Differential diagnosis of <3 cm renal tumors by ultrasonography: a rapid, quantitative, elastography self-corrected contrast-enhanced ultrasound imaging mode beyond screening

Abstract

Objectives:

To assess the combined diagnostic strategy of contrast-enhanced ultrasound (CEUS) and acoustic radiation force impulse (ARFI) in the precise differential diagnosis of clear cell renal cell carcinoma (CCRCC) and urothelium carcinoma of the renal pelvis (UCRP) with other small renal tumors (SRTs) ＜3 cm in size.

Methods:

The elastography self-corrected CEUS (ESC) mode was established to perform the quantitative differential diagnosis of SRTs (<3 cm). The kidney shear wave velocity (SWV) value recorded by ARFI showed substantial variability in patients with CCRCC (high elasticity value) and UCRP (low elasticity value) compared with other renal masses, thus providing critical self-correction information for the ultrasound differential diagnosis of SRTs.

Results:

In this work, the ESC observations and the corresponding ESC criteria show a remarkable 94.6% accuracy in reference to the gold standards, thus allowing the quantitative, early triple distinction of CCRCC with UCRP and other SRTs in patients with suspicious SRTs.

Conclusions:

This ARFI self-corrected CEUS diagnostic strategy is far beyond a screening method and may have the potential to identify a window of therapeutic opportunity in which emerging therapies might be applied to patients with CCRCC and UCRP, reducing overtreatment and medical costs.

Advances in knowledge:

In our study, a new rapid and non-invasive elastography self-corrected CEUS (ESC) ultrasound imaging mode was developed, which was useful in the triple distinction of CCRCC, UCRP, and other SRTs with 94.6% accuracy. ESC is a promising method in the differential diagnosis of SRTs with accuracy and practicability far beyond a single screening model.

Introduction

Renal cell carcinoma (RCC) is the most lethal genitourinary malignancy, accounting for 3% of all adult tumors and 90–95% of renal cancers.¹ The incidence of renal tumor is steadily increasing per year, with the greatest increase in localized small renal tumors (SRTs) by the increasing use of cross-sectional imaging. SRTs are defined as a tumor smaller than 3 cm in maximum axial diameter. It is now widely accepted that most SRTs have a more favorable natural history than larger ones, and thus early detection and differential diagnosis of SRTs will significantly improve the prognosis.^2,3

Within the 1997 Heideberg classification, the most common subtypes of RCC are CCRCC (70–80%), PRCC (14–17%), and CCC (4–8%). Each subtype of RCC is associated with a different prognosis and clinical therapy. CCRCC has the worst prognosis, with a 5-year survival rate of 44–69% and a greater likelihood of metastasis than other subtypes, and thus more likely to choose nephron sparing surgery.^4–6 PRCC and CCC have better prognosis, with a 5-year survival rate of 78–92%, and close monitoring is considered for some cases in clinic. For certain types of PRCC, there are also some medical treatment choices such as molecularly targeted therapies available.^7,8 Besides RCC, urothelium carcinomas of the renal pelvis (UCRP) account for approximately 5% of renal tumors and for its high spread and recurrence rate, nephroureterectomy with excision of the bladder cuff is considered the standard therapy for most UCRP. For benign renal tumors regular follow-up is employed. Generally, the most fundamental consensus in the field of urology has been to distinguish CCRCC and UCRP in their early stages with other SRTs, because the prognosis is worse and thus surgical operations are usually involved in the treatment of them.^9–11 Thus, it is extremely important to clarify SRTs into three types (CCRCC, UCRP, and others) via preoperative imaging inspection.¹²

