ttp://www.bsava.com
PAPER
Morphology of splenocaval congenital portosystemic shunts in dogs and cats R. N. White and A. T. Parry Willows Referral Service, Shirley, Solihull, West Midlands B90 4NH R. N. White’s current address is School of Veterinary Medicine and Science, University of Nottingham, Sutton Bonington Campus, Leicestershire LE12 5RD
OBJECTIVE: To describe the anatomy of congenital portosystemic shunts involving the splenic vein communicating with the caudal vena cava at the level of the epiploic foramen. MATERIALS AND METHODS: A retrospective review of a consecutive series of dogs and cats managed for congenital portosystemic shunts. RESULTS: Ninety-eight dogs and eight cats met the inclusion criteria of a congenital portosystemic shunt involving the splenic vein communicating with the prehepatic caudal vena cava plus recorded intra-operative mesenteric portovenography or computed tomography angiography and gross observations at surgery. All cases (both dogs and cats) had a highly consistent shunt that involved a distended gastrosplenic vein that communicated with the caudal vena cava at the level of the epiploic foramen via an anomalous left gastric vein. CLINICAL SIGNIFICANCE: The morphology of the shunt type described appeared to be a result of an abnormal communication between the left gastric vein and the caudal vena cava and the subsequent development of preferential blood flow through an essentially normal portal venous system. The abnormal communication (shunt) was through the left gastric vein and not the splenic vein, as might have been expected. This information may help with surgical planning in cases undergoing shunt closure surgery. Journal of Small Animal Practice (2016) 57, 28–32 DOI: 10.1111/jsap.12414 Accepted: 23 September 2015; Published online: 13 November 2015
INTRODUCTION Methods described for the imaging of congenital portosystemic shunts (PSSs) include ultrasonography (Lamb 1996, Szatmári & Rothuizen 2006), magnetic resonance angiography (Sequin et al. 1999, Bruehschwein et al. 2010, Mai & Weisse 2011), computed tomography angiography (CTA) (Frank et al. 2003, Zwingenberger et al. 2005, Nelson & Nelson 2011, White & Parry 2013), findings on intra-operative mesenteric portovenography (IOMP) (White et al. 2003, White & Parry 2013), direct gross observations at surgery (White & Parry 2013, White & Parry 2015) and the examination of corrosion casts made at postmortem examination from individuals suffering from an extrahepatic portosystemic shunt (EHPSS) (Szatmári & Rothuizen 2006). Using these imaging techniques, it has proved possible to broadly classify congenital PSS as either intrahepatic or extrahepatic (Payne et al. 1990, Martin 1993, Levy et al. 1995, Lamb & White 1998, Tillson & Winkler 2002, Hunt 2004) with further 28
sub-classification of EHPSSs commonly being restricted to either porto-caval or porto-azygos (Szatmári et al. 2004). Recently, congenital EHPSSs involving the left phrenic and azygos veins, and right gastric vein were independently described in detail using a combination of CTA, IOMP, and gross anatomical findings (White & Parry 2013, White & Parry 2015). These studies concluded that the left gastric vein commonly represented the anomalous vessel (shunt) communicating with the systemic vein. In addition, the morphology of each shunt type described was shown to be the result of an abnormal communication between the left gastric vein and the systemic vein, and the subsequent development of preferential blood flow through essentially normal portal vessels within the portal venous system. The purpose of this study was to define the morphology of congenital EHPSSs that have previously been named “splenocaval shunts” in dogs and cats using the same methodology in a series of consecutive clinical cases.
Journal of Small Animal Practice
•
Vol 57
•
January 2016
•
© 2015 British Small Animal Veterinary Association
Portosystenic shunts involving the splenic vein
MATERIALS AND METHODS This retrospective study reviewed dogs and cats seen by the authors between 1997 and 2014 for the investigation and management of congenital PSS. The main inclusion criterion was that all cases must have a congenital PSS that involved the splenic vein communicating with the prehepatic vena cava at the level of the epiploic foramen. In addition, all cases must have undergone recorded IOMP and direct gross observations at the time of surgery. After 2009, some cases also underwent preoperative CTA. Data on breed, signalment (age, sex, neutering status), imaging investigation, type of PSS and gross surgical findings were collected and reviewed. Shunts that involved the splenic vein communicating with the prehepatic vena cava at the level of the epiploic foramen were separated from the main body of shunts collected and reviewed. CTA was performed under anaesthesia using a 16-slice multidetector unit (Brightspeed; General Electric Medical Systems) as described previously (White & Parry 2013, White & Parry 2015). Briefly, images were acquired using a 0·625 or 1·25-mm slice thickness, depending on the size of the animal, 120 kVp and variable mAs. Scanned field of view and displayed field of view were selected according to the size of the animal. The pitch was 0·938. Pre- and postintravenous contrast (600 mg I/kg iopromide, Ultravist; Bayer PLC) images were obtained using a standard algorithm and a 512×512 matrix, and viewed using a window and level optimised for soft tissue (window 400 HU, level 50 HU). Contrast was injected at a speed of 2·0 mL/second using a pressure injector. To optimise contrast enhancement, a transverse slice over the porta hepatis was selected and repetitively examined whilst contrast injection was performed. At the onset of opacification of the portal vessels, a complete abdominal CTA examination was performed using proprietary bolus tracking software with an automated trigger threshold of 120 HU to start the scan. The trigger region of interest was positioned over the portal vein at the level of the porta hepatis in all dogs, in the central aspect of the vessel to allow for respiratory motion. Studies were assessed in their native format, using multi-planar reformatting and using surface-shaded volume rendering. Vascular maps were obtained and postprocessing was limited to removal of arterial vessels and unnecessary portions of the caudal vena cava (CVC) from the maps. All CTA studies were reviewed by both authors. In addition, a number of normal CTA studies in dogs were reviewed for the purposes of cross reference. IOMP was carried out during surgery by using a mobile image intensification unit obtaining ventrodorsal images of the cranial abdomen (White et al. 2003, White & Parry 2015). Images were obtained before the manipulation of the shunt and during the temporary full ligation of the shunting vessel. Angiograms were recorded and reviewed by both authors. The gross anatomy of the shunt was recorded in the surgical report for each case. Information recorded included the course of the distended vasculature, any obvious tributary vessels and its entrance into the CVC. Using the combined data of IOMP, gross findings during surgery and CTA, the morphology of EHPSSs involving the splenic Journal of Small Animal Practice
•
Vol 57
•
January 2016
•
vein communicating with the prehepatic vena cava at the level of the epiploic foramen was compared. On the basis of this combined data, the anatomy of this shunt type was described and evaluated in both the dog and cat.
