ULTRASOUND NAKAGAMI IMAGING in TISSUE CHARACTERIZATION

TIỀM NĂNG CỦA TẠO HÌNH SIÊU ÂM NAKAGAMI TRONG KHẢO SÁT MÔ

Hình  siêu âm thang xám B-mode là công cụ quan trọng trong khảo sát  lâm sàng các cấu trúc nội tại của mô. Các giá trị điểm ảnh thang xám của hình B-mode chỉ ra độ mạnh của echoes backscattered  [siêu âm tán xạ ngược] do các thay đổi đột ngột trong trở kháng âm của mô. Bởi vì cường độ B-scan phụ thuộc vào nhiều yếu tố, chẳng hạn như tín hiệu và xử lý ảnh, hệ thống cài đặt và hoạt động người sử dụng [1], [2], [3], siêu âm B-scan chỉ cung cấp mô tả chủ yếu định tính về hình thái học, mà không định lượng thuộc tính mô.

Nhiều nhà nghiên cứu nỗ lực phát triển các kỹ thuật tạo hình siêu âm chức năng nhằm cải thiện nhiều hạn chế của B-scan trong chẩn đoán lâm sàng. Trong số đó, phân tích các tín hiệu siêu âm thô tần số vô tuyến (RF) tán xạ ngược  trở về từ mô [raw ultrasound backscattered radio-frequency (RF) signals ] là một cách tiếp cận dễ dàng và hiệu quả trong khảo sát mô [tissue characterization]. Dữ liệu siêu âm RF đã được chứng tỏ là thông tin có giá trị vốn phụ thuộc vào hình dạng, kích thước, mật độ, và các đặc tính  khác của tán xạ [scatterers] trong một loại mô [4], [5], [6]. Dựa trên ngẫu nhiên của siêu âm tán xạ ngược, phân bố thống kê toán học có thể được áp dụng cho mẫu dạng hàm mật độ xác suất (pdf, probability density function) của siêu âm tán xạ ngược để đánh giá các thuộc tính của scatterers trong mô.

Các mô hình [model] Nakagami ban đầu được đề xuất để mô tả các thống kê của radar echoes [12] sau đó được áp dụng cho các phân tích thống kê của tín hiệu tán xạ ngược backscattered [13], [14], [15], [16] và thu hút sự chú ý của các nhà nghiên cứu. Phân phối Nakagami rất phù hợp với backscattered pdf và với các tham số Nakagami tương ứng mà chúng biến thiên với thống kê backscattered [15]. So với phân bố non-Rayleigh khác, phân phối Nakagami có ít tính toán phức tạp và có thể mô tả tất cả các điều kiện tán xạ trong siêu âm y khoa, gồm phân bố pre-Rayleigh, Rayleigh, và post-Rayleigh. Tham số Nakagami đã được chứng minh  phân biệt tốt nhiều  thuộc tính tán xạ  khác nhau [17], [18], [19].

Các nghiên cứu gần đây liên quan đến phương pháp tiếp cận Nakagami tập trung vào sự phát triển của tạo hình siêu âm Nakagami. Vắn tắt, tạo hình siêu âm Nakagami  được thiết kế bằng cách sử dụng bản đồ tham số Nakagami [Nakagami parametric map].

Hình  Nakagami cho phép bác sĩ và chuyên viên quang tuyến xác định trực quan thuộc tính scatterer trong lâm sàng. Khái niệm của hình Nakagami có nguồn gốc từ giáo sư Shankar [25] và một số nghiên cứu sơ bộ khác [26], [27]. Dựa trên các nghiên cứu thí điểm, chúng tôi đề xuất tiêu chí chuẩn để thiết kế tạo hình Nakagami bằng cách sử dụng dữ liệu siêu âm RF [28], và xác nhận tính hữu dụng của nó trong khảo sát mô bằng thử nghiệm [29] và mô phỏng [30]. Trong 5 năm qua, một loạt các nghiên cứu thực hiện bởi các nhóm  khác nhau, đã chứng minh rằng tạo hình Nakagami cung cấp các liên quan với cách sắp xếp tán xạ và nồng độ trong mô, bổ sung cho B-scan quy ước trong khảo sát đặc tính mô và chẩn đoán lâm sàng. Tạo hình Nakagami  đã được khảo sát trong  phát hiện đục thủy tinh thể [29], phân loại  u vú [31], [32], ước lượng dòng máu chảy [33], đánh giá dây thanh [34], giám sát tổn thương do nhiệt gây ra  [35], [36], đánh giá xơ hoá mô [37], [38], [39], và dự toán nhiệt độ [40].