Among the medical imaging techniques for the detection of SRTs, ultrasound is currently an important screening method. While it is difficult to diagnose the pathological subtype with routine ultrasound imaging. Contrast-enhanced ultrasound (CEUS) and the acoustic radiation force impulse (ARFI) quantification technique are two increasingly developed ultrasound imaging techniques.^13–17 CEUS has unique advantages in the diagnosis of renal tumors with the lack of nephrotoxicity, the absence of ionizing radiation, and the ability to delineate the microvascular architecture in real-time.^18–20 While the diagnosis accuracy of SRTs is still limited because the characterization of small lesions is less typical than the larger ones. Meanwhile, ARFI elastography technique provides information about the mechanical property of the lesion and has been successfully applied to focal lesions of the kidney, breast, prostate, thyroid gland, and liver.^21–23 Specifically, ARFI technique is composed of “Virtual Touch tissue imaging (VTI)” technology and “virtual touch tissue quantification (VTQ)” technology. VTI, which usually appears as a sectional image similar to traditional two-dimensional ultrasound image, can intuitively reflects the relative hardness of tissue through its grayscale. VTQ quantifies the elastic modulus of the tissue by the parameter “shear wave velocity (SWV),” which is equal to the square root of the ratio of shear modulus (G), a constant of the medium, to density (ρ) of the medium. In this work, the SWV value obtained via the ARFI-VTQ technique was one of the most important quantificational parameter in the new combining differential diagnostic criteria for SRTs. Recently, the combination of CEUS and ultrasound elastography has been successfully applied in the preoperative diagnosis of breast lesions.²⁴ However, few recent studies have investigated the diagnostic efficacy of CEUS combined with ARFI for diagnosis of SRTs.²⁵

In current study, CEUS and ARFI were combined based on their separate evaluation in the early diagnosis of SRTs. Specifically, the CEUS characteristics of SRTs and their correlations with ARFI were analyzed, and the auxiliary diagnostic values of the ARFI were evaluated in improving the diagnosis accuracy of SRTs. In particular, we proposed a scoring system named elastography self-corrected CEUS (ESC) ultrasound imaging strategy for the precise differential diagnosis of CCRCC and UCRP with other SRTs. The purpose of this study is to determine whether the combining diagnosis system can help identify the three groups of SRTs, reducing overtreatment and medical costs.

Methods and materials

Subjects

Among the renal tumors examined between March 2017 and May 2018, 35 patients (mean age, 47; range, 4–86) with 37 renal tumors with diameters < 3 cm (mean size, 23.6 mm; range, 10–30 mm) were inspected by both CEUS and ARFI. All patients denied a history of chronic kidney diseases and were normal on renal function examination, including urea nitrogen and creatinine. The retrospective study was approved by the institutional ethics committee of our hospital. Informed consent was obtained from all patients before their examination.

Instruments and statistic methods

Instruments

A Mylab-90 color Doppler ultrasonic instrument (Esaote, Italy) equipped with a real-time CEUS device (2.5–5.5 MHz CA431 probe, mechanical index 0.8) was used to perform the CEUS imaging, and Sonovue (Bracco SpA, Milan, Italy) was used as ultrasound contrast agents (Bracco).

ACUSON-S2000 color ultrasonic instruments (Siemens, Germany) with virtual touch tissue quantification system (1-4MHz 4C1 convex array probe, mechanical index 1.0 ~ 1.7) was used for ARFI imaging.

Ultrasound Examination

All the operators are senior registered radiologists who have both national radiologist medical licence and license for large medical equipment, and with over 10 years’ experience on ultrasonography and, which can reduce bias from misjudgments for lack of experience. To reduce information bias for non-standard operation, the operators have received national-level standardized training on CEUS and ARFI before research according to guidelines recommend by Chinese Ultrasonic Medical Association and European Federation of Societies for Ultrasound in Medicine and Biology.

The optimal tumor observation section was selected as the CEUS target section and observed in the real-time imaging matching mode via contrast pulse sequence (CPS). First, 1.2 ml of a Sonovue contrast agent was injected via the left elbow vein group, followed immediately by the rapid push injection of 5 ml of normal saline. After that, the built-in clock on the ultrasound instrument was started. For each procedure, the patient was asked to breathe calmly. Then, the echo intensities in the focus and peripheral renal parenchyma were observed for 5 min. Dynamic CEUS images were stored in the built-in hard disk for subsequent analysis.

30 min after CEUS examination, ARFI-VTQ examination was carried out as follows. The sampling frame size was 1×0.5 cm and the maximum sampling depth was 8 cm. After region of interest (ROI) was selected with the best observation section, we held the probe motionlessly and pressed the VTQ measuring key, with the patient breathing calm. Each ROI needed to be measured effectively for five times with the results of “X.XX m/s” deleted as an invalid data. Static images were stored in the hard disk of the instrument system.