RESULTS In total, 169 dogs and 12 cats were found that had shunts that involved the splenic vein communicating with the prehepatic vena cava at the level of the epiploic foramen. Of these, 98 dogs and 8 cats met the inclusion criteria – having both recorded IOMP and complete descriptions of the direct gross observations at the time of surgery. The median age of dogs that met the inclusion criteria was 10 months (range 2 to 108 months). Of these dogs, 54 were male and 44 were female. Affected breeds were Yorkshire terrier (n=14), West Highland white terrier (n=12), Cairn terrier (n=10), Jack Russell terrier (n=8), shih-tzu (n=8), bichon frise (n=6), miniature schnauzer (n=6), crossbred (n=5), Norfolk terrier (n=5), pug (n=4), Maltese terrier (n=3), Border terrier (n=2), Pekinese (n=2), cavalier King Charles spaniel (n=2), Chihuahua (n=2), Lhasa apso (n=2), papillon (n=2), beagle (n=1), English setter (n=1), giant schnauzer (n=1), Shetland sheepdog (n=1) and Staffordshire bull terrier (n=1). The median age of cats that met the inclusion criteria was 7 months (range 3 to 42 months). Of these cats, three were male and five were female. Affected breeds were domestic short hair (n=2), Persian (n=2), British short hair (n=1), Havanese (n=1), Siamese (n=1) and tonkinese (n=1). In addition to IOMP that was performed in all cases, CTA was performed in seven dogs and one cat. The morphology of shunts involving the splenic vein communicating with the prehepatic CVC at the level of the epiploic foramen was consistent in all the cases identified. The following description was based on the findings of CTA, IOMP and gross findings at the time of surgery. Figure 1 shows a diagram of a normal portal vasculature for cross reference. The anomalous vessel arose from an enlarged gastrosplenic and tributary left gastric vein at the level of the angular notch (incisura angularis) on the dorsal wall of the pyloric aspect of the stomach in both the dog (Fig 2A to C) and the cat (Fig 3A to C). The enlarged left gastric vein continued as the anomalous vessel passing in a dorso-medial direction towards the prehepatic CVC where it entered the cava on the left side at the level of the epiploic foramen (Fig 4). There was very little variation in the morphology of the shunting vessel; the sole variability appeared related to the relative lengths of the contributory vessels: the gastrosplenic vein, the tributary left gastric vein and its continuation as the anomalous shunting vessel before its communication with the prehepatic CVC.
DISCUSSION The results of this study revealed that EHPSSs involving the splenic vein communicating with the prehepatic CVC at the level
© 2015 British Small Animal Veterinary Association
29
R. N. White & A. T. Parry
FIG 1. A diagram showing the normal portal vasculature in dogs and cats (Modified from Miller et al. 1964). Gd v gastroduodenal vein, Gs v gastrosplenic vein
FIG 2. (A) A figure showing the salient features of the IOMP and gross observations at surgery obtained from dogs with a left gastrocaval shunt. The shunt consisted of a distended gastrosplenic vein that communicated with the caudal vena cava (CVC) at the level of the epiploic foramen via an anomalous left gastric vein. The tributary left gastric vein and splenic vein (shaded grey) were not apparent in all cases on IOMP. (B) An example of a ventrodorsal IOMP of a 12-month-old male miniature schnauzer. Note the enlarged gastrosplenic vein draining into an enlarged, anomalous left gastric vein before entering the prehepatic CVC at the level of the epiploic foramen. (C) This image shows a surface-shaded volume rendered computed tomography angiography (CTA) of the same dog shown in (B). Similar to (B), note the enlarged gastrosplenic vein draining into an enlarged, anomalous left gastric vein before entering the prehepatic CVC at the level of the epiploic foramen. CP-DV Cranial pancreatic-duodenal vein, LGV Left gastric vein
of the epiploic foramen were consistent, allowing their anatomical description in both dogs and cats. There was very little variation between the cases examined: in each the portal blood was 30
shunted initially through an enlarged gastrosplenic vein, then through its tributary left gastric vein (passing along the dorsal aspect of the pyloric antrum), before continuing, at the level of
Journal of Small Animal Practice
•
Vol 57
•
January 2016
•
© 2015 British Small Animal Veterinary Association
Portosystenic shunts involving the splenic vein
FIG 3. (A) A figure showing the salient features of the IOMP and gross observations at surgery obtained from cats with a left gastrocaval shunt. Similar to dogs, the shunt consisted of a distended gastrosplenic vein that communicated with the caudal vena cava (CVC) at the level of the epiploic foramen via an anomalous left gastric vein. The tributary left gastric vein and splenic vein (shaded grey) were not apparent in all cases on IOMP. (B) An example of a ventrodorsal IOMP of a 10-month-old female domestic short hair cat. Note the enlarged gastrosplenic vein draining into an enlarged, anomalous left gastric vein before entering the prehepatic CVC at the level of the epiploic foramen. (C) This image shows a surface-shaded volume rendered computed tomography angiography (CTA) of an equivalent shunt to that shown in (B). Similar to (B), note the enlarged gastrosplenic vein draining into an enlarged, anomalous left gastric vein before entering the prehepatic CVC at the level of the epiploic foramen. CP-DV Cranial pancreatic-duodenal vein, G-S V Gastrosplenic vein
FIG 4. An intraoperative view of a 13-month-old male Cairn terrier. The anomalous vessel (shunt) can be seen entering the prehepatic caudal vena cava at the level of the epiploic foramen
