25只犬和1只猫幽门-胃、十二指肠-胃及真性胃-胃套叠的影像学特征。 机翻

[内容请查看原文]

Introduction

Intussusceptions are defined as the prolapse of a region of the gastrointestinal tract into the lumen of an immediately adjacent region, typically in the direction of peristalsis. The most common site of intussusception in dogs is at the ileocolic junction [1, 2]; however, they can also occur in other locations and can also occur in a retrograde fashion [3]. Intussusceptions involving the stomach are less common. There are various types of gastric‐related intussusceptions, namely, true‐gastrogastric (true‐GG), duodenogastric (DG), pylorogastric (PG), and gastroesophageal intussusceptions [4, 5, 6, 7, 8]. In veterinary medicine, the term gastrogastric intussusception has at times been used interchangeably with PG and DG intussusceptions. A true‐GG intussusception involves prolapse of the lesser or greater curvature of the stomach into the gastric lumen without involvement of the pyloric region or duodenum. This definition is based on prior literature and consensus review within this study [4].

Intussusceptions are more common in animals younger than 1 year of age [5, 9, 10, 11, 12]. Although a definitive cause is often not identified, there are numerous suspected predisposing factors. These include gastroenteritis (nonspecific or viral‐caused), leptospirosis, parasitism, intestinal‐associated masses, foreign bodies, and recent general anesthesia (both in patients with or without laparotomy) [10, 11, 13, 14, 15]. They may also occur due to alterations in adjacent intestinal characteristics, such as a change in consistency, flaccidity, or alteration in diameter. Anything that encourages a kink or a fold in the bowel wall, such as an obstruction, an intraluminal mass, intestinal malrotation, or adhesion, may also be considered a predisposing factor for intussusception development [13, 15, 16, 17, 18, 19, 20].

Multiple case reports of gastric intussusception in dogs exist in the literature, with true‐GG and PG intussusceptions being most common [4, 5, 7, 21, 22, 23, 24, 25, 26, 27]. There are two published cases of DG intussusception—both of which reported some degree of concurrent pyloric involvement [8, 28]. Prior case reports have a broad age range from 17 weeks to 15 years, are equally distributed between males and females, and do not appear to have any predilection for specific breed or size of dog. Vomiting was reported in all cases, with other common clinical signs including abdominal pain, cardiovascular abnormalities, lethargy, and anorexia. Co‐morbidities included nephropathies (n = 4), azotemia (n = 4), Helicobacter spp. infection (n = 2), foreign body ingestion (n = 4), recent laparotomy (n = 2), intestinal‐associated masses (n = 2), pancreatitis (n = 2), and intestinal parasites (n = 1). Various combinations of abdominal ultrasound (US), radiographs, exploratory laparotomy, necropsy, endoscopy, and computed tomography (CT) were used for diagnosis.

Radiographic findings in these case reports have described a well‐defined soft tissue mass within the gastric lumen in some PG, DG, and true‐GG cases [4, 5, 8, 21, 25, 26, 27]. US findings have included visualization of invagination of the pylorus and/or duodenum or gastric wall with associated gastric wall thickening and the classic targetoid pattern indicative of intussusception [4, 7, 21, 22, 23, 25, 26, 27, 28]. CT findings have included clear visualization of the invaginated region and heterogeneous hypoattenuation of the invaginated gastric wall [4, 7].

This retrospective study focuses on intussusceptions involving the stomach (true‐GG and PG) and duodenum (DG), excluding gastroesophageal occurrences. There is no retrospective study comparing the imaging features of PG, DG, and true‐GG intussusceptions in dogs and cats. Therefore, the aims of the current study were to describe these diagnostic imaging characteristics. We hypothesized that some imaging features may be used to differentiate these intussusception types, which could be clinically useful for accurate diagnosis of this disease process and treatment options. A small number of case reports have reported spontaneous resolution of the intussusception, and thus patient management may differ depending on the intussusception type [24, 27].

[内容请查看原文]

Materials and Methods

This was a retrospective multicenter descriptive study. Cases were included if they had diagnostic imaging performed and a gastric intussusception (true‐GG, PG, DG) diagnosed either via US or CT, at the time of surgery, or at necropsy. No cases were excluded due to insufficient imaging or incomplete records. Gastroesophageal intussusception cases were not included. Ethical approval was not required for this study, as it did not involve animal intervention or experimental procedures.