Image analysis

CEUS qualitative analysis

Ultrasound doctors who have over 10 years of experience on CEUS with both national radiologist medical licence and license for large medical equipment reviewed the CEUS dynamic images and observed the following index with the renal cortex at the same depth as the reference. The interobserver blind design was used in the CEUS research to reduce selection bias involved by subjective diagnosis tendency. ① Time-to-wash-in, time-to-peak, and time-to-wash-out of the contrast agent: early, equal, or fast compared to the renal cortex; ② Peak intensity: hyper-enhancement, iso-enhancement, or hypo-enhancement compared to the renal cortex; ③ Enhancement homogeneity: homogeneous or heterogeneous; ④ Pseudocapsule: presence or absence of a ring-shaped significant enhancement appeared around the tumor; Meanwhile, peritumoral vescular shadows should be eliminated.

Scoring of CEUS

Based on our previous study,²⁶ the characteristics of tumor vasculature were focused to assess as follows:

1. Peak intensity compared with kidney cortex (hyper-enhancement, iso-enhancement, or hypo-enhancement), a score of 2 or 0 was given ;

2. Time to peak compared with kidney cortex (fast or slow), a score of 2 or 0 was given;

3. Homogeneity (heterogeneous or homogeneous), a score of 1 or 0 was given;

4. Pseudocapsule (presence or absence), a score of 1 or 0 was given;

5. Time to wash-in (earlier or later than kidney cortex), a score of 1 or 0 was given.

The diagnostic and scoring criteria for CEUS characteristics is summarized in Table 1.

Table 1.

CEUS scoring criterion in renal tumors

CEUS characteristics	0	1	2
Peak intensity	hypo-enhancement	-	hyper-enhancement or iso-enhancement
Time to peak	Slow	-	Fast or equal
Homogeneity	homogeneous	heterogeneous	-
Psuedocapsule	absence	presence	-
Time to wash-in	Slow	Fast or equal	-

Analysis of ARFI-VTQ value

The ARFI-VTQ examination was analyzed by two ultrasound doctors with over 10 years of experience who are blind to CEUS examinations, histological information, and other imaging findings. The interobserver blind design was also used in the ARFI-VTQ research to reduce selection bias involved by subjective diagnosis tendency. When there is a major disagreement, senior director radiologist arbitration is introduced. The average SWV value of five measurements for each ROI was adopted into the statistics as a representative value.

Statistical analysis

The CEUS characterizaiton includes time to wash-in, time to peak, peak intensity, enhancement homogeneity, and presence of pseudocapsules. The number of positive and negative △SWV were statistically described for each type of renal tumors. The peak intensity and positive/negative △SWV were examined via Fisher’s exact texts, using a significance level of p < 0.05. Then, the accuracy rates of CEUS alone and together with ARFI were summarized, and the misdiagnosed cases were analyzed.

Results

CEUS manifestations of <3 cm renal tumors according to histopathology

All CEUS imaging characteristics of renal lesions < 3 cm are shown in Table 2, including 22 malignant tumors and 15 benign tumors. There were 22 cases of malignant tumors, including 9 CCRCCs, 2 CCCs, 2 PRCCs, 8 UCRPs, and 1 Xp11.2/TFE3 fusion gene-related renal cell cancer. Besides, 15 cases of benign tumors were involved, including 11 AMLs, one metanephric adenoma, one cyst fibrosis, one renal parenchyma infectious inflammation, and one renal column hypertrophy.

Table 2.

CEUS manifestations of renal focal lesions < 3 cm

Tumor		No.	Imaging characteristics
			Time			Peak intensity	Homogeneity	Pseudocapsule
			wash- in	peak	wash-out
			fast/ slow	fast/ slow	fast/ slow	high/ equal/low	Yes/no	Yes/no
Malignant		22	5/17	10/12	13/9	7/2/13	12/10	5/17
CCRCC		9	4/5	7/2	4/5	6/2/1	5/4	4/5
LMRCC		4	0/4	0/4	0/4	0/1/3	4/0	1/3
	CCC	2	1/1	0/2	2/0	1/0/1	2/0	0/2
	PRCC	2	0/2	0/2	2/0	0/0/2	2/0	1/1
UCRP		8	0/8	1/7	4/4	0/0/8	2/6	0/8
Xp11.2/TFE3 fusion gene-related RCC		1	0/1	1/0	1/0	0/0/1	1/0	0/1
Benign		15	2/13	2/13	9/6	0/1/14	10/5	3/12
AML		11	2/9	2/9	7/4	0/0/11	8/3	3/8
Metanephric adenoma		1	0/1	0/1	1/0	0/0/1	0/1	0/1
Cyst organization		1	0/1	0/1	0/1	0/0/1	1/0	0/1
Renal parenchyma		1	0/1	0/1	0/1	0/1/0	1/0	0/1
Renal parenchyma aseptic inflammation		1	0/1	0/1	1/0	0/0/1	0/1	0/1

CCRCC, clear cell renal cell carcinoma; UCRP, urothelium carcinoma of the renal pelvis; CCC, chromophobe cell carcinoma; PRCC, papillary renal cell carcinoma; AML, angiomyolipoma. No., Number.

