Accuracy of 64-multidetector computed tomography in diagnosis of adnexal tumors

Malignancies of the female reproductive system are among serious causes of mortality and morbidity, and adnexal cancers are in fifth place among the tumors with the highest mortality in the female population [1]. While the diagnosis may be delayed because of unspecified symptoms, appropriate treatment plan will be achievable with deliberate staging of the tumor and will follow by a better outcome [2]. The presence of an adnexal mass is the leading indications for gynecologic surgery, but the characterization of clinically diagnosed ovarian masses is frequently not possible until surgery and histopathologic examination have been performed. In most institutions the type of surgery (laparotomy vs. laparoscopy) depends on the probability of malignancy, which is based mostly on imaging appearance [3, 4].

Putting together with a thorough observation, physical examination and characteristics of the mass will give valuable information about its nature [5, 6]. Afterward, several invasive and non-invasive paraclinical evaluations can provide additional information [7, 8]. Computed tomography (CT) has been used primarily in patients with ovarian malignancies to reveal the stage of tumor, detect persistent or recurrent disease and demonstrate tumor response to therapeutic approach [9, 10].

Computed tomography of abdomen and pelvic can depict the masses as well as probable local or regional invasions. Additionally, it can differentiate gastrointestinal tract, urinary tract and reproductive malignancy from each other using contrast materials. Multi-detector computed tomography (MDCT) makes multiplanar evaluation of pelvic and abdominal structures available as well as two or three dimensional illustrations [11]. Further given details about the extension of the tumors particularly improves the treatment plan and outcome. The newly introduced 64-slice MDCT can provide high quality images of surrounding organs like diaphragm, paracolic gutters and intestine which defines patients who will benefit neo-adjuvant chemotherapy before debulking [12].

Although the diagnostic accuracy of spiral CT and its axial views for nature and extension of the adnexal masses were reported [13], the present study aimed to evaluate the accuracy of 64-multidetector computed tomography in detection, differentiation and staging of adnexal tumors.

During the study period 95 women with a primary diagnosis of ovarian mass (mean age, 48.63 ± 13.93 years) were included in the study. The frequency of pathologic findings reported by MDCT, surgery and histopathology are demonstrated in Figure 1. MDCT diagnosed 25 (26.3%) masses to be benign and 70 (73.7%) masses to be malignant. The sensitivity, specificity, positive and negative predictive value and accuracy of MDCT for diagnosing a malignant mass was 92.8%, 88.0%, 95.5%, 81.4% and 91.5% respectively comparing to histopathological evaluation of the specimens (Figure 2).

Table 1 presents the diagnostic performance of MDCT for detecting involvement of organs (pelvis, lymph nodes and peritoneum) in comparison to surgical and histopathological findings. In comparison with surgical and histopathologic findings, accuracy of MDCT was 90.7% and 90.7% for detecting pelvic involvement, 95.35% and 92.3% for detecting lymph nodes distribution, and 89.2% and 89.2% for peritoneal involvement (Figure 3 & 4). In addition, sensitivity of MDCT was 91.1% for detecting ascites (compared to report of the surgical findings), while specificity of it for detecting a malignant cell inside it was 43.3% (compared to histopathologic findings).

The disagreement in staging of the tumor between MDCT and surgery was found in 12 cases (table 2). Adnexal masses were overstaged in three patients and understaged in nine patients. Such disagreement observed in 10 cases in comparison to histopathologic findings; three patients were overstaged and seven understaged. However, there were significant agreements between MDCT and surgical findings (κ = 0.891) and between MDCT and histopathologic findings (κ = 0.858).

The results of the present study describe the significant agreement between MDCT, surgery and histopathology in determining stages of adnexal masses. Also, it has been demonstrated that MDCT have high efficacy and accuracy in defining the nature of a pelvic mass and detecting extension of malignant tumors which could be very useful in planning of treatment.

Adnexal masses are usually detected by clinical examination or sonography. Once an ovarian mass is detected, determination of a degree of suspicion for malignancy is important and is based largely on imaging appearance. Ultrasound (US) is considered the primary imaging modality for the assessment and characterization of adnexal masses. But, although the reported sensitivities of the technique are high (85-100%), the specificities are variable (50-100%) [15, 16]. Several studies have suggested that CT can play an important role in characterizing ovarian masses, emphasizing the comparability of CT to other imaging modalities such as magnetic resonance imaging (MRI) or US [17, 18]. Sensitivity and specificity of contrast-enhanced helical CT is reported to be 88-90% and 88-89% (respectively) for distinguishing malignant and benign adnexal masses [19]. The rates are 89-91% (sensitivity) and 88-93% (specificity) when using MRI [20, 21], while different kinds of ultrasonography have a sensitivity of 35-99% with lower rates of specificity [22, 23].

Tsili et al. reported the sensitivity of the16-slice MDCT to be 90% and accuracy of 89.1% for detecting malignant tumors in patients with an adnexal mass [24]. However, higher sensitivity (90.5%) and accuracy (92.9%) were reported by the same MDCT imaging method later [25]. Improved results in the present study might be due to thinner slices (64-multi-slices) of MDCT which got enhanced by reconstructed images.

The sensitivity, specificity, positive and negative predictive value and accuracy of CT has some potential limitations which are the topic of today researches to be improved. The sensitivity of CT for detecting peritoneal metastasis is reported about 85-93%, while it decreases to 20-25% when the metastasis is lesser than 1 cm in diameter [26]. This may be minimized by thinner slices, loss of artifacts due to partial volume effect and multiplanar reformatting which makes it possible to evaluate bending planes as well [27].

Results of MDCT imaging were compatible with histopathological findings in 84.6%. When there was a different MDCT mostly under-estimated the stage of the tumor. Accuracy of MDCT was higher in advanced stages compared to earlier stages (I and II). This maybe explained by high capability of MDCT to illustrate peritoneal seeding and involvement of abdominal visceral organs. Similar results have been reported by Tsili et al. about the compatibility between results of MDCT and histopathological evaluations [24] which is reported to be 85%. The main difference is that the majority of our study population had malignant tumors.

This study concludes that despite the possibilities of overstaging and understaging the adnexal masses, MDCT provides accurate information on detection, differentiation and staging of adnexal tumors and allows planning therapeutic approach.