the angular notch of the stomach, as the anomalous vessel entering the CVC at the level of the epiploic foramen. Any variation in observed shunt morphology was minor and appeared to be associated with variation in the lengths of the vessels through which the portal blood was being shunted, i.e. the gastrosplenic vein, the tributary left gastric vein or the anomalous vessel entering the CVC. No measurements of vessel length were undertaken in this study, and further studies will be required if this variation in shunt morphology is to be further investigated. Naming conventions of EHPSSs have not always been clear or specific (Johnson et al. 1987, Payne et al. 1990, Levy et al. 1995, Szatmári et al. 2004, Nelson & Nelson 2011). Shunts entering the prehepatic CVC have frequently been described as portocaJournal of Small Animal Practice
•
Vol 57
•
January 2016
•
val without providing further information regarding shunt origin or morphology (Berent & Tobias 2012). Reasons for this lack of detailed description relate predominantly to the method by which the shunt is imaged. In most instances, all imaging modalities will confirm a PSS, but the ability of each to allow for accurate description of the anatomy of the shunt is open to observer interpretation and wide variation. Recently, with the advent of CTA, attempts have been made to classify portocaval EHPSSs in more detail on the basis of their origin, course and morphology (Nelson & Nelson 2011, White & Parry 2013, White & Parry 2015). In this study, the anatomy of the PSS was characterised in the majority of individuals using the results of the IOMP and gross anatomical findings at surgery. The findings were consistent in all cases allowing the classification of these shunts as described. This shunt type has been previously described as either a splenocaval or a gastrosplenocaval EHPSS (Hunt 2004, Hunt et al. 2004, Szatmári et al. 2004, Szatmári & Rothuizen 2006, Nelson & Nelson 2011, Fukushima et al. 2014). In addition, a number of reports have acknowledged the possible involvement of the left gastric vein in this shunt type (Szatmári et al. 2004, Cullen et al. 2006). The use of preoperative CTA in a number of cases after 2009 confirmed, and never contradicted, the anatomical descriptions obtained from both IOMP and gross anatomical findings at surgery. The main additional information provided by CTA was in defining the draining tributary veins of the portal vasculature that could not be seen on the IOMP or, in some cases, at the time of surgery. By anatomical convention, vascular shunts are most commonly named using the name of the portal vessel from which the shunt emanates and the name of the systemic vein it joins and supplies (Payne et al. 1990). Using this convention, by combining our results of IOMP, gross intra-operative surgical findings and CTA, we conclude that this previously described
© 2015 British Small Animal Veterinary Association
31
R. N. White & A. T. Parry
“splenocaval” PSS might be more accurately described as a left gastrocaval shunt. Nelson & Nelson (2011) also described an “atypical splenocaval” shunt with a short, wide shunting vessel with no evidence of continuation of the portal vein extending cranially from the shunt insertion into the prehepatic CVC. This shunt variation was not seen in either the dog or the cat in this study. It is hoped that further studies will identify this shunt type and allow it to be classified in a more detailed manner. The findings of this study support the previous suggestion that, embryologically, it is the development of the left gastric vein that is critical in the formation of this shunt type (White & Parry 2015). In this study, the anomalous vessel that joined with the prehepatic CVC appeared to be a continuation of the left gastric vein. This vein is normally programmed to connect with the splenic vein and, subsequently, the portal vein in the developing embryo (Noden & de Lahunta 1985). This inappropriate communication between an anomalous portion of the left gastric vein and the prehepatic CVC at the level of the epiploic foramen resulted in the development of the consistent shunt type described. In both dogs and cats, there was a highly consistent PSS which involved a distended gastrosplenic vein that communicated with the prehepatic CVC via an anomalous left gastric vein. The site of entrance of the shunt into the CVC was also highly consistent in every case – at the level of the epiploic foramen. The morphology of this shunt appeared to be the result of the abnormal communication between the anomalous left gastric vein and the CVC and the development of preferential hepatopetal blood flow through essentially normal gastrosplenic and left gastric veins. Conflict of interest None of the authors of this article has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper. References Berent, A. C. & Tobias, K. M. (2012) Hepatic Vascular Anomalies. In: Veterinary Surgery: Small Animals. Eds K. M. Tobias and S. A. Johnston. Saunders Elsevier, St Louis, MO, USA. pp 1624-1658 Bruehschwein, A., Foltin, I., Flatz, K., et al. (2010) Contrast-enhanced magnetic resonance angiography for diagnosis of portosystemic shunts in 10 dogs. Veterinary Radiology and Ultrasound 51, 116-121 Cullen, J. M., van den Ingh, T. S. G. A. M., Bunch, S. E., et al. (2006) Morphological classification of circulatory disorders of the canine and feline liver. In: WSAVA Standards for Clinical and Histological Diagnosis of Canine and Feline Liver Disease. Saunders Elsevier, Edinburgh, UK. pp 41-59
32