Trước khi sử dụng tạo hình siêu âm Nakagami như một công cụ trong chẩn đoán lâm sàng, vẫn còn một số thử thách cần phải giải quyết. Một trong những vấn đề khó chịu là artifact. Có 2 loại artifacts xảy ra trong tạo hình Nakagami. Loại đầu tiên là artifact do nhiểu ồn gây ra, được tạo ra do hiệu ứng nhiểu ồn trong vùng mô không sinh âm [anechoic]. Khu vực anechoic (ví dụ: nang) không có scatterers; do đó, các tín hiệu nhận là chỉ là nhiểu ồn, có thể ngăn cản ước lượng tham số Nakagami và tạo bóng mờ (shade) trong hình Nakagami. Gần đây, chúng tôi đã đề xuất thuật toán hỗ trợ nhiểu ồn tương quan (NCA, noise-assisted correlation algorithm) để giải quyết vấn đề của artifact do nhiểu ồn gây ra [41], [42]. Loại  artifact Nakagami thứ hai, hiệu ứng tham số mơ hồ [parameter ambiguity effect], liên quan đến  ý nghĩa vật lý không rõ ràng của các tham số Nakagami vì hiệu ứng phân kỳ chùm [beam divergence effect]. Nhớ lại rằng tập trung bộ biến tử đầu dò là điều kiện tiên quyết cho tham số Nakagami để định lượng độ nhạy các biến thiên trong số liệu thống kê tán xạ ngược [backscattered]. Tuy nhiên, vì hiệu ứng tập trung bộ biến tử đầu dò [transducer-focusing effect] đồng thời đi kèm với hiệu ứng phân kỳ chùm, việc ước lượng tham số Nakagami gần với sự thống nhất, không phân biệt mật độ tán xạ [density scatterers] cao - hoặc thấp – trong mô. Hiện đang cố gắng để phát triển tạo hình multifocus Nakagami để loại bỏ hiệu ứng tham số mơ hồ này trong hình Nakagami.

Theo những bằng chứng hiện tại, tạo hình siêu âm Nakagami có tiềm năng lớn trong lâm sàng. Đặc biệt, hình ảnh Nakagami chỉ cần một máy siêu âm echo xung tiêu chuẩn, và do đó tương thích với hầu hết máy siêu âm hiện dùng. Trong tương lai, hình B-mode  quy ước và hình Nakagami có thể được kết hợp trong cùng một máy nhằm  đồng thời mô tả các hình thái mô và đánh giá các thuộc tính tán xạ.

_______________________________________________________

Ultrasound grayscale B-mode images are important clinical tools for clinically examining the internal structures of tissues. The grayscale pixel values of the B-mode image indicate the strengths of echoes backscattered because of abrupt changes in the acoustic impedance of tissues. Because the B-scan intensity is dependent on several factors, such as signal and image processing, system settings, and user operations [1], [2], [3], the ultrasound B-scan only provides a primarily qualitative description of the morphology, without quantifying tissue properties.

Many researchers make efforts to develop functional ultrasound imaging techniques to improve the limitations of the B-scan in clinical diagnosis. Among all possibilities, analyzing the raw ultrasound backscattered radio-frequency (RF) signals returned from tissues is an easy and effective approach for tissue characterization. The ultrasound RF data have been shown to contain valuable information that is dependent on the shape, size, density, and other properties of the scatterers in a tissue [4], [5], [6]. Based on the randomness of ultrasonic backscattering, mathematic statistical distributions can be applied to model the shape of the probability density function (pdf) of the backscattered echoes to evaluate the properties of scatterers in tissues.

Rayleigh distribution is the first model used to describe the statistics of the ultrasound backscattered signals. The pdf of the backscattered envelope follows Rayleigh distribution when the resolution cell of the ultrasonic transducer contains a large number of randomly distributed scatterers [7], [8]. However, it should be noted 
that the scatterers in most biological tissues have numerous possible arrangements. If the resolution cell contains scatterers that have randomly varied scattering cross-sections with a comparatively high degree of variance, the envelope statistics are pre-Rayleigh distributions. If the resolution cell contains periodically located scatterers in addition to randomly distributed scatterers, the envelope statistics are post-Rayleigh distributions. This is the reason why non-Rayleigh statistical models, such as the Rician [8], K [9], homodyned K [10], and generalized K [11] models, were developed to encompass various backscattering conditions in biological tissues.