Data represent number of patients with each CEUS manifestation.

ARFI-VTQ manifestations of renal tumors <3 cm

Of the 37 cases of SRTs, there were 11 and 26 cases with positive and negative △SWV, respectively Table 3. Fisher’s exact test of peak intensity and △SWV is illustrated in Table 4. The distributions of positive and negative △SWV are significantly different among the different peak intensities (p = 0.002). The majority of high-enhanced tumors (6/7, 85.7%) have a positive △SWV, indicating that the tumor hardness is higher than in the cortex. In contrast, the majority of low-enhanced tumors (18/21, 85.7%) have a negative △SWV or, namely, a lower tumor hardness than the cortex. The ARFI images of typical cases of CCRCC, UCRP and CCC are shown in Figure 1, Figure 2 and Figure 3 separately, including the VTI image, VTQ image and the physical picture.

Table 3.

Positive and negative △SWV of renal focal lesions < 3 cm

Type		Number	△SWV (m/s)
			+	-
Malignant		22	9	13
CCRCC		9	8	1
LMRCC		4	0	4
	CCC	2	0	2
	PRCC	2	0	2
UCRP		8	0	8
Xp11.2/TFE3 fusion gene-related RCC		1	1	0
Benign		15	2	13
AML		11	2	9
Metanephric adenoma		1	0	1
Cyst organization		1	0	1
Renal parenchyma		1	0	1
Renal parenchyma aseptic inflammation		1	0	1

SWV, shear wave velocity;

CCRCC, clear cell renal cell carcinoma; UCRP, urothelium carcinoma of the renal pelvis; CCC, chromophobe cell carcinoma; PRCC, papillary renal cell carcinoma; AML, angiomyolipoma.

Data represent number of patients with positive and negative △SWV of SRTs.

Table 4.

Comparison between peak intensity and △SWV of small renal tumors

Peak intensity	△SWV(m/s）		Total
	+	-
High enhancement	6	1	7
Equal enhancement	2	7	9
Low enhancement	3	18	21
Total	11	26	37

Figure 1.

A case of clear cell renal carcinoma of Fuhrman grade I and less than 3 cm. (a) An isoechoic tumor at the right renal hilum. The ARFI-VTI image shows that the tumor (arrow) is darker than the cortex and that the periphery is sharp. (b) The gross specimen (arrow) shows a Sallow tough tumor with a diameter of 2.9 cm. ARFI-VTQ quantification: (c) The SWV of the tumor is 3.35 m/s at a depth of 6.2 cm; (d) The SWV of the cortex next to the tumor is 2.35 m/s at a depth of 5.4 cm.

Figure 2.

A case of papillary urothelial carcinoma. (a) A tumor of medium-to-high echo at the middle and upper part of the left renal sinus (arrow). The ARFI-VTI image shows that the tumor is lighter than the sinus with a sharp periphery. (b) The gross specimen (arrow) shows a cauliflower, off-white soft tumor with a diameter of 3.0 cm; (c) ARFI-VTQ quantification: The SWV of the tumor is 0.96 m/s at a depth of 5.6 cm; (d) The SWV of the cortex next to the tumor is 2.31 m/s at a depth of 3.4 cm.

Figure 3.

A case of chromophobe cell carcinoma. (a) 2D-US shows a hypoechoic tumor at the lower part of the left renal sinus of 2.1 cm with a sharp periphery except with the lower boundary unclear from the renal parenchyma; (b) ARFI-VTI image shows the tumor (arrow) is similar in darkness to the cortex with the boundary unclear and internal echo distribution homogeneous; (c) ARFI-VTQ quantification: The SWV of the tumor is 1.75 m/s at the depth of 5.5 cm; (d) The SWV of the cortex next to the tumor is 2.95 m/s at the depth of 3.0 cm.