Frank, P., Mahaffey, M., Egger, C., et al. (2003) Helical computed tomography portography in ten normal dogs and ten dogs with a portosystemic shunt. Veterinary Radiology and Ultrasound 44, 392-400 Fukushima, K., Kanemoto, H., Ohno, K., et al. (2014) Computed tomographic morphology and clinical features of extrahepatic portosystemic shunts in 172 dogs in Japan. Veterinary Journal 199, 376-381 Hunt, G. B. (2004) Effect of breed on anatomy of portosystemic shunts resulting from congenital diseases in dogs and cats: a review of 242 cases. Australian Veterinary Journal 82, 746-749 Hunt, G. B., Kummeling, A., Tisdall, P. L., et al. (2004) Outcomes of cellophane banding for congenital portosystemic shunts in 106 dogs and 5 cats. Veterinary Surgery 33, 25-31 Johnson, C. A., Armstrong, P. J. & Hauptmann, J. G. (1987) Congenital portosystemic shunts in dogs: 46 cases (1979-1986). Journal of the American Veterinary Medical Association 191, 1478-1483 Lamb, C. R. (1996) Ultrasonographic diagnosis of congenital portosystemic shunts in dogs: results of a prospective study. Veterinary Radiology and Ultrasound 37, 281-288 Lamb, C. R. & White, R. N. (1998) Morphology of congenital intrahepatic portocaval shunts in dogs and cats. Veterinary Record 142, 55-60 Levy, J. K., Bunch, S. E. & Komtebedde, J. (1995) Feline portosystemic vascular shunts. In: Kirk’s Current Veterinary Therapy XII. Small Animal Practice. Ed J. D. Bonagura. W.B. Saunders, Philadelphia, PA, USA. pp 743-749 Mai, W. & Weisse, C. (2011) Contrast-enhanced portal magnetic resonance angiography in dogs with suspected congenital portal vascular anomalies. Veterinary Radiology and Ultrasonography 52, 284-288 Martin, R. A. (1993) Congenital portosystemic shunts in the dog and cat. Veterinary Clinics of North America Small Animal Practice 23, 609-623 Miller, M. E. (1964) The venous system. In: Anatomy of the Dog. Ed M. E. Miller. W.B. Saunders, Philadelphia, PA, USA. pp 389-429 Nelson, N. C. & Nelson, L. L. (2011) Anatomy of extrahepatic portosystemic shunts in dogs as determined by computed tomography angiography. Veterinary Radiology and Ultrasonography 52, 498-506 Noden, D. M. & de Lahunta, A. (1985) Cardiovascular system III: venous system and lymphatics. In: The Embryology of Domestic Animals – Developmental Mechanisms and Malformations. Williams & Williams, Baltimore, MA, USA. pp 257-269 Payne, J. T., Martin, R. A. & Constantinescu, G. M. (1990) The anatomy and embryology of portosystemic shunts in dogs and cats. Seminars in Veterinary Medicine and Surgery (Small Animal) 5, 76-82 Seguin, B., Tobias, K. M., Gavin, P. R. et al. (1999) Use of magnetic resonance angiography for diagnosis of portosystemic shunts in dogs. Veterinary Radiology and Ultrasound 40, 251-258 Szatmári, V. & Rothuizen, J. (2006) Ultrasonographic identification and characterization of congenital portosystemic shunts and portal hypertensive disorders in dogs and cats. In: WSAVA Standards for Clinical and Histological Diagnosis of Canine and Feline Liver Disease. Saunders Elsevier, Edinburgh, UK. pp 15-39 Szatmári, V., Rothuizen, J., van den Ingh, T. S., et al. (2004) Ultrasonographic findings in dogs with hyperammonemia: 90 cases (2000-2002). Journal of the American Veterinary Medical Association 224, 717-727 Tillson, D. M. & Winkler, J. T. (2002) Diagnosis and treatment of portosystemic shunts in the cat. Veterinary Clinics of North America Small Animal Practice 32, 881-899 White, R. N. & Parry, A. T. (2013) Morphology of congenital portosystemic shunts emanating from the left gastric vein in dogs and cats. Journal of Small Animal Practice 54, 459-467 White, R. N. & Parry, A. T. (2015) Morphology of congenital portosystemic shunts involving the right gastric vein in dogs. Journal of Small Animal Practice 56, 430-440. doi: 10.1111/jsap.12355 White, R. N., Macdonald, N. J. & Burton, C. A. (2003) Use of intraoperative mesenteric portovenography in congenital portosystemic shunt surgery. Veterinary Radiology and Ultrasound 44, 514-521 Zwingenberger, A. L., Schwarz, T. & Saunders, H. M. (2005) Helical computed tomographic angiography of canine portosystemic shunts. Veterinary Radiology and Ultrasound 46, 27-32
Journal of Small Animal Practice
•
Vol 57
•
January 2016
•
© 2015 British Small Animal Veterinary Association
PAPER
Morphology of splenocaval congenital portosystemic shunts in dogs and cats R. N. White and A. T. Parry Willows Referral Service, Shirley, Solihull, West Midlands B90 4NH R. N. White’s current address is School of Veterinary Medicine and Science, University of Nottingham, Sutton Bonington Campus, Leicestershire LE12 5RD
OBJECTIVE: To describe the anatomy of congenital portosystemic shunts involving the splenic vein communicating with the caudal vena cava at the level of the epiploic foramen. MATERIALS AND METHODS: A retrospective review of a consecutive series of dogs and cats managed for congenital portosystemic shunts. RESULTS: Ninety-eight dogs and eight cats met the inclusion criteria of a congenital portosystemic shunt involving the splenic vein communicating with the prehepatic caudal vena cava plus recorded intra-operative mesenteric portovenography or computed tomography angiography and gross observations at surgery. All cases (both dogs and cats) had a highly consistent shunt that involved a distended gastrosplenic vein that communicated with the caudal vena cava at the level of the epiploic foramen via an anomalous left gastric vein. CLINICAL SIGNIFICANCE: The morphology of the shunt type described appeared to be a result of an abnormal communication between the left gastric vein and the caudal vena cava and the subsequent development of preferential blood flow through an essentially normal portal venous system. The abnormal communication (shunt) was through the left gastric vein and not the splenic vein, as might have been expected. This information may help with surgical planning in cases undergoing shunt closure surgery. Journal of Small Animal Practice (2016) 57, 28–32 DOI: 10.1111/jsap.12414 Accepted: 23 September 2015; Published online: 13 November 2015