The Nakagami model initially proposed to describe the statistics of radar echoes [12] was then applied to the statistical analysis of backscattered signals [13], [14], [15], [16] and attracted the attention of researchers. The Nakagami distribution was highly consistent with the backscattered pdf and with the corresponding Nakagami parameter varying with the backscattered statistics [15]. Compared to other non-Rayleigh distributions, the Nakagami distribution has less computational complexity and can describe all of the scattering conditions in a medical ultrasound, including pre-Rayleigh, Rayleigh, and post-Rayleigh distributions. The Nakagami parameter has been shown to perform well in distinguishing various scatterer properties [17], [18], [19]. A number of Nakagami compounding distributions, which involve the Nakagami-Gamma [20], [21], Nakagami-lognormal [22], Nakagami-inverse Gaussian [22], Nakagami-generalized inverse Gaussian [23], and Nakagami Markov random field models [24], have also been developed to better fit the statistical distribution of backscattered envelopes.

Recent studies related to the Nakagami approach focus on the developments of ultrasound Nakagami imaging. In brief, the ultrasound Nakagami image is constructed using the Nakagami parametric map. The construction of the Nakagami image allows physicians and radiologists to visually identify scatterer properties in clinical situations. The concept of Nakagami imaging originated from Professor Shankar [25] and certain other preliminary studies [26], [27]. Based on these pilot studies, we proposed a standard criterion to construct a Nakagami image using the ultrasound RF data [28], and confirm its usefulness in tissue characterizations by using experiments [29] and simulations [30]. In the past five years, a series of studies conducted by different research groups have demonstrated that the Nakagami image provides clues associated with scatterer arrangements and concentrations in tissues, which complement the conventional B-scan for tissue characterization and clinical diagnoses. The Nakagami image has already been explored in a number of medical applications, including cataract detection [29], breast tumor classification [31], [32], blood flow estimation [33], vocal fold characterization [34], monitoring ultrasound-induced thermal lesions [35], [36], tissue fibrosis assessment [37], [38], [39], and temperature estimation [40].

Before using ultrasound Nakagami imaging as a reliable tool to assist in clinical diagnosis, we still have some challenging problems that need to be resolved. One of the annoying problems is artifact. Two types of artifacts occur in the Nakagami image. The first type of Nakagami artifact is the noise-induced artifact, generated because of the effects of noise in an anechoic area of tissue. The anechoic area (e.g., cyst) has no scatterers; therefore, its received signals are solely noise, which disrupt the Nakagami parameter estimation to produce unreasonable shading in the Nakagami image. Recently, we have proposed the noise-assisted correlation algorithm (NCA) to resolve the problem of noise-induced artifacts [41], [42]. The second type of Nakagami artifact, the parameter ambiguity effect, refers to ambiguity in the physical meaning of the Nakagami parameter because of the beam divergence effect. Recall that transducer focusing is the prerequisite for the Nakagami parameter to sensitively quantify variations in backscattered statistics. However, because the transducer-focusing effect simultaneously accompanies the beam divergence effect, the estimation of the Nakagami parameter is close to unity, irrespective of high- or low-density scatterers in tissue. Now we are trying to develop multifocus Nakagami imaging to remove the parameter ambiguity effect in the Nakagami image.

According to the current evidences, ultrasound Nakagami imaging has great potential in clinical applications. In particular, the Nakagami image requires only a standard pulse-echo system for construction, and is therefore compatible with most clinical ultrasound systems. In the future, the conventional B-mode image and the Nakagami image may be combined in the same system platform for simultaneously describing the tissue morphology and evaluating the scatterer properties.

 

Abstract

Previous studies have demonstrated the usefulness of the Nakagami parameter in characterizing breast tumors by ultrasound. However, physicians or radiologists may need imaging tools in a clinical setting to visually identify the properties of breast tumors. This study proposed the ultrasonic Nakagami image to visualize the scatterer properties of breast tumors and then explored its clinical performance in classifying benign and malignant tumors. Raw data of ultrasonic backscattered signals were collected from 100 patients (50 benign and 50 malignant cases) using a commercial ultrasound scanner with a 7.5 MHz linear array transducer. The backscattered signals were used to form the B-scan and the Nakagami images of breast tumors. For each tumor, the average Nakagami parameter was calculated from the pixel values in the region-of-interest in the Nakagami image. The receiver operating characteristic (ROC) curve was used to evaluate the clinical performance of the Nakagami image. The results showed that the Nakagami image shadings in benign tumors were different from those in malignant cases. The average Nakagami parameters for benign and malignant tumors were 0.69 ± 0.12 and 0.55 ± 0.12, respectively. This means that the backscattered signals received from malignant tumors tend to be more pre-Rayleigh distributed than those from benign tumors, corresponding to a more complex scatterer arrangement or composition. The ROC analysis showed that the area under the ROC curve was 0.81 ± 0.04 and the diagnostic accuracy was 82%, sensitivity was 92% and specificity was 72%. The results showed that the Nakagami image is useful to distinguishing between benign and malignant breast tumors.