[内容请查看原文]

Data Recording and Analysis

[内容请查看原文]

Medical Record Data

For all patients, the following data were recorded from their medical records: patient age at diagnosis, sex, breed, presenting clinical signs, concurrent diseases, and reported imaging findings (if available). If available, surgery and necropsy reports were also reviewed.

[内容请查看原文]

Image Analysis

All imaging studies were retrieved and reviewed independently by two ACVR board‐certified veterinary radiologists for a consensus on the intussusception type: PG intussusception was defined as invagination of the pylorus into the gastric lumen; true‐GG intussusception was defined as invagination of the greater or lesser curvature of the stomach into the gastric lumen; DG intussusception was defined as invagination of the duodenum into the gastric lumen (Figure 1). The reviewers were not blinded to any surgical or necropsy follow‐up. If there were discrepancies between reviewers, a consensus was reached by discussion until there was mutual agreement.

[图：Schematic images of a pylorogastric intussusception (A), a duodenogastric intussusception (B), and a true‐GG intussuscep]

Radiographs were evaluated for the degree of gastric distension, gastric contents, presence of gas in the pylorus/fundus/duodenum on different projections, serosal detail, and the presence or absence of an intraluminal soft tissue mass. US images and cine‐loops and CT images were evaluated for wall thickness, wall layering, the volume of mesenteric fat within the intussusception, gastric distension, gastric contents, peritoneal effusion, adjacent steatitis, involvement of the pancreas, and presence or absence of mucosal ulceration. Wall thickness measurements were given quantitative categories as follows: mild (0.5–0.7 cm), moderate (0.8–1 cm), moderate to marked (1–1.5 cm), or marked (>1.5 cm). Other parameters were graded subjectively. On CT images, the degree of contrast enhancement of the gastric wall was also evaluated.

The images were reviewed using commercially available imaging software (Horos—Nimble Co LLC d/b/a Purview in Annapolis, MD, USA). Group allocation was based on US or CT findings. In cases with radiographs only, the allocation was based on surgical/necropsy findings. If available, surgical/necropsy follow‐up was reported for those diagnosed via US/CT.

[内容请查看原文]

Results

Twenty‐five dogs and one cat met the inclusion criteria. They ranged in age from 7 weeks to 10 years (mean 3.04 years, median 1.75 years), with 12/26 being ≤1 year of age. Ten were castrated male dogs, nine intact male dogs, three male dogs with unknown castration status, two female spayed dogs, one female intact dog, and one female spayed cat. Multiple breeds were represented: five mixed‐breed dogs, three French Bulldogs, two Siberian Huskies, two Labrador Retrievers, two Jack Russell terriers, and one of each: Pit Bull Terrier, Bernese Mountain Dog, Shar‐Pei, Dachshund, Bichon Frise, Boxer, Chihuahua, English Bulldog, Cavapoo, Golden Retriever, Boston Terrier, and Domestic Shorthair cat. Table 1 shows the breakdown for the intussusception types.

Twenty‐one patients presented with clinical signs of vomiting (acute in eight), including six with hematemesis. Additional clinical signs included lethargy (n = 4), diarrhea with hyporexia/anorexia (n = 3), anorexia (n = 1), altered mentation (n = 1), tachycardia (n = 1), and coagulopathy (n = 1). One dog was presented for renal biopsy and dialysis, and one for surgical correction of an extrahepatic portosystemic shunt. Table 1 shows the breakdown for the intussusception types and reported comorbidities.

Imaging modalities performed included abdominal US, plain radiographs, positive contrast upper gastrointestinal (upper GI) radiographic studies, and CT. Twelve dogs had plain radiographs, and US was performed. The time between radiographs and US ranged from <1 to 12 h (median 1.5 h). Two dogs had plain radiographs only. Two dogs had upper GI radiographic studies; one was followed by a US examination. Nine patients (including the single cat) had a US only. One dog had radiographs, US, and CT, and one dog had a CT only. The imaging findings are summarized in Table 2 and outlined in detail below.

Fourteen patients were diagnosed with PG intussusception, nine with true‐GG intussusception (six greater curvature [true‐GG‐greater], three lesser curvature [true‐GG‐lesser]), and three with DG intussusception for a total of 26 cases. Twenty‐one of these cases were diagnosed and allocated into a group via US, two via CT, and three at surgery. Surgical or necropsy findings were available for 17 of the 23 cases diagnosed via US or CT and contradicted the imaging findings in two PG cases with regard to the intussusception type. Initial inter‐observer agreement for the 23 cases diagnosed via imaging was good (91%), with simple discussion resulting in complete agreement.

