. The volume angle was 120°. The power Doppler settings were as follows: power, 100%; gain, − 5.8; frequency, low; quality, normal; wall motion filter (WMF), low 1 (30 Hz); pulse repetition frequency (PRF), 0.3 kHz. The 3D volumes were saved from the ultrasound machine onto an external hard disk. At the start of the study, all volumes were copied onto DVD and sent to the external examiners by mail (seven DVDs per examiner). During the copy process, data from a few random cases were incidentally lost in some of the DVDs.

Six independent experienced ultrasound examiners (T.Bi., G.C., E.E., F.L., A.T. and D.V.) were asked to evaluate the ultrasound volumes and to report—amongst other information—the presence or absence of a pedicle artery5 at power Doppler imaging before and at GIS. According to Timmerman's definition for real-time 2D examination, the pedicle artery is a distinct vascular pedicle running from the myometrium into the endometrial lining5. Because of its tortuosity, a pedicle artery often cannot be visualized over its whole length in a single plane. During real-time 2D examination, the sonographer can place the probe at the optimal angle and wait for the optimal time to obtain a perfect image of the pedicle artery over its whole length. This is not always achievable using offline 3D examination. The visualization of a distinct color signal in the middle of the endometrium at offline 3D examination was therefore considered as sufficiently indicative of the presence of a pedicle artery. The examiners were asked if there was a significant color signal inside the endometrium or the polyp at unenhanced ultrasound and at GIS. This is referred to as a ‘pedicle artery’ in the present paper. The examiners knew the woman's age and menopausal status but did not have access to any other clinical information. The examiners were blinded to the histological results. They were not blinded between the different volumes of the same patient.

Twenty-five of the 111 women had a histologically confirmed endometrial polyp and are included in our analysis.

We calculated the percentage agreement between examinations performed before and at GIS with regard to identifying a pedicle artery, as well as the corresponding Cohen's kappa value. Agreement was considered poor if κ≤0.20, fair if 0.21≤κ≤0.40, moderate if 0.41≤κ≤0.60 and substantial if 0.61≤κ≤0.8019. These guidelines are, however, arbitrary. Marginal homogeneity between data obtained before and at GIS was determined using the Exact McNemar's test. Statistical analysis was performed using SAS 9.2 (SAS Institute Inc., Cary, NC, USA).

RESULTS

The mean (SD) age and age range of the subjects was 49.1 (10.4) years (range, 25–86 years) in the total group and 50.6 (7.8) years (range, 35–68 years) in the polyp group. The mean (SD) weight and weight range of the patients was 71.6 (16.1) kg (range, 42–153 kg) in the total group and 74.7 (15.6) kg (range, 42–111 kg) in the polyp group. Thirty percent of patients in the total group were premenopausal vs. 31% in the polyp group. The median parity was 2 (range, 0–10) in the total group and 2 (range, 0–4) in the polyp group.

Our results are presented in Figure 1. At unenhanced ultrasound a pedicle artery was seen in 27–46% of cases: in 44, 36, 46, 39, 27 and 45% of cases by Observers 1, 2, 3, 4, 5 and 6, respectively. After gel infusion, the offline examiners reported a pedicle artery in 30–46% of cases: in 44, 32, 46, 30, 36 and 41% cases by Observers 1, 2, 3, 4, 5 and 6, respectively. The agreement between unenhanced ultrasound and GIS for each observer with regard to the presence of a pedicle artery ranged from 59 to 91%: 84% (Cohen's kappa (95% CI) = 0.64 (0.38–0.97); Exact McNemar's test, P = 1.00), 80% (0.56 (0.21–0.90); P = 1.00), 75% (0.50 (0.15–0.85); P = 1.00), 83% (0.62 (0.28–0.96); P = 0.63), 91% (0.79 (0.52–1); P = 0.50) and 59% (0.17 (0–0.59); P = 1.00) for Observers 1, 2, 3, 4, 5 and 6, respectively. There was no tendency for a pedicle artery to be identified less often at GIS than before gel instillation.

Figure 1.

Agreement between power Doppler ultrasound findings before and at gel-instillation sonography (GIS), for six different observers. Pedicle −, absence of pedicle artery; Pedicle +, presence of pedicle artery.

DISCUSSION

Our results suggest that GIS does not affect the power Doppler signal in patients with endometrial polyps. GIS with color imaging may therefore reliably be used in future studies aiming, for example, to differentiate between benign and malignant polyps or between asymptomatic polyps and polyps likely to cause bleeding. Such studies would have been precluded if our data had demonstrated that the color signal is suppressed by gel instillation.

