Sonographic appearance of anal cushions of hemorrhoids

Abstract

AIM

To evaluate the diagnostic value of different sonographic methods in hemorrhoids.

METHODS

Forty-two healthy volunteers and sixty-two patients with grades I-IV hemorrhoids received two different sonographic examinations from January 2013 to January 2016 at the First and Second Hospitals of Xinjiang Medical University in a prospective way. We analyzed the ultrasonographic findings of these participants and evaluated the outcomes. Resected grades III and IV hemorrhoid tissues were pathologically examined. The concordance of ultrasonographic results with pathology results was assessed with the Cohen’s kappa coefficient.

RESULTS

All healthy volunteers and all patients had no particular complications related to sonography. There were no statistically significant differences between the participants regarding age (P = 0.5919), gender (P = 0.4183), and persistent symptoms (P > 0.8692). All healthy control participants had no special findings. However, 30 patients with hemorrhoids showed blood signals around the dentate line on ultrasonography. When grades I and II hemorrhoids were analyzed, there were no significant differences between transrectal ultrasound (TRUS), transperianal ultrasound (TPUS), and transvaginal ultrasound (TVUS) (P > 0.05). Grades III and IV hemorrhoids revealed blood flow with different directions which could be observed as a “mosaic pattern”. In patients with grades III and IV hemorrhoids, the number of patients with “mosaic pattern” as revealed by TRUS, TPUS and TVUS was 22, 12, and 4, respectively. Patients with grades III and IV disease presented with a pathologically abnormal cushion which usually appeared as a “mosaic pattern” in TPUS and an arteriovenous fistula in pathology. Subepithelial vessels of resected grades III and IV hemorrhoid tissues were manifested by obvious structural impairment and retrograde and ruptured changes of internal elastic lamina. Some parts of the Trietz’s muscle showed hypertrophy and distortion. Arteriovenous fistulas and venous dilatation were obvious in the anal cushion of hemorhoidal tissues. After pathological results with arteriovenous fistulas were taken as the standard reference, we evaluated the compatibility between the two methods according to the Cohen’s kappa co-efficiency calculation. The compatibility (Cohein kappa co-efficiency value) between “mosaic pattern” in the TPUS and arteriovenous fistula in pathology was very good (ĸ = 0.8939). When compared between different groups, TRUS presented the advantage that the mosaic pattern could be confirmed in more patients, especially for group A. There was a statistical difference when comparing group A with group B or C (P < 0.05 for both). There were obvious statistical differences between group A and group B with regard to the vessel diameter and blood flow velocity measured by TRUS (P < 0.05).

CONCLUSION

Patients with grades III and IV hemorrhoids present with a pathologically abnormal cushion which usually appears as a “mosaic pattern” in sonography, which is in accord with an arteriovenous fistula in pathology. There are clearly different hemorrhoid structures shown by sonography. “Mosaic pattern” may be a parameter for surgical indication of grades III and IV hemorrhoids.

Key Words: Hemorrhoids, anal cushion, transperianal, transrectal, transvaginal, sonography

Core tip: There are still controversial opinions on the etiology of hemorrhoids. As to patients with grades III and IV hemorrhoids, a special signal of blood flow with different directions could be observed, as a mosaic pattern, which was confirmed as arteriovenous fistula in pathology. Mosaic pattern could be a parameter for surgical indication of grades III and IV hemorrhoids. If this abnormal cushion which appears as a “mosaic pattern” in sonography is confirmed, it could help to interpret important etiological aspects of hemorrhoids.