[内容请查看原文]

Pylorogastric Intussusception (14 Cases)

A PG intussusception was diagnosed in 14 dogs. Nine had plain abdominal radiographs (one plain radiograph only, one followed by an upper GI contrast study, seven followed by US), 11 had abdominal US (seven following radiographs, four US only), and one dog had a CT only. The PG intussusception was diagnosed via imaging in 12 (11 US, 1 CT) and two via surgery. Twelve out of 14 dogs diagnosed via imaging went to surgery or necropsy. The PG intussusception was confirmed in nine dogs, was described as a DG intussusception in two dogs, and one resolved by the time of surgery. In one case, the intussusception had resolved upon re‐checking the US the following day. The two cases that were described as DG intussusceptions at necropsy/surgery remained in the PG group for the purpose of this study due to the group allocation being on the basis of US findings.

On plain abdominal radiographs, the stomach was moderately to markedly distended in five and mildly distended in four. In 8/9 dogs, the stomach contained gas and a small/medium (n = 1), medium (n = 2), or large (n = 5), rounded, smoothly margined, soft tissue opacity mass (Figure 2A,B). This was located at the level of the body in four, body/fundus in two, and encompassed the majority of the stomach in two. This mass was seen in the ventrodorsal (VD) projection only (n = 2), left lateral (LL) projection only (n = 2), LL and VD projections (n = 2), and on all projections (n = 1). Only one dog had gas present in the pylorus and duodenum in the LL projection, and this was the dog without an intraluminal soft tissue mass. There was gas present in the fundus in the RL projection in all dogs. There was no radiographically evident gastrointestinal foreign material. A fat‐opaque crescent‐shaped or ring‐like area was superimposed with the center of the soft tissue mass in 3/9 dogs (33%) (Figure 3A).

[图：Four‐year‐old male castrated mixed breed dog that presented for acute vomiting. A pylorogastric intussusception was diag]

[图：A 18‐month‐old intact male mixed breed dog that presented for an acute onset of vomiting. A pylorogastric intussusceptio]

In the dog that had upper GI positive contrast radiographs performed following the plain abdominal radiographs, there was a large, rounded filling defect in the region of the pylorus seen in the VD and RL projections that was broad‐based to the pyloric region. Contrast media was not seen aborad to the stomach (Figure 4).

[图：Three‐view abdominal radiographs of a 6‐year‐old male castrated Dachshund that presented for acute vomiting. Left (A) an]

An abdominal US was performed in eleven dogs, and the pylorus was found to be intussuscepted into the gastric body region. Due to its anatomic relationship with the pylorus, the duodenum was displaced into the center of the intussusception, without passing through the lumen of the pylorus (Figure 1A). US findings included varying degrees of wall thickening (0.8–2.2 cm) of the intussusceptum in 10 cases, with maintenance of normal wall layering. One dog had normal wall thickness. All cases of thickening were predominantly of the submucosal layer, which was hypoechoic and demonstrated faint radial echogenic striations and prominent vasculature (slightly tortuous and increased in size) (Figure 5). A small to moderate amount of mesenteric fat was present within the intussusception in all cases, and the pancreaticoduodenal lymph node was present within the intussusception in five cases (Figures 1 and 2). A small portion of the pancreas was located within the intussusception in one dog, and at the periphery of the intussusception in another. Mild or moderate focal steatitis was present in the region of the intussusception in eight dogs, was described as diffuse in one, and absent in two. There was mild or mild to moderate gastric distension in nine cases, with moderate or moderate to marked distension in two cases. There was no intraluminal material or mucosal ulceration noted in any dog. A small volume of anechoic peritoneal fluid was noted in one dog and moderate in another.

[图：Five‐month‐old male intact Golden Retriever that presented for vomiting, anorexia, and diarrhea. A pylorogastric intussu]

On CT examination, the stomach was in the left cranial abdomen with marked gastric wall thickening (2.2 cm) of the intussusceptum that was predominantly of the submucosa with a small number of hyperattenuating non‐enhancing regions within the thickened submucosa. The wall layer definition was normal (Figure 6). There was minimal to no contrast enhancement of the affected submucosa (pre‐contrast HU value 26, post‐contrast HU value 28), and contrast enhancement of the affected mucosa was slightly reduced compared to the post‐contrast values of the unaffected gastric mucosa (75 vs. 95 HU).