Diagnosis values of single CEUS and elastography self-corrected CEUS (ESC) mode

The diagnosis accuracy of renal tumors by CEUS was analyzed referring to the pathology result. It was found that 5/37 cases (13.5%) were misdiagnosed (Table 5), resulting in an accuracy rate of 86.5%. The diagnosis of three cases among them was corrected by ARFI-VTQ and the accuracy of diagnosis was improved to 94.6%. In one case of Xp11.2/TFE3 fusion gene-related renal cell cancer and one case of metanephric adenoma, no diagnostic conclusion was made. With the combination of CEUS and ARFI-VTQ, which means that a diagnosis of CCRCC or UCRP should only be made when both techniques were approved. In particular, one case of CCRCC was hyperechoic and manifested as “later to enter, slightly earlier to exit, and equal peak intensity” compared with the cortex, but did not show blood supply abundance and thus was misdiagnosed as non-CCRCC. Together with ARFI-VTQ, its SWV was higher than the renal cortex (△SWV = 0.75), indicating that it may be CCRCC, which is consistent with the pathology.

Table 5.

The △SWV behavior of renal tumors in all the five CEUS misdiagnosed cases according to the pathological diagnosis as golden standard

Diagnosis by CEUS	Diagnosis by VTQ (+or - △SWV)	Pathological diagnosis
Non-CCRCC	CCRCC (+)	CCRCC
UCRP	CCRCC (+)	CCRCC
CCRCC	LMRCC (-)	CCC
LMRCC	Non-LMRCC (+)	Xp11.2/TFE3 fusion gene-related RCC
LMRCC	Non-LMRCC (-)	Metanephric adenoma

CCRCC, clear cell renal cell carcinoma; UCRP, urothelium carcinoma of the renal pelvis; CCC, chromophobe cell carcinoma; PRCC, papillary renal cell carcinoma; AML, angiomyolipoma.

Discussions

Although the single identification of CCRCC or UCRP with other SRTs may be possible, current imaging techniques and biomarkers do not allow SRTs to be reliably differentiated into the three important differential types: CCRCC, UCRP, and other SRTs.^27,28 To the best of our knowledge, this is the first study to investigate the utility of elastography for the triple differentiation among CCRCC, UCRP, and other SRTs.

The preoperative differentiation of CCRCCs, UCRP, and other SRTs is important due to different clinical-surgical management, follow-up strategies, and prognoses.²⁹ Especially, radical ureteronephrectomy is always carried out in cases of UCRP, whereas nephrectomy, increasingly with minimally invasive approaches, is performed for CCRCCs.³⁰ In particular, for patients with UCRP, the bladder should be surveyed routinely because of a bladder recurrence rate of 15–50%. Preoperative biopsy is invasive which results in complications such as hematoma, infection, and seeding. Among non-invasive means, contrast-enhanced CT or MRI are restricted for some patients, especially those with a very low glomerular filtration rate because of concerns for renal damage or nephrogenic systemic fibrosis. Ultrasound is the most common imaging technique for SRTs, which is radiation-free and non-invasive.³¹ In this study, we focus on the differentiation among CCRCC, UCRP, and other SRTs, and the result showed that the diagnostic accuracy of single CEUS was 86.5%, and after the ARFI elastography self-correction, the diagnostic accuracy of the proposed ESC mode increased to a remarkable 94.6%, which was the best among all US diagnostic strategies and was also comparable to the best literature results of imaging methods for the differentiation of <3 cm SRTs.^32–35 It indicated that the diagnostic efficiency of CEUS was unsatisfactory, and meanwhile, ARFI can be a very suitable self-correction factor following CEUS. This present investigation on this self-calibration mode is supported by the pathological and anatomical foundations as shown in Figure 4a. Histologically, CCRCC is composed of solid and densely packed tumors with hypercellularity compared with UCRP and other SRTs, which determined both higher △SWV values and CEUS enhancement values. The cytoplasma of chromophobe cells in CCC is translucent and fine reticular, and there are few interstitial cells with hyaline degeneration. A large number of small vacuoles formed by monolayer membrane in the cytoplasm are observed under electron microscopy, so the tumor is relatively soft. In PRCC, hemorrhage, necrosis, and cystic changes are common, and thus the texture is soft. UCRP originates from the transitional epithelium of the pelvis, in which focal necrosis always occur to the central part, and the interstitium can develop mild fibrosis with inflammatory response, and thus the tumor is also soft. The elasticity characters of CCC, PRCC, and UCRP in this study are consistent with the pathology feature. ARFI-VTQ of the renal parenchyma inflammation indicated moderate texture in the peripheral part and soft in the central part, consistent with central liquefaction of abscess. Therefore, the proposed ESC mode can complement the advantages of two ultrasound imaging techniques with different imaging mechanisms as shown in Figure 4b. It should be noted that performing VTQ is not complicated. Patients in our hospital do not need to pay an extra fee for this examination. Therefore, we still suggest spending several minutes to perform VTQ for each detectable small lesion to obtain its elasticity characteristics after routine CEUS test.