INTRODUCTION

The etiology of hemorrhoids remains somewhat controversial, with the main theories purported as either prolapse of anatomical anal cushions[1] or areas of abnormal vascular hyperplasia with an intrinsically hyperactive vascular sphincter[2,3]. The anal cushions, representing the anal mucosa and submucosa above the dentate line, have been implicated as important structures in hermetic closure of the anal canal, and Gibbons et al[4] have shown that internal anal sphincter closure alone is insufficient to maintain anal occlusion at rest. There are several hypotheses, but no definite evidence regarding the etiology or origin of hemorrhoids. Most surgeons confirm the diagnosis and indication of hemorrhoids according to their anal examination result. Ultrasound is an inexpensive, safe technique that can dynamically and noninvasively evaluate the anorectal area using transperianal, transrectal, and transvaginal examinations. In addition, these techniques are helpful for diagnosing rectal cancer, fecal incontinence, anal sphincter abnormalities, sphincter atrophy, and perianal fistula. However, there have been fewer reports concerning hemorrhoid diagnosis or abnormalities using endoanal sonography. Several different techniques of ultrasonography were used for the evaluation of the functional anatomy of the pelvic floor, of which transperianal ultrasound (TPUS), transvaginal ultrasound (TVUS), and transrectal ultrasound (TRUS) are frequently used[5]. As a noninvasive method, TPUS was initially used in 1983 as an examination for the anorectal area in newborn babies with imperforate anus; more recently, other studies have reported the importance of this examination for the assessment of the anorectal area[6,7], and TRUS can clearly show the morphological characteristics of the hemorrhoid vascular network and pathological anal cushion. Therefore, there is clinical importance attached to the selection of treatment options. Endoanal ultrasound (EUS) or TRUS was introduced 25 years ago by urologists to evaluate the prostate[8]. Later, TRUS was used to stage rectal tumors[9] and to diagnose benign disorders of the anal sphincters and pelvic floor[10,11]. Sultan et al[12] first reported an abnormal manifestation of female hemorrhoids using TVUS to visualize the anal canal, and it was used to measure the hemorrhoid cushions by Nicholls et al[13]. The purpose of this prospective study was to measure the anal cushion area using sonographic examination in a group of patients with hemorrhoids of different grades (I-IV) and to compare the results with a control group of age-matched healthy volunteers. This research also aimed to evaluate the diagnostic value of sonographic examination of the hemorrhoid cushion via different sonographic methods to obtain early detection and early intervention. In this study, we analyzed color Doppler ultrasound characteristics of 62 symptomatic patients with hemorrhoids, with the aim to investigate the clinical value of ultrasound and to improve the early diagnostic ability of ultrasound for hemorrhoids.

MATERIALS AND METHODS

Study subjects

Forty-two healthy control volunteers and sixty-two patients with grades I-IV hemorrhoids received two different sonographic examinations (TRUS and TPUS for males, and an additional TVUS for female patients) from January 2013 to January 2016 at the First and Second Hospitals of Xinjiang Medical University. The participants ranged in age from 21 to 76 years with a mean age of 53.6 years. We have prospectively performed ultrasonography for all these participants. All included cases for analysis were followed for at least 4 wk. Rectal cancer, hemorrhoid thrombosis, anal fissure, anal fistula, fecal incontinence, ulcerative colitis, Crohn’s disease and any bleeding risk condition were excluded. Study participants were divided into three groups. Group A consisted of patients with stages III and IV disease (38 patients, 26 males and 12 females); the mean age was 42 ± 2.4 years, and the duration of symptoms was 7.8 ± 2.4 mo. Group B consisted of patients with stages I and II disease (24 patients, 14 males and 10 females); the mean age was 42 ± 2.4 years, and the duration of symptoms was 6.6 ± 3.8 mo. Group C consisted of normal healthy participants (42 volunteers, 24 males and 18 females), with a mean age of 42 ± 2.4 years. TRUS was considered as the standard modality to determine abnormal findings of hemorrhoids. Resected grades III-IV hemorrhoid tissues were pathologically examined.

Diagnostic criteria

Hemorrhoids were diagnosed based on the diagnostic criteria established previously[14]. Grade I hemorrhoids are characterized by prominent vasculature with engorgement but no prolapse. Grade II hemorrhoids prolapse only with straining but spontaneously reduce. Grade III hemorrhoids prolapse beyond the dentate line with straining and require manual reduction. Grade IV hemorrhoids prolapse beyond the dentate line with straining but cannot be reduced manually.

