Crosssectional imaging

Natural Female Hormone Balance Program

Natural Menopause Relief Secrets

Get Instant Access

Jeremiah Healy Simon Padley

Modern pelvic imaging is largely achieved using the cross-sectional modalities of ultrasonography, computed tomography and magnetic resonance imaging. These techniques can all elegantly display pelvic anatomy but have different strengths and weaknesses for the evaluation of gynaecological malignancies. Imaging may be used at presentation for diagnosis, in the assessment of disease extent prior to operation, and postoperatively for residual or recurrent disease evaluation. Imaging is also central to the assessment of response to adjuvant chemotherapy.

Carcinoma of the endometrium

In approximately 90% of cases, endometrial cancer is well-differentiated adenocarcinoma which typically presents with postmenopausal bleeding. Histopathological diagnosis is usually confirmed by hysteroscopy and curettage, which despite a small false negative rate of 2-6% remains the 'gold standard' for differentiating benign from malignant endometrium. At presentation 75% of women have tumour confined to the endometrium (stage IA) for whom the 5-year survival rate is 80% following total hysterectomy (Creasman et al., 1987). The International Federation of Obstetrics and Gynaecology (FIGO) committee recommends staging by total hysterectomy, bilateral salpingo-oophorectomy and lymphadenectomy.

Role of imaging

Imaging aids detection of endometrial carcinoma and helps select those postmenopausal women with bleeding who need dilatation and curettage for histopathological diagnosis. It can provide preoperative information about depth of myometrial invasion, an important prognostic indicator. In endometrial carcinoma the incidence of nodal metastases increases from 3% for stage IB tumours (invasion of less than half of the myometrium) to 40% for stage IC tumours (invasion of more than one half of the myometrium).

Imaging also allows assessment of the local extent of the tumour, such as cervical involvement (stage II) and local spread (stage III). Imaging can demonstrate enlarged pelvic and para-aortic lymph nodes which may assist surgical planning or identify patients with advanced disease (stage IV) in whom nonsurgical treatment may be more appropriate. Imaging is particularly useful in patients unfit for surgery or in whom clinical examination is difficult.

Pelvic Imaging Measurement

Transvaginal ultrasound. Normal hyperreflective endometrial appearances during the proliferate phase of the cycle. The endometrium is homogeneous in reflectivity and clearly delineated in outline

Figure 1

Transvaginal ultrasound. Normal hyperreflective endometrial appearances during the proliferate phase of the cycle. The endometrium is homogeneous in reflectivity and clearly delineated in outline

Ultrasonography

An office endometrial biopsy is the initial investigation of postmenopausal bleeding; pelvic ultrasonography may be used to select patients in whom hysteroscopy, dilatation and curettage are indicated. The normal endometrium is visualized as a highly reflective band in the centre of the uterus. Transvaginal sonography allows more precise measurement of endometrial thickness as the higher-frequency ultrasound probe gives images of improved spatial resolution.

Normal endometrial thickness and appearance vary with hormonal status (prepubertal, menstrual age or postmenopausal) and phase of the menstrual cycle. Early in the menstrual cycle the endometrium is visualized as a thin reflective line. In the proliferative phase the endometrium appears as a triple line (Figure 1), and in the secretory phase of the endometrium is at its greatest thickness, with homogeneously increased reflectivity and greater through transmission.

In women with postmenopausal bleeding a thin endometrium measuring 4 mm or less obviates the need for endometrial biopsy. Endometrial carcinoma is characterized by increased endometrial thickness, often with heterogeneous reflectivity and an irregular, ill-defined margin. However, there is overlap between the sonographic appearances of endometrial polyps, hyperplasia (associated with hormone replacement and tamoxifen therapy) and carcinoma in postmenopausal women with bleeding.

Transabdominal sonography is considered unreliable for staging endometrial carcinoma, but the reported accuracy of transvaginal ultrasound for assessing the depth of myometrial invasion ranges from 78% to 87%. However, overstaging is well recognized if the tumour is polypoid or when blood clot distends the endometrial cavity. The evidence for the benefit of Doppler ultrasonography in detecting and locally staging endometrial carcinoma is equivocal. Reports suggest no significant differences in Doppler flow measurements between benign and malignant uterine tumours.