INTRODUCTION Methods described for the imaging of congenital portosystemic shunts (PSSs) include ultrasonography (Lamb 1996, Szatmári & Rothuizen 2006), magnetic resonance angiography (Sequin et al. 1999, Bruehschwein et al. 2010, Mai & Weisse 2011), computed tomography angiography (CTA) (Frank et al. 2003, Zwingenberger et al. 2005, Nelson & Nelson 2011, White & Parry 2013), findings on intra-operative mesenteric portovenography (IOMP) (White et al. 2003, White & Parry 2013), direct gross observations at surgery (White & Parry 2013, White & Parry 2015) and the examination of corrosion casts made at postmortem examination from individuals suffering from an extrahepatic portosystemic shunt (EHPSS) (Szatmári & Rothuizen 2006). Using these imaging techniques, it has proved possible to broadly classify congenital PSS as either intrahepatic or extrahepatic (Payne et al. 1990, Martin 1993, Levy et al. 1995, Lamb & White 1998, Tillson & Winkler 2002, Hunt 2004) with further 28
sub-classification of EHPSSs commonly being restricted to either porto-caval or porto-azygos (Szatmári et al. 2004). Recently, congenital EHPSSs involving the left phrenic and azygos veins, and right gastric vein were independently described in detail using a combination of CTA, IOMP, and gross anatomical findings (White & Parry 2013, White & Parry 2015). These studies concluded that the left gastric vein commonly represented the anomalous vessel (shunt) communicating with the systemic vein. In addition, the morphology of each shunt type described was shown to be the result of an abnormal communication between the left gastric vein and the systemic vein, and the subsequent development of preferential blood flow through essentially normal portal vessels within the portal venous system. The purpose of this study was to define the morphology of congenital EHPSSs that have previously been named “splenocaval shunts” in dogs and cats using the same methodology in a series of consecutive clinical cases.
Journal of Small Animal Practice
•
Vol 57
•
January 2016
•
© 2015 British Small Animal Veterinary Association
Portosystenic shunts involving the splenic vein
MATERIALS AND METHODS This retrospective study reviewed dogs and cats seen by the authors between 1997 and 2014 for the investigation and management of congenital PSS. The main inclusion criterion was that all cases must have a congenital PSS that involved the splenic vein communicating with the prehepatic vena cava at the level of the epiploic foramen. In addition, all cases must have undergone recorded IOMP and direct gross observations at the time of surgery. After 2009, some cases also underwent preoperative CTA. Data on breed, signalment (age, sex, neutering status), imaging investigation, type of PSS and gross surgical findings were collected and reviewed. Shunts that involved the splenic vein communicating with the prehepatic vena cava at the level of the epiploic foramen were separated from the main body of shunts collected and reviewed. CTA was performed under anaesthesia using a 16-slice multidetector unit (Brightspeed; General Electric Medical Systems) as described previously (White & Parry 2013, White & Parry 2015). Briefly, images were acquired using a 0·625 or 1·25-mm slice thickness, depending on the size of the animal, 120 kVp and variable mAs. Scanned field of view and displayed field of view were selected according to the size of the animal. The pitch was 0·938. Pre- and postintravenous contrast (600 mg I/kg iopromide, Ultravist; Bayer PLC) images were obtained using a standard algorithm and a 512×512 matrix, and viewed using a window and level optimised for soft tissue (window 400 HU, level 50 HU). Contrast was injected at a speed of 2·0 mL/second using a pressure injector. To optimise contrast enhancement, a transverse slice over the porta hepatis was selected and repetitively examined whilst contrast injection was performed. At the onset of opacification of the portal vessels, a complete abdominal CTA examination was performed using proprietary bolus tracking software with an automated trigger threshold of 120 HU to start the scan. The trigger region of interest was positioned over the portal vein at the level of the porta hepatis in all dogs, in the central aspect of the vessel to allow for respiratory motion. Studies were assessed in their native format, using multi-planar reformatting and using surface-shaded volume rendering. Vascular maps were obtained and postprocessing was limited to removal of arterial vessels and unnecessary portions of the caudal vena cava (CVC) from the maps. All CTA studies were reviewed by both authors. In addition, a number of normal CTA studies in dogs were reviewed for the purposes of cross reference. IOMP was carried out during surgery by using a mobile image intensification unit obtaining ventrodorsal images of the cranial abdomen (White et al. 2003, White & Parry 2015). Images were obtained before the manipulation of the shunt and during the temporary full ligation of the shunting vessel. Angiograms were recorded and reviewed by both authors. The gross anatomy of the shunt was recorded in the surgical report for each case. Information recorded included the course of the distended vasculature, any obvious tributary vessels and its entrance into the CVC. Using the combined data of IOMP, gross findings during surgery and CTA, the morphology of EHPSSs involving the splenic Journal of Small Animal Practice
•
Vol 57
•
January 2016
•
vein communicating with the prehepatic vena cava at the level of the epiploic foramen was compared. On the basis of this combined data, the anatomy of this shunt type was described and evaluated in both the dog and cat.