Instruments and ultrasound procedures

Patients received a cleansing enema 1 h before examination. The patients were placed in the left lateral position. Performance of the examination was accomplished using either a linear 12- to 7-MHz transducer or a transvaginal 8- to 4-MHz probe (ATL 3000 or 5000; ATL Ultrasound, Bothell, WA, United States), which was placed against the patient body. TPUS was used to provide a standardized image of the anal canal displaying the mucosa and submucosa, the hypoechoic internal anal sphincter, and the hyperechoic external anal sphincter, and the images provided are comparable with those obtained with an endoluminal probe. The anal cushion area was measured as described by Nicholls et al[13]. Generally, the measurement was made in the mid-anal canal where the image of the anal canal is most enduring and reproducible measurements can be made. Before TRUS and TVUS, the rigid probes were covered for hygienic reasons with a condom filled with ultrasound gel. Then, during the endoluminal process, the condom was covered with a gel on the outside and gently introduced into the rectum for a distance of about 3 to 5 cm. Landmarks used are the prostate, vagina, and puborectalis (PR) muscle. Then the probe was slowly withdrawn and entered the anal canal. Pressure of the probe and gravity can lead to different thickness measurements (that is, lying on the left side, left thickness measurements may be smaller). In female patients, the transvaginal probe was placed low in the vagina and angled posteriorly to image the anal canal as described by Stewart and Wilson[15]. Then, the ultrasound probe was used to examine the rectal cavity in a clockwise direction. We analyzed the ultrasonographical findings of these participants and evaluated the outcomes. After finding the irregular widening of the hypoechoic area in the second layer, the probe was placed in the five-layer longitudinal muscle of the rectal wall, and we used color Doppler ultrasound to examine the size, shape, distribution, and number of blood vessels, and the vessel shape and vascular hemodynamic index of the hemorrhoid. To achieve maximum sensitivity, Doppler settings were set at low frequency and filter, and the artifact in the image was removed. Starting from the highest level (4-6 cm above the anorectal junction) and continuing to the anal verge (1 cm below the line of the internal sphincter muscle), the transanal probe was slowly pulled distally. Doppler window was placed at the haemorrhoidal artery, and the baseline and pulse repetition frequency were changed according to blood flow velocity to obtain continuous blood flow images for measurement of peak systolic velocity and resistance index. Each artery diameter was measured three times, and mean value was calculated. Parameters of power Doppler settings were: frequency, 6.1 MHz; power Doppler gain, 40 dB; dynamic power Doppler setting parameter: range, 20 dB; edge, 1; time smooth, 7; special smooth, 2; color map, 5; filter, 5; pulse repetition frequency, 21.5 kHz; scale, 3.8 m/s. The vessel diameter and blood flow velocity were detected by TRUS in groups A and B.

Ethic and informed consent

This research was approved by the Ethic Committee of Human Subject Research of the First and Second Affiliated Hospitals of Xinjiang Medical University. All healthy participants and patients signed an informed consent form before this study. There were no commercial conflicts or other problems related to the participants. All sonographic examinations were free to all the participants.

Statistical analysis

Statistical analyses were performed using SPSS software version 19.0. The intention-to-treat principle was applied in this study. Numerical data are expressed as median and ranges. Student’s t-test was used to compare the treatment results, and the chi square (χ²) test was used for the comparison of proportions. A P-value < 0.05 was considered statistically significant. The diagnostic reliability of sonography in determining significant mosaic pattern (more than 70%) was assessed. Pathological results with arteriovenous fistulas were taken as the standard reference. To evaluate the compatibility between the two methods in grading the mosaic pattern, Cohen kappa co-efficiency [very good (ĸ > 0.8), good (ĸ = 0.61-0.8), moderate (ĸ = 0.41-0.6), low (ĸ = 0.21-0.4), and very low (ĸ ≤ 0.2)] was calculated.

RESULTS

Comparison of baseline data

All healthy participants and all patients had no particular complications related to sonography. There were no statistical differences between the three groups with regard to age and gender, or between the two patient groups with regard to persistent symptoms (P > 0.05). All healthy participants had no special findings on sonography (Figures 1 and 2). The general characteristics of the three groups showed no significant differences (P > 0.05) (Table 1).

Table 1 General characteristics of the three groups.

	Total No.	Gender, male/female	Age (yr)	Duration of symptoms (mo)
Group A (stages III and IV)	38	26/12	42 ± 2.4	7.8 ± 2.4
Group B (stages I and II)	24	14/10	44 ± 2.5	6.6 ± 3.8
Group C (healthy controls)	42	24/18	41 ± 2.2	-
χ2/t-value	1	0.28751	1
	2	0.65502	2
	3	0.02713	3
P value	1	0.59191	1
	2	0.41832	2
	3	0.86923	3

¹Comparison between group A and group B;

²Comparison between group A and group C; ³Comparison between group B and group C.

