Two-dimensional shear wave elastography can improve the diagnostic accuracy of ultrasonography in acute appendicitis

Abstract

Aim

We aimed to evaluate the effectiveness of combining 2-dimensional shear wave elastography (2D-SWE) with ultrasonography (US) in diagnosing acute appendicitis in patients with suspected acute appendicitis.

Methods

Clinical and laboratory findings, gray-scale US and 2D-SWE imaging features, operation information, and pathology results of 48 patients diagnosed with acute appendicitis who presented with right lower quadrant pain were prospectively evaluated. We compared the findings to the US and SWE imaging features of 79 asymptomatic patients.

Results

Mean Alvarado score and appendix diameter were statistically significantly higher for acute appendicitis (p < 0.001). In patients with acute appendicitis, mesenteric lymphadenopathy and fat stranding were also more frequent (p < 0.001). The mean velocity and kPa values for appendix and mesenteric fat were statistically significantly higher in acute appendicitis (p < 0.001).

Conclusion

In the diagnosis of acute appendicitis, 2D-SWE increases the diagnostic performance of gray-scale US in the differentiation of inflamed and normal appendixes.

Introduction

Acute appendicitis (AA) is the most common cause of abdominal emergencies, with a lifetime prevalence of approximately 7–10%, and it is the most common condition requiring abdominal surgery [1–3]. The diagnosis of AA is primarily based on clinical history and physical examination. Classically, abdominal pain starting from the periumbilical region and progressing to the right lower quadrant and accompanying symptoms, such as leukocytosis and anorexia, are the best diagnostic clues [4]. However, symptoms may also occur with nonspecific clinical manifestations or overlap with other diseases. Therefore, this may lead to delays in diagnosis or misdiagnosis [5].

Alvarado scoring is the most commonly used system to shorten the time required for the diagnosis of AA and to reduce the number of negative appendectomies [6–8]. The Alvarado scoring system is based on the presence or absence of multiple scoring parameters and symptoms, including anorexia, pain migrating to the right iliac fossa, nausea, vomiting, rebound, high fever, leukocytosis, and neutrophilia [9]. An Alvarado score of < 7 is considered “low” and a score of ≥ 7 is considered “high” and determines the presence of AA [8, 10]. However, clinical signs and symptoms may be similar to other conditions, such as gastroenteritis, intussusception, and mesenteric adenitis. In addition, gynecological pathologies, such as ovarian cyst rupture, may mimic appendicitis in female patients [11]. Imaging methods play a key role in establishing an accurate and rapid diagnosis when the clinical picture is uncertain [12]. The management of patients with clinically suspected AA includes a risk management decision, including whether to operate (perforation risk and associated morbidity). Therefore, the appropriate use of modern imaging techniques that increase diagnostic accuracy can help confirm clinically suspected AA and reduce the rate of false-negative appendectomies [13, 14].

Multidetector computed tomography (MDCT) is considered the gold standard for evaluating patients with suspected AA and is often recommended as the primary imaging tool for emergencies [5]. However, its routine use is being questioned due to increasing concern about the long-term risk of malignancy from CT-associated ionizing radiation [15, 16]. Recently, researchers have noted a trend toward increasing confidence in the diagnostic contribution of US and decreasing the use of CT for radiation shielding [17]. On ultrasonographic imaging, direct signs of appendicitis are enlargement of the anterior–posterior (AP) diameter of the appendix, thickened appendix wall, and distention of the appendix lumen. Indirect signs include periappendicular inflammatory changes, presence of appendicolith, extraluminal air, presence of mesenteric lymph nodes, and reactive thickening of the adjacent intestinal wall [18–20]. However, not all of the above features are always useful in diagnosing appendicitis. Appendicitis, reactive thickening of the adjacent intestinal wall, and mesenteric lymphadenopathy may show poor diagnostic value [21]. The occurrence of periappendicular inflammatory changes is usually of varying degrees. In general, we tend to rely on direct indications of an increased appendiceal anteroposterior diameter and wall thickness. However, there are differences between the measurements reported in the literature, sometimes causing diagnostic confusion in cases with intermediate findings [22]. False-positive cases have been reported in CT, in which the appendiceal diameter increased and streaks were observed in the periappendicular fat planes, but histopathological examination did not show appendicitis-supporting inflammation [23]. The false-positive CT report rate was found to be between 1–10% in previous studies [23]. This highlights the need for a new imaging modality to show the presence of inflammation.

Two-dimensional shear wave elastography (2D-SWE) is an emerging imaging tool. It can be combined with sonography toincrease the accuracy of sonography in challenging diagnostic situations. In our study, we evaluated the role of 2D-SWE in patients with suspected AA and evaluated the effectiveness of combining 2D-SWE with sonography in diagnosing AA.