[图：Six‐year‐old male castrated English Bulldog that presented for vomiting and lethargy. A pylorogastric intussusception wa]

[内容请查看原文]

True‐GG Intussusception (Nine Cases)

A true‐GG intussusception was diagnosed in nine cases; six were of the greater curvature and three of the lesser curvature. Of these, all nine patients had a US performed, and six had plain radiographs (four followed by US, one followed by upper GI contrast and US, and one followed by US and CT). Eight were diagnosed via US and one via US and CT. Three dogs went to surgery, and in all three, the intussusception had resolved, with one dog demonstrating adhesions in the region of the cardia and lesser curvature. Two dogs underwent necropsy, and in both, the intussusception had resolved. In one, there was severe gastric submucosal edema of the previously noted region of intussusception, and in the other, there was severe necrosis of the lesser curvature, which was the region seen to be intussuscepted on US/CT.

In the six dogs in which plain radiographs were performed, the stomach was mildly distended in two, moderately distended in one, markedly distended in one, and not distended in two (Figure 8A). An intraluminal soft tissue opaque mass was seen in three dogs, one large and two medium in size, which were broad‐based to the gastric wall (Figure 7). Two were located in the fundic region, and the other encompassed most of the stomach (body to fundus). Two of these masses were best seen in the RL projection, and one in the VD projection. A small volume of gas was present in the fundus in the RL projection in four dogs, with a rim of gas outlining the soft tissue mass in one. An LL projection was performed in 5/6 dogs, and gas was present in the pylorus in two. Serosal detail was reduced in three cases; however, two of these were young dogs (<10 weeks). There was no gas in the duodenum, and no radiographically evident gastrointestinal foreign material was present. In the dog that had upper GI contrast radiographs performed following plain abdominal radiographs, the findings were nonspecific. At the time of image acquisition, only a small volume of barium was present in the stomach, with the majority having progressed into the small intestine.

[图：A 4‐month‐old male castrated mixed‐breed dog that presented for hematemesis. (A) Ventrodorsal radiograph of the cranial ]

[图：(A) Four‐month‐old male intact Pit Bull Terrier that presented for a recent onset of vomiting. A true‐GG intussusception]

All cases had an abdominal US performed, one of which had a repeat US 3 days later. The greater curvature of the gastric wall was found to be intussuscepted into the gastric lumen in six cases and intussusception of the lesser curvature in three. Increased wall thickness ranged from mild to marked (0.8–1.8 cm) and was predominantly isolated to the intussusceptum (Figure 8B). One dog initially had mild to moderate generalized thickening (0.9 cm) of the gastric wall that was most severe in the fundic region. A repeat US examination 3 days later showed marked thickening of the intussusceptum (1.5 cm). Wall layering was maintained in eight cases and was slightly reduced in one case. The wall thickening was predominantly affecting the submucosal layer, which was hypoechoic and demonstrated faint radial echogenic striations. Mucosal ulceration was present in one case. There was mild gastric distension in three cases and moderate in one. Gastric contents were primarily echogenic fluid, with fluid and gas in two cases, and no intraluminal foreign material was noted in any case. Steatitis was present in seven cases and was mild focal (n = 3), mild diffuse (n = 1), moderate focal (n = 2), or moderate diffuse (n = 1). Peritoneal effusion was noted in four cases (small amount in three and moderate amount in one case).

CT examination was performed in one dog following radiographs and abdominal US. The stomach was in a normal position without dilation. The lesser curvature was invaginated into the gastric lumen with wall thickening extending from the cardia to pylorus, measuring up to 2.7 cm in thickness (Figure 8C). The majority of the wall thickening was of the submucosa. Wall layer definition was normal. There was minimal to no contrast enhancement of the affected submucosa (pre‐contrast HU 30, post‐contrast HU 33), and contrast enhancement of the affected mucosa was slightly reduced compared to the post‐contrast values of the unaffected gastric mucosa (75 vs. 100 HU). The gastric lymph node was moderately enlarged and uniformly contrast‐enhancing. A small volume of peritoneal effusion was noted, as well as regional fat stranding consistent with steatitis along the lesser curvature.