Figure 1 Normal five-layer structure of a healthy participant.

Figure 2 Typical blood flow in patients with stage I or II (left) and stage III or IV (right) hemorrhoids.

“Mosaic pattern” in TPUS

TPUS was considered as the standard modality to determine abnormal findings of hemorrhoids. Grades III and IV hemorrhoids revealed blood flow with different directions which could be observed as a “mosaic pattern” (Figures 3 and 4, Table 2). In patients with grade III or IV hemorrhoids, color Doppler ultrasound showed multidirectional, turbulent flow consistent with an arteriovenous flow pattern, which seemed similar to the mosaic pattern mentioned in other studies of arteriovenous malformation[16,17]. In patients with grade III and IV hemorrhoids, the number of patients with the “mosaic pattern” as revealed by TRUS, TPUS, and TVUS was 22, 12, and 4, respectively. When compared between different groups, TRUS presented the advantage that the “mosaic pattern” could be confirmed in more patients, especially for group A. Patients with grade III and IV disease presented with a pathologically abnormal cushion which usually appeared as a “mosaic pattern” in TPUS and an arteriovenous fistula in pathology. The compatibility (Cohen’s kappa co-efficiency value) between the TPUS and pathology was calculated as 0.8939 (Table 3). The vessel diameter measured by TRUS in group A and group B was 2.6 ± 0.4 and 0.6 ± 0.4, respectively. Blood flow velocity measured by TRUS in group A and group B was 56.4 ± 4.3 and 20.8 ± 3.4, respectively (Table 4). There were obvious statistical differences between group A and group B with regard to the vessel diameter and blood flow velocity measured by TRUS (P < 0.05).

Table 2 The mosaic pattern according to different methods for the three groups.

	TRUS	TPUS	TVUS in female	Pathology	P value
Group A (stages III and IV)	22/38	12/38	4/12		0.03794
				20/38	0.24885
					0.80936
Group B (stages I and II)	4/24	2/24	2/10	-	0.66254
					0.79835
					0.70556
Group C (healthy controls)	0/42	0/42	0/18	-	-
χ2/t-value	8.64491	3.31421	0.04771	-	-
	30.69812	13.22522	4.33892
	4.81173	1.33053	1.44793
P value	0.00331	0.06871	0.82701	-	-
	0.00002	0.00032	0.03732
	0.02833	0.24873	0.22893

¹Comparison between group A and group B;

²Ccomparison between group A and group C;

³Comparison between group B and group C;

⁴Comparison between TRUS and TPUS;

⁵Comparison between TRUS and TVUS;

⁶Comparison between TPUS and TVUS. TRUS: Transrectal ultrasound; TPUS: Transperianal ultrasound; TVUS: Transvaginal ultrasound.

Table 3 Cohen’s kappa co-efficiency between mosaic pattern in transrectal ultrasound and arteriovenous fistula in pathology in stages III and IV hemorrhoids.

Pathology	Mosaic pattern in transrectal US		Total
	Positive	Negative
Positive	20	0	20
Negative	2	16	18
Total	22	16	38

The compatibility (Cohen’s kappa co-efficiency value) between the transperianal US and pathology was calculated as 0.8939. The distribution of the Cohen kappa (κ) co-efficiency values: very high (κ > 0.8), high (κ = 0.61-0.8), moderate (κ = 0.41-0.6), low (κ = 0.21-0.4), very low (κ ≤ 0.2).

Figure 3 Mosaic pattern in the anal cushion of stages III and IV hemorrhoids. Blood flow with different directions could be observed as a “mosaic pattern” (white arrow). High-speed low-resistance arterial flow spectrum and arterialized venous spectrum could be observed as a bright colored area.

Table 4 Vessel diameter and blood flow velocity detected by transrectal ultrasound.

	n	Vessel diameter (mm)	Blood flow velocity (cm/s)
Group A (stages III and IV)	22	2.6 ± 0.4	56.4 ± 4.3
Group B (stages I and II)	20	0.6 ± 0.4	20.8 ± 3.4
χ2/t-value		16.1835	29.5566
P value		0	0

Figure 4 Mosaic pattern in the pathological anal cushion. A: “Mosaic pattern” was a special blood flow with different directions (white arrow); B: Special blood flow of “mosaic pattern” (white arrow); C: Special blood flow of “mosaic pattern” instructed with red and blue color (white arrow); D: Special blood flow of “mosaic pattern” instructed with multiple colors (white arrow).