LOWER LIMB VARICOSE VEIN and UVDE

Background: To evaluate the effectiveness of ultrasonic venous duplex examination (UVDE) for lower limb varicose vein.

Materials and methods: Sixty-five patients with varicose vein of the lower limb during a 2-year period were enrolled in this study. There were 21 men and 44 women with an age range from 20 to 80 years and a mean age of 60 years. All patients received UVDE to determine the causes of varicose vein, including valvular incompetence, incompetent perforating vein, deep vein thrombosis, and congenital abnormality.

Results: In these 65 patients with 80 abnormal lower limbs, valvular incompetence was observed in 40 lower limbs (50%), valvular incompetence combined with incompetent perforating vein was observed in 22 lower limbs (27.5%), incompetent perforating vein only was observed in 13 lower limbs (16.3%), deep vein thrombosis in four lower limbs (5%), and congenital abnormality in one lower limb (1.2%).

Conclusion: Ultrasonic venous duplex examination (UVDE) is a safe and effective technique for evaluating lower limb varicose veins before planning the treatment course.

ª 2013, Elsevier Taiwan LLC and the Chinese Taipei Society of Ultrasound in Medicine. All rights reserved.

Discussion

Among the methods used to treat lower limb varicose vein, surgical treatment is the standard management [6e8].

However, curative resection is frequently precluded because of medical comorbidities that render patients inoperable and cause cosmetic problems for young individuals after surgery. Under these circumstances, alternative minimally invasive therapies such as percutaneous injection sclerotherapy, endovenous radiofrequency ablation and laser therapy were developed [9e11]. The treatment course for varicose vein depends on the etiology and severity of the varicose vein. To prevent local recurrence, patients with lower limb varicose vein need to be evaluated to assist in determining the etiology.

Because of technological advances in US, several studies have supported UVDE as an efficient and valuable technique for diagnosing the presence and determining the etiology of varicose veins [3e5]. The deep venous system is not involved in patients with primary varicose vein. Patients with secondary varicose vein develop this condition because of damage to the deep venous system, usually caused by deep vein thrombosis [12]. UVDE is a noninvasive, simple and reproducible diagnostic tool that can demonstrate the cause and location of lower limb varicose vein before planning the treatment course and post-treatment follow-up [2e5].

Incompetence of the greater or lesser saphenous vein or both has been reported to be the most common cause of lower limb varicose vein [1]. Our study results also demonstrated that 50% of the lower limbs with varicose veins resulted from sapheno-femoral valvular incompetence. On B-mode and color flow imaging, the dynamic motion of venous valve was well demonstrated. On spectral Doppler imaging, the direction of flow during the Valsalva’s maneuver also could be assessed well [1,12].

In addition to incompetence of the venous valve, incompetent perforating vein appears to have a role in the cause of lower limb varicose vein.

Manfred et al [2] used ascending venography and color-coded duplex sonography for detection of incompetent perforating vein and demonstrated more incompetent perforating veins were found by ascending venography. In our study, the incidence of perforating vein with reflux in the perforators with diameters 4 mm was high (8/10, 80%). Phillips et al [13]suggested that the difficulty in demonstrating reflux with US in all incompetent perforating veins is because of the small volume and low velocity of flow involved. This may explain, in part, why the two perforators with diameter 4 mm, but without reflux demonstration on UVDE was probably incompetent.

One of the limitations of the present study was that it was performed as a retrospective single-center study with limited patients. Another limitation was that 40% of patients received conservative treatment after UVDE, the location and diameter of incompetent perforating vein could not be confirmed, which possibly led to the bias.

In conclusion, UVDE is a safe and effective examination for detection of valvular  incompetence and the anatomic location of incompetent perforating vein of lower limb varicose vein before planning the treatment course.