Materials and methods

Study design and setting

Our study was a single-center study approved by the ethics committee of our university (25.01.2017/2017/33). Informed consent was signed by patients according to the World Medical Association Declaration of Helsinki, revised in 2000 in Edinburgh.

Patient data

Ninety-eight patients who presented to the emergency department with right lower quadrant pain were evaluated. In total, 50 patients were excluded from the study. In 28 of those 50 patients appendix could not be visualized with a linear transducer for reasons such as abdominal obesity, intense gas distension, or patient incompatibility in the sonographic examination; 12 of those 50 patients refused the surgery; and 10 female of those 50 patients whose appendix was normal and sonographic findings in favor of right ovarian cyst rupture. Demographic data, clinical examination and laboratory values, imaging characteristics, operation information, and pathology results of the remaining 48 patients were scanned from the hospital data system and recorded. 79 asymptomatic control patients were included in the study. As the control group, we included patients who had no complaints, had a body mean index less than 25, whose appendix could be visualized using a superficial probe in sonographic examination, had no signs of appendicitis in sonographic examination, had a normal appendix, and did not have right lower quadrant pain.

US imaging

Gray-scale ultrasonography and 2D-SWE examinations were performed with an Aplio 500 ultrasound scanner (Canon Medical System Corporation, Tokyo, Japan) equipped with a linear transducer (4–14 MHz). Ultrasonography and 2D-SWE examinations were performed by the only radiologist (M.S.D.) with 16 years of ultrasonography experience. In the US examination, the appendix was examined using the graduated compression technique in the transverse and longitudinal planes. When the appendix ending with a tubular blunt end was visualized, the appendix anteroposterior diameter, increased echogenicity in the surrounding adipose tissue, and presence of mesenteric lymphadenopathy were evaluated. The appendix and periappendicular fatty tissue were evaluated using 2D-SWE. The default device settings were used for the evaluation (smoothing factor, persistence, frame rate (7 frames/second), and range of elastic modulus values (0–180 kPa)). The patient held their breath for 5 s, and cine loops were captured. The cine loops were replayed until the color-coded elasticity reached a steady state. Stiff areas are shown in red, and soft areas are shown in blue on the 2D SWE color map. The radiologist selected a single 2D-SWE frame by visual inspection. 2D-SWE [speed (shear velocity) mode (m/s) and elasticity mode (kPa)] measurements were performed (Fig. 1). All patients with a positive diagnosis underwent surgery, and the imaging findings were confirmed by the surgical findings.

Fig. 1.

A 18-year-old symptomatic girl was admitted to the emergency department with complaints of nausea, vomiting, and non-localized abdominal pain. On physical examination, there was no defense or rebound in the right lower quadrant of the abdomen. Body temperature was 37.8 °C and leukocytosis (> 10.000 white blood cells, leukocytosis with left shift) was present. The patient's Alvarado score was six. In the sonographic examination performed, there were reactive lymph nodes (arrows) with a short axis smaller than 1 cm in the mesenteric fat planes in the right lower quadrant of the abdomen (a). Appendix diameter was 8 mm and did not respond to compression (b), there was mesenteric fat stranding adjacent to the appendix (mesenteric fat stranding width was 9 mm) and minimal free fluid (arrow head) (c). Quantitative elasticity values were measured in the anterior, posterior, medial, and lateral walls of the distal appendix, and in the mesenteric fat planes adjacent to the appendix on the propagation (d), speed(m/s) (e) and elasticity(kPa) (f) mode in the axial plane, using the round ROI. The mean of four measurements from the appendix wall was calculated (T1 + T2 + T3 + T4 / 4). The mean quantitative elasticity values were measured as 2.9125 m/s and 26.075 kPa

Statistical analysis

All procedures were conducted using the Statistical Package for Social Sciences software (IBM SPSS Statistics 21.0, IBM Corporation, Armonk, NY, USA). The Kolmogorov–Smirnov test was used to determine whether the scale variables’ distributions were normal. For continuous numerical variables, descriptive statistics are reported as the mean and standard deviation. Categorical variables are represented by the number of cases and (percent). The Student’s t test was used to compare continuous numerical data, and chi-square tests were used to compare categorical variables. Receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic performance. The optimal cut-off point was derived using the Youden index if the area under the curve (AUC) was found to be significant. DeLong’s test was used to compare the differences in AUC values. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of the diagnostic performance indicators were calculated. A p value of less than 0.05 was considered statistically significant.