[内容请查看原文]

Duodenogastric Intussusception (Three Cases)

A DG intussusception was diagnosed in three dogs. One had abdominal radiographs only, and two had abdominal US only. The DG was diagnosed via US in two dogs. The dog that only had radiographs performed had the DG intussusception diagnosed at surgery. Surgery was also performed in the two dogs with US diagnosis; and both were confirmed as small DG intussusceptions; one also had pyloric mucosal hypertrophy.

In the case that had radiographs performed (RL and VD only), there was moderate gastric distension with gas outlining a large, rounded, smoothly margined, soft tissue opacity mass located in the body and fundic region. A moderately sized, well‐defined radiopaque foreign body (consistent with a ball) was present within the gastric fundus (Figure 9A). The pylorus and duodenum were not visualized, and serosal detail was normal.

[图：(A) Right lateral radiograph of the cranial abdomen of a 3‐year‐old female spayed boxer that presented for acute onset v]

Of the two cases diagnosed via US, the wall thickness and layering were normal, and only a small portion of the duodenum was intussuscepted. There was only a very small amount of fat within the intussusception in one case. No pancreaticoduodenal lymph node or pancreatic involvement was seen in either case. Gastric distension was moderate in both cases. In one case, there was intraluminal material present, extending from the pylorus into the duodenum with associated duodenal plication—consistent with a linear foreign body (Figure 9B). No mucosal ulceration, steatitis, or peritoneal effusion was observed in either case.

[内容请查看原文]

Discussion

PG intussusception was the most common classification in our study (14/26). Of the reported cases in the literature, a true‐GG was most common [4, 7, 21, 26]. However, on the basis of their description, one of these cases was a PG intussusception [21]. In addition, some reports diagnose a DG intussusception but describe concurrent invagination of both the duodenum and pylorus into the gastric lumen [8, 28]. Two cases in our study were described as gastrogastric in their imaging reports but were PG upon image review. This is likely due to the incorrect adoption of the generalized gastrogastric terminology for all intussusceptions of the stomach. However, this study provides clear differentiation between a PG or DG intussusception and true‐GG intussusceptions, particularly with the involvement of the duodenum seen at the center of a PG or DG intussusception, compared to the true‐GG intussusception that only involves the gastric wall. This highlights the importance of using appropriate descriptive nomenclature when describing and concluding/diagnosing the intussusception type. The ultimate diagnosis may affect patient management options, particularly because the majority of our true‐GG cases spontaneously resolved—potentially indicating that these may not require immediate surgical intervention.

Just under half (12/26 or 46%) of the cases in this study were less than 1 year of age. This is consistent with prior literature regarding enteric intussusceptions, which have been previously reported to be more common in young patients [5, 9]. However, it is worth noting that the oldest patient was 10 years old at the time of diagnosis, so this should not be excluded from the differential list in older patients. The average age of the PG group was higher than that of the true‐GG and DG groups, which could mean that PG intussusception may be more likely to occur in older patients. Most patients in this study were male (22/26, 85%), which could suggest an increased risk in male dogs—this contrasts with the even male to female distribution in previous case reports. In human pediatrics, there is a slight male predominance for enteric intussusceptions [29]. It is unknown whether there is a factor that predisposes males to such conditions, such as hormonal or anatomic. Case selection bias, given the nature of this study, cannot be excluded as a reason for this disparity. Similar to the previous reports on gastric‐related intussusceptions in dogs and cats, vomiting was the most common clinical sign [4, 5, 7, 8, 21, 22, 23, 24, 25, 26, 27, 28]. It is unclear what condition takes precedence—whether the retrograde peristaltic contractions associated with vomiting cause the intussusception, or a primary intussusception with secondary vomiting.

PG intussusceptions had the most consistent radiographic findings with a variably sized soft tissue opaque mass within the gastric lumen, and a lack of gas in the pyloric region in the LL projection. This mass was best seen in the LL and VD projections, likely due to the displacement of intraluminal fluid and gas (Figures 2A,B and 3A). These findings are similar to four previous PG case reports [5, 21, 25]. One dog did not have a mass present on radiographs with normal gas‐filling of the pylorus and duodenum but had a PG intussusception diagnosed on US 12 h later. The intussusception in this case likely occurred during that 12‐h timeframe. Only one of the DG intussusception cases had abdominal radiographs, and this patient had similar radiographic findings to the PG cases (Figure 9A). Despite this low number for comparison, it is worth noting that these two intussusception types could appear similar radiographically. Other differentials for intraluminal gastric mass‐like lesions or gastric wall thickening include hypertrophic pyloric stenosis, chronic gastropathies, gastritis, neoplastic processes (adenocarcinoma/adenoma, lymphoma, leiomyosarcoma/leiomyoma, and malignant histiocytoma), eosinophilic sclerosing fibroplasia (cats), and granulomas (e.g., pythiosis) [30]. It may be challenging to differentiate the mass associated with a PG or DG intussusception from a neoplastic mass based on radiographs alone; however, a neoplastic mass would be less likely in a young dog. The lack of gas in the pyloric region could help determine lesion location, with a PG or DG intussusception being ruled out in those where the pylorus has a normal appearance in an LL projection.