Automated Breast Ultrasound: Characterisation of Breast Papillary Neoplasm

Characterisation of breast papillary neoplasm on automated breast ultrasound, Q-L Zhu,  J Zhang, X-J Lai, H-Y Wang, M-S Xiao, and Y-X Jiang

 

Abstract

Intraductal papillary neoplasms of the breast form a wide spectrum of pathological changes with benign intraductal papilloma and papillary carcinoma. They can occur anywhere within the breast ductal system. This review illustrates some characteristic appearances of breast papillary neoplasms on coronal planes reconstructed by automatic breast volume scan. Such manifestations are not uncommon in papillary neoplasms, and familiarity will enable confident diagnosis.

CBD Dilatation

Abstract

 

Objectives—To evaluate changes in the common duct diameter on sonography over time in patients with and without cholecystectomy.

Methods—We retrospectively evaluated the common duct diameter, central biliary dilatation, and interval change in 1079 patients who underwent sonography at least 2 years apart over a 6-year period. A board-certified radiologist, blinded to clinical and laboratory data, measured the duct diameter. A total of 893 patients (568 female and 325 male) were divided into 3 groups: group 1, remote cholecystectomy before sonography (mean, 9.7 years before sonography; n = 117); group 2, interval cholecystectomy between the first and second sonographic examinations (n = 56); and group 3, no cholecystectomy (n = 720). All groups were stratified by age, and group 3 was also stratified by the absence (n = 528) or presence (n=192) of gallstones.

Results—Duct diameters at baseline and follow-up averaged 4.5 and 5.2, 3.6 and 4.9, and 3.5 and 3.9 mm in groups 1, 2, and 3, respectively. Group 1 ducts were larger at baseline than in the other groups (P < .001). At follow-up, group 2 ducts showed a greater interval diameter increase than the other groups (P < .001). In a subanalysis of each group based on age, there was a mild increase in duct size with increasing age, although not clinically significant and within normal limits. In group 3 patients who never had gallstones, there was a significant small increase in duct size over decades (P < .001). The baseline duct sizes for patients with gallstones were not significantly different from those who never had gallstones (P = .15). 

Conclusions—Patients with remote cholecystectomy have larger common duct diameters than those with no or interval cholecystectomy. Most asymptomatic patients with or without cholecystectomy have a normal common duct diameter.

CHEST ULTRASONOGRAPHY in RESPIRATORY DISEASES

ULTRASOUND TREATMENT MAY STOP ACUTE KIDNEY INJURY

ULTRASOUND TREATS VENOUS ULCER

APPENDICEAL INTUSSUSCEPTION into CECUM: CASE REPORT PUBLISHED on DIAGNOSTIC IMAGING, June 2013

NHÂN CA LỒNG RUỘT RUỘT THỪA tại MEDIC

VALUE of CT after ULTRASOUND for DIAGNOSING APPENDICITIS

PLACENTA ACCRETA and CESAREAN DELIVERIES

GALLBLADDER POLYPS: FACTORS AFFECTING SURGICAL DECISION.

Abstract

AIM: To determine the factors affecting the decision to perform surgery, and the efficiency of ultrasonography (USG) in detecting gallbladder polyps (GP).

METHODS: Data for 138 patients who underwent cholecystectomy between 1996 and 2012 in our clinic with a diagnosis of GP were retrospectively analyzed. Demographic data, clinical presentation, principal symptoms, ultrasonographic and histopathological findings were evaluated. Patients were evaluated in individual groups according to the age of the patients (older or younger than 50 years old) and polyp size (bigger or smaller than 10 mm) and characteristics of the polyps (pseudopolyp or real polyps). χ2  tests were used for the statistical evaluation of the data.

RESULTS: The median age was 50 (26-85) years and 91 of patients were female. Of 138 patients who underwent cholecystectomy with GP diagnosis, only 99 had a histopathologically defined polyp; 77 of them had pseudopolyps and 22 had true polyps. Twenty-one patients had adenocarcinoma. Of these 21 patients, 11 were male, their median age was 61 (40-85) years and all malignant polyps had diameters > 10 mm (P< 0.0001). Of 138 patients in whom surgery were performed, 112 had ultrasonographic polyps with diameters < 10 mm. Of the other 26 patients who also had polyps with diameters > 10 mm, 22 had true polyps. The sensitivity of USG was 84.6% for polyps with diameters > 10 mm (P < 0.0001); however it was only 66% in polyps with diameters < 10 mm.

CONCLUSION: The risk of malignancy was high in the patients over 50 years old who had single polyps with diameters >10 mm.