Sonohysterography (transvaginal ultrasound scanning during the instillation of saline into the uterine cavity) can be helpful in further evaluating focal endometrial thickening and also appears to be very sensitive in identifying significant uterine pathology. Tumour spread beyond the uterus, especially nodal status, cannot be precisely evaluated using transvaginal ultrasound, which is a limitation in its overall usefulness in staging endometrial carcinoma.

From a recent meta-analysis transvaginal sonography appears to be very sensitive at detecting endometrial carcinoma (94.3%) but has a poor specificity (52.4%), giving it an overall accuracy of only 57%. However, it does have a very high negative predictive value (98.6%) (Timmermann et al., 1997).

Computed tomography

Dynamic contrast-enhanced computed tomography (CT) may show endometrial tumours as hypodense masses in the endometrial cavity or myometrium. In locally advanced disease it may show an eccentric mass extending into the parametrial tissues. Occasionally the uterus may be fluid-filled secondary to tumour obstruction of the endocervical canal or owing to tumour necrosis. However, CT is not as good as transvaginal sonography or magnetic resonance imaging (MRI) for staging early disease, with a staging accuracy of 76% for stage I and stage II disease.

The strength of CT lies in its ability to evaluate more advanced disease where there is parametrial or pelvic side-wall tumour spread. It is also useful in identifying both local and retroperitoneal lymph node enlargement, which if present should encourage lymphadenectomy at surgery. It may also identify distant metastases to the lungs, liver, bone and brain, which will aid in the planning of appropriate therapy.

Magnetic resonance imaging

On T2-weighted imaging the uterus has a clearly delineated zonal anatomy which provides a useful guide in staging the depth of myometrial invasion in endometrial carcinoma (Figure 2). In the normal patient the endometrium has a high signal intensity and is separated from the peripheral myometrium by the inner myometrium or junctional zone which has a very low signal intensity because of its low water content and scanty extracellular matrix. The outermost layer identified on MRI, the peripheral myometrium, has a moderately high signal intensity which is slightly greater than that of striated muscle.

On T2-weighted sequences endometrial carcinoma has a relatively high signal intensity, which widens the endometrial canal or causes its signal intensity to be more heterogeneous. These changes may be the only abnormality in stage IA carcinoma. On ^-weighted sequences the signal intensity of the endometrial carcinoma is usually equal to (or less than) that of myometrium, and the tumour may be difficult to delineate (Figure 3). On ^-weighted sequences the normal low signal of the junctional zone is replaced by hyperintense tumour in stage IB tumours (Figure 4). In stage IC tumours there is extension through the junctional zone into the outer half of the myometrium (Figure 5).

On T2-weighted imaging the cervix also has a zonal anatomy, with high-signal central secretions, an inner low-signal band and an outer intermediate-signal band (see Figure 3). Disruption of the normal appearance allows identification of high-signal tumour extension into the cervical stroma, indicating stage II disease.

The accuracy for T2-weighted sequences in assessing depth of myometrial invasion has been reported to be 68-85%. Overestimation of myometrial invasion is more common than its underestimation, especially in bulky tumours within a small uterus where the zonal anatomy is distorted, or if the junctional zone is atrophied, which commonly occurs in the menopause.

Saggital Tampon Vaginal Caviti Mri

Sagittal T2-weighted MR scan through a normal uterus. The normal signal characteristics of the uterus are demonstrated, with a relatively increased signal in the peripheral myometrium, a reduced signal from the junctional zone or inner myometrium and a high signal from the endometrium. The endometrial high signal continues into the cervical canal. The body of the cervix is generally of homogeneously low signal intensity owing to the lower water content. B, urinary bladder; M, small bowel mesentery

Figure 2

Sagittal T2-weighted MR scan through a normal uterus. The normal signal characteristics of the uterus are demonstrated, with a relatively increased signal in the peripheral myometrium, a reduced signal from the junctional zone or inner myometrium and a high signal from the endometrium. The endometrial high signal continues into the cervical canal. The body of the cervix is generally of homogeneously low signal intensity owing to the lower water content. B, urinary bladder; M, small bowel mesentery

Contrast-enhanced T1-weighted images of the uterus improve the accuracy of assessment of depth invasion because endometrial carcinoma enhances to a lesser degree than the surrounding myometrium. This also aids detection of extension into the cervix and spread into the parametrial tissues.