A thin, fat opaque ring, typically in the orientation of a circle or oval, was inconsistently seen superimposed with the mass (Figure 3A). We suggest that this finding may indicate the degree of mesenteric/peri‐gastric fat involved in the intussusception, as this radiographic finding correlated with the degree of intussuscepted adipose tissue on US (Figure 3B). This characteristic finding was not commonly observed in our study; however, it is a finding that may help increase the suspicion that the mass is a PG intussusception.

The radiographic findings of dogs with true‐GG intussusception were more variable, similar to the previously reported cases [4, 27]. Three dogs demonstrated a soft tissue opaque intraluminal mass; however, this was more broad‐based in appearance and less well‐defined compared to the PG and DG findings (Figures 2A,B, 7A,B and 9A). They were best seen in the RL and VD projections compared to the PG and DG, which showed the mass on the LL and VD projections. Pyloric gas was not seen in two true‐GG dogs in the LL projection, which may be secondary to the intussusceptum with associated mural edema extending to the level of the pylorus, or the small volume of gastric gas that was present in these patients. The varied radiographic appearance is likely due to the location of the intussusception (greater vs. lesser curvature), a variable degree of mural edema, and reduced volume of gastric gas. These findings are based on only three true‐GG cases that had radiographs performed; thus, the application of these findings to the greater population should be considered with caution.

Upper GI radiographic contrast studies allowed for better visualization of a PG intussusception but were inconclusive for the true‐GG intussusception. In the PG case, the large soft tissue opacity filling defect in the pyloric region was well outlined, and the lack of contrast media entering the duodenum was likely due to the intussusception causing a pyloric outflow obstruction. The inconclusive findings in the true‐GG case were likely secondary to inappropriate timing of radiographs following contrast administration, as the majority of contrast media were present in the intestines and only a small volume remained in the stomach at the time the images were acquired. Unfortunately, the timing/volume of contrast media administration was not available for review.

US helped diagnose the presence of invaginated tissue into the gastric lumen in all subtypes. Wall thickening of the intussusceptum was common and, at times, included the adjacent wall of the intussuscipiens. Increased wall thickness was most prominent in PG and true‐GG cases, and the majority of the thickening was of the submucosal layer (Figure 5). This is likely due to venous congestion with secondary edema of the heavily vascularized submucosa [30]. The true‐GG case that had a recheck US several days after initial diagnosis exhibited progressive thickening of the wall, which could represent increasing intussusception severity or increasing edema of the wall. This finding suggests the necessity of prompt diagnosis and follow‐up. The radial striations of the submucosa likely represented blood vessels or the separation of collagen fibers [31, 32]. Determining which gastric wall layer is thickened can help narrow down the differential list, as other etiologies of gastric wall thickening, such as muscular hypertrophy, would affect the muscularis layer [33], and uremic gastritis tends to affect the mucosal layer [34]. Additionally, all cases had maintenance of wall layer definition, which differs from many neoplastic and infectious etiologies [35].

The classic targetoid pattern of concentric rings as seen in PG and DG cases (Figures 2C and 9B) was not seen in true‐GG intussusceptions, likely due to this being more of an invagination of the gastric wall rather than telescoping of different portions of a tubular portion of the gastrointestinal tract (Figure 8B). Doppler images were not available for all patients, but evaluation of possible vascular compromise to the affected gastric wall could be useful, as this may affect surgical intervention and patient prognosis.

