This was a study of diagnostic accuracy that was conducted in accordance with the Declaration of Helsinki.
1.1 Design of diagnostic accuracy research
CEUS examinations were performed in consecutive patients diagnosed with suspected HAE at the First Affiliated Hospital of Xinjiang Medical University, and the CEUS results were compared with those of the gold standard diagnostic method, postoperative pathology. Figure 1 shows the diagnostic pathway of the study. The sensitivity, specificity, positive predictive value (PV+), negative predictive value (PV−), positive likelihood ratio (LR+), and negative likelihood ratio (LR−) of the CEUS procedure were calculated to evaluate the diagnostic accuracy with respect to HAE. We then compared the baseline data of 148 patients assigned to the HAE group based on pathological results with those of the non-HAE group.
2.2 Research subjects
This study included 148 patients with suspected HAE: 83 females and 65 males, with an average age 38 ± 13 years (range, 18–74 years), who visited the First Affiliated Hospital of Xinjiang Medical University between January 2013 and February 2023. The diagnostic results of all the patients were confirmed with postoperative pathology. All 148 patients enrolled in the study signed informed consent forms for CEUS. The contrast agent was administered in accordance with the guidelines for the use of contrast agents established by the European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB)[11, 12].
1.2.1 Inclusion criteria
(1) Patients suspected of having HAE based on a preliminary clinical diagnosis; and (2) patients in good health who were able to cooperate and complete the study requirements.
1.2.2 Exclusion criteria
(1) Patients with detailed external hospital information or confirmed as having malignant liver tumors, liver hemangiomas, and other liver-occupying lesions on CEUS examination;
(2) HAE cases confirmed with previous surgery;
(3) incomplete medical record data required for this study;
(4) patients who have not got the gold standard (i.e., postoperative pathologic evaluation) results.
1.2.3 Sample size calculation
Sample size was calculated according to the following formula:
The sample size required for a single-sample diagnostic trial was calculated based on a literature review[8] and a preliminary investigation. It was predicted that the sensitivity and specificity of CEUS in diagnosing HAE were 95.0% and 20.7%, respectively (α = 0.05, β = 0.1). We used SPSS 26.0 software (IBM SPSS, Inc., Armonk, NY, USA) to calculate the sample size required (n = 62 cases). To reduce errors and improve repeatability, we decided to increase the estimated sample size, and a total of 148 patients with suspected HAE were finally included.
1.3 Gold standard for diagnosis of HAE and CEUS examination
1.3.1 Gold standard
The results of postoperative pathology after surgery are the preferred gold standard for the diagnosis of HAE[13, 14].
1.3.2 CEUS examination
The ultrasound diagnostic equipment used was the Acuson Sequoia 512 (Siemens Healhthcare, Mountain View, CA, USA) and the Logiq E9 color Doppler ultrasound diagnostic instrument (GE Healthcare, Milwaukee, WI, USA), with an abdominal convex array probe at a probe frequency of 2.0–5.0 MHz.
We used a microbubble ultrasound contrast agent containing sulfur hexafluoride (SonoVue®, Bracco Imaging, Milan, Italy). The suspension was injected intravenously through a 20G peripheral venous catheter into the median cubital vein, with a 2.4 ml push within 3 s. After the push, 5 ml of physiological saline was used for flushing[11, 12].
1.3.3 Ultrasound contrast method
First, routine ultrasound scanning of the liver was performed to observe and record the location, size, shape, boundary, and internal echoes of the lesion. The blood flow distribution in the lesion was evaluated in the color Doppler flow imaging mode. The instrument was then set to the liver ultrasound contrast mode. Dual real-time images were selected and the lesion was set separately at the optimal position in the basic image on the right side of the screen. Liver CEUS is divided into three phases according to the fourth guideline and clinical practice recommendations for CEUS examination of the liver[15]: the arterial phase, from 10–20 s to 30–45 s; the portal phase, from 30–45 seconds to 120 s; and the delay phase, > 120 s. The CEUS images were reviewed, and the liver enhancement, the regression pattern, and the perfusion of the ultrasound contrast agent were observed continuously in the arterial phase, portal phase, and delay phase. Each scanning process was observed continuously for 5 min after the injection of contrast agent. If there were more than two lesions in the same patient, the interval between the two examinations was 10 min. A satisfactory image was selected, the diameter of the HAE lesion was measured in CEUS images of the lesion obtained before and after the administration of contrast reagent in the same scanning plane and at the same depth[10].
1.4 Quality control
All patients were examined and diagnosed by physicians with extensive experience in CEUS. An expert in alveolar hydatid disease of the liver provided full supervision and guidance of the process throughout the study. The study were examined and evaluated based on the 30 criteria for diagnostic accuracy research reports [16], which ran through the research design, data collection and organization, material analysis, and manuscript preparation. Finally, one person recorded the results of the CUES examinations, collected the diagnostic results of the gold standard method, established a database, and calculated each parameter with which the accuracy of CEUS was evaluated.
1.5 Statistical analysis
We used Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) to establish a database, and a cross table was generated to study the accuracy of the CEUS in the diagnosis of HAE based on the final database and research objectives. SPSS 26.0 software and MedCalc 19.0.4 (Ostend, Belgium) were used for data organization and statistical analyses. Measurement data that did not conform to a normal distribution are presented as medians (interquartile intervals) (M [Q1, Q3]), and nonparametric tests were used for intergroup comparisons. Counting data are presented as frequencies (%) and a χ2 test was used for their comparison. A difference was considered statistically significant at a P < 0.05. The sensitivity, specificity, PV+, PV−, LR+, LR−, and 95% confidence intervals (95% CIs) of CEUS in the diagnosis of HAE were calculated. The inspection level was set at α = 0.05 based on the 95% CI, and a plot of the area under the receiver operating characteristic (ROC) curve (AUC). We evaluated the diagnostic ability of CEUS based on the AUC value, sensitivity, and specificity. An AUC > 0.7 indicated an excellent diagnostic performance.