RESULTS In total, 169 dogs and 12 cats were found that had shunts that involved the splenic vein communicating with the prehepatic vena cava at the level of the epiploic foramen. Of these, 98 dogs and 8 cats met the inclusion criteria – having both recorded IOMP and complete descriptions of the direct gross observations at the time of surgery. The median age of dogs that met the inclusion criteria was 10 months (range 2 to 108 months). Of these dogs, 54 were male and 44 were female. Affected breeds were Yorkshire terrier (n=14), West Highland white terrier (n=12), Cairn terrier (n=10), Jack Russell terrier (n=8), shih-tzu (n=8), bichon frise (n=6), miniature schnauzer (n=6), crossbred (n=5), Norfolk terrier (n=5), pug (n=4), Maltese terrier (n=3), Border terrier (n=2), Pekinese (n=2), cavalier King Charles spaniel (n=2), Chihuahua (n=2), Lhasa apso (n=2), papillon (n=2), beagle (n=1), English setter (n=1), giant schnauzer (n=1), Shetland sheepdog (n=1) and Staffordshire bull terrier (n=1). The median age of cats that met the inclusion criteria was 7 months (range 3 to 42 months). Of these cats, three were male and five were female. Affected breeds were domestic short hair (n=2), Persian (n=2), British short hair (n=1), Havanese (n=1), Siamese (n=1) and tonkinese (n=1). In addition to IOMP that was performed in all cases, CTA was performed in seven dogs and one cat. The morphology of shunts involving the splenic vein communicating with the prehepatic CVC at the level of the epiploic foramen was consistent in all the cases identified. The following description was based on the findings of CTA, IOMP and gross findings at the time of surgery. Figure 1 shows a diagram of a normal portal vasculature for cross reference. The anomalous vessel arose from an enlarged gastrosplenic and tributary left gastric vein at the level of the angular notch (incisura angularis) on the dorsal wall of the pyloric aspect of the stomach in both the dog (Fig 2A to C) and the cat (Fig 3A to C). The enlarged left gastric vein continued as the anomalous vessel passing in a dorso-medial direction towards the prehepatic CVC where it entered the cava on the left side at the level of the epiploic foramen (Fig 4). There was very little variation in the morphology of the shunting vessel; the sole variability appeared related to the relative lengths of the contributory vessels: the gastrosplenic vein, the tributary left gastric vein and its continuation as the anomalous shunting vessel before its communication with the prehepatic CVC.
DISCUSSION The results of this study revealed that EHPSSs involving the splenic vein communicating with the prehepatic CVC at the level
© 2015 British Small Animal Veterinary Association
29
R. N. White & A. T. Parry
FIG 1. A diagram showing the normal portal vasculature in dogs and cats (Modified from Miller et al. 1964). Gd v gastroduodenal vein, Gs v gastrosplenic vein
FIG 2. (A) A figure showing the salient features of the IOMP and gross observations at surgery obtained from dogs with a left gastrocaval shunt. The shunt consisted of a distended gastrosplenic vein that communicated with the caudal vena cava (CVC) at the level of the epiploic foramen via an anomalous left gastric vein. The tributary left gastric vein and splenic vein (shaded grey) were not apparent in all cases on IOMP. (B) An example of a ventrodorsal IOMP of a 12-month-old male miniature schnauzer. Note the enlarged gastrosplenic vein draining into an enlarged, anomalous left gastric vein before entering the prehepatic CVC at the level of the epiploic foramen. (C) This image shows a surface-shaded volume rendered computed tomography angiography (CTA) of the same dog shown in (B). Similar to (B), note the enlarged gastrosplenic vein draining into an enlarged, anomalous left gastric vein before entering the prehepatic CVC at the level of the epiploic foramen. CP-DV Cranial pancreatic-duodenal vein, LGV Left gastric vein
of the epiploic foramen were consistent, allowing their anatomical description in both dogs and cats. There was very little variation between the cases examined: in each the portal blood was 30
shunted initially through an enlarged gastrosplenic vein, then through its tributary left gastric vein (passing along the dorsal aspect of the pyloric antrum), before continuing, at the level of
Journal of Small Animal Practice
•
Vol 57
•
January 2016
•
© 2015 British Small Animal Veterinary Association
Portosystenic shunts involving the splenic vein
FIG 3. (A) A figure showing the salient features of the IOMP and gross observations at surgery obtained from cats with a left gastrocaval shunt. Similar to dogs, the shunt consisted of a distended gastrosplenic vein that communicated with the caudal vena cava (CVC) at the level of the epiploic foramen via an anomalous left gastric vein. The tributary left gastric vein and splenic vein (shaded grey) were not apparent in all cases on IOMP. (B) An example of a ventrodorsal IOMP of a 10-month-old female domestic short hair cat. Note the enlarged gastrosplenic vein draining into an enlarged, anomalous left gastric vein before entering the prehepatic CVC at the level of the epiploic foramen. (C) This image shows a surface-shaded volume rendered computed tomography angiography (CTA) of an equivalent shunt to that shown in (B). Similar to (B), note the enlarged gastrosplenic vein draining into an enlarged, anomalous left gastric vein before entering the prehepatic CVC at the level of the epiploic foramen. CP-DV Cranial pancreatic-duodenal vein, G-S V Gastrosplenic vein