US was useful in detecting concurrent pathology, such as a linear foreign body and intestinal parasites (Figure 9B). These likely led to altered peristalsis, which could have contributed to the intussusception as was suspected in prior cases [4, 26, 28]. Regional steatitis was present in multiple cases, likely secondary to inflammation/edema of the gastric wall; however, in some PG and DG cases, the pancreas was either very close to or partially involved in the intussusception, which could cause secondary pancreatic compromise and pancreatitis that could also cause regional steatitis. Concurrent pancreatitis has also been described in case reports [27, 28].

Despite US being useful for determining the intussusception type, definitive differentiation between PG and DG was challenging in some cases. The sonographic appearance of the pyloroduodenal junction has been described in small adult dogs, which revealed that 85.7% of these dogs had hyperechogenicity in that region [36]. This could be used to determine the intussusception type by definitive visualization of the pyloroduodenal junction, but given that this hyperechogenicity is not present in all dogs, differentiation of PG and DG may remain challenging. Although there were only a small number of DG cases, there appeared to be less mesenteric fat within the intussusception when compared to PG intussusceptions, possibly related to the more restricted space within the intussusception in the DG subtype. DG intussusceptions may also be less likely to demonstrate severe wall thickening, which could reflect a less severe or less clinically significant intussusception. These findings could possibly be used to determine the intussusception type. The image review was not blinded to surgical or necropsy results. This could introduce the potential for confirmation bias; however, care was taken to systematically evaluate the images and allocate them to a group based on those findings alone. Despite being aware of the outcomes, two PG cases were found to be DG at surgery/necropsy. If a gastrotomy is not performed at the time of reduction of the intussusception, it is possible that a PG could be misdiagnosed as a DG or vice versa due to a similar appearance on external visualization at surgery.

CT was useful in confirming the intussusception type (Figure 6A–C). It also gave information about the degree of contrast enhancement of the gastric wall, which was reduced in the markedly thickened intussusceptum. This was likely secondary to edema/congestion, which has been previously described [32], and possible early necrosis, which could affect prognosis since recommendations would drastically change if a significant portion of the gastric wall was devitalized. Multiphasic CT was not performed in either case but could be useful to determine the degree of vascular compromise in the affected gastric wall. Specifically, to detect delayed/compromised arterial enhancement of the gastric wall, wall thickening/congestion to suggest venous compromise, and reduced enhancement in the delayed phase, suggesting ischemia and necrosis, which may affect the choice of treatment for that patient. It should be noted that the reduced contrast enhancement is an observation from two cases only and may not reflect changes present in all cases. The moderately enlarged and uniformly contrast‐enhancing gastric lymph node in one case likely represented a reactive change, but definitive diagnosis was not obtained via sampling.

Endoscopy was not performed in any of our cases; however, it could be useful for assessing mucosal viability, identification of gastric foreign material, and possibly identifying the intussusception and any underlying lesions. In one case report, a PG intussusception was diagnosed via endoscopy, and they reported dynamic movement of the intussusception, which had spontaneously resolved by the time of surgery, so it is possible that endoscopy could be used to assess if there is dynamic movement of the intussusception that could indicate an increased chance of spontaneous resolution [24]. However, in another case report, endoscopy was able to identify a soft tissue mass, but it was only diagnosed as an intussusception at surgery, and they recommended abdominal US in future cases [8].

Of the five true‐GG cases with surgical or necropsy follow‐up, all had resolved, and one PG case that underwent surgery had resolved. One PG case had resolved on follow‐up US examination the following day. We suspect that the resolution of these cases may be secondary to muscular relaxation of the gastric wall during sedation/general anesthesia or euthanasia. This has also been described in prior case reports of PG and true‐GG intussusception [22, 23, 25, 30]. In all cases, there was evidence of edema, hemorrhage, necrosis, or adhesions that suggested the region had been previously invaginated and was unlikely to be an incidental finding. We suspect spontaneous resolution is more common in true‐GG cases due to the decreased constriction of the invagination. With this type of intussusception, it is more of a focal infolding/invagination of the gastric wall rather than the hallmark “telescoping” of one of the more tubular regions, such as with PG or DG. This finding may significantly impact the prognosis and treatment of true‐GG cases.

One DG case with evidence of intussusception on US had pyloric mucosal hypertrophy and only a very small intussusception present at surgery. It is possible that the intussusception had mostly spontaneously resolved or that the appearance of pyloric mucosal hypertrophy could artifactually increase the appearance of the degree of duodenal involvement on US. Thus, in these patients, extra care should be taken when evaluating this region, as these conditions can occur simultaneously.