“Arteriovenous fistulas” in pathology

Subepithelial vessels of resected grades III-IV hemorrhoid tissues were manifested by obvious structural impairment and retrograde and ruptured changes of internal elastic lamina. Some parts of the Trietz’s muscle showed hypertrophy and distortion. Arteriovenous fistulas and venous dilatation were obvious in the anal cushion of hemorhoidal tissues (Figures 5, 6 and 7). After pathological results with arteriovenous fistulas were taken as the standard reference, we evaluated the compatibility between the two methods according to the Cohen’s kappa co-efficiency calculation. The compatibility (Cohen’s kappa co-efficiency value) between the “mosaic pattern” in TPUS and arteriovenous fistula in pathology was very good (Table 3).

Figure 5 Subepithelial vessels of resected grades III and IV hemorrhoid tissues were manifested by obvious structural impairment and retrograde and ruptured changes of internal elastic lamina. Arteriovenous fistulas and venous dilatation were obvious in the anal cushion of hemorhoidal tissues (white arrow, magnification × 80); A: Artery, V: Vein, L: Lymphatic duct.

Figure 6 Blood cells in the arteriovenus fistula of anal cushion were seen (white arrow, magnification × 80). M: Muscle; G: Glands.

Figure 7 Arteriovenus fistula of anal cushion was obvious (white arrow, magnification × 400). Some parts of the Trietz’s muscle showed hypertrophy and distortion.

Comparison of findings of the three different sonographic modalities

There was no special findings in sonography of healthy participants (Figure 1). A pilot series of experiments enabled the sonographers to attain technical expertise in all methods before commencement of the study. Ultrasonographic measurements using a 2-sonographer protocol were obtained for all 104 study patients; the examinations were blinded and paired and used transperineal and transrectal techniques combined with transvaginal technique in female patients. No patients were excluded from the analysis. We calculated the number of patients in which the “mosaic pattern” could be observed in each group, and we analyzed the outcomes between the different groups of patients, and between different techniques (Table 2). For patients with stages III and IV hemorrhoids, TRUS seemed to be the most effective technique. When compared with TPUS, there was a significant difference (P < 0.05). But between TRUS and TVUS, and between TPUS and TVUS, there was no significant difference (P > 0.05 for both). When stages I and II hemorrhoids were examined, there were no significant differences between TRUS, TPUS, and TVUS (P > 0.05). When compared between different groups, the technique of TRUS presented advantages; the “mosaic pattern” could be confirmed in more patients, especially in group A, and there was a statistical difference when comparing group A with group B or C (P < 0.05 for both). A statistical difference using TPUS was also observed between group A and group B or C, but group A and B had no statistical difference between the results of the TVUS examination. In patients with hemorrhoids, the submucosa may be thickened and show a more echolucent appearance, suggesting the presence of fluid (i.e., blood). The amount of fluid was correlated with the degree of severity of hemorrhoids. Treatment of hemorrhoids with rubber band ligation or infrared coagulation did not alter anal configuration. The mucosal hemorrhoidal plexus was better visualized by vaginal endosonography. Thirty patients with hemorrhoids had blood signals around the dentate line on ultrasonography, and on the same scanning plane, blood flow with different directions could be observed as a mosaic pattern in grades III and IV hemorrhoids. There were underlying characteristics of the sonographic changes of hemorrhoids (Figures 1-4). Patients with grade III and IV hemorrhoids presented with a pathologically abnormal cushion which usually appeared as a “mosaic pattern” in sonography and an arteriovenous fistula in pathology (Figures 1-6). Two-dimensional ultrasonography revealed point sheet hypoechoic areas and intertwined flowing fine punctate anechoic areas; these were observed inside the inner layer of the thickened internal sphincter, which resembled a mixed cystic and solid honeycomb echo area. Color Doppler ultrasonography indicated that the anal cushion area of hemorrhoids contained blood signals. On the same scanning plane, blood flow with different directions could be observed as a “mosaic pattern”. High-speed low-resistance arterial flow spectrum and arterialized venous spectrum could be observed as a bright colored area (Figures 3 and 4).