FIG 4. An intraoperative view of a 13-month-old male Cairn terrier. The anomalous vessel (shunt) can be seen entering the prehepatic caudal vena cava at the level of the epiploic foramen
the angular notch of the stomach, as the anomalous vessel entering the CVC at the level of the epiploic foramen. Any variation in observed shunt morphology was minor and appeared to be associated with variation in the lengths of the vessels through which the portal blood was being shunted, i.e. the gastrosplenic vein, the tributary left gastric vein or the anomalous vessel entering the CVC. No measurements of vessel length were undertaken in this study, and further studies will be required if this variation in shunt morphology is to be further investigated. Naming conventions of EHPSSs have not always been clear or specific (Johnson et al. 1987, Payne et al. 1990, Levy et al. 1995, Szatmári et al. 2004, Nelson & Nelson 2011). Shunts entering the prehepatic CVC have frequently been described as portocaJournal of Small Animal Practice
•
Vol 57
•
January 2016
•
val without providing further information regarding shunt origin or morphology (Berent & Tobias 2012). Reasons for this lack of detailed description relate predominantly to the method by which the shunt is imaged. In most instances, all imaging modalities will confirm a PSS, but the ability of each to allow for accurate description of the anatomy of the shunt is open to observer interpretation and wide variation. Recently, with the advent of CTA, attempts have been made to classify portocaval EHPSSs in more detail on the basis of their origin, course and morphology (Nelson & Nelson 2011, White & Parry 2013, White & Parry 2015). In this study, the anatomy of the PSS was characterised in the majority of individuals using the results of the IOMP and gross anatomical findings at surgery. The findings were consistent in all cases allowing the classification of these shunts as described. This shunt type has been previously described as either a splenocaval or a gastrosplenocaval EHPSS (Hunt 2004, Hunt et al. 2004, Szatmári et al. 2004, Szatmári & Rothuizen 2006, Nelson & Nelson 2011, Fukushima et al. 2014). In addition, a number of reports have acknowledged the possible involvement of the left gastric vein in this shunt type (Szatmári et al. 2004, Cullen et al. 2006). The use of preoperative CTA in a number of cases after 2009 confirmed, and never contradicted, the anatomical descriptions obtained from both IOMP and gross anatomical findings at surgery. The main additional information provided by CTA was in defining the draining tributary veins of the portal vasculature that could not be seen on the IOMP or, in some cases, at the time of surgery. By anatomical convention, vascular shunts are most commonly named using the name of the portal vessel from which the shunt emanates and the name of the systemic vein it joins and supplies (Payne et al. 1990). Using this convention, by combining our results of IOMP, gross intra-operative surgical findings and CTA, we conclude that this previously described
© 2015 British Small Animal Veterinary Association
31
R. N. White & A. T. Parry
“splenocaval” PSS might be more accurately described as a left gastrocaval shunt. Nelson & Nelson (2011) also described an “atypical splenocaval” shunt with a short, wide shunting vessel with no evidence of continuation of the portal vein extending cranially from the shunt insertion into the prehepatic CVC. This shunt variation was not seen in either the dog or the cat in this study. It is hoped that further studies will identify this shunt type and allow it to be classified in a more detailed manner. The findings of this study support the previous suggestion that, embryologically, it is the development of the left gastric vein that is critical in the formation of this shunt type (White & Parry 2015). In this study, the anomalous vessel that joined with the prehepatic CVC appeared to be a continuation of the left gastric vein. This vein is normally programmed to connect with the splenic vein and, subsequently, the portal vein in the developing embryo (Noden & de Lahunta 1985). This inappropriate communication between an anomalous portion of the left gastric vein and the prehepatic CVC at the level of the epiploic foramen resulted in the development of the consistent shunt type described. In both dogs and cats, there was a highly consistent PSS which involved a distended gastrosplenic vein that communicated with the prehepatic CVC via an anomalous left gastric vein. The site of entrance of the shunt into the CVC was also highly consistent in every case – at the level of the epiploic foramen. The morphology of this shunt appeared to be the result of the abnormal communication between the anomalous left gastric vein and the CVC and the development of preferential