Dynamic contrast enhancement, with rapid acquisition of images after a bolus injection of gadolinium DTPA, assists myometrial staging by the identification of a subendometrial zone. The region enhances before the rest of the myometrium and corresponds histologically to the innermost part of the junctional zone (see Figure 4). Overall the sensitivity and accuracy of MRI for detecting deep myometrial invasion are significantly higher than transvaginal sonography or CT, ranging from 82% to 94%.

With the advent of rapid imaging sequences that allow acquisition of information during a single breath-hold, the ability of MRI to detect more advanced disease in the retroperitoneum and liver is improving. However, older studies report a poor sensitivity (17%) and positive predictive value (50%) for MRI detection of advanced disease.

Uterine sarcoma

Uterine sarcoma is a rare tumour, accounting for only 2-6% of gynaecological malignancies. Uterine sarcomas may be broadly divided into three subtypes. The first is the mixed mullerian sarcoma, which occurs most frequently in postmenopausal women. The other two types are endometrial stromal sarcomas and leiomyosarcomas, which usually present in younger women. Mixed mullerian sarcomas tend to be bulky, arising centrally within the uterus, and are frequently associated with necrosis (Figure 6). However, they have no imaging characteristics that aid their differentiation from endometrial carcinoma.

Ovaran Cancer Mri

Axial Tj-weighted gradient echo MR scan through the body of the uterus. The signal from fat has been suppressed. There is an endometrial carcinoma bulging into the endometrial cavity (Ca). More posteriorly a low signal intensity uterine fibroid is present (arrows)

Leiomyosarcomas also tend to be bulky tumours indistinguishable from benign fibroids on imaging. Malignancy should be suspected if the tumour shows marked contour irregularity, significant necrosis or evidence of metastatic disease.

Figure 3

Axial Tj-weighted gradient echo MR scan through the body of the uterus. The signal from fat has been suppressed. There is an endometrial carcinoma bulging into the endometrial cavity (Ca). More posteriorly a low signal intensity uterine fibroid is present (arrows)

Leiomyosarcomas also tend to be bulky tumours indistinguishable from benign fibroids on imaging. Malignancy should be suspected if the tumour shows marked contour irregularity, significant necrosis or evidence of metastatic disease.

Carcinoma of the cervix

Invasive carcinoma of the cervix is most frequently squamous in histologic type (80-90% of cases), developing at the squamocolumnar junction. Adenocarcinoma and adenosquamous carcinomas tend to arise deeper in the endocervical canal and may remain occult until they are advanced. Histological examination establishes the diagnosis and determines whether disease is preinvasive (cervical intraepithelial neoplasia) or invasive.

Role of imaging

The FIGO committee recommends a clinical staging system with vaginal examination under anaesthesia, even though this strategy has obvious limitations in advanced disease. Although not officially part of the FIGO system of staging, cross-sectional imaging techniques are increasingly used to assess volume of the disease, parametrial spread (stage IIB, IIIB disease) and lymph node enlargement. More than 90% of patients with stage IB 1 tumours (< 4 cm) and disease-free lymph nodes will be cured. In patients with stage IB2 disease (lesions more than or equal to 4 cm in diameter) the prognosis is halved if lymph node involvement is present.

Stage IB endometrial carcinoma. (A) Sagittal T2-weighted image demonstrates an endometrial carcinoma replacing the normal low-intensity junctional zone and bulging into the endometrial cavity. (B) The post-gadolinium contrast-enhanced T1-weighted sagittal image demonstrates an intact and normally enhancing junctional zone (arrow), indicating that the tumour is confined to the inner myometrium without extension into the outer myometrium

Demonstration of lymphadenopathy may also modify treatment plans by directing the extent of lymphadenectomy Patients with parametrial spread (stage IIB, IIIB disease) may be treated with radiotherapy rather than surgery, or be treated with chemotherapy to achieve downstaging prior to surgery

Figure 4

Stage IB endometrial carcinoma. (A) Sagittal T2-weighted image demonstrates an endometrial carcinoma replacing the normal low-intensity junctional zone and bulging into the endometrial cavity. (B) The post-gadolinium contrast-enhanced T1-weighted sagittal image demonstrates an intact and normally enhancing junctional zone (arrow), indicating that the tumour is confined to the inner myometrium without extension into the outer myometrium

Demonstration of lymphadenopathy may also modify treatment plans by directing the extent of lymphadenectomy Patients with parametrial spread (stage IIB, IIIB disease) may be treated with radiotherapy rather than surgery, or be treated with chemotherapy to achieve downstaging prior to surgery

Ultrasonography

Cervical tumours are usually hypoechoic on transvaginal and transrectal ultrasound scans, which are useful for assessing tumour size and determining infiltration into cervical tissues. Some reports suggest that local staging of early cervical cancer is equivalent to MRI and better than spiral CT (Yang et al., 1996).