Some of the differences between unenhanced ultrasound and GIS (Figures 2–4) might be explained by transient changes in the flow in the pedicle artery (for example because of myometrial contractions) because the unenhanced and GIS volumes were acquired a few minutes apart. Differences might also be explained by the endometrium not always being easily delineated at unenhanced ultrasound, and therefore a flow signal just outside the endometrium might be more difficult to differentiate from a true pedicle artery at unenhanced ultrasound than at GIS. Artifacts leading to false-positive findings of a pedicle artery may also explain differences in findings before and at GIS. Although the examiners were unaware of the histology results, lack of blinding to the different volumes in the same patient might have had some influence on the results of the intraobserver agreement.

Figure 2.

Agreement before (a) and after (b) gel-instillation sonography regarding the presence of a pedicle artery.

Figure 3.

Disagreement before and after gel-instillation sonography (GIS) regarding the presence of a pedicle artery: at unenhanced ultrasound a pedicle artery was visible (a), whereas at GIS the polyp was visualized but not its vascular pedicle (b).

Figure 4.

Disagreement before and after gel-instillation sonography (GIS) regarding the presence of a pedicle artery: no pedicle artery was seen at unenhanced ultrasound (a), whereas at GIS there was a color signal visible within the polyp (b).

Gel has recently been proposed as an alternative to saline for fluid instillation sonography18, 20. Because gel has a high viscosity, GIS is associated with less backflow through the cervix and less transtubal flow than is saline contrast sonohysterography. This results in a more stable filling of the uterine cavity. The present study, using gel, could not confirm the impression, by Epstein and Valentin17, that color Doppler signals in focal lesions in the uterine cavity tended to disappear during saline contrast sonohysterography. In the present study we used power Doppler imaging because it is less influenced by the insonation angle and therefore more sensitive to flow in small tortuous vessels.

Differences in the contrast medium (gel or saline), in the temperature of the contrast medium or in the catheter used might explain the discrepancy. We used gel at 37 °C, while saline at room temperature might have caused vasoconstriction or myometrial contractions. However, in a study using gel at about 4 °C, a pedicle artery was detected in 41% of polyps21. During GIS a small instillation volume of about 3 mL is usually sufficient to obtain an acceptable image. During saline contrast sonohysterography, higher instillation volumes are necessary to compensate for backflow. These fluctuations in the instillation rate might lead to unstable intracavitary pressures, potentially impeding flow in small vessels. In the present study, we used a neonatal suction catheter of 2-mm diameter without a balloon. Epstein and Valentin17 also did not use a balloon catheter. The use of a balloon catheter preventing backflow results in higher intracavitary pressures and is therefore likely to affect blood flow in a polyp more than a catheter without a balloon. Transient myometrial contractions might affect the flow in feeding vessels of polyps, and so Doppler ultrasound findings might vary over time. Depending on the examiner, a pedicle artery was identified in 27–46% of polyps at unenhanced ultrasound examination in our study. In the initial report by Timmerman et al.5, 76% of polyps manifested a pedicle artery. The discrepancy might be explained by differences in the Doppler sensitivity of the ultrasound systems used and in ultrasound technique. First, in the present study we used a GE Voluson E8 whereas Timmerman et al. used an Acuson Sequoia ultrasound system, which produces a superior color Doppler image. Second, in the present study a large volume including the whole uterus was acquired, whereas in the study of Timmerman et al., the color-box was confined to the endometrium—a small color box giving a better detection of small vessels. Finally, Timmerman et al. used real-time ultrasound examination, whereas we analyzed 3D volumes. Flow in pedicle arteries is more likely to be detectable during a longer real-time examination than during a short acquisition of a volume, if we assume intermittent changes in flow, as a result of myometrial contractions or fluctuating blood pressure.

There was substantial variability in the detection rate of a pedicle artery in endometrial polyps between our observers. Five out of six observers showed moderate to substantial agreement, while one had a low kappa value. The interobserver variability may be influenced by a number of factors, including the level of expertise, the understanding of the ultrasound definitions, the duration of examination per case or technical limitations such PC-screen resolution. Better agreement before the start of the study, achieved through teaching and consensus sessions, might reduce interobserver variability. However, the strength of our study was that the same cases were examined independently by six different examiners who were blinded to the histology results. The fact that a kappa of > 0.50 was obtained for the large majority of examiners makes it reasonable to conclude that color signals within the endometrium are not suppressed by gel infusion.

To sum up, our study suggests that gel infusion does not affect the power Doppler signal in patients with endometrial polyps. Gel may therefore become the distension medium of choice in further studies on endometrial vascularity aiming to differentiate between benign and malignant polyps or between asymptomatic polyps and polyps likely to cause bleeding.

Ancillary