DISCUSSION

Hemorrhoids are a commonly occurring human disease, and the causes are not well known to date. There are still controversial opinions on the etiology of hemorrhoids. There are several hypotheses, but no definite evidence regarding the etiology or origin of hemorrhoids. Ultrasonographic examination is an accepted technique for local staging of both benign anorectal disease and disorders, and malignant anal and perianal neoplasms[18,19]. Endoanal ultrasound (EAUS) of the anal sphincters is achieved by the simple expedient of replacing the balloon system used for rectal scanning with a hard cone[20]. The normal rectal wall is 2 to 3 mm thick and is composed of a 5-layer longitudinal muscle, joint longitudinal muscle and external sphincter, as is the rest of the digestive tract[21]. The anal canal is 2 to 4 cm long and is closed in the normal situation. Therefore, excellent images can be obtained by TRUS, as the anus lies tight around the probe. It is very important that the rectum is empty and clear prior to examination because residual stool can cause a deterioration of the image quality and impede interpretation[22]. The external anal sphincter is a voluntary muscle arising from the levator ani and puborectalis muscle to form a circular structure around the anal canal. The anatomy of the external anal sphincter remains controversial and is usually described as having three parts: a deep part joining with the puborectalis muscle, a superficial part attached to the superficial transverse perinei muscle, and a subcutaneous part continuing below the internal anal sphincter. The length of the anal canal, corresponding to the posterior longitudinal length of the external anal sphincter, was substantially greater in men than in women (3.5 cm vs 3.05 cm, on average), consistent with numerous previous reports[23,24]. The anatomical gap produced by this asymmetrical configuration would explain, on the basis of a lower resistance of the anterior rectal wall, the greater incidence of functional female pelvic floor disorders[5,25]. This anatomical description cannot always be observed using TRUS[26-28]. Vaginal endosonography, used to visualize the perianal area and especially the perineum, is an alternative when rectal endosonography is not possible because the anus is asymmetrical, causing air artifacts, extreme anal stenosis, or pain[29]. TPUS is also used to image the perianal area in patients with perianal fistula[30,31] or imperforate anus[31]. However, results using EUS seem somewhat better[32,33].

The highlight of this study was a new finding of a “mosaic pattern” in sonography which may be concordant with an arteriovenous fistula in pathology. Patients with grades III and IV hemorrhoids presented with a pathologically abnormal cushion which usually appeared as a “mosaic pattern” in sonography and an arteriovenous fistula in pathology. “Mosaic pattern” was a special blood flow with different directions. This special pattern was firstly reported by Aslan H[16], and described as “clover-leafed shape” in a pilot study reported by Zbar et al[34]. Color Doppler ultrasonography also indicated that “mosaic pattern” in the anal cushion area of hemorrhoids contained special blood signals. On the same scanning plane, blood flow with different directions can be observed as an “mosaic pattern”. High-speed low-resistance arterial flow spectrum and arterialized venous spectrum can be observed as a bright colored area. This result is encouraging. This new finding of a pathologically abnormal cushion, observed as “mosaic pattern” in sonography and an arteriovenous fistula in pathology, was confirmed in our study. After pathological results with arteriovenous fistulas were taken as the standard reference, we evaluated the compatibility between the two methods according to the Cohen’s kappa co-efficiency calculation. The compatibility (Cohen’s kappa co-efficiency value) between “mosaic pattern” in the transperianal US and arteriovenous fistula in pathology was very good (Kappa = 0.8939). It could help to interpret important etiological aspects of hemorrhoids, and could influence traditional surgical methods. The procedure of prolapse to hemorrhoids (PPH) or tissue selecting therapy (TST) for hemorrhoid treatment could be replaced by sonographic vascular therapy or other related techniques which are safer and less expensive.

TPUS is a simple, accessible, inexpensive, safe, and painless technique that dynamically and noninvasively evaluates anorectal structures. A previous study showed that two-dimensional TPUS had a specificity and sensitivity of 85% and 64%, respectively[35]. In addition, TRUS has been used for almost every possible disease in the anal region, and by delineating the anatomy it has increased insight into anal pathology. Clinical indications for TRUS in benign anorectal diseases are fecal incontinence for the detection of defects and atrophy, perianal fistulas, and abscesses for the demonstration of fistula tracts. Thekkinkattil et al[36] reported differences of the anal cushion area for some anorectal diseases using TVUS. These types of anal cushion differences were also confirmed in this study. We suggest that abnormalities and pathological changes of the anal cushion are one of the important reasons for symptomatic hemorrhoids.