Parametrial extension is suggested when there is soft tissue stranding laterally from the tumour, but because of their limited focal range neither transvaginal nor transrectal techniques can evaluate parametrial spread in bulky tumours or detect pelvic lymphadenopathy.

Computed tomography

Computed tomography has no role in the early stages of cervical cancer in terms of local spread. Local tumour may be identified on dynamic post-contrast images as an area of low attenuation in normally enhancing cervical stroma. A CT scan may identify parametrial spread as poor definition of the cervix with the surrounding fat or soft tissue stranding in the parametrial fat. False positive diagnoses are frequent, because hyperaemia and pericervical inflammatory changes (which are very common) can have similar appearances. Reliance on CT therefore often leads to overestimates of the extent of local disease, resulting in varying reports of accuracy in local staging of disease of 58% to 88%. More reliable signs of local spread include obliteration of the pericervical fat planes or an eccentric parametrial soft tissue mass (Vick et al., 1984).

Stage IC endometrial carcinoma. Tumour fills the endometrial cavity and extends into the outer myometrium, seen on a sagittal r2-weighted image

Figure 5

Stage IC endometrial carcinoma. Tumour fills the endometrial cavity and extends into the outer myometrium, seen on a sagittal r2-weighted image

More recent work suggests that helical CT may improve the differentiation between normal parametrial vessels and ligaments from adenopathy and tumour extention. Pelvic side-wall extension is characterized by tumour reaching the obturator internus or piriformis muscles. Bladder and rectal involvement is suggested if there is loss of the fat plane and asymmetric wall thickening of the adjacent viscera, nodular thickenings or contiguous masses extending into these organs, or by the demonstration of a fistula (Figure 7). Overall accuracy of diagnosing tumour extension to the pelvic side wall is over 90%.

Computed tomography will identify more distant spread by revealing retroperitoneal lymph node enlargement, hydronephrosis and liver metastases, which are infrequently present at the time of presentation.

Recurrent disease is also accurately assessed by CT, which has a high sensitivity and specificity for detecting recurrent pelvic tumour or pelvic and para-aortic lymph node enlargement. However, CT is not accurate in the differentiation of radiation fibrosis from postsurgical change or from recurrent disease. Vesicovaginal and rectovaginal fistulas can also be difficult to delinate on CT.

Magnetic resonance imaging

Magnetic resonance imaging, like CT, is not part of the formal FIGO staging system. However, it is frequently able to make a significant contribution to the staging of cervical cancer. This is especially so in bulky tumours, endocervical tumours, or tumours difficult to assess on examination under anaesthesia. In these cases MRI will aid evaluation of local tumour extent and can identify para-aortic and pelvic lymph node enlargement.

Invasive cervical carcinoma that is less than stage IB is not usually identified on MRI, but the site of the cervical biopsy is usually easily identified because of oedema and blood products. Tumours are usually of

Cervix Foley

Figure 6

Sagittal T2-weighted image through the pelvis. There is a Foley catheter (F) and a tampon (t) in situ. A large mixed signal intensity tumour replaces the normal uterus; at surgery this proved to be a partially necrotic uterine sarcoma intermediate or high signal intensity on T2-weighted images, surrounded by the very low signal intensity of the cervical stroma (Figure 8). Preservation of this low-intensity rim around the tumour on T2-weighted imaging and a smooth interface with the surrounding parametrium (Figure 9) are reliable signs of tumour confined to the cervix (stage IB).

Extension into the vagina is readily identified on sagittal or coronal MRI; it may be into the upper third (stage IIA) or the lower third (stage IIIA). The accuracy of MRI for assessing vaginal extension is reported to be 72-93%.