 

Core tip: Early stage gallbladder cancers can often be detected as polyps in imaging studies. The aim of this study was to determine the factors affecting surgery by analyzing the incidence of malignancy of gallbladder polyps (GP) and the efficiency of ultrasonography in detecting GP. Of 138 patients with GP on imaging, 99 had polyps and 21 had histopathologically confirmed adenocarcınoma. Of these 21 patients, all malignant polyps were solitary and had a diameter >10 mm. In our study, the risk of malignancy correlated wıth age over 50 years old, solitary polyp and polyp diameter >10 mm.

 

Key words: Gallbladder; Polyps; Ultrasound; Cholecystectomy; Malignancy

 

© 2013 Baishideng. All rights reserved.

 

GASTRIC DUPLICATION CYST @ MEDIC

Male patient 26yo from An Giang province complaints of nausea. Endoscopy revealed a submucosa mass at posterior antrum, which was confirmed later by CT,  but cannot rule out a heterotopic pancreas.

Ultrasound detected a cystic mass from greater curvature which has cyst wall like gastric wall and no debris inside.

Transversal section at antrum= Water filling helps viewing clearly root of gastric duplication cyst which continued greater curvature of stomach. The cyst wall has 2 layers: hypoechoic muscularis and hyperechoic lining mucosa [muscular rim].  

Our present case has these criteria : (a) the wall of the cyst is contiguous with the stomach wall; (b) the cyst is surrounded by smooth muscle, which is continuous with the muscle of the stomach; and (c) the cyst wall is lined by epithelium of gastric or any other type of gut mucosa.

The gastric duplication cyst has clearly muscular rim with size of 27.1x15.6mm without debris inside.

The hypoechoic muscularis of the gastric antrum appears to be contiguous with the hypoechoic rim of the cystic lesion (white arrow).

A video clip of gastric duplication cyst on ultrasound examination.

So it is a noncommunicating gastric duplication cyst with the gastric lumen. Patient underwent surgery to remove the gastric duplication cyst on 20/8/2013 with the normal pathological report. After 3 months of operation he remains well and endoscopy shows a gastric scar on posterior surface of his stomach.

_________________________________

Explanation (from Chavira, http://pediatricimaging.wikispaces.com/Chavira-001-Gastric+Duplication+Cyst)

Gastrointestinal duplication cysts occur in 1 in 4500 births, of these 75% are intraabdominal, 20% thoracic, and 5% thoracoabdominal. Gastric duplication cysts, in particular, are one of the most rare comprising 7% of all gastrointestinal cysts. Other locations include the jejunum/ileum (53%), esophagus (18%), colon (13%), and duodenum (6%). Gastric duplication cysts are often a missed diagnosis because of their rarity and vague clinical presentation.

Duplication cysts arise due to a congenital developmental abnormality within the gastrointestinal tract, of which the etiology is not completely understood. Multiple theories have been developed to explain their existence including partial twinning, in utero ischemic events, and abnormal endoderm and notochord separation. Bremer's theory of abnormal recanalization has also been described to explain their occurrence. A brief review of the embryology related to the development of the alimentary tract helps to better understand this theory.

During the 6th week of gestation, epithelial cells proliferate and the lumen of the GI tract becomes occluded. Vacuoles are then created which fuse with one another to recanalize the lumen. Abnormal development of the gastrointestinal lumen can occur when a group of vacuoles coalesce with one another but do not form a connection with the developing lumen, with continued growth a wall of bowel will then form between them, which may lead to the formation of a duplication cyst not in communication with the gastric lumen. Conversely, if a small group of vacuoles coalesce and one opens into the main lumen, the duplication cyst may form in communication with the gastric lumen. 

Image 1. Normal and abnormal development of the gastrointestinal lumen are depicted in the above image.

It is important to note that although each theory offers an explanation for the formation of duplication cysts, there is no singly theory that can adequately explain the occurrence of all forms of duplication cysts.

The presentation of duplication cyst depends on the size and location. The typical presentation is in a child less than 2 years of age with females being more commonly affected than males. However, adult cases have been reported. The typical symptoms are nausea, vomiting, and a palpable abdominal mass. However, in some cases patients may be entirely asymptomatic. The features characteristic of a duplication cyst include a well-developed coat of smooth muscle, an epithelial lining the same as that of the alimentary tract, and an attachment or prior attachment to a part of the alimentary tract. The location of the cyst is typically along the greater curvature and most are not in communication with the gastric lumen.

The diagnosis of a duplication cyst is usually made with CT and US imaging. Ultrasound is becoming the preferred diagnostic imaging tool, as in some cases it may reveal the "double wall sign" outlining the echogenic inner layer and hypoechoic outer muscular layer which is characteristic of a duplication cyst. In some cases, however, imaging is not sufficient and the diagnosis is confirmed intraoperatively.