hepatopetal blood flow through essentially normal gastrosplenic and left gastric veins. Conflict of interest None of the authors of this article has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper. References Berent, A. C. & Tobias, K. M. (2012) Hepatic Vascular Anomalies. In: Veterinary Surgery: Small Animals. Eds K. M. Tobias and S. A. Johnston. Saunders Elsevier, St Louis, MO, USA. pp 1624-1658 Bruehschwein, A., Foltin, I., Flatz, K., et al. (2010) Contrast-enhanced magnetic resonance angiography for diagnosis of portosystemic shunts in 10 dogs. Veterinary Radiology and Ultrasound 51, 116-121 Cullen, J. M., van den Ingh, T. S. G. A. M., Bunch, S. E., et al. (2006) Morphological classification of circulatory disorders of the canine and feline liver. In: WSAVA Standards for Clinical and Histological Diagnosis of Canine and Feline Liver Disease. Saunders Elsevier, Edinburgh, UK. pp 41-59
32
Frank, P., Mahaffey, M., Egger, C., et al. (2003) Helical computed tomography portography in ten normal dogs and ten dogs with a portosystemic shunt. Veterinary Radiology and Ultrasound 44, 392-400 Fukushima, K., Kanemoto, H., Ohno, K., et al. (2014) Computed tomographic morphology and clinical features of extrahepatic portosystemic shunts in 172 dogs in Japan. Veterinary Journal 199, 376-381 Hunt, G. B. (2004) Effect of breed on anatomy of portosystemic shunts resulting from congenital diseases in dogs and cats: a review of 242 cases. Australian Veterinary Journal 82, 746-749 Hunt, G. B., Kummeling, A., Tisdall, P. L., et al. (2004) Outcomes of cellophane banding for congenital portosystemic shunts in 106 dogs and 5 cats. Veterinary Surgery 33, 25-31 Johnson, C. A., Armstrong, P. J. & Hauptmann, J. G. (1987) Congenital portosystemic shunts in dogs: 46 cases (1979-1986). Journal of the American Veterinary Medical Association 191, 1478-1483 Lamb, C. R. (1996) Ultrasonographic diagnosis of congenital portosystemic shunts in dogs: results of a prospective study. Veterinary Radiology and Ultrasound 37, 281-288 Lamb, C. R. & White, R. N. (1998) Morphology of congenital intrahepatic portocaval shunts in dogs and cats. Veterinary Record 142, 55-60 Levy, J. K., Bunch, S. E. & Komtebedde, J. (1995) Feline portosystemic vascular shunts. In: Kirk’s Current Veterinary Therapy XII. Small Animal Practice. Ed J. D. Bonagura. W.B. Saunders, Philadelphia, PA, USA. pp 743-749 Mai, W. & Weisse, C. (2011) Contrast-enhanced portal magnetic resonance angiography in dogs with suspected congenital portal vascular anomalies. Veterinary Radiology and Ultrasonography 52, 284-288 Martin, R. A. (1993) Congenital portosystemic shunts in the dog and cat. Veterinary Clinics of North America Small Animal Practice 23, 609-623 Miller, M. E. (1964) The venous system. In: Anatomy of the Dog. Ed M. E. Miller. W.B. Saunders, Philadelphia, PA, USA. pp 389-429 Nelson, N. C. & Nelson, L. L. (2011) Anatomy of extrahepatic portosystemic shunts in dogs as determined by computed tomography angiography. Veterinary Radiology and Ultrasonography 52, 498-506 Noden, D. M. & de Lahunta, A. (1985) Cardiovascular system III: venous system and lymphatics. In: The Embryology of Domestic Animals – Developmental Mechanisms and Malformations. Williams & Williams, Baltimore, MA, USA. pp 257-269 Payne, J. T., Martin, R. A. & Constantinescu, G. M. (1990) The anatomy and embryology of portosystemic shunts in dogs and cats. Seminars in Veterinary Medicine and Surgery (Small Animal) 5, 76-82 Seguin, B., Tobias, K. M., Gavin, P. R. et al. (1999) Use of magnetic resonance angiography for diagnosis of portosystemic shunts in dogs. Veterinary Radiology and Ultrasound 40, 251-258 Szatmári, V. & Rothuizen, J. (2006) Ultrasonographic identification and characterization of congenital portosystemic shunts and portal hypertensive disorders in dogs and cats. In: WSAVA Standards for Clinical and Histological Diagnosis of Canine and Feline Liver Disease. Saunders Elsevier, Edinburgh, UK. pp 15-39 Szatmári, V., Rothuizen, J., van den Ingh, T. S., et al. (2004) Ultrasonographic findings in dogs with hyperammonemia: 90 cases (2000-2002). Journal of the American Veterinary Medical Association 224, 717-727 Tillson, D. M. & Winkler, J. T. (2002) Diagnosis and treatment of portosystemic shunts in the cat. Veterinary Clinics of North America Small Animal Practice 32, 881-899 White, R. N. & Parry, A. T. (2013) Morphology of congenital portosystemic shunts emanating from the left gastric vein in dogs and cats. Journal of Small Animal Practice 54, 459-467 White, R. N. & Parry, A. T. (2015) Morphology of congenital portosystemic shunts involving the right gastric vein in dogs. Journal of Small Animal Practice 56, 430-440. doi: 10.1111/jsap.12355 White, R. N., Macdonald, N. J. & Burton, C. A. (2003) Use of intraoperative mesenteric portovenography in congenital portosystemic shunt surgery. Veterinary Radiology and Ultrasound 44, 514-521 Zwingenberger, A. L., Schwarz, T. & Saunders, H. M. (2005) Helical computed tomographic angiography of canine portosystemic shunts. Veterinary Radiology and Ultrasound 46, 27-32
Journal of Small Animal Practice
•
Vol 57
•
January 2016
•
© 2015 British Small Animal Veterinary Association