Extension of tumour through the cervical stroma and into the surrounding parametrial fat is best seen on Trweighted images (stage IIB) (Figure 10). Magnetic resonance imaging will detect 95% of stage IB tumours and has an overall accuracy of detecting parametrial invasion of 70-90%. Similarly, stage IIIB tumour, which extends to the muscles of the pelvic side wall or obstructs the lower end of the uterus, is clearly seen on T1- and T2-weighted imaging. The use of MRI also allows detection of stage IVA disease by demonstration of extension into the bladder or rectum (Figure 11). Obliteration of the intervening fat planes can be identified on T1-weighted images and disruption of the low signal intensity bladder or rectal wall can be identified on T2-weighted images. Lymph node enlargement is best detected on T1-weighted images, which can be achieved using breath-hold axial T1-weighted images or coronal T1-weighted images to survey the retroperitoneum. Magnetic resonance imaging performs similarly to CT in its ability to detect lymph node metastases, with accuracy ranging from 76% to 88%.

Like CT, MRI can be used to detect recurrent disease, but it is not always possible confidently to distinguish recurrent disease from postsurgical and radiotherapy changes which can appear as high-signal areas on T2-weighted imaging for a prolonged time (Figure 12).

Figure 7

Computed tomography. (A) Contrast-enhanced scan through the cervix. There is bilateral parametrial invasion and thickening of the posterior bladder wall due to direct extension of carcinoma of the cervix into the adjacent tissue. Note the soft tissue density stranding in the paracervical fat, best seen on the right (arrows). (B) Scan through the level of the kidneys demonstrates partial obstruction of the right kidney and complete obstruction of the left kidney with bilateral hydronephrosis. No contrast has passed into the collecting system on the left because of the greater degree of obstruction (stage III)

Dynamic contrast-enhanced MRI with quantitative analysis of the time-enhancement curves may allow separation of recurrent disease from irradiated tissue. There are reports that initial staging of cervical cancer

Iib Cervical Mri

Stage I carcinoma of the cervix. (A) The axial T2-weighted image through the cervix demonstrates the relatively highsignal carcinoma (curved arrow) surrounded by the low-signal normal cervical stroma. Normal high signal is demonstrated in the cervical canal (straight arrow). A nabothian follicle is also present as a round area of high intensity adjacent to the canal. (B) Sagittal image in the same patient. The normal cervical canal is demonstrated (short arrow) with a small carcinoma in the anterior lip of the cervix (curved arrows) may be improved using this technique (Yamashita et al., 1996).

Magnetic resonance imaging is very good for the evaluation of vaginal fistulas (Figure 13), and furthermore can be used to distinguish fistulas associated with recurrent disease from those that are a complication of surgery or radiotherapy (Healy et al., 1996).

Figure 8

Stage I carcinoma of the cervix. (A) The axial T2-weighted image through the cervix demonstrates the relatively highsignal carcinoma (curved arrow) surrounded by the low-signal normal cervical stroma. Normal high signal is demonstrated in the cervical canal (straight arrow). A nabothian follicle is also present as a round area of high intensity adjacent to the canal. (B) Sagittal image in the same patient. The normal cervical canal is demonstrated (short arrow) with a small carcinoma in the anterior lip of the cervix (curved arrows) may be improved using this technique (Yamashita et al., 1996).

Magnetic resonance imaging is very good for the evaluation of vaginal fistulas (Figure 13), and furthermore can be used to distinguish fistulas associated with recurrent disease from those that are a complication of surgery or radiotherapy (Healy et al., 1996).

Gestational trophoblastic disease

Gestational trophoblastic disease is caused by abnormal proliferation of trophoblastic elements in the fertilized ovum, which has the potential for malignant transformation. Complete hydatidiform mole is the most common subtype of this disease, accounting for 80% of cases. The uterine cavity becomes distended by a grape-like proliferation of chorionic villi. Sonographically the uterus contains a complex multicystic/solid mass filling the uterine cavity, with anechoic spaces reflecting hydropic swelling of the villi. The process can be invasive with penetration of the uterine myometrium by trophoblastic elements. Malignant transformation into choriocarcinoma can also occur.

Ultrasonography cannot reliably distinguish between cpmplex hydatidiform mole confined to the uterus and invasive forms of the disease, such as choriocarcinoma, but can sometimes show myometrial invasion. Computed tomography and MRI are of limited value in routine management, but do have a role in demonstrating metastatic disease to the lungs or central nervous system (Newlands et al., 1995).