There were several concurrent conditions that could have contributed to altered GI peristalsis, including a gastric foreign body, linear foreign body, pyloric mucosal hypertrophy, recent gastrotomy, sliding hiatal hernia, giardiasis, duodenal parasites, and pancreatitis. Whether some of these were coincidental or causative is unknown; however, a number of case reports had similar comorbidities [4, 5, 7, 8, 25, 26, 28]. There were many other concurrent conditions, including Lyme disease, acute nephritis, extrahepatic portosystemic shunt, low hematocrit, lymphoma versus acute myeloid leukemia, splenic abnormalities, liver failure, hemophilia A, and amyloidosis. It is unclear how or if these contributed to the formation of the intussusception.

There are several limitations to this study. The retrospective and multi‐institutional nature of this study is a limitation due to the variation in the diagnostic imaging performed and the availability of complete medical records. Given the lack of surgical or necropsy confirmation in some cases, it is unclear if some cases could have fallen into a different category (e.g., PG being surgically diagnosed as DG and vice versa), or if they could have spontaneously resolved. The radiologists in this study were not blinded to surgical/necropsy outcomes, which introduces the possibility of confirmation bias. The small DG group limits the application of the comparative statements we make within this manuscript. Ultrasonographic Doppler evaluation was not performed in most cases, which would have provided additional insight into vascular compromise and tissue viability.

There was no observable difference in the imaging findings and the reported duration of clinical signs or outcome. Given the retrospective and multi‐center nature of this study, more detailed data regarding the clinical signs were not available. It is not known if the more chronic clinical signs were due to a chronic intussusception or an underlying disease that caused an acute intussusception with escalation of clinical signs that prompted the owner to seek medical care.

There was only one cat affected in this study. It is unknown if this is due to case selection bias from the retrospective and multi‐center nature of this study or if they are less likely to be affected by gastric intussusception due to differing anatomy. The feline stomach is typically “J‐shaped” compared to the dog's normal gastric shape and is oriented more parallel to the spine than in dogs, with a thinner gastric wall [37]. These differences may cause the cat's stomach to be less susceptible to these intussusception types.

Abdominal radiographs are often the first line of imaging in patients presenting with abdominal signs; thus, in many patients with these gastric intussusceptions, this is likely to be the initial imaging performed and should consist of the three standard views. This study has shown that there are some radiographic features that could be highly suggestive of a PG intussusception. However, given that some are nonspecific or additional confirmation may be required, an abdominal US could follow to determine the presence and type of intussusception. An additional advantage of US is its use in the evaluation of underlying disease that may be undetected on radiographs. CT could be considered if the US findings are unclear or if an US by a competent individual is unavailable. Endoscopy could be used as an adjunct to evaluate the gastric mucosa, presence of foreign material, or underlying disease but is confined to evaluation of the luminal surface.

In conclusion, specific radiographic imaging features can increase suspicion of PG or DG intussusception. Dogs with true‐GG intussusception have nonspecific radiographic changes. US is useful for evaluating and classifying the intussusception type; however, in some cases, definitive differentiation between PG and DG intussusceptions can still be challenging. Abdominal CT is also helpful in determining the intussusception type and assessing mural perfusion, which may alter prognosis should necrosis be present. It is possible that true‐GG intussusceptions are more likely to spontaneously resolve than PG or DG cases. Understanding the imaging features of gastric‐related intussusceptions is important for patient prognosis and management.

Author Contributions

Anthony Fischetti: resources, writing – review and editing. Jaime Rechy: resources, writing – review and editing. Kelsey Cline: resources, writing – review and editing. Christine Gremillion: resources, writing – review and editing. Pablo Espinosa‐Mur: resources, writing – review and editing. Wilfried Mai: resources, writing – review and editing. Linda Dillenbeck: conceptualization, investigation, writing – original draft, visualization, resources. Alexandra Gibalski: writing – review and editing, resources. Megan Stadler: conceptualization, investigation, writing – review and editing, visualization. Dylan Burton: conceptualization, investigation, writing – original draft, visualization.

Disclosure

Preliminary results of this study have been presented at the annual ACVR Conference in October 2025. CARE checklist that was adapted to a case series.

Conflicts of Interest

The authors declare no conflicts of interest.

Acknowledgments

The authors thank Katherine Phillips, Yannick Ruel, Charles Artiles, Amy Habing, and Jean Reichle for their contributions.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

[内容请查看原文]

Associated Data

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.