Treatment for gastric duplication cyst is surgical. The excision of a gastric duplication cyst can usually be done with minimal loss of adjacent stomach. However, if the duplication cyst and gastric lumen are in communication, partial gastrectomy is required. Complications that may occur if untreated include perforation and ulceration. There is also a risk of malignant transformation.

_________________________________

 From Review of Radiology, DAHNERT.

Gastric duplication cyst = intramural gastric cyst lined with secretory epithelium

Incidence: 7% of all alimentary tract duplications

Path: noncommunicating spherical cyst (majority); may communicate with aberrrant pancreatic duct; ectopic pancreatic tissue found in 37%

Symptomatic age: infancy; in 75% detected before age 12; M:F = 1:2

• pain (from overdistension of cyst, rupture with peritonitis, peptic ulcer formation, internal pancreatitis)

• vomiting, anemia, fever

• symptoms mimicking congenital hypertrophic pyloric stenosis (if duplication in antrum / pylorus)

Most common site: greater curvature (65%)

- paragastric cystic mass up to 12 cm in size, indenting greater curvature

- seldom communicates with main gastric lumen at one or both ends

- may enlarge + ulcerate

- Tc-99m uptake



US:

+ cyst with two wall layers (inner echogenic layer of mucosa, outer hypoechoic layer of muscle)

+ clear / debris-containing fluid


Cx:

(1) Partial / complete small bowel obstruction

(2) Relapsing pancreatitis (with ductal communication)

(3) Ulceration, perforation, fistula formation

DDx: pancreatic cyst, pancreatic pseudocyst, mesenteric cyst, leiomyoma, adenomatous polyp, hamartoma, lipoma, neurofibroma, teratoma

Discussion

Gastrointestinal duplication is a relatively rare anomaly that may occur at any level from oral cavity to rectum with ileum being the most common site. Duplication cysts of the stomach are quite rare, and most of them have been reported in children [1, 5, 6]. Duplication cysts of ileum are usually located on mesenteric border [7], whereas the usual location for gastric duplication cysts is along the greater curvature [4, 6, 7]. The duplication cyst is entirely separated from the adjacent bowel but shares a common wall [8].The essential criteria for diagnosis of a gastric duplication cyst are (a) the wall of the cyst is contiguous with the stomach wall; (b) the cyst is surrounded by smooth muscle, which is continuous with the muscle of the stomach; and (c) the cyst wall is lined by epithelium of gastric or any other type of gut mucosa [1, 4, 9].

Our present cases fulfilled these criteria excluding other diagnoses.

Gastric duplication cysts comprise 4% of all gastrointestinal duplications. Various other congenital anomalies such as alimentary tract duplications, esophageal diverticulum, or spinal cord abnormalities are encountered in up to 50% patients [8]. These malformations are believed to be congenital, formed before the differentiation of epithelial lining, and therefore named for the organ with which they are associated [3, 10]. Duplications result from the disturbances in embryonic development, and various theories have been proposed for the actual mechanism. Bremer proposed the theory of errors of recanalization and fusion of longitudinal folds. He suggested that duplication cysts originated from the fusion of longitudinal folds allowing the passage of a bridge of submucosa and muscle at the second and third month of intrauterine life [5]. McLetchie suggested that adhesion of notochord and embryonic endoderm might not elongate as quickly as its surrounding structures, causing traction diverticulum leading to duplication cyst formation [5]. Other theories of enteric duplication include abortive twinning, persistent embryological diverticula, and hypoxic or traumatic events [5]. There is no single theory that is satisfactory for all types of duplications [5].

Greater than 80% of gastric duplications are cystic and do not communicate with lumen of the stomach. The remainders are tubular with some communication [5]. The structure is defined as tubular when the lumen is contiguous and cystic when the lumen is not contiguous with stomach lumen [6]. The mucosal lining of duplication may be histologically similar to the segment of gut to which is topographically related.

However, some duplications may include lining from other segment of alimentary or respiratory tract. The presence of respiratory epithelium in the cysts of thorax, tongue, liver, and stomach suggests that the undifferentiated epithelium of foregut might undergo transition to differentiated specialized epithelium during embryonic period [5].