Figure 4.

The proposed pathological basis and advantages of the ESC mode. (a) The pathological and anatomical foundations of the self-calibration mode. (b) Advantages of ESC mode composed of ARFI and CEUS.

The results of this study are consistent with previous ones. In a study conducted by Tan et al, the elasticity of RCC was significantly higher than those for angiomyolipoma.³⁶ In another recent study, Onur et al researched on more subtypes of benign and malignant renal masses, including oncocytoma, transitional cell carcinoma, metastasis, and lymphoma, and found malignant renal tumors significantly stiffer than benign tumors.³⁷ However, in most of the reports on the elasticity evaluation of small renal masses, the effort of progressive subdivision of cancer subtypes was abandoned. At the same time, in our work, the elasticity value was not used merely as a main diagnostic parameter but as a critical self-correction value to serve in the ESC mode and actually plays an important role in treatment decision.

However, in most of the reports on the elasticity evaluation of small renal masses, the effort of progressive subdivision of cancer subtypes was abandoned. At the same time, one problem that cannot be ignored is that usually, the elastic test area is not clear or cannot be informed by other imaging means, so that the significance of an ultrasound diagnosis as an early diagnostic means will be lost. In our work, the strain index value was not used merely as a main diagnostic parameter but as a critical self-correction value to serve in the elastography self-corrected CEUS mode and actually plays an important role in treatment decision changes. The ESC mode with a 94.6% accuracy is comparable to the best literature results of imaging methods for the differentiation of <3 cm SRTs and thus can be expected to contribute to the development of a new early diagnostic criterion of <3 cm SRTs. Regarding to the facile implantation of scientific double blind US test, we used two US machine to perform the independent CEUS and ARFI-VTQ measurement. Actually, we believe that in practical clinical US imaging, the CEUS and ARFI-VTQ can sufficiently be acquired on the exact same focused area by one operator using the same US machine to further minimize the possible system errors and operation errors, and reduce the operation time of the as-developed ESC mode to improve patient compliance and tolerance.

To better understand the advantage of ESC system, a case of CCRCC misdiagnosed by CEUS and corrected by ESC was shown in Figure 5. In this case, the CCRCC protruded to the renal sinus and was manifested on CEUS as a blood supply deficiency with “later to exit, later to peak, and slightly lower peak intensity” compared with the cortex, and thus was misdiagnosed as non-CCRCC. However, its SWV was higher than the renal cortex (△SWV = 0.75 m/s), indicating it to be a CCRCC in accordance with the pathology. One CCC was manifested on CEUS as an abundant blood supply with “synchronous perfusion, earlier to exit, earlier to peak and high peak intensity” and was misdiagnosed as a CCRCC. Then, the SWV of the tumor was slightly lower than the renal cortex (△SWV = −0.23 m/s), indicating the tendency of a low-grade malignant tumor, which was proved to be pRCC by pathological result. The case of Xp11.2/TFE3 fusion gene-related RCC was manifested on CEUS as a blood supply deficiency and thus was considered a low-grade malignant tumor, but ARFI-VTQ showed a slightly higher SWV than in the renal cortex, which is not a typical manifestation of a low-grade malignant tumor, but the conclusion was not drawn. Another case of a rare benign tumor, namely metanephric adenoma, was manifested on CEUS as a blood supply deficiency and had a lower SWV than in the renal cortex; and thus was misdiagnosed as LMRCC by both imaging techniques.

Figure 5.