Surgical management is discussed in Chapter 13.

Uterine Carcinoma Mri

Carcinoma of the cervix (stage IBii). There is extension of a bulky cervical tumour into the uterus (arrows), and preservation of the normal low-signal vaginal wall in the posterior fornix (small arrows)

Carcinoma of the ovary

Figure 9

Carcinoma of the cervix (stage IBii). There is extension of a bulky cervical tumour into the uterus (arrows), and preservation of the normal low-signal vaginal wall in the posterior fornix (small arrows)

Carcinoma of the ovary

Ovarian cancer accounts for about 4% of all female cancers but is the most frequent cause of death from gynaecological malignancy. Primary ovarian tumours arise from one of three ovarian components, namely the surface epithelium, germ cells or stroma of the ovary. The age distribution varies according to tumour histology. Epithelial ovarian cancer is a disease of postmenopausal women, with approximate age at diagnosis of 60 years, whereas ovarian germ cell tumours and sex cord stromal tumours are most prevalent in the second and third decades. The majority of patients do not present until the tumour is large and there has been spread beyond the ovary, usually with intraperitoneal dissemination.

No satisfactory screening test exists for ovarian cancer, and thus early detection is very difficult. The value of cancer antigen 125 (CA125) in screening for ovarian cancer was assessed in 22 000 females by a study that highlights the difficulty in the early detection of these tumours. Amongst this large cohort 11 cancers were identified, of which 7 were advanced stage III and IV tumours. Seven patients with normal screening CA125 levels subsequently presented with ovarian cancer, two within a year of the normal result (Jacobs et al., 1993): see Chapter 3.

Once the diagnosis of ovarian cancer has been established FIGO recommends a staging system based on laparotomy with specified parameters for sampling high-risk areas such as the omentum and peritoneal reflections.

Role of imaging

At presentation imaging is important for identifying the organ of origin of a palpable pelvic mass, and aids characterization. Although imaging acts as a guideline for the differentiation of malignant from benign tumours, it is not sufficiently accurate to preclude the need for tissue diagnosis.

Cervical Parametrium

Axial T2-weighted image through a cervical carcinoma. There is direct extension into the left parametrium with strands of tissue extending into the paracervical fat (arrow), indicating a stage IIB tumour

Figure 10

Axial T2-weighted image through a cervical carcinoma. There is direct extension into the left parametrium with strands of tissue extending into the paracervical fat (arrow), indicating a stage IIB tumour

Characterization is usually achieved initially using ultrasound. Cross-sectional imaging may be performed before surgery if malignancy is clinically suspected in order to assess para-aortic and pelvic lymph nodes. There are a number of situations in which further preoperative imaging is required:

• when the diagnosis is uncertain

• when the gynaecologist requires overall assessment of the extent of disease

• for guidance of percutaneous aspiration/biopsy— care is required in cases of 'seeding' tumour

• to assess invasion of adjacent organs or structures in advanced local disease.

Ultrasonography

Ultrasonography is a valuable method of detecting and characterizing ovarian masses. This examination may be performed via a transabdominal or transvaginal approach and benefits from being quick, inexpensive and noninvasive. Transvaginal Ultrasonography is especially sensitive for detecting small tumours in postmenopausal females and is said to provide additional diagnostic information in 70% of cases, particularly in distinguishing malignant from benign pathology. Unilocular simple cysts with a thin wall, measuring less than 5 cm, are likely to be benign and can be followed up with imaging (Higgins et al., 1989; Fleischer, 1991).

Malignant ovarian masses range from entirely cystic to solid masses. Sonographic features that suggest malignancy include soft tissue vegetations on the cyst wall, irregular cyst wall thickening, a partially solid mass, a homogeneously solid mass, the presence of ascites, and peritoneal nodules. However, there is a significant overlap between the appearances of benign and of malignant masses, preventing accurate characterization by imaging criteria alone.

Was this article helpful?

0 0
Natural Cures For Menopause

Natural Cures For Menopause

Are Menopause Symptoms Playing Havoc With Your Health and Relationships? Are you tired of the mood swings, dryness, hair loss and wrinkles that come with the change of life? Do you want to do something about it but are wary of taking the estrogen or antidepressants usually prescribed for menopause symptoms?

Get My Free Ebook


Post a comment