Gastric duplications typically become symptomatic during childhood. 67% are diagnosed within the first year of life, and less than 25% are discovered after age 12 [4]. The duplication cysts of the stomach are usually diagnosed intraoperatively in adults [10]. In our first patient, the preoperative CT and MRI findings were interpreted as being most consistent with a pancreatic neoplasm, and diagnosis of GDC was suspected only during surgery. The clinical presentation of gastric duplication cysts can be highly variable and nonspecific ranging from vague abdominal pain to nausea, vomiting, epigastric fullness, weight loss, anemia, dysphagia, dyspepsia with abdominal tenderness and epigastric mass on physical examination [4,10]. Because most cases occur along the greater curvature of the stomach, the cysts can potentially compress the adjacent organs such as pancreas, kidney, spleen, and adrenal gland. Accordingly, the differential diagnosis would include lesions arising from these organs [2]. The cysts may also be manifested by complications such as infection, gastrointestinal bleeding, perforation, ulceration, fistula formation, obstruction, compression, or carcinoma arising in the cysts [7, 8]. Up to 10% of gastric duplications may contain ectopic pancreatic tissue which may lead to pancreatitis and mimic a pancreatic pseudocyst [3, 8].

Because of the rarity of adult gastric duplications, it is difficult to outline their natural history with certainty.

As with the native gastric mucosa, the cyst lining may undergo erosions, ulceration, and regenerative changes. In noncommunicating cysts, increased fluid production may result in pressure-induced necrosis of the mucosa. These changes may lead to bleeding into the cyst or perforation into the peritoneal cavity.

Duplication cysts have the potential for neoplastic transformation. The production of oncofetal antigens raises the problem of a precancerous condition in long standing intestinal duplications [8]. Out of 11 reported cases of malignancy arising within the duplication cysts, 8 were adenocarcinomas [4]. Five of the carcinomas originated from gastric duplications. Adenomyoma arising from a gastric duplication has also been reported [4]. Malignancies arising from duplication cysts are likely to be present at advanced stages because of their unusual symptoms and difficulty of diagnosis [4].

Although it is difficult to diagnose GDC preoperatively, recent imaging modalities have provided some informative findings. CT scan and endoscopic ultrasound (EUS) are the best ways to identify GDC [8]. Classically, radiographic studies show an intramural filling defect indenting the gastric contour [8]. Contrast-enhanced CT scan typically demonstrates GDC as a thick-walled cystic lesion with enhancement of the inner lining [2]. Calcification is occasionally observed on CT. These findings are of diagnostic significance for GDCs [2]. However, since mucinous cystic tumors of the pancreas also show similar radiological features, GDCs adjoining the pancreas are indistinguishable from pancreatic mucinous cystic tumors based on these CT findings. Moreover, because the wall is sometimes thin, enhancement of the inner cyst wall is not always demonstrated. Generally, MRI can provide additional information about the cyst content compared to CT scan. However, the nature of the fluid in the GDC was reported to differ in each case according to bleeding, chronic inflammation, or infection. Therefore, MRI seems to be of less significance than expected in diagnosing GDCs [2].

EUS is useful in distinguishing between the intramural and extramural lesions of the stomach. When EUS demonstrates a cyst with an echogenic internal mucosal layer and a hypoechoic intermediate muscular layer, the diagnosis of GDC is highly likely[2]. The role of EUS-guided FNA in GDC is uncertain because (a) the cytological features of GDC may closely resemble those of mucinous pancreatic neoplasms, and (b) GDCs with elevated levels of CEA and CA19-9 have been reported, mimicking mucinous pancreatic neoplasms [4, 8, 11].

Complete removal is the treatment choice to avoid the risk of possible complications such as obstruction, torsion, perforation, hemorrhage, and malignancy [9, 10]. A noncommunicating GDC is classically treated by complete excision of the cyst and resection of the shared wall between stomach and the duplication cyst [8]. Communicating GDC usually requires no intervention when both gastric lumens are patent [8]. Drainage and marsupialization of the cyst have been suggested. However, marsupialization into the stomach exposes the unprotected mucosa of the cyst to gastric contents with the risk of ulceration [4]. Drainage procedures such as cystojejunostomy may be complicated by stenosis of the anastomosis or blind loop syndrome and therefore discouraged [4]. Furthermore, leaving the cyst in place is ill-advised given the potential for malignant transformation [4].

Conclusion

In summary, this unusual developmental anomaly should be included in the differential diagnosis of cystic masses of the gastrointestinal tract, and the possibility of malignancy should also be considered. While the diagnosis of gastrointestinal tract duplications may be suggested by imaging studies, more often the correct diagnosis is not established prior to surgery. Due to the risk of malignant transformation and other complications, GDCs should be treated surgically by complete resection.