Ultrasound can provide high-resolution images that can show hemodynamic information, and allow one to observe morphological and hemodynamic changes through a unique perspective. Therefore, ultrasound can play an important role in the diagnosis and differential diagnosis of hemorrhoids. The researcher can observe the rectal cavernous region and blood supply to the hemorrhoid through color endosonography, perineal sonography, and even transvaginal sonography. The vascular plexus full of blood flow can also be observed[37]. The reduction of the hemorrhoid blood supply could reduce blood reflux burden, symptoms of bleeding, and swelling. In recent years, hemorrhoid artery ligation has already been used as a minimally invasive surgery[38]. Color biplane endosonographic imaging studies have shown that the vast majority of hemorrhoidal arteries are located within the anorectal junction at 2 cm above the rectal mucosa, and this is regarded as the best area for hemorrhoid artery ligation[39]. Ten years before, Aigner et al[40] concluded that increased caliber and arterial blood flow of the terminal branches of the superior rectal artery are correlated with the appearance of hemorrhoids. They suggested that the hypervascularization of the anorectum contributes to the growth of hemorrhoids rather than being a consequence of hemorrhoids. Their observations confirmed that morphological changes are clearly detectable with the use of transperineal color Doppler ultrasound in patients with symptomatic hemorrhoids. Aigner F believed that transperineal color Doppler ultrasound is an appropriate method to assess these findings in patients with hemorrhoids. Four years before, Miyamoto et al[41] reported encouraging results using power Doppler imaging transanal ultrasound and three-dimensional power doppler angiography to visualize the haemorrhoidal plexus and the course of the haemorrhoidal artery in vivo. They found that blood flow significantly increased following advancement of the grade of haemorrhoid, and they also concluded that the distribution of haemorrhoidal arteries varies widely in both the number and the position. In their research, they demonstrated that the median number of haemorrhoidal arteries was five (range, 3-9) and that they were found in various positions, being located, in 62.1% of patients, at 1 o’clock but more commonly at the 3, 7 and 11 o’clock positions. Around the hemorrhoid area, using color biplane endosonographic imaging, we found this new abnormal structure characterized as a “mosaic pattern” inside the abnormal anal cushions, and we found an arteriovenous vascular plexus or fistula using pathology. We also observed a specific sonographic appearance characterized as arterialized venous blood flow of the arteriovenous fistula under pulsed sonography. The hemorrhoid in the same patient may have a single or several vessels in certain anal points or may have specific characteristics of an arteriovenous fistula at other points. Hemorrhoid vessels may directly penetrate the submucosa in other patients. It is unclear if these characteristics define the severity of the hemorrhoid, and further research is needed to confirm this. The discovery of an arteriovenous fistula in pathological anal cushions provides important diagnostic, therapeutic, and etiologic information for hemorrhoids. We strongly believe that the “mosaic pattern” inside the abnormal anal cushion on endosonography was most likely to fit with a pathological diagnosis of an arteriovenous fistula. However, we need further investigation and research to clarify these changes.