A case of CCRCC misdiagnosed by CEUS and corrected by ARFI-VTQ. (a) A tumor of medium-to-high echo at the middle and upper part of the right kidney, with a size of 1.7 cm, a sharp boundary and homogeneous echo; (b) CDFI: a weak blood flow signal in the surrounding area; CEUS: (c) The tumor shows slow enhancement compared to the cortex, and the perfusion is uneven; (d) The peak intensity is equal to the cortex; (e) The tumor shows synchronous washout with the cortex. (f) The VTI image shows that the tumor is darker than the cortex, indicating a tough texture; ARFI-VTQ quantification: (g) The SWV of the tumor is 3.25 m/s at a depth of 4.6 cm; (h) The SWV of the hypoechoic area is 2.50 m/s at a depth of 4.6 cm.

To better illustrate the joint ESC US imaging diagnostic model and criteria, a proposed standardized scale based on the ESC US imaging mode is shown in Figure 6, in which the impersonal valuation of every single patient case was realized. The color of every point represents the confirmed subtypes of the tumors based on gold-standard criteria. The CEUS grading system in this work was identified according to our previous study³⁸ and literature reports,^26,39 in which the peak intensity and time to peak were highly valued to differentiate the subtypes of SRTs. The quantized elastic value of all the tumors was displayed by using the △SWV difference, which can partly reflect the pathological nature of tumors through the elasticity character.⁴⁰ Apparently, all the points can be separated into three groups: UCRP with both low CEUS score (0 ~ 2 points) and low △SWV (−2.82 ~ −1.79 m/s), CCRCC with both high CEUS score (5 ~ 6 points) and high △SWV (−0.36 ~ −1.94 m/s) and other SRTs with moderate CEUS score and moderate △SWV. In summary, the most concerned and highly malignant CCRCC and UCRP, which have quite different therapeutic strategies, can be reliably differentiated from other low-grade malignant and benign SRTs. This ESC standard is a condensed summary of practical experience and discovery. The visual and concise standards summarized and proposed in this work can guide further promotion of the ESC model in clinical applications.

Figure 6.

The proposed standardized scale based on the ESC US imaging mode.

Our study has limitations. Although the subtypes included in our study is more comprehensive compared to previous studies, the population was not large enough for several rare subtypes, and our results will require validation in future studies. In addition, we did not statistically evaluate the interobserver and intraobserver variability of the CEUS and ARFI-VTQ assessment of renal masses, which will be evaluated in future studies. However, a high interobserver agreement ratio for ARFI examination was reported in previous studies,^41,42 which alleviated the potential interobserver bias limitation.

Despite these limitations, our methods and standards have good prospects for clinical application. All renal tumors involved had been evaluated and followed up with gold standards, all malignant tumors with postoperative pathological results and benign ones with two consistent medical imaging results at least, for the triple differentiation of CCRCC, UCRP, and other SRTs; thus, the accuracy achieved is convincing. Besides, ESC system is a promising technique for reducing the use of radioactive imaging examinations and invasive biopsy, especially for those who are contraindicant with CT or MRI for renal insufficiency or forbidden from biopsy for coagulation function defect. Therefore, differentiation among CCRCC, UCRP, and other SRTs with a non-invasive ultrasound imaging method could be quite beneficial in clinical practice. Of note, the diagnostic value of the proposed non-invasive ESC mode and the corresponding ECS standardized scale goes far beyond screening, which is comparable to the best literature results of imaging methods for the differentiation of <3 cm SRTs. In summary, ESC system may become a high-priority alternative imaging technique in the differential diagnosis of SRTs.

Conclusion

In this work, an non-invasive elastography self-corrected CEUS imaging system (ESC) was developed with a quantitative standard for the differential diagnosis of SRTs (<3 cm) in their early stage beyond screening, in which the core concern is the triple distinction of CCRCC and UCRP with other SRTs. Simple CEUS may perform misdiagnoses (86.5% accuracy) and is only suitable for screening. The ARFI-VTQ technique with quantitative elastography values can be a very suitable self-correction factor following CEUS. The present investigation of this self-calibration mode is supported by solid pathological and anatomical foundations and can complement the advantages of the two ultrasound imaging techniques with different imaging mechanisms. After ARFI correction, the accuracy of the ESC mode increased to a remarkable 94.6%. Through grading and scaling, this rapid, non-invasive and quantitative diagnostic procedure could be readily performed by using routine US equipment. The elastography self-corrected CEUS imaging system described here for the differential diagnosis of SRTs can also easily be extended to other preclinical systems.