Results

Patients’ characteristics

A total of 127 patients (10–66 years; 65 males, 62 females) enrolled in our study. Of these, 48 had AA, while 79 were in the healthy control group. In terms of age and gender, there was no difference between the groups (p > 0.05). The mean Alvarado score was statistically significantly higher for AA (p < 0.001) (Table 1).

Table 1.

Clinical characteristics and ultrasound findings of patients and healthy controls

	Acute appendicitis (n = 48)	Healthy control group (n = 79)	p value
Gender*			0.067
Male	30 (62.5)	35 (44.3)
Female	18 (37.5)	44 (55.7)
Age**	25.71 ± 14.93	27.11 ± 13.58	0.587
Alvarado score**	6.46 ± 1.47	0.20 ± 0.81	< 0.001
Appendix diameter**	9.36 ± 3.06	5.16 ± 0.70	< 0.001
Mesenteric lymphadenopathy*			< 0.001
Yes	20 (41.7)	4 (5.1)
No	28 (58.3)	75 (94.9)
Mesenteric fat stranding*			< 0.001
No	1 (2.1)	78 (98.7)
Mild (1 = 9 mm)	19 (39.6)	1 (1.3)
Moderate (10 = 19 mm)	16 (33.3)	-
Severe (> 20 mm)	12 (25.0)	-
Appendix elastography**
kPa	39.11 ± 11.11	16.51 ± 3.71	< 0.001
m/sec	3.34 ± 0.54	2.27 ± 0.27	< 0.001
Mesenteric fat elastography**
kPa	33.35 ± 11.90	16.29 ± 4.20	< 0.001
m/sec	3.08 ± 0.59	2.19 ± 0.43	< 0.001

kPa kilopascals, m/sec meters per second

Bold values indicated that statistically significant (p < .05)

*Chi-Square Tests, data are presented as counts, with percentages in brackets

**Independent sample t test, data are presented as mean ± standard

Imaging findings

The mean appendix diameters were statistically significantly higher for AA (p < 0.001). In patients with AA, mesenteric lymphadenopathy and fat stranding were also more frequent (p < 0.001) (Table 1). Appendix and mesenteric fat SWE values were statistically significantly higher for patients with AA (p < 0.001). The mean velocity and kPa values of the normal appendix were 2.27 ± 0.27 m/sn and 16.51 ± 3.7 kPa, respectively. The mean velocity and kPa values for AA were 3.34 ± 0.54 m\sn and 39.11 ± 11.11 kPa, respectively. The degree of mesenteric inflammation was successfully delineated using 2D-SWE. The mean velocity and kPa values of normal mesentery were 2.19 ± 0.43 m/sn and 16.29 ± 4.20 kPa, respectively. The mean velocity and kPa values for the fat stranding mesentery were 3.08 ± 0.59 m\sn and 33.35 ± 11.90 kPa, respectively. The ROC analysis results are shown in Table 2 and Fig. 2. Appendix diameter, appendix elastography, and mesenteric fat elastography showed excellent diagnostic performance. KPa values of the appendix showed the highest AUC value, and there was no significant difference between the groups (p > 0.05).

Table 2.

Diagnostic performance of Appendix diameters and elastography parameters for differentiating acute appendicitis from heathy control

	Appendix diameter	Appendix kPa	Appendix m/sec	Mesenteric kPa	Mesenteric m/sec
AUC (95% CI)	0.996 (0.991–1)	0.998 (0.994–1)	0.984 (0.968–0.999)	0.966 (0.936–0.996)	0.929 (0.880–0.979)
p value	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
Cut-off	> 6.45	> 23.25	> 2.60	> 23.65	> 2.51
Sensitivity	100	100	97.9	89.6	89.6
Specificity	93.7	96.2	91.1	96.2	84.8
PPV	90.6	94.1	87.0	93.4	78.2
NPV	100	100	98.6	93.8	93.1
Accuracy	96.1	97.6	93.7	93.7	86.6

AUC Area under the curve, 95% CI 95% confidence interval, PPV positive predictive value, NPV negative predictive value, m/sec meters per second

Bold values indicated that statistically significant (p < .05)

Fig. 2.