In our study, arteriovenous fistulas and venous dilatation were very obvious in the anal cushion of hemorhoidal tissues. After pathological results with arteriovenous fistulas were taken as the standard reference, we evaluated the compatibility between the two methods according to the Cohen’s kappa co-efficiency calculation. The compatibility (Cohen’s kappa co-efficiency value) between “mosaic pattern” in the transperianal US and arteriovenous fistula in pathology was very good. We believe that correctly identifying the hemorrhoidal artery and arteriovenous vascular plexus and avoiding blind treatment at the 3, 7, and 11 o’clock positions has become an urgent problem when dealing with bleeding anal diseases. The aim of this study was to provide a highly sensitive and accurate method for positioning vessels in the hemorrhoid area. We believed that not all the grades III and IV hemorrhoids should receive surgery. We considered that mosaic pattern should be a key parameter in determining surgical operation. In addition, we hope to guide clinicians in their preoperative assessment by selecting minimally invasive, safe and effective treatment strategies and decreasing surgical complications and the recurrence rate. At the same time, we aimed to provide a broader range of ideas for reasonable treatment, the innovation of new surgical techniques, and drug development which targets hemorrhoids. However, ultrasonography still has its limitations, although TVUS of the anorectum is well tolerated and can accurately detect abnormalities of the anal sphincter and surrounding structures[42]. Limitations for the application of EAUS include strictures and acute painful conditions. These caveats are largely caused by the size of regular rectal probes which have an average diameter of 17 to 20 mm. The introduction of EAUS probes with a diameter below 1 cm (actually 7 mm) represents a major improvement, particularly for imaging all layers of the anal canal even under acute pain[43]. The second pitfall of EAUS, also caused by the size of the probes, relates to changes in the anatomy of the anal canal resulting from the stretching of the anal canal and compression of the mucosal tissues. Thus, the EAUS picture of the anal canal does not reflect the anatomical situation, which makes it difficult to judge the status of the anoderm and to locate the hemorrhoidal tissue. The use of a small endoanal probe of only 7 mm in diameter enabled us to overcome these limitations. The limitation of EAUS vs endoanal magnetic resonance imaging, however, is the poor inherent contrast on images, which makes characterization of the external anal sphincter difficult, but it could be overcome by using a higher frequency transducer (10 MHz)[44].

We can conclude from this prospective study that there were clearly different structures of hemorrhoids observed by sonography compared with sonography of the control group. Findings of special mosaic pattern in sonography greatly influence clinical decisions in the treatment of hemorrhoids. Sonography may be useful for the early detection and early intervention of hemorrhoids. Mosaic pattern can play a key role in determining the best hemorrhoids management. Mosaic pattern could be a parameter for surgical indication of stages III and IV hemorrhoids. If our new finding of a pathologically abnormal cushion, which appeared as a “mosaic pattern” in sonography and an arteriovenous fistula in pathology, is confirmed in the future, it could help to interpret important etiological aspects of hemorrhoids and would influence traditional surgical methods. The PPH or TST for hemorrhoids could be replaced by sonographic vascular therapy or other related techniques which are safer and less expensive.

ACKNOWLEDGMENTS

The authors would like to thank Professor Adali and Professor Xiang yan at Xinjiang Medical University for their technical support and careful statistical assistance. The authors also thank Dr. Turgunjan Tuerxun at Xinjiang Medical University and Pro Adili Abudoureyimu at ATLAN International Language Center for their careful language assistance.

COMMENTS

Background

There are still controversial opinions on the etiology of hemorrhoids. Most surgeons confirm the diagnosis and indication of hemorrhoids according to their anal examination result. Ultrasound is an inexpensive, safe technique that can dynamically and noninvasively evaluate the anorectal area, using transperianal, transrectal, and transvaginal examinations. However, there have been few reports concerning hemorrhoid diagnosis or abnormalities using endoanal sonography. There are clearly different hemorrhoid structures shown by sonography in patients.

Research frontiers

The main purpose of this prospective study was to measure the anal cushion area using sonographic examinations in a group of patients with hemorrhoids of different grades (I-IV) and to compare them with a control group of age-matched healthy volunteers. This research also aimed to evaluate the diagnostic value of different sonographic methods in hemorrhoids.

Innovations and breakthroughs

The authors found the special structure in hemorrhoid patients by ultrasound and it was confirmed by pathology. Patients with grade III and IV disease presented with a pathologically abnormal cushion which usually appeared as a mosaic pattern in transperianal US and an arteriovenous fistula in pathology.

Applications

Mosaic pattern could be a parameter for surgical indication of stages III and IV hemorrhoids. Mosaic pattern can play a key role in determining the best hemorrhoid management. If they new finding of a pathologically abnormal cushion, which appeared as a “mosaic pattern” in sonography and an arteriovenous fistula in pathology, is confirmed in the future, it could help to interpret important etiological aspects of hemorrhoids and would influence traditional surgical methods.

Terminology

“Mosaic pattern” is a special blood flow with different directions. Patients with grades III and IV hemorrhoids present with a pathologically abnormal cushion which usually appears as a “mosaic pattern” in sonography and an arteriovenous fistula in pathology.

Peer-review

It is a very interesting paper. These data could be very useful for the diagnosis and choice of treatment for hemorrhoids.