Receiver operating characteristic (ROC) analyses of the 2-dimensional shear wave elastography (2D-SWE) with ultrasonography (US) in diagnosing acute appendicitis

Discussion

Complications such as perforation that causes delay in diagnosis and the prevention of negative laparotomy are two important conditions to be avoided in AA. There is consensus that improvements in imaging techniques reduce the incidence of these two undesirable conditions [24]. US is the imaging modality of choice for diagnosis because it is inexpensive, safe, and easily accessible. US has been reported to have a sensitivity of 75% and a specificity of 90% in AA diagnosis, and it has become the choice of modality due to the absence of radiation risk [25]. However, US is an operator-dependent modality. Despite the high prevalence of AA, diagnosis remains problematic, and false-negative US scans are reducing the acceptance of US by surgeons and emergency room physicians for definitive diagnosis. Trout et al. [26] conducted a study that evaluated 1009 patients with US. They reported that 246 pediatric AA cases diagnosed with AA were false positive in 35 cases, and the accuracy rate of US was 85–91%. The most important reason for false negatives was cases in which the appendix could not be visualized. However, they found that in false-positive cases, at least two or less diagnostic criteria (non-compressible appendix or appendix diameter > 6 mm) were present in 71.4% of the cases. In true positive cases, a diameter of the appendix greater than 6 mm was the most common radiological finding, and these cases had at least two or more radiological positive criteria [26]. In our study, we believe that 2D-SWE can increase the diagnostic accuracy of US by showing inflammation in cases in which the diameter of the appendix is increased or cannot be compressed in the US examination. The advantage of 2D-SWE is that it allows tissue elasticity to be measured in a given target tissue using the elastic modulus scale. Further, conventional gray-scale US images can be obtained using the 2D-SWE system [27]. In our study, we found that combining 2D-SWE with conventional US findings increased its accuracy in diagnosing AA. The most important finding of this study was that the mean elastic modulus of an inflamed appendix was higher than that of a normal appendix, which is consistent with the hypothesis of this study. An elastic modulus of 23.2 kPa exhibited 97% sensitivity and 91% specificity for the most rigid region of the appendix.

Cha et al. [27], also find median elastic modulus value of the appendix was significantly higher in patients with appendicitis (25.0 kPa) than that in patients without appendicitis (10.4 kPa) (0.001). Their SWE values in AA group are similer to our study population. They chose 12.5 kPa as the cutoff value for the stiffest region of the appendix exhibited 93% sensitivity, and 100% specificity. According to the study of Işık et al. [28] in a pediatric patient group including 48 patients and 18 healthy volunteers the cutoff value was discovered to be 14.3 kPA (95% CI 0.753–0.995) in the acute appendicitis group, with a 100% sensitivity and 79.17% specificity. Their SWE values are lower than the our population and other adult study group. They explain this could be because of differences in the study research group. In pediatric population, sonography examination is much more difficult than adult population, and mostly, patients are presented at the early stage of the inflammation rather than late stage, which could explain these lower rates of specificity and higher rates of sensitivity.

According to the study of Göya et al. [29], the sensitivity of US in the diagnosis of AA was 83.3%, and the specificity was 80%. The diameter of the appendix was reported to be less than 6 mm in 15% of the cases, and the authors indicated that US alone may be insufficient in the diagnosis of appendicitis cases. With strain SWE, the diagnosis of AA could be performed successfully, even in cases with an appendix diameter less than 6 mm. There is no single sonographic criterion with high sensitivity and specificity in the differentiation of normal and inflamed appendixes in US. Therefore, a combination of ultrasonography criteria is often used for diagnosis. Some researchers have investigated the contribution of appendiceal wall hyperemia to the diagnosis of AA with Doppler US. However, Doppler US could only increase sensitivity up to 87%, since there were false negative cases in which hyperemia of the appendix wall was not observed [30]. Our study supports the idea that inflammation in the appendix wall and surrounding mesenteric adipose tissue can be demonstrated by 2D-SWE and can be used as an additional sonographic criterion for clinical diagnosis in cases such as normal-sized appendicitis or distal appendicitis on gray-scale US.

Limitations

A limitation of this study is that 2D-SWE applications to support the diagnosis of AA could not be performed in cases where the appendix could not be visualized due to obesity, retrocecally located appendix, or patient incompatibility, which are the most important causes of false negativity in US. Further, the small sample size decreases the reliability and reproducibility of threshold values. One of our most important limitations is that the control group consisted of only healthy individuals. We did not evaluate how appendix elasticity can be affected in cases with other pathologies, such as diverticulitis, terminal ileitis, or pelvic inflammatory diseases that cause right lower quadrant pain.

Conclusion

The findings of our study suggest that the 2D-SWE technique can be effectively used as an imaging method for the diagnosis of AA using objective numeric values. The 2D-SWE technique should be part of the B mode of US, as it could be integrated into the current algorithms to improve diagnostic accuracy. Further comparative studies in larger populations are required to confirm the use of the 2D-SWE technique for the diagnosis of AA in routine clinical practice.

Funding

This study and authors not received any funding from other sources.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethics approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. The study protocol was approved by the local ethics committee.

Informed consent

Approval from the Institutional Review Board was obtained and in keeping with the policies for a retrospective review